Inorganic Reactions and Methods- Oligomerization and Polymerization Formation of Intercalation Compounds

Inorganic Reactions and Methods Volume 17

Inorganic Reactions and Methods Editor Professor A.P. Hagen Department of Chemistry The University of Oklahoma Norman, Oklahoma 73019

Editorial Advisory Board Professor N. Bartlett Department of Chemistry University of California at Berkeley Berkeley, California 94720 Professor F.A. Cotton Department of Chemistry Texas A&M University College Station, Texas 77840 Professor E.O. Fischer Anorganisch-chemischesLaboratorium der Technischen UniversitM D-8046 Garching Lichtenbergestrasse4 Federal Republic of Germany Professor P. Hagenmuller Laboratoire de Chemie du Solide du C.N.R.S. 351 cours de la Liberation F-33405 Talence France Professor M.F. Lappert The Chemical Laboratory University of Sussex Falmer, Brighton, BN1 9A3 England

Professor A.G. MacDiarmid Department of Chemistry University of Pennsylvania Philadelphia, Pennsylvania 19174 Professor M. Schmidt lnstitut fur Anorganische Chemie der Universitat D-8700 Wurzburg Am Hubland Federal Republic of Germany Professor H. Taube Department of Chemistry Stanford University Stanford, California 94305 Professor L.M. Venanzi Laboratorium fur Anorganische Chemie der ETH CH-80006 Zurich Universitatsstrasse5 Switzerland Professor Sir Geoffrey Wilkinson, F.R.S. Department of Chemistry Imperial College South Kensington London, SW7 2AY England

@ 1990 VCH Publishers, Inc , New York

Distribution: VCH Verlagsgesellschafi mbH, P.O. Box 1260/1280, D-6940 Weinheim, Federal Republic of Germany USA and Canada: VCH Publishers, Inc., 303 N.W. 12th Avenue, Deerfield Beach, FL 33442-1705, USA

Inorganic Reactions and Methods

Volume 17 Oligomerization and Polymerization Formation of Intercalation Compounds Founding Editor

J.J. Zuckerman Editor

A.P. Hagen

8WILEY-VCH

@ 1990 VCH Publishers, Inc.

This work is subject to copyright. All rights are reserved, whether the whole or part of the material is concerned, specifically those of translation, reprinting, re-use of illustrations, broadcasting, reproduction by photocopying machine or similar means, and storage in data banks. Authorization to photocopy items for internal or personal use, or the internal or personal use of specific clients, is granted by VCH Publishers, Inc. for libaries and other users registered with the Copyright Clearance Center (CCC) Transactional Reporting Service, provided that the base fee of $1.00 per copy, plus $0.25 per page is paid directly to CCC, 27 Congress Street, Salem, MA 01970. Registered names, trademarks, etc. used in this book, even when not specifically marked as such, are not to be considered unprotected by law.

ISBN 0-471-18667-8 ISBN 3-527-26275-X VCH Verlagsgesellschaft

Contents of Volume 17

How to Use this Book Preface to the Series Editorial Consultants to the Series Contributors to Volume 17

15.

Oligomerization and Polymerization

15.1.

Introduction

15.1.l. 15.1.1.1. 15.1.1.2. 15.1.1.3. 15.1.1.3.1. 15.1.1.3.2. 15.1.1.3.3. 15.1.1.3.4. 15.1.2. 15.1.2.1. 15.1.2.2. 15.1.3. 15.1.3.1. 15.1.3.2. 15.1.3.2.1. 15.1.3.2.2.

Definitions and Scope. Definitions of Oligomers and Polymers. General Synthesis Methods. Scope. Small-Ring Systems-Examples of Synthesis. Cages-Exam pIes of Synthesis. Short Chai ns-Exam ples of Synthesis. Long Chains-Examples of Synthesis. Characteristic Differences Between Small and Large Molecules as Related to Their Synthesis. Special Characteristics of Small Cyclic and Linear Molecules. Special Characteristics of Long-Chain Molecu Ies. Chai n-C hai n, Cha i n-R ing and R ing-R ing Equilibria. Bond Lability in Inorganic Systems. Some Important Relationships Based on Theory. The Chain Population. The Ring Population and Network Macromolecules.

1

7 12 13 15 17 17 18 19 20 21 22 24 V

vi

Contents of Volume 17

~~

15.1 15.1.3.3. 15.1.3.3.1. 15.1.3.3.2. 15.1.3.3.3. 15.1.3.3.4.

15.2. 15.2.1. 15.2.2. 15.2.2.1. 15.2.2.2. 15.2.2.2.1. 15.2.2.2.2. 15.2.2.2.3. 15.2.2.2.4. 15.2.2.2.5. 15.2.2.2.6. 15.2.2.2.7. 15.2.2.2.8. 15.2.2.2.9. 15.2.2.2.10. 15.2.2.2.11. 15.2.2.2.12. 15.2.2.2.13. 15.2.2.2.14.

Introduction Scrambling in Inorganic Systems. All-Neso Systems. Oligomers and Polymers in QZ,-QT, Systems with Z Bridging. Exchange of Segments Between Cyc Iic Mo I ecu les. Chains and Rings from Scrambling Between Two Kinds of Central Moieties.

Ring-Ring and Ring-Polymer Interconversions Introduction. lnterconversion of Sulfur or Selenium Rings and Chains Theory . Ex per imental Techniques Viscosity Measurements. Ultrasonic Absorption in Liquid Selenium. Quenching Experiments. Diamagnetic Suscepti bi Iity Measurements. Paramagnetic Suscepti bi Iity Measurements. Electron Spin Resonance Absorption (ESR). Optical Absorption. Vibrational Spectroscopy of Sulfur. Nuclear Magnetic Resonance (NMR) Measurements on Selenium. Dielectric Constant Measurements of Su Ifur. Specific Heat Measurements on Sulfur and Selenium. Density Measurements. Thermal Conductivity in Liquid Selenium. Electrical Conductivity of Sulfur.

26 26 31 33 34

37 37 39 39 46 46 47 47 48 48 50 50 51 51 51

52 52 53 53

Contents of Volume 17

15.2.3. 15.2.3.1. 15.2.3.2. 15.2.3.3. 15.2.3.4. 15.2.3.5. 15.2.4. 15.2.4.1. 15.2.4.1.l. 15.2.4.1.2. 15.2.4.1.3. 15.2.4.1.4. 15.2.4.2. 15.2.4.2.1. 15.2.4.2.2. 15.2.4.2.3. 15.2.4.3. 15.2.4.4. 15.2.4.5. 15.2.5. 15.2.5.1. 15.2.5.1.l. 15.2.5.1.2. 15.2.5.2. 15.2.5.3. 15.2.5.4. 15.2.5.5. 15.2.5.5.1. 15.2.5.5.2. 15.2.6 15.2.6.1.

15.2.6.2. 15.2.6.3. 15.2.6.4. 15.2.7.

In Oligomeric Catenates of P, As, Sb and Bi. General Trends. Phosphorus. Arsenic. Antimony. Stabilized Rings and Chains. In Silicon-Silicon Systems. Formation of Cyclic Silicon-Silicon Systems. Perarylcyclosilanes. Permethylcyclosilanes. Other Peralkylcyclosilanes. A Iky Ia ry lcyclosi Ianes . Reactions of Cyclosilanes. Substitutional Reactions. Ring-Cleavage Reactions. Photo1ysis. Ring-Ring and Chain-Ring Interconversion. Cage Polysilanes. Silane High Polymers. in Boron-Nitrogen Systems. Synthesis of Ring Compounds Borazines and Related Compounds. Coordinately Saturated B-N Rings. Factors Affecting Stability of Ring Species. Preparation of Linear Boron-Nitrogen 01i gomers. lnterconversions in Boron-Nitrogen Systems. Preparation of Boro n-N it r oge n Polymers. Linear Boron-Nitrogen Chains. Polycyclic Chains. in Boron-Oxygen and Boron-Sulfur Systems An ionic Boron-Oxyg e n Compounds. Molecular Boron-Oxygen Species. Boron-Su Ifur Com pounds. Ring-ring and Ring-chain interconversions. in Cage Polyboranes and Carboranes.

vi i

53 53 54 55 58 59 61 61 61 62 66 66 69 69 71 74 75 79 80 83 83 83 86 87 88 89 90 90 91 93 93 94 95 96 97

viii

15.2 15.2.7.1. 15.2.7.2. 15.2.7.3. 15.2.8. 15.2.8.1. 15.2.8.2. 15.2.8.3. 15.2.8.4. 15.2.8.5. 15.2.9. 15.2.9.1. 15.2.9.2. 15.2.9.3. 15.2.9.4. 15.2.9.5. 15.2.9.6. 15.2.10. 15.2.10.1. 15.2.10.2. 15.2.10.3. 15.2.10.4. 15.2.11. 15.2.11. I . 15.2.11.2. 15.2.11.3. 15.2.11.4.

Contents of Volume 17

Ring-Ring and Ring-Polymer Interconversions Polyborane Oligomers and Polymers; Syntheses. Preparation of Carborane-Siloxane Polymers. Preparation of Other CarboraneIncorporated Inorganic Polymers. Si Iicon-Oxygen Systems Formation of Cyclic Oligomers. Cyclosiloxane Structural Diversity. Ring-Ring Equilibration of Cyc Iosi loxanes. R ing-C hai n Equ i I ibrat io n : Polymerization of Cyclosiloxanes. End-Capping Reactions. in Silicon-Nitrogen and Silicon-Sulfur Systems. Silicon-Nitrogen Syntheses. Si I icon-N it rogen Ri ng-Ri ng Interconversions. Cyclic Silicon-Nitrogen Oligomers Polymerization. Si Iicon-Su Ifur Syntheses. Si I icon-Su If ur Ring-Ri ng Interconversions. Cyclic Si Iicon-Sulfu r 01igomer Polymerization. in Phosphor us-Oxyg e n Systems Reorganized Mixtures of Chains and Rings. Preparation of Cyclic Polymers. Chain Oligomers and Macromolecules. Related Mixed Systems. in Phosphorus-Nitrogen Oligomers and Polymers Halocyclophosphazene Polymerization Organocyclophosphazenes: Ring-Ring and Ring-Polymer Interconversions Mechanisms and Thermodynamics of Phosphazene Equilibrations. Polyorganophosphazenes via the Reactions of Polydihalophosphazenes.

98 99 103 105 107 108 112 114 115

116 116 118 137 139 142 142 143 144 145 148 150 151 151 154 159

161

Contents of Volume 17

ix

~

15.2.12. 15.2.12.1. 15.2.12.2. 15.2.12.3. 15.2.12.4. 15.2.12.5. 15.2.12.6. 15.2.12.7. 15.2.12.8. 15.2.12.9. 15.2.12.10. 15.2.13. 15.2.13.1. 15.2.13.2. 15.2.13.2.1. 15.2.13.2.2. 15.2.13.2.3. 15.2.13.3. 15.2.13.3.1. 15.2.13.3.2. 15.2.13.3.3. 15.2.13.3.4. 15.2.13.3.5. 15.2.13.3.6. 15.2.13.4. 15.2.13.5. 15.2.13.6. 15.2.14. 15.2.14.1. 15.2.14.2.

in Su Ifur-Ni trogen 01igomers and Polymers Formation of the Monomer SN Formation of Cyclic Oligomers (SN), and (SN),. Polymerization of (SN), to (SN),. Halogenated Derivatives of (SN),. Formation of Cyclic Oligomers (NSX). (X = CI, F; n = 3,4). Formation of the Cyclic Oligomers “S(O)XI,. Formation of the Cyclic Oligomer (SNH),. Formation of Thionyl lmide Polymers (HNSO),. S-N Oligomers with Terminal Organic Groups. S-Polymers with Organic Groups in the Backbone. Polymers Containing Metal-Oxygen and Metal-Carbon Bonds Introduction. M-C-Containing Polymers Metallocenes. Polycarbosilanes, Polysilarylenes, Polycarbosiloxanes and PoIy (carbo rane-s i Ioxanes) . Other Metallorganic Compounds. M-0-Contai ni ng Polymers Inorganic PoIys iIicas. PoIy meta1Iosi Ioxan es. Poly(si Iyl arylene si loxanes). Polyarylsilsesquioxanes. Polymetal Phosphinates and Polyphosphonatolanes Miscellaneous.

172 172 173 174 175 176 177 178 178 179 180 180 180 181 181 184 185 187 187 187 188 190 190 191

0

II

M-0-C-Containing Polymers. M-0-R (Polyet her)-Contai n i ng Polymers. M-0-N-Contain i ng Polymers (Polyamidoximes and Polyoximes). Coordination Polymers Introduction. General Synthetic Routes.

191 192 193 194 194 194

X

15.2 15.2.14.3. 15.2.14.4. 15.2.14.5. 15.2.14.5.1. 15.2.14.5.2. 15.2.14.5.3. 15.2.14.5.4. 15.2.14.5.5.

Contents of Volume 17

Ring-Ring and Ring-Polymer lnterconversions lonomers. Property-St r ucture Pr i nci pIes . Selected Polymers. Metal Phosphinates. Platinum and Pd Coordination Polymers. Uranyl Ion Complexation. Bis-P-diketone-Based Polymers. Bisthiopiocol inam ide-Based Polymers.

200 200 201 20 1 202 203 204 205

16.

Formation of Intercalation Compounds

207

16.1.

Introduction

208

16.2.

The Formation of Clathrates

209

16.2.1. 16.2.2. 16.2.2.1. 16.2.2.2. 16.2.2.3. 16.2.3. 16.2.3.1.

16.2.3.2.

16.3. 16.3.1. 16.3.1.1. 16.3.1.2. 16.3.1.3.

Definitions The Formation of Clathrates Having a Water Host Lattice Clathrates of the Gas Hydrate Type. Clathrates of the Liquid Hydrate Type Other Clathrate Hydrates. The Formation of Clathrates Having a Silicon, Germanium and Tin Host Lattice The Formation of Silicon and Germanium Clathrates by Thermal Decomposition of Silicides and Germanides. The Formation of Silicon, Germanium and Tin Clathrates from the Elements.

The Formation of Tunnel Structures Ion Exchange Zeolites with Intersecting Tunnel Frameworks. Pyrochlores A, xxM20, A2M,0,, Structure and lntergrowths with Pyrochlores.

209 210 21 1 214 215 219

22 1 222

223 223

223 226 228

Contents of Volume 17

16.3.1.4. 16.3.1.5. 6.3.1 -6. 6.3.2. 6.3.2.1. 6.3.2.2. 6.3.2.2.1. 16.3.2.2.2. 16.3.2.3. 16.3.2.3.1. 16.3.2.3.2. 16.3.2.3.3.

16.4. 16.4.1. 16.4.2. 16.4.2.1. 16.4.2.1.1. 16.4.2.1.2. 16.4.2.1.3. 16.4.2.1.4. 16.4.2.1 -5. 16.4.2.2. 16.4.2.2.1. 16.4.2.2.2. 16.4.2.3. 16.4.2.4. 16.4.2.4.1 16.4.2.4.2.

Alkali Antimonates K,Sb,O, KC$

xi

and

b501 4'

Ammonium and Potassium 12 Molybdophosphates A3P04Mo,2036. Structures with Parallel Tunnels by Electrolysis Rate of Growth of Deposit. Mechanism of Electrochemical Preparation of Tungsten Bronzes Nature of the Reduced Species. Reactions at the Electrodes. Experimental Procedure and Data the Electrolyte Cell. Effect of the Melt Composition and Temperature. Examples of Oxides with Tunnel Structures Obtained by Molten Salt Electrolysis.

The Formation of Sheet Structures Introd ucti on. Graphite and Boron Nitride Intercalation Compounds Electron Acceptors Hal ides. Oxyhalides. Oxides. Acids. Halogens and Interhalogens. Electron Donors Synthesis of Graphite and Carbon Intercalation Compounds Containing One metal. Graphite Intercalation Compounds Containing Two Metals (Ternary Com pou nds). Mixed Metal-Molecule-Graphite Ternaries. Al ka Ii-M etal-Hydrogen42 raph ite Ternaries. Hydrogenation of the binary phases MC, and MC, (M = K, Rb, Cs). Intercalation of Alkali-Metal Hydrides into Graphite.

228 228 229 229 230 230 230 23 1 232 232 232 234

237 237 238 238 238 243 244 245 246 247 247 25 1 255 257 257 257

xii

16.4 16.4.2.4.3. 16.4.2.4.4. 16.4.2.5. 16.4.2.5.1. 16.4.2.5.2. 16.4.2.5.3. 16.4.2.5.4. 16.4.2.6. 16.4.2.6.1. 16.4.2.6.2. 16.4.2.6.3. 16.4.2.7. 16.4.2.7.1. 16.4.2.7.2. 16.4.2.7.3. 16.4.2.8. 16.4.2.8.1. 16.4.2.8.2. 16.4.2.8.3. 16.4.3. 16.4.3.1. 16.4.3.2. 16.4.3.3. 16.4.3.4. 16.4.3.5. 16.4.3.6. 16.4.4. 16.4.5. 16.4.6. 16.4.7. 16.4.7.1. 16.4.7.2.

Contents of Volume 17

The Formation of Sheet Structures Intercalation of Quaternary Compounds with Hydrogen and Two Alkali Metals into Graphite. Physiosorption of Hydrogen and Deuterium in the Second-Stage Compounds MC,,(M = K, Rb or Cs) Graphite Oxides Definition and Nomenclature of Graphite Oxides. Preparation of Graphite Oxides. Reactivity of Graphite Oxides. Graphite Oxide Membranes. Graphite Fluorides Poly(monocarbon monofluoride), (CF)n Poly (dicar bon monof I uo r ide) , (C,F) . Poly(tetracarbon monofluoride), (C4F)n. Residue and Surface Compounds Introduction and Definitions. Preparation of Residue Compounds. Preparation of Surface Compounds. Electrochemical Formation. Strengths and Limitations of the Electrochemical Technique. Cathodic Reduction of Graphite. Anodic Oxidation of Graphite. Dichalcogenides Layered Dichalcogenides. Alkali-Metal intercalates. Electrochemical Formation Molecular Intercalates. Hydrated and Solvated Alkali-Metal Phases. Intercalation-Substitution Compounds and Ionic Mobility Trichalcogenohypophosphates. Transition-Metal Oxyhalides. Lamellar Oxides. Lamellar Silicates and Related Materials. Lamellar Silicates. lntercalation Products.

-

259 259 259 260 26 1 263 265 265 265 269 270 27 1 27 1 272 274 276 276 280 281 284 284 288 295 300 306 308 309 313 318 323 324 328

Contents of Volume 17

xiii

~

16.4.8. 16.4.8.1. 16.4.8.2. 16.4.8.3. 16.4.8.4. 16.4.8.5. 16.4.8.6. 16.4.8.7.

The Chemical Reactivity of LowDimensional Solids The Electronic Transfer and the Formation of Cationic Intercalation Compounds. Structural Charges Induced by Electronic Transfer. Strong Ion-Electron Coupling. Microdomains in MPS, Phases. Electronic Transfer and Ac Parameter Expansion. Acidobasic Topochemical Reactions. “Internal Surfaces” Grafting Chemistry.

List of Abbreviations Author Index Compound Index Subject Index

331 332 333 335 337 338 339 340 341 000 000 000

How to Use this Book 1. Organization of Subject Matter 1.1, Logic of Subdivision and Add-on Chapters This volume is part of a series that describes all of inorganic reaction chemistry. The contents are subdivided systematically and so are the contents of the entire series: Using the periodic system as a correlative device, it is shown how bonds between pairs of elements can be made. Treatment begins with hydrogen making a bond to itself in H, and proceeds according to the periodic table with the bonds formed by hydrogen to the halogens, the groups headed by oxygen, nitrogen, carbon, boron, beryllium and lithium, to the transition and inner-transition metals and to the members of group zero. Next it is considered how the halogens form bonds among themselves and then to the elements of the main groups VI to I, the transition and inner-transition metals and the zero-group gases. The process repeats itself with descriptions of the members of each successive periodic group making bonds to all the remaining elements not yet treated until group zero is reached. At this point all actual as well as possible combinations have been covered. The focus is on the primary formation of bonds, not on subsequent reactions of the products to form other bonds. These latter reactions are covered at the places where the formation of those bonds is described. Reactions in which atoms merely change their oxidation states are not included, nor are reactions in which the same pairs of elements come together again in the product (for example, in metatheses or redistributions). Physical and spectroscopic properties or structural details of the products are not covered by the reaction volumes which are concerned with synthetic utility based on yield, economy of ingredients, purity of product, specificity, etc. The preparation of short-lived transient species is not described. While in principle the systematization described above could suffice to deal with all the relevant material, there are other topics that inorganic chemists customarily identify as being useful in organizing reaction information and that do not fit into the scheme. These topics are the subject of eight additional chapters constituting the last four books of the series. These chapters are systematic only within their own confines. Their inclusion is based on the best judgment of the Editorial Advisory Board as to what would be most useful currently as well as effective in guiding the future of inorganic reaction chemistry. xv

xvi

How to Use this Book

1.2. Use of Decimal Section Numbers The organization of the material is readily apparent through the use of numbers and headings. Chapters are broken down into divisions, sections and subsections, which have short descriptive headings and are numbered according to the following scheme: 1. Major Heading 1.1. Chapter Heading 1.1.1. Division Heading l . l . l , l , Section Heading 1.1.1,l.1. Subsection Heading Further subdivision of a five-digit “slice” utilizes lower-case Roman numerals in parentheses: (i), (ii), (iii), etc. It is often found that as a consequence of the organization, cognate material is located in different chapters but in similarly numbered pieces, i.e., in parallel sections. Section numbers, rather than page numbers, are the key by which the material is accessed through the various indexes.

1.3. Building of Headings 1.3.1. Headings Forming Part of a Sentence

Most headings are sentence-fragment phrases which constitute sentences when combined. Usually a period signifies the end of a combined sentence. In order to reconstitute the context in which a heading is to be read, superiorrank titles are printed as running heads on each page. When the sentences are put together from their constituent parts, they describe the contents of the piece at hand. For an example, see 2.3 below. 1.3.2. Headings Forming Part of an Enumeration

For some material it is not useful to construct title sentences as described above. In these cases hierarchical lists, in which the topics are enumerated, are more appropriate. To inform the reader fully about the nature of the material being described, the headings of connected sections that are superior in hierarchy always occur as running heads at the top of each page.

2. Access and Reference Tools 2.1. Plan of the Entire Series (Front Endpaper) Printed on the inside of the front cover is a list, compiled from all 18 reaction volumes, of the major and chapter headings, that is, all headings that

How to Use this Book

xvii

are preceded by a one- or two-digit decimal section number. This list shows in which volumes the headings occur and highlights the contents of the volume that is at hand by means of a gray tint.

2.2. Contents of the Volume at Hand All the headings, down to the title of the smallest decimal-numbered subsection, are listed in the detailed table of contents of each volume. For each heading the table of contents shows the decimal section number by which it is preceded and the number of the page on which it is found. Beside the decimal section numbers, successive indentations reveal the hierarchy of the sections and thereby facilitate the comprehension of the phrase (or of the enumerative sequence) to which the headings of hierarchically successive sections combine. To reconstitute the context in which the heading of a section must be read to become meaningful, relevant headings of sections superior in hierarchy are repeated at the top of every page of the table of contents. The repetitive occurrences of these headings is indicated by the fact that position and page numbers are omitted.

2.3. Running Heads In order to indicate the hierarchical position of a section, the top of every page of text shows the headings of up to three connected sections that are superior in hierarchy. These running heads provide the context within which the title of the section under discussion becomes meaningful. As an example, the page of Volume 1 on which section 1.4.9.1.3 “in the Production of Methanol” starts, carries the running heads: 1.4. The Formation of Bonds between Hydrogen and O,S,Se,Te,Po 1.4.9. by Industrial Processes 1.4.9.1. Involving Oxygen Compounds

whereby the phrase “in the Production of Methanol” is put into its proper perspective.

2.4. List of Abbreviations Preceding the indexes there is a list of those abbreviations that are frequently used in the text of the volume at hand or in companion volumes. This list varies somewhat in length from volume to volume; that is, it becomes more comprehensive as new volumes are published. Abbreviations that are used incidentally or have no general applicability are not included in the list but are explained at the place of occurrence in the text.

How to Use this Book

xviii

2.5. Author Index The author index is compiled by computer from the lists of references. Thus it tells whose publications are cited and in that respect is comprehensive. It is not a list of authors, beyond those cited in the references, whose results are reported in the text. However, as the references cited are leading ones, consulting them, along with the use of appropriate works of the secondary literature, will rapidly lead to the complete literature related to any particular subject covered. Each entry in the author index refers the user to the appropriate section number.

2.6. Compound Index The compound index lists individual, fully specified compositions of matter that are mentioned in the text. It is an index of empirical formulas, ordered according to the following system: the elements within a given formula occur in alphabetical sequence except for C, or C and H if present, which always come first. Thus, the empirical formula for Ti(SO,), is BH,-NH, Be,CO, CsHBr, Al(HCO,),

O,S,Ti BH6N CBe,O, Br,CsH C,H,AlO,

The formulas themselves are ordered alphanumerically without exception; that is, the formulas listed above follow each other in the sequence BH,N, Br,CsH, CBe,O,, C,H,AlO,, O,S,Ti. A compound index constructed by these principles tells whether a given compound is present. It cannot provide information about compound classes, for example, all aluminum derivatives or all compounds containing phosphorus. In order to open this route of access as well, the compound index is augmented by successively permuted versions of all empirical formulas. Thus the number of appearances that an empirical formula makes in the compound index is equal to the number of elements it contains. As an example, C,H,AlO,, mentioned above, will appear as such and, at the appropriate positions in the alphanumeric sequence, as H3A109*C3,AlO,*C,H, and O,*C,H,Al. The asterisk identifies a permuted formula and allows the original formula to be reconstructed by shifting to the front the elements that follow the asterisk. Each nonpermuted formula is followed by linerarized structural formulas that indicate how the elements are combined in groups. They reveal the connectivity of the compounds underlying each empirical formula and serve to

How to Use this Book

xix

distinguish substances which are identical in composition but differ in the arrangement of elements (isomers). As an example, the empirical formula C,H *,O might be followed by the linearized structural formulas (CH3CH2),0, CH, (CH2),0CH3, (CH,),CHOCH,, CH,(CH,),OH, (CH,),CHCH,OH and CH, CH2(CH,)CHOH to identify the various ethers and alcohols that have the element count C,H,,O. Each linearized structural formula is followed in a third column by keywords describing the context in which it is discussed and by the number($ of the section(s) in which it occurs.

2.7. Subject index The subject index provides access to the text by way of methods, techniques, reaction types, apparatus, effects and other phenomena. Also, it lists compound classes such as organotin compounds or rare-earth hydrides which cannot be expressed by the empirical formulas of the compound index. For multiple entries, additional keywords indicate contexts and thereby avoid the retrieval of information that is irrelevant to the user’s need. Again, section numbers are used to direct the reader to those positions in the book where substantial information is to be found.

2.8. Periodic Table (Back Endpaper) Reference to periodic groups avoids cumbersome enumerations. Section headings in the series employ the nomenclature. Unfortunately, however, there is at the present time no general agreement on group designations. In fact, the scheme that is most widely used (combining a group number with the letters A and B) is accompanied by two mutually contradictory interpretations. Thus, titanium may be a group IVA or group IVB element depending on the school to which one adheres or the part of the world in which one resides. In order to clarify the situation for the purposes of the series, a suitable labeled periodic table is printed on the inside back cover of each volume. All references to periodic group designations in the series refer to this scheme.

Preface to the Series Inorganic Reactions and Methods constitutes a closed-end series of books designed to present the state of the art of synthetic inorganic chemistry in an unprecedented manner. So far, access to knowledge in inorganic chemistry has been provided almost exclusively using the elements or classes of compounds as starting points. In the first 18 volumes of Inorganic Reactions and Methods, it is bond formation and type of reaction that form the basis of classification. This new route of access has required new approaches. Rather than sewing together a collection of review articles, a framework has had to be designed that reflects the creative potential of the science and is hoped to stimulate its further development by identifying areas of research that are most likely to be fruitful. The reaction volumes describe methods by which bonds between the elements can be formed. The work opens with hydrogen making a bond to itself in H, and proceeds through the formation of bonds between hydrogen and the halogens, the groups headed by oxygen, nitrogen, carbon, boron, beryllium and lithium to the formation of bonds between hydrogen and the transition and inner-transition metals and elements of group zero. This pattern is repeated across the periodic system until all possible combinations of the elements have been treated. This plan allows most reaction topics to be included in the sequence where appropriate. Reaction types that do not arise from the systematics of the plan are brought together in the concluding chapters on oxidative addition and reductive elimination, insertions and their reverse, electron transfer and electrochemistry, photochemical and other energized reactions, oligomerization and polymerization, inorganic and bioinorganic catalysis and the formation of intercalation compounds and ceramics. The project has engaged a large number of the most able inorganic chemists as Editorial Advisors creating overall policy, as Editorial Consultants designing detailed plans for the subsections of the work, and as authors whose expertise has been crucial for the quality of the treatment. The conception of the series and the details of its technical realization were the subject of careful planning for several years. The distinguished chemists who form the Editorial Advisory Board have devoted themselves to this exercise, reflecting the great importance of the project. It was a consequence of the systematics of the overall plan that publication of a volume had to await delivery of its very last contribution. Thus was the defect side of the genius of the system revealed, as the excruciating process of extracting the rate-limiting manuscripts began. Intense editorial effort was xx i

xxii

Preface to the Series

required in order to bring forth the work in a timely way. The production process had to be designed so that the insertion of new material was possible up to the very last stage, enabling authors to update their pieces with the latest developments. The publisher supported the cost of a computerized bibliographic search of the literature and a second one for updating. Each contribution has been subjected to an intensive process of scientific and linguistic editing in order to homogenize the numerous individual pieces, as well as to provide the highest practicable density of information. This had several important consequences. First, virtually all semblances of the authors’ individual styles have been excised. Second, it was learned during the editorial process that greater economy of language could be achieved by dropping conventionally employed modifiers (such as very) and eliminating italics used for emphasis, quotation marks around nonquoted words, or parentheses around phrases, the result being a gain in clarity and readability. Because the series focuses on the chemistry rather than the chemical literature, the need to tell who has reported what, how and when can be considered of secondary importance. This has made it possible to bring all sentences describing experiments into the present tense. Information on who published what is still to be found in the reference lists. A further consequence is that authors have been burdened neither with identifying leading practitioners, nor with attributing priority for discovery, a job that taxes even the talents of professional historians of science. The authors’ task then devolved to one of describing inorganic chemical reactions, with emphasis on synthetic utility, yield, economy, availability of starting materials, purity of product, specificity, side reactions, etc. The elimination of the names of people from the text is by far the most controversial feature. Chemistry is plagued by the use of nondescriptive names in place of more expository terms. We have everything from Abegg’s rule, Adkin’s catalyst, Admiralty brass, Alfven number, the Amadori rearrangement and Adurssov oxidation to the Zdanovskii law, Zeeman effect, Zincke cleavage and Zinin reduction. Even well-practiced chemists cannot define these terms precisely except for their own areas of specialty, and no single source exists to serve as a guide. Despite these arguments, the attempt to replace names of people by more descriptive phrases was met in many cases by a warmly negative reaction by our colleague authors, notwithstanding the obvious improvements wrought in terms of lucidity, freedom from obscurity and obfuscation and, especially, ease of access to information by the outsider or student. Further steps toward universality are taken by the replacement of element and compound names wherever possible by symbols and formulas, and by adding to data in older units their recalculated SI equivalents. The usefulness of the reference sections has been increased by giving journal-title abbreviations according to the Chemical Abstracts Service Source Index, by listing in each reference all of its authors and by accompanying references to patents and journals that may be difficult to access by their Chemical Abstracts cita-

Preface to the Series

xxiii

tions. Mathematical signs and common abbreviations are employed to help condense prose and a glossary of the latter is provided in each volume. Dangerous or potentially dangerous procedures are highlighted in safety notes printed in boldface type. The organization of the material should become readily apparent from an examination of the headings listed in the table of contents. Combining the words constituting the headings, starting with the major heading (one digit) and continuing through the major chapter heading (two digits), division heading (three digits), section heading (four digits) to the subsection heading (five digits), reveals at once the subject of a “slice” of the plan. Each slice is a selfcontained unit. It includes its own list of references and provides definitions of unusual terms that may be used in it. The reader, therefore, through the table of contents alone, can in most instances quickly reach the desired material and derive the information wanted. In addition there is for each volume an author index (derived from the lists of references) and a subject index that lists compound classes, methods, techniques, apparatus, effects and other phenomena. An index of empirical formulas is also provided. Here in each formula the element symbols are arranged in alphabetical order except that c, or C and H if present, always come first. Moreover, each empirical formula is permuted successively. Each permuted formula is placed in its alphabetical position and cross referenced to the original formula. Therefore, the number of appearances that an empirical formula makes in the index equals the number of its elements. By this procedure all compounds containing a given element come together in one place in the index. Each original empirical formula is followed by a linearized structural formula and keywords describing the context in which the compound is discussed. All indexes refer the user to subsection rather than page number. Because the choice of designations of groups in the periodic table is currently in a state of flux, it was decided to conform to the practice of several leading inorganic texts. To avoid confusion an appropriately labeled periodic table is printed on the back endpaper. From the nature of the work it is obvious that probably not more than two persons will ever read it entire: myself and the publisher’s copy editor, Dr. Lindsay S. Ardwin. She, as well as Ms. Mary C. Stradner, Production Manager of VCH Publishers, are to be thanked for their unflagging devotion to the highest editorial standards. The original conception for this series was the brainchild of Dr. Hans F. Ebel, Director of the Editorial Department of VCH Verlagsgesellschaft in Weinheim, Federal Republic of Germany, who also played midwife at the birth of the plan of these reaction volumes with my former mentor, Professor Alan G. MacDiarmid of the University of Pennsylvania, and me in attendance, during the Anaheim, California, American Chemical Society Meeting in the Spring of 1978. Much of what has finally emerged is the product of the inventiveness and imagination of Professor Helmut Grunewald, President of VCH Verlagsgesellschaft. It is a pleasure to

xxiv

Preface to the Series

acknowledge that I have learned much from him during the course of our association. Ms. Nancy L. Burnett is to be thanked for typing everything that had to do with the series from its inception to this time. Directing an operation of this magnitude without her help would have been unimaginable. My wife Rose stood by with good cheer while two rooms of our home filled up with 10,OOO manuscript pages, their copies and attendant correspondence. Finally, and most important, an enormous debt of gratitude toward all our authors is to be recorded. These experts were asked to prepare brief summaries of their knowledge, ordered in logical sequence by our plan. In addition, they often involved themselves in improving the original conception by recommending further refinements and elaborations. The plan of the work as it is being published can truly be said to be the product of the labors of the advisors and consultants on the editorial side as well as the many, many authors who were able to augment more general knowledge with their own detailed information and ideas. Because of the unusually strict requirements of the series, authors had not only to compose their pieces to fit within narrowly constrained limits of space, format and scope, but after delivery to a short deadline were expected to stand by while an intrusive editorial process homogenized their own prose styles out of existence and shrank the length of their expositions. These long-suffering colleagues had then to endure the wait for the very last manuscript scheduled for their volume to be delivered so that their work could be published, often after a further diligent search of the literature to insure that the latest discoveries were being cited and that claims for facts now proved false were eliminated. To these co-workers (270 for the reaction volumes alone), from whom so much was demanded but who continued to place their knowledge and talents unstintingly at the disposal of the project, we dedicate this series. J. J. ZUCKERMAN Norman, Oklahoma July 4, 1985 The scientific community is appreciative of the JJZ vision for a systematic inorganic chemistry. Many of the contributions had been edited prior to his death; therefore, his precise syntax will remain an important part of the series. A.P. HAGEN Norman, Oklahoma June 1, 1990

Editorial Consultants to the Series Professor H.R. Allcock Pennsylvania State University Professor J.S. Anderson University of Aberystwyth Professor F.C. Anson California Institute of Technology Dr. M.G. Barker University of Nottingham Professor D.J. Cardin Trinity College Professor M.H. Chisholm Indiana University

Professor J.R. Etourneau Laboratoire de Chemie du Solide du C.N.R.S. Professor G.L. Geoffroy Pennsylvania State University Professor L.S. Hegedus Colorado State University Professor W.L. Jolly University of California at Berkeley Professor C.B. Meyer University of Washington Professor H. Noth Universitiit Munchen

Professor C. Cros Laboratoire de Chemie du Solide du C.N.R.S.

Professor H. Nowotny University of Connecticut

Dr. B. Darriet Laboratoire de Chemie du Solide du C.N.R.S.

Dr. G.W. Parshall E.I.du Pont de Nemours

Professor E.A.V. Ebsworth University of Edinburgh

Professor M. Pouchard Laboratoire de Chemie du Solide du C.N.R.S.

Professor J.J. Eisch State University of New York at Binghamton

Professor J. Rouxel Laboratoire de Chimie Minirale au C.N.R.S.

xxv

xxvi

Editorial Consultants to the Series

Professor R. Schmutzler Technische Universitat Braunschweig

Dr. N. Sutin Brookhaven National Laboratory

Professor A.W. Searcy University of California at Berkeley

Professor R.A. Walton Purdue University

Professor D. Seyferth Massachusetts Institute of Technology

Dr. J.H. Wernick Bell Laboratories

Contributors to Volume 17 Professor H. R. Allcock Department of Chemistry Pennsylvania State University University Park, Pennsylvania 16802 (Sections 15.1.1, 15.1.2, 15.2.11) Professor Dr. J. 0. Besenhard Anorganisch-Chemisches Institut Westfalische Wilhelms-Universitat Miinster 4400 Munster, BRD (Sections 16.4.2.5,- 16.4.2.8) Professor R. K. Bunting Department of Chemistry Illinois State University Normal, Illinois 61761-6901 (Sections 15.2.5, 15.2.6) Professor T. Chivers Department of Chemistry The University of Calgary Alberta, T2N 1N4 Canada (Section 15.2.12)

Dr. C. E. Carraher, Jr. Department of Chemistry Florida Atlantic University Boca Raton, Florida 33431 (Sections 15.1.13, 15.2.14) Dr. C. Cros Laboratoire de Chemie du Solide du CNRS 351, cours de la Liberation 33405 Talence, Cedex (Sections 16.2.1- 16.2.3)

Dr. J. P. Doumerc Laboratoire de Chemie du Solide du CNRS Universite de Bordeaux I 33405 Talence, Cedex (Section 16.3.2)

Dr. L. B. Ebert Corporate Research Exxon Research and Engineering Company Clinton Township Rt 22 East Annandale, New Jersey 08801 (Sections 16.4.1-16.4.2.1) Professor J. P. Francois Department of Chemistry Limburgs Universitair Centrum 3610 Diepenbeek, Belgium (Sections 15.2.1, 15.2.2)

Dr. C. L. Frye Health & Environmental Sciences Dow Corning Corporation Midland, Michigan 48686-0994 (Section 15.2.8) Dr. D. Gubrard Laboratoire de Chimie du Solide Mineral Universite de Nancy I B.P. 239-54506 Vandoeuvse les Nancy Cedex (Section 16.4.2.4) xxvii

xxviii

Contributors to Volume 17

Professor A. Herold Laboratoire de Chimie du Solide Mineral Universitk de Nancy I B.P. 239-54506 Vandoeuvse les Nancy Cedex (Sections 16.4.2.2- 16.4.2.4) Professor Dr. U. Klingebiel Institut fur Anorganische Chemie der Universitat Gottingen Tammannstr. 4 3400 Gottingen, BRD (Section 15.2.9) Dr. P. Lagrange Laboratoire de Chemie du Solide Mineral Universite de Nancy I B.P. 239-54506 Vandoeuvse les Nancy Cedex (Section 16.4.2.2.2)

dr. ir. C. H. Massen Department of Physics Eindhoven University of Technology 5600 Eindhoven, The Netherlands (Sections 15.2.1, 15.2.2) Professer T. P. Onak Department of Chemistry and Biochemistry California State University, Los Angeles Los Angeles, California 90032 (Section 15.2.7) Prof. J. A. Poulis Department of Physics Eindhoven University of Technology 5600 Eindhoven, The Netherlands (Sections 15.2.1, 15.2.2)

Professor B. Raveau Laboratoire de Cristallographie et Chimie du Solide Universite de Caen 14032 Caen, Cedex (Section 16.3.1) Professor A. L. Rheingold Department of Chemistry University of Delaware Newark, Delaware 19716 (Section 15.2.3) Professor J. Rouxel Laboratoire de Chimie des Solides Institut de Physique et Chimie des Materiaux Universite de Nantes 44072 Nantes, Cedex (Section 16.4.3) Dr. R. Setton Centre de Recherche sur les Solides B Organisation Cristalline Imparfaite, CNRS 45071 Orleans, Cedex 2 (Section 16.4.2.3) Dr. L. C. van Poucke Department of Chemistry Limburgs Universitair Centrum 5600 Eindhoven, The Netherlands (Sections 15.2.1, 15.2.2) Professor J. R. Van Wazer Department of Chemistry Vanderbilt University Nashville, Tennessee 37235 (Sections 15.1.3, 15.2.10) Dr. R. West Department of Chemistry University of Wisconsin Madison, Wisconsin 53706 (Section 15.2.4)

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc.

15. Oligomeriration and

Polymerization

15.1. Introduction 15.1.l.Definitions and Scope. 15.1.1.1, Definitions of Oligomers and Polymers.

The inorganic elements show a strong tendency to form covalently linked rings, cages and chains’-’. Although not as widely developed as carbon-based macromolecules, inorganic polymers promise far greater structural and chemical diversity and the synthesis of inorganic high polymers is a subject of growing interest. Much of, the synthetic work with inorganic rings and high polymers focuses on the main-group elements of groups IIIB-VIB. However, the transition elements form heteroatom rings, cages and chains, especially in combination with oxygen. Moreover, in the future an extensive oligomer and polymer chemistry may develop from metal-metal bonded species. In the following sections, the terms monomer, oligomer and polymer are used extensively. A monomer is a small molecule that can combine to form larger molecules-either cyclic or linear. Each monomer molecule becomes incorporated into a repeating unit in the ring or chain, and the larger molecule so formed consists of an extended array of repeating units. Oligomers are small rings or short chains formed by very limited polymerization sequences. In a typical oligomer, from 2 to perhaps 20 repeating units are linked together. Oligomers are definitely not polymers, although they are often confused with polymers. They have all the physical characteristics of small molecules and few, if any, of the attributes of high polymers. The term polymer or macromolecule is usually reserved for very large molecules, often with 100 or more repeating units. Most of the macromolecules that are useful in technology are in fact high polymers with 1,000-20,000 or more repeating units linked end to end. Copolymers are macromolecules in which more than one type of monomer is incorporated. Mixed-substituent polymers are macromolecules in which more than one type of side group has been introduced by substitutive techniques. The term polymer is very broad and describes a wide variety of different structures. If A and B are two inorganic elements and R is a side group unit (either inorganic or organic), a linear polymer structure can be represented by I or 11, where n is the degree of polymerization. End groups are usually not specified, except for linear oligomers. Typical cyclic oligomers are shown in I11 or IV.

4’Y 1 A-B-

,,-

R

\ /

R A ’

\I

B \A

B

I 2

I1

I l A-B-A

I/

R

A

I11

R-B

I

R

I

I

l

I

B-R

A-B-A

B

/ \ / \

R

R Y R

R

l

I

l

R R R IV

15.1. Introduction 15.1.1. Definitions and Scope. 15.1.1.1. Definitions of Oligomers and Polymers.

3

-cd

A linear high polymer can also be represented by the spaghetti-like strand shown in V and, in these terms, we can recognize branched polymers (VI) and random crosslinked polymers (VII). Ladder polymers (VIII) are also known:

d@?J/

e V

VI

VII

VIII

Cyclolinear polymers are macromolecules formed by the linkage of intact ring systems, as depicted in IX, whereas cyclomatrix polymers (X) are formed by the three-dimensional crosslinkage of ring systems. \ /

B’

A

I

R

R

R

\ B-A /

\R..

\ B-A /

-A / \

\

/

\

R

R

/

\B/

B/A\

FA\ B-A R R IX

/B-A\

R

/

‘B I

1

B

R”

\

X

Linear, branched and cyclolinear polymers are usually soluble. Lightly crosslinked polymers (VII) are swelled by, but are insoluble in liquids. Highly crosslinked or cyclomatrix polymers are insoluble in all media. Oligomers dissolve in liquids to give non-viscous solutions. High polymers dissolve to give highly viscous solutions. Unlike macromolecules of biological origin, synthetic polymers do not possess a discrete molecular weight. The chain construction process yields polymers with different chain lengths in the same reaction mixture, polymers are usually characterized by a molecular weight distribution. A broad molecular weight distribution is an indication of an inefficient chain growth mechanism. For example, competing side reactions such as chain transfer, depolymerization to oligomers or crosslinking broaden the molecular weight distribution; rapid initiation of chain growth, followed by efficient chain propagation, yields a narrow molecular weight distribution. As will be seen in later sections, inorganic oligomers are vital in the synthesis of macromolecules and are the main products of polymer decomposition at high tempera-

4

15.1. Introduction 15.1.1. Definitions and Scope.

tures. Most of the known inorganic polymers are synthesized from cyclic oligomers either by ring-opening reactions or by ring-coupling processes. (H.R. ALLCOCK)

F. G. A. Stone, W. A. G. Graham, eds., Inorganic Polymers, Academic Press, New York, 1962. H. R. Allcock, Heteroatom Ring Systems and Polymers, Academic Press, New York, 1967. I. Haiduc, The Chemistry of Inorganic Ring Systems, Wiley-Interscience, New York, 1970. D. A. Armitage, Inorganic Rings and Cages, Arnold, 1972. A. L. Rheingold, ed., Homoatomic Rings, Chains, and Macromolecules of Main Group Elements, Elsevier, Amsterdam, 1977. 6. H. R. Allcock, F. W. Lampe, Contemporary Polymer Chemistry, Prentice Hall, Englewood Cliffs, New Jersey, 1981, 1990. 7. M. Zeldin, K. J. Wynne, H. R. Allcock, eds., Inorganic and Organometallic Polymers, A.C.S. Symposium Series 360, American Chemical Society, Washington, D.C. 1988.

1. 2. 3. 4. 5.

15.1.1.2. General Synthesis Methods.

Four general methods are available for the synthesis of oligomers and polymers. (i) Condensation. Condensation reactions involve the linkage of two or more monomer molecules, accompanied by a loss of a small molecule such as H,O, NH, or HC1. Such reactions can be illustrated by the processes shown in Eqs. (a)-(c), where E is an inorganic element:

R

R

I

I + HO-E-OH I

HO-E-OH

I

R

R

- HzO

R

R

R I

R

R

I I HO-E-0-E-0-E-OH I I R

R

R

I I

-..\i R

I

HO-E-OH

R I 7 HO-E H2O R

R

I I HO-E-0-E-OH I I

I

R

0-E

(a)

R

:i, T

0-E-OH R

\

-HzO

L

chain growth

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc.

4

15.1. Introduction 15.1.1. Definitions and Scope.

tures. Most of the known inorganic polymers are synthesized from cyclic oligomers either by ring-opening reactions or by ring-coupling processes. (H.R. ALLCOCK)

F. G. A. Stone, W. A. G. Graham, eds., Inorganic Polymers, Academic Press, New York, 1962. H. R. Allcock, Heteroatom Ring Systems and Polymers, Academic Press, New York, 1967. I. Haiduc, The Chemistry of Inorganic Ring Systems, Wiley-Interscience, New York, 1970. D. A. Armitage, Inorganic Rings and Cages, Arnold, 1972. A. L. Rheingold, ed., Homoatomic Rings, Chains, and Macromolecules of Main Group Elements, Elsevier, Amsterdam, 1977. 6. H. R. Allcock, F. W. Lampe, Contemporary Polymer Chemistry, Prentice Hall, Englewood Cliffs, New Jersey, 1981, 1990. 7. M. Zeldin, K. J. Wynne, H. R. Allcock, eds., Inorganic and Organometallic Polymers, A.C.S. Symposium Series 360, American Chemical Society, Washington, D.C. 1988.

1. 2. 3. 4. 5.

15.1.1.2. General Synthesis Methods.

Four general methods are available for the synthesis of oligomers and polymers. (i) Condensation. Condensation reactions involve the linkage of two or more monomer molecules, accompanied by a loss of a small molecule such as H,O, NH, or HC1. Such reactions can be illustrated by the processes shown in Eqs. (a)-(c), where E is an inorganic element:

R

R

I

I + HO-E-OH I

HO-E-OH

I

R

R

- HzO

R

R

R I

R

R

I I HO-E-0-E-0-E-OH I I R

R

R

I I

-..\i R

I

HO-E-OH

R I 7 HO-E H2O R

R

I I HO-E-0-E-OH I I

I

R

0-E

(a)

R

:i, T

0-E-OH R

\

-HzO

L

chain growth

15.1. introduction 15.1.1. Definitions and Scope. 15.1.1.2. General Synthesis Methods.

5

~~

These reactions are only a sample of the many condensation steps that are possible. Monomer molecules may react with chains of any length. Moreover, linear oligomers or polymers may link together, and the end groups of open-chain molecules can react with rings to cleave them, or with the middle units of its own chain to split out rings. Note that at every step in the chain growth process, an option exists for cyclization or further chain growth. In practice, because of the bond angles involved and the effect of these on the statistics of ring closure’, cyclization is especially favored if the product is a 6-, 8-, lo-, or 12-membered ring. The probability of cyclization then declines as the chain length increases because of the decreasing chance of the chain ends finding each other. This is the reason that cyclic oligomers are the principal initial products formed in most inorganic condensation reactions. It also explains why attempts to prepare high polymers by this technique often fail. Another reason is that cyclization (a unimolecular process) is favored over linear chain growth in solution, especially at low concentrations. The driving force for condensation is the removal of the small-molecule side product. Most condensation processes are reversible and readily establish complex equilibria. Only when the water, for example, is removed from the system can high conversions to cyclic oligomers or long-chain polymers be achieved. Condensation processes are used to prepare oligomers of organosiloxanes, silazanes, silthianes, borazines, borazanes, metaborates, metaphosphates, phosphazenes, thiazyls, alazanes, metal oxides and so on. (ii) Addition. Addition reactions are less common in inorganic than in organic chemistry. This process usually involves a linear coupling of unsaturated molecules to form saturated larger molecules:

The driving force for such reactions is the loss of multiple-bond enthalpy. Addition reactions of this type may be involved in the formation of sulfur-nitrogen oligomers. An alternative form of addition process that may be more common in inorganic systems is the coupling of fragments that contain only one skeletel atom:

(iii) Ring-Opening Polymerization. This is the main method of preparing inorganic high polymers. Schematically, and using the same symbolism as before, a ring-opening polymerization is depicted by:

Such reactions are well known for siloxanes, silazanes, metaphosphates, phosphazenes, thiazenes, sulfur, cyclosilanes, etc. In principle, no strong driving force exists for

6

15.1. Introduction 15.1.1. Definitions and Scope. 15.1.1.2. General Synthesis Methods.

polymerizations of this kind because the enthalpy per repeating unit is similar in the polymer and in the cyclic oligomer. Hence, equilibrating polymeric series are common in systems of this kind. In these, oligomers and a broad range of polymers interact and interconvert at elevated temperatures. This is one reason many inorganic high polymers depolymerize to cyclic oligomers at elevated temperatures. As the lower oligomeric members of the series volatilize from the system, the equilibr&e readjusts to convert more of the polymer to oligomers. Such systems are especially sensitive to entropic factors. For example, the preponderance of the TAS term at elevated temperatures favors the formation of many small oligomer molecules at the expense of the high polymers. Catalysts are often needed to facilitate ring-opening polymerization reactions. Although the S,-S, equilibration is believed to be a free-radical process, most ringopening polymerizations follow an ionic pathway: B/

A

I

A

I

A \ B / A-Y

\B

I + x + y -I

A

\B/

I

A

1-1

\BY

B/

A

\B-X

I

A

A

\B/

A B/ B’‘

A\B/A--Y

A--Y

A

Xd+

B ‘

I A \B/

B*-

I

A

Different polymers or oligomers can be prepared by the equilibration of oligomers with different side groups. Copolymerization of oligomers is also possible. These techniques are particularly applicable to organosiloxane and organophosphazene systems. (iv) Substitution. The skeleton of an oligomeric or polymeric system is constructed by condensation, addition or ring-opening techniques, but variations in side-group structure must generally be introduced by substitutive methods:

The chemistry of cyclic oligomeric and high polymeric phosphazenes is dominated by substitution reactions, as is the chemistry of boron-nitrogen heterocycles. However, substitutive techniques play only a minor role in the chemistry of siloxane, silazane, phosphate or metaborate oligomers and polymers. It is not yet well developed in sulfur-nitrogen systems but may become more important in the future.

1. H R. Allcock, J . Macromol. Sci., Rev. Macromol. Chem., C4, 149 (1970).

(H R. ALLCOCK)

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc.

15.1.1. Definitions and Scope. 15.1.1.3.Scope. 15.1.1.3.1. Small-Ring Systems-Examples of Synthesis.

7

15.1.1.3. Scope. 15.1.1.3.1. Small-Ring Systems-Examples

of Synthesis.

(i) Condensation Syntheses. Condensation is the principle method for the synthesis of inorganic small-ring systems. One mechanism involves loss of water, and others take place with loss of hydrohalides, silicon halides, metal halides, ammonia or amines, alkanes or hydrogen, and nitrogen. A well-known example of cyclic oligomer formation by loss of water is the preparation of cyclic organosiloxanes during the hydrolysis of diorganodichlorosilanes:

R

o/

-HzO

I

Si

R’

R

Si

R \0

I/R+ Si R ‘

‘0’

R

R

I I R-Si-0-Si-R I I 0 0 I I R-Si-0-Si-R

I

R

I

R

(a) The ease of condensation in this reaction decreases with increasing size of the side group, R. This type of condensation reaction is also used for the preparation of cyclic metaborates and cyclic metaphosphates:

OM

I

OM

I

3 HO-B-OH

B

-H20 >

o/ \o I

B / \

MO

I

B / \ 0 OM

where M is a metallic cation. Obviously, unless the functionality of boric or phosphoric acids is reduced by partial salt formation, the products will be a three-dimensional crosslinked matrix rather than cyclic or linear species. The same restriction of functionality can be achieved by the formation of organic species. For example, PhB(OH), yields (OBPh), on condensation. The driving force for such condensations is the removal of water either by volatilization or by spontaneous separation from a hydrophobic phase. Unless the water is removed, complex mixtures of monomer, cyclic species and linear fragments are formed. The loss of HCl or H F provides the momentum for a large number of oligomerization reactions. The hydrohalide is removed either by volatilization or by insolubilization as R salt with added amine.

8

15.1. Introduction 15.1.1. Definitions and Scope. 15.1.1.3. Scope.

Borazine formation is a well-known example’ e.g.:

PhNH,

+ BCl,

-

PhH,N: BCl,

B ‘NPh

PhN’

heat

I

1

C l B y BCl

(dl

_.

Ph

Cyclic dimeric or tetrameric boron-nitrogen compounds can be prepared by the same route. Boron-sulfur heterocycles can be generated:

Boron-phosphorus cyclic trimers can be prepared similarly:

Me,BBr

+ HPMe,

-

Me2 D

Et3N

Me,Br :PHMe,

MezB

1

/=

\

BMez

I

Me2P,B/PMe,

0

The hydrohalide elimination route has also been used for the synthesis of cyclosilazanes and cyclosilthianes:

H R,Si -N-SiR,

R2

HN/Si\ R,SiCI,

+ NH,

1

R,Si

NH

I

N ‘’

SiR, H

+

I

I

I I

NH , etc.

HN R,Si-N-SiR, H

Cg)

Perhaps the most widely studied of the dehydrochlorination syntheses are the reactions of halophosphoranes with ammonia to form cyclophosphazenes’. This is the classic route from which nearly all cyclo- and polyphosphazene chemistry is derived. Both halogeno- and organocyclophosphazenes have been prepared by this method:

x

x

15.1.1. Definitions and Scope. 15.1.1.3. Scope. 15.1. I .3.1. Small-Ring Systems--Examples of Synthesis.

9

where X = C1 or Br;

R,PCl,

+ NH,Cl=

R, /R P\ N4 N R, I II,R etc. P R/ R

+\

(.i1

The hydrohalide is volatilized from the reaction system, driving the condensation to completion. Cyclophosphazanes are formed when tricoordinate phosphorus halides react with amines:

c1

\

pcl3+PhNHzz

Ph

P-N

I

,N-P Ph

/

I

(k) \

c1

The hydrohalide elimination route is also widely employed for the synthesis of cyclic sulfur-nitrogen compounds, such as tetrasulfur tetranitride (cyclotetrathiazyl), thiazyl halides, and sulfanuric halides3:

OSF,+NH,=

0% / F S N4 \N

p

F I \S 04 %N/

(1)

‘F

Silicon halide elimination provides a driving force for the construction of cyclic oligomers, with the volatility of the silicon halide being used to push the condensation process to completion4:

BClR,

+ Me,PSiMe,

Me2 P R,B’ ‘BR,

I

I

Me2P\B/PMe2

(m)

R2 Precipitation of insoluble metal halide from a solution of the reactants is often used to drive oligomerization reactions to completion. Such reactions may be of two types: type a, in which a metal derivative of one reactant interacts with a halide of another, or type b, where a reactive metal removes halogen from two reactant molecules. An example of situation (a) is the synthesis of the cyclodisilazane: Me2 Si MeN’ “Me Li Li

+ Me,SiCl,

Me, Si

/

MeN,

\

./ NMe s1

Me2

(n)

15.1. Introduction 15.1.1. Definitions and Scope. 15.1.1.3. Scope.

10

and an example of reaction type (b) is provided by the formation of c y c l ~ s i l a n e s ~ ’ ~ :

Me,SiCl,

Me, Si

\ SiMe,

Me,Si/

+ Na-K

1

1

Me,%\

,SiMe,

etc.

Si Me, Related cyclotin compounds of formula (R,Sn),,,,,,,,,,,,,, can be prepared by the dehydrogenation of R,SnH, in the presence of amines, or by the elimination of MgCl, or NaCl when RSnC1, reacts with magnesium or sodium7. Ammonium chloride elimination has been used for the construction of cyclophosphazenes8:

N-Silylphosphoranimines eliminate substituted silanes when heated, to yield either cyclic or high polymeric p h o s p h a z e n e ~ ~ * ’ ~ :

-

- Me3SiX

Me,SiN=PR,-X

(NPR,),

The formation of cyclic species is favored when the leaving group X is a halogen. Similarly, the pyrolysis of bis(alky1amino)diorganosilazanes brings about cyclization by loss of an amine: R,Si(NHR‘),

heat

-R”H; (R,SiNR’),

Pyrolysis techniques that involve the elimination of hydrogen or alkanes provide facile routes to boron- or aluminum-containing small-ring systems. For example, borazines are synthesized by the pyrolysis of mixtures of boranes and amines:

3 MeNH,

3 NH,

+f

H B M e d \NMe

B2H6

+ 3 BMe,

=

I

I

HB

BH

‘N/ Me

Me B HN/ ‘NH

I

I

MeB\

N H

/

BMe

Similarly, aluminum-nitrogen rings are formed when triorganoaluminum compounds react pyrolytically with amines:

15.1.1. Definitions and Scope. 15.1.1.3. Scope. 15.1.1.3.1. Small-Ring Systems-Examples

11 of Synthesis.

Finally, elimination of nitrogen from phosphinous azides leads to the formation of cyclophosphazenes'

These latter reactions may proceed through the intermediate formation of a monomer rather than through sequential condensation steps. (ii) Addition Reactions. Fewer addition reactions have been identified in the synthesis of inorganic oligomeric rings than in organic chemistry. Usually, unsaturated inorganic monomers are presumed to participate in ring formation only as short-lived intermediates generated by condensation methods. However, in at least one case, a monomer has been isolated as a quasistable compound. Thus, monomeric thiazyl fluoride can be isolated by several techniques and trimerizes to the cyclic trithiazyl fluoride during storage in a sealed glass vessel'':

F

Transient monomeric intermediates may be formed during the pyrolytic synthesis of borazines. The pathway shown in Eq. (x) has been postulated for the reaction of ammonia with diborane:

J.

HN' I

HB\

(4 H B 'NH I BH N

/

H

As mentioned above, the decomposition of phosphinous azides may yield monomers that trimerize or tetramerize. (iii) Ring-Equilibration Syntheses. Cyclic oligomers can be prepared by ring-ring equilibration reactions. For example, cyclodithiazyl, (SN), , is prepared by heating (SN),. A wide range of cyclic organosiloxanes is produced by the equilibrium of (Me'SiO), or ,. Some cyclophosphazenes undergo ring-ring equilibration; e.g. (N=PMe,),, is converted by heating to an equilibrium mixture of the cyclic trimer and tetramer. The cyclic trimer [NP(OCH,CF,),] equilibrates to higher cyclic analogs at elevated temperatures. Sulfur rings (S,) yield a range of higher and lower cyclic oligomers when heated in the melt. The cyclosilthiane (Me,SiS),, forms an equilibrium system with (Me,SiS), . Thus,.equilibration, often at high temperatures, allows conversion of one ring size to another. The main practical problem in these syntheses is the development of

12

15.1. Introduction 15.1.1. Definitions and Scope. 15.1.1.3. Scope.

separation procedures for the (sometimes complex) mixtures of cyclic oligomers that are formed. Separation techniques that require high temperatures are often unsatisfactory because equilibration reactions continue during separation. (iv) Substitution Reactions. Nearly all the cyclic oligomers that possess halogen atoms linked to the skeleton are capable of undergoing nucleophilic substitution reactions with replacement of the halogen. Many and diverse cyclic oligomeric organophosphazenes are prepared by such routes2'13, as discussed in $15.2.11. The linkage of inorganic ring systems by difunctional reagents provides a method for the synthesis of cyclolinear or cyclomatrix polymers. (H.R. ALLCOCK)

1. H. R. Allcock, Heteroatom Ring Systems and Polymers, Academic Press, New York, 1967. 2. H. R. Allcock, Phosphorus-Nitrogen Compounds, Academic Press, New York, 1972. 3. F. Seel, G. Simon, Z . Naturjorsch, Teil 8,19, 354 (1964). 4. H. Noth, W. Schragle, Angew. Chem., Int. Ed. Engl., 1,457 (1962). 5. N. Graf zu Stollberg, Angew. Chem., lnt. Ed. Engl., 2, 150 (1963). 6. E. Carberry, R. West, J . Organomet. Chem., 6, 583 (1966). 7. W. P. Newman in Homoatomic Rings, Chains, and Macromolecules of Main Group Elements, A. L. Rheingold, ed., Elsevier, Amsterdam, 1977, Ch. 11. 8. H. H. Sisler, S . E. Frazier, R. G. Rice, M. G. Sanchez, lnorg. Chem., 5, 326 (1966). 9. P. Wisian-Nielson,R. H. Nielson, J . Am. Chem. SOC.,102, 2898 (1980). 10. R. H. Nielson, P. Wisian-Nielson, Chem. Rev., 88, 541 (1988). 11. D. L. Herring, C. M. Douglas, Inorg. Chem., 4, 1012 (1965). 12. 0.Glemser,in Preparative Inorganic Reactions, Vol. 1, W. L. Jolly, ed., Wiley-Interscience,New York, 1964, Ch. 9. 13. H. R. Allcock, Acc. Chem. Res., 12, 351 (1979). 15.1.1.3.2. Cages-Examples

of Synthesis.

Borane and carborane cage structures are the most widely studied oligomeric cages'. The principal starting material for cage synthesis is diborane, which on pyrolysis under various conditions yields open cages such as B,H,, B,H,,, B,H,, and BioH14. The best known closed-cage boranes are anions of formula [BnHn12-,where n is 6-12. Of particular interest is the closed-cage anion B,,H:;, which is prepared from B,,Hl4 and a base, such as a tertiary amine. The important closed-cage carborane B,,C,H,, is also prepared from B1,H14 by treatment with diethylsulfide and an acetylene: nido-B,,H,,

+ 2 Et,S

-

B,,H,,(SEt,),

!

HC-CH

+ H, (4

B 1OCZH, 4 The two carbons enter the icosahedral cage adjacent to each other (0-carborane). At temperatures 2 450°C,they migrate away from each other across the surface of the cage to yield m- and p-carboranes. Carborane cages can be linked by siloxane units to give poly(carborane-siloxanes).

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc. 12

15.1. Introduction 15.1.1. Definitions and Scope. 15.1.1.3. Scope.

separation procedures for the (sometimes complex) mixtures of cyclic oligomers that are formed. Separation techniques that require high temperatures are often unsatisfactory because equilibration reactions continue during separation. (iv) Substitution Reactions. Nearly all the cyclic oligomers that possess halogen atoms linked to the skeleton are capable of undergoing nucleophilic substitution reactions with replacement of the halogen. Many and diverse cyclic oligomeric organophosphazenes are prepared by such routes2'13, as discussed in $15.2.11. The linkage of inorganic ring systems by difunctional reagents provides a method for the synthesis of cyclolinear or cyclomatrix polymers. (H.R. ALLCOCK)

1. H. R. Allcock, Heteroatom Ring Systems and Polymers, Academic Press, New York, 1967. 2. H. R. Allcock, Phosphorus-Nitrogen Compounds, Academic Press, New York, 1972. 3. F. Seel, G. Simon, Z . Naturjorsch, Teil 8,19, 354 (1964). 4. H. Noth, W. Schragle, Angew. Chem., Int. Ed. Engl., 1,457 (1962). 5. N. Graf zu Stollberg, Angew. Chem., lnt. Ed. Engl., 2, 150 (1963). 6. E. Carberry, R. West, J . Organomet. Chem., 6, 583 (1966). 7. W. P. Newman in Homoatomic Rings, Chains, and Macromolecules of Main Group Elements, A. L. Rheingold, ed., Elsevier, Amsterdam, 1977, Ch. 11. 8. H. H. Sisler, S . E. Frazier, R. G. Rice, M. G. Sanchez, lnorg. Chem., 5, 326 (1966). 9. P. Wisian-Nielson,R. H. Nielson, J . Am. Chem. SOC.,102, 2898 (1980). 10. R. H. Nielson, P. Wisian-Nielson, Chem. Rev., 88, 541 (1988). 11. D. L. Herring, C. M. Douglas, Inorg. Chem., 4, 1012 (1965). 12. 0.Glemser,in Preparative Inorganic Reactions, Vol. 1, W. L. Jolly, ed., Wiley-Interscience,New York, 1964, Ch. 9. 13. H. R. Allcock, Acc. Chem. Res., 12, 351 (1979). 15.1.1.3.2. Cages-Examples

of Synthesis.

Borane and carborane cage structures are the most widely studied oligomeric cages'. The principal starting material for cage synthesis is diborane, which on pyrolysis under various conditions yields open cages such as B,H,, B,H,,, B,H,, and BioH14. The best known closed-cage boranes are anions of formula [BnHn12-,where n is 6-12. Of particular interest is the closed-cage anion B,,H:;, which is prepared from B,,Hl4 and a base, such as a tertiary amine. The important closed-cage carborane B,,C,H,, is also prepared from B1,H14 by treatment with diethylsulfide and an acetylene: nido-B,,H,,

+ 2 Et,S

-

B,,H,,(SEt,),

!

HC-CH

+ H, (4

B 1OCZH, 4 The two carbons enter the icosahedral cage adjacent to each other (0-carborane). At temperatures 2 450°C,they migrate away from each other across the surface of the cage to yield m- and p-carboranes. Carborane cages can be linked by siloxane units to give poly(carborane-siloxanes).

13

15.1.1. Definitions and Scope. 15.1.1.3. Scope. 15.1.1.3.3. Short Chains-Examples of Synthesis

Oligomeric cages are also known for metal oxide systems. For example, the [MO,O,,]~- ion exists as a cage-type cluster with a molybdenum-oxygen framework. The anion is prepared by treating molybdenum trioxide with a base such as the cyclophosphazene [NP(NMe,),] z. These systems can be compared with metal-metal clusters and with nonmetallic clusters, such as those found in P,. Considerable interest exists in the incorporation of cage structures into macromolecular chains. (H.R.ALLCOCK) 1. E. L. Muetterties, Ed., Boron Hydride Chemistry, Academic Press, New York, 1975. 2. H. R. Allcock, E. C. Bissell, E. T. Shawl, Inorg. Chem., 12,2963 (1973).

15.1.1.3.3. Short Chains-Examples

of Synthesis.

Short-chain linear oligomers are usually prepared by the same techniques as those used for the preparation of cyclic oligomers. Linear oligomers are often formed as precursors to the cyclized species. Most condensation reactions yield a mixture of cyclic and linear oligomers with the ratios of the two depending on the reaction conditions and especially on the reactant ratios. The main chemical distinguishing features between small rings and short chains are the end groups in the latter. Normally, cyclization of a short chain occurs by loss of endgroup components: B-A-X A’

I

B\

- I - XY A-B--Y

B/ A,

A ‘B B

I

/A

The alternative (and much rarer) mode of cyclization involves the coupling of the freeradical end groups on a short chain, as in the sulfur equilibrate. Three main methods exist for producing short-chain oligomers. They are (i) stepwise chain growth from monomers, usually by condensation mechanisms; (ii) skeletal cleavage of cyclic oligomeric compounds; or (iii) extensive skeletal cleavage of long polymer chains. Because of the random nature of the cleavage in (iii), methods (i) and (ii) are preferred if discrete linear oligomers are to be prepared. In method (ii) it is sometimes possible to cleave the oligomeric ring by a reversal of the reaction that led to cyclization in the first place. (i) Stepwise Chain Growth from Monomers. Nearly all the methods reviewed earlier for the synthesis of cyclic oligomers can be applied to the preparation of short chains. The following are a few examples. Linear oligomers are formed when silanediols, dihydrogen borates or dihydrogen phosphates undergo loss of water and condensation. The degree of linear oligomerization depends on reaction conditions and, generally, a spectrum of different chain lengths may be formed. Limits to the chain growth can be imposed by the introduction of chain-

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc. 13

15.1.1. Definitions and Scope. 15.1.1.3. Scope. 15.1.1.3.3. Short Chains-Examples of Synthesis

Oligomeric cages are also known for metal oxide systems. For example, the [MO,O,,]~- ion exists as a cage-type cluster with a molybdenum-oxygen framework. The anion is prepared by treating molybdenum trioxide with a base such as the cyclophosphazene [NP(NMe,),] z. These systems can be compared with metal-metal clusters and with nonmetallic clusters, such as those found in P,. Considerable interest exists in the incorporation of cage structures into macromolecular chains. (H.R.ALLCOCK) 1. E. L. Muetterties, Ed., Boron Hydride Chemistry, Academic Press, New York, 1975. 2. H. R. Allcock, E. C. Bissell, E. T. Shawl, Inorg. Chem., 12,2963 (1973).

15.1.1.3.3. Short Chains-Examples

of Synthesis.

Short-chain linear oligomers are usually prepared by the same techniques as those used for the preparation of cyclic oligomers. Linear oligomers are often formed as precursors to the cyclized species. Most condensation reactions yield a mixture of cyclic and linear oligomers with the ratios of the two depending on the reaction conditions and especially on the reactant ratios. The main chemical distinguishing features between small rings and short chains are the end groups in the latter. Normally, cyclization of a short chain occurs by loss of endgroup components: B-A-X A’

I

B\

- I - XY A-B--Y

B/ A,

A ‘B B

I

/A

The alternative (and much rarer) mode of cyclization involves the coupling of the freeradical end groups on a short chain, as in the sulfur equilibrate. Three main methods exist for producing short-chain oligomers. They are (i) stepwise chain growth from monomers, usually by condensation mechanisms; (ii) skeletal cleavage of cyclic oligomeric compounds; or (iii) extensive skeletal cleavage of long polymer chains. Because of the random nature of the cleavage in (iii), methods (i) and (ii) are preferred if discrete linear oligomers are to be prepared. In method (ii) it is sometimes possible to cleave the oligomeric ring by a reversal of the reaction that led to cyclization in the first place. (i) Stepwise Chain Growth from Monomers. Nearly all the methods reviewed earlier for the synthesis of cyclic oligomers can be applied to the preparation of short chains. The following are a few examples. Linear oligomers are formed when silanediols, dihydrogen borates or dihydrogen phosphates undergo loss of water and condensation. The degree of linear oligomerization depends on reaction conditions and, generally, a spectrum of different chain lengths may be formed. Limits to the chain growth can be imposed by the introduction of chain-

l4

15.1. Introduction 15.1.1. Definitions and Scope. 15.1.1.3.Scope

d.

terminator or end-capping residues in the appropriate molar ratio to define the average chain length: R

I

R

HO-E-OH

I

I

+ HO-E-OH I

R

R

R

R

I

I

- HzO

HO-E-0-E-OH

I

R

R

I

V

-

O

H

R

I

R

I

HO-E-0-E-0-E-OH

I

R

I

R

R

I

I

etc.

R

The synthesis of linear phosphazene oligomers has been studied in some detail from the reaction of PCl, with ammonium chloride in tetrachloroethane or chlorobenzene’. The following chain growth mechanism is believed to operate:

cpc~,l~pc~,I + NH,

Cl,P=NH

+ [PCl,][PCl,] + NH, [Cl,P=N-PCl,][PCl,]

[Cl,P=N--PCl,][PCl,]

Cl,P=NH

Cl,P=N-PCl,=NH

(c> (4 (e)

Thus, the linear oligomeric chains are end capped by HCl or by PC1,. Loss of HCl causes cyclization. Clearly, these reaction mixtures contain a broad spectrum of linear oligomers, although the lower members of the series are favored if one reagent (for example, PCl,) is in excess. Linear oligomeric silylmethylenes are accessible by reaction sequences such as: Me ClSiCH,Cl Me

Mg

Me ClSiCH,MgCl Me

-

Me ClSiCHzCl Me -MgCl2

M~ M~ ClSiCH,SiCH,CI etc. Me Me

(f)

Since sequential reactions are involved, some control over the chain length can be exercised. Linear oligotin compounds are prepared by reactions of the type’: 2 R,SnNEt,

+ HSn(R’,)H aR,SnSn(RL)SnR,

- -

Oligomers of SiF, are prepared by the reaction: n SiF,

+ n Si

1150°C

n [SiF,]

(g)

15.1.1. Definitions and Scope. 15.1.1.3. Scope. 15.1.1.3.4. Long Chains-Examples

15

of Synthesis.

The initial product of oligomerization of SiF, is believed to be a diradical. (ii) Cleavage of Cyclic Oligomers. Typical ways in which to convert cyclic oligomers into short linear chains are: Me

Me

I I Me-Si-0-Si-Me I I 0 0 I I Me-Si0-SiI I Me

-

MeMe

I I/ Me-Si-0-Si-OSiMe,

+ NaCOSiMe,] Me

I

0

I

Me-Si-0-Si-ONa / \ 1 Me Me Me

Me

C1\

Me

/C1

P N4 N ' I/ ,cl + PCI, cl\ I P P C1/ \N / \ C1

-

Cl,P=N-P

c1

c1 I =N-P =NPCl, I c1 c1 I I

The linear products can be formed in high yield if a 1: 1 ratio of cyclic compound and cleavage reagent is maintained. Hydrogen chloride, AlCl,, and a variety of other halides are useful ring cleavage agents for silicon-containing rings. Water or base can be used to cleave metaborate or metaphosphate rings. A variety of colored short-chain anions can be prepared from S, rings in basic or reducing media, and many metal sulfides contain oligosulfur anions. (H.R ALLCOCK)

1. M. Becke-Goehnng, E. Fluck, Angew. Chem., Int. Ed. Engl., 1, 281 (1962). 2. W. P. Newman in Homoatomic Rings, Chains, and Macromolecules ofiMain group Elements, A. L. Rheingold, ed., Elsevier, Amsterdam, 1977, Ch. 11.

15.1.1.3.4. Long Chains-Examples

of Synthesis.

Two main techniques are employed for the synthesis of long, polymeric inorganic chains-condensation reactions and ring-opening polymerization. With few exceptions, condensation processes do not yield high polymers. The ease of skeletal scrambling reactions and depolymerization favors the formation of low polymers with very broad molecular weight distributions. Hence, the ring-opening polymerization technique is preferred for high-polymer synthesis. (i) Condensation Synthesis of Polymers. Poly(dimethy1phosphazene) with a molecular weight of 50,000 (650 repeating units) can be prepared by the condensation polymerization of a silylphosphinimine1s2: OCH,CF, I

Me,%-N=PMe,

19O"C,4 h - Me3SiOCH2CF3

' (NPMe,),

The elimination of Me,SiOCH,CF, is a new technique, but it is not yet known why this process favors polymer formation, especially since the elimination of Me,SiBr from Me,SiN=P(Br)Me, results in a quantitative yield of the cyclic tetramer.

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc. 15.1.1. Definitions and Scope. 15.1.1.3. Scope. 15.1.1.3.4. Long Chains-Examples

15

of Synthesis.

The initial product of oligomerization of SiF, is believed to be a diradical. (ii) Cleavage of Cyclic Oligomers. Typical ways in which to convert cyclic oligomers into short linear chains are: Me

Me

I I Me-Si-0-Si-Me I I 0 0 I I Me-Si0-SiI I Me

-

MeMe

I I/ Me-Si-0-Si-OSiMe,

+ NaCOSiMe,] Me

I

0

I

Me-Si-0-Si-ONa / \ 1 Me Me Me

Me

C1\

Me

/C1

P N4 N ' I/ ,cl + PCI, cl\ I P P C1/ \N / \ C1

-

Cl,P=N-P

c1

c1 I =N-P =NPCl, I c1 c1 I I

The linear products can be formed in high yield if a 1: 1 ratio of cyclic compound and cleavage reagent is maintained. Hydrogen chloride, AlCl,, and a variety of other halides are useful ring cleavage agents for silicon-containing rings. Water or base can be used to cleave metaborate or metaphosphate rings. A variety of colored short-chain anions can be prepared from S, rings in basic or reducing media, and many metal sulfides contain oligosulfur anions. (H.R ALLCOCK)

1. M. Becke-Goehnng, E. Fluck, Angew. Chem., Int. Ed. Engl., 1, 281 (1962). 2. W. P. Newman in Homoatomic Rings, Chains, and Macromolecules ofiMain group Elements, A. L. Rheingold, ed., Elsevier, Amsterdam, 1977, Ch. 11.

15.1.1.3.4. Long Chains-Examples

of Synthesis.

Two main techniques are employed for the synthesis of long, polymeric inorganic chains-condensation reactions and ring-opening polymerization. With few exceptions, condensation processes do not yield high polymers. The ease of skeletal scrambling reactions and depolymerization favors the formation of low polymers with very broad molecular weight distributions. Hence, the ring-opening polymerization technique is preferred for high-polymer synthesis. (i) Condensation Synthesis of Polymers. Poly(dimethy1phosphazene) with a molecular weight of 50,000 (650 repeating units) can be prepared by the condensation polymerization of a silylphosphinimine1s2: OCH,CF, I

Me,%-N=PMe,

19O"C,4 h - Me3SiOCH2CF3

' (NPMe,),

The elimination of Me,SiOCH,CF, is a new technique, but it is not yet known why this process favors polymer formation, especially since the elimination of Me,SiBr from Me,SiN=P(Br)Me, results in a quantitative yield of the cyclic tetramer.

16

15.1. Introduction 15.1.1. Definitions and Scope. 15.1.1.3. Scope.

(ii) Ring-Opening Polymerization. The conversion of small rings to very long chains is a class of reactions that probably would not have been predicted on theoretical grounds because the small rings should be favored for entropic reasons. Nevertheless, such polymerizations do occur and several cases are known in which rings with three or four repeating units are converted to chains that contain 15,000 or 20,000 repeating units linked end to end. Six examples are shown in Eqs. (b)-(g):

Me

Me

I I Me-Si-0-Si-Me I I 0 0 I I Me-Si0-SiI I Me

heat

(.-Ti? Me n

Me

Me

Me\ /

trace of base or acid

/Me Si

\

HN NH Me\ I I Me si si/ / \” \ Me Me H Me

Me

\

\

/

Me

.P\Si/Me

I \Me Me0?’ Me-% Si-Me \ Me/ \,si, Me Me’ ‘Me

s-s-s I I

S

s-s-s

I I

s

160°C

fS+,

The polymerization of (NPCI,), in the molten state at 250°C yields a very long chain polymer [Eq. (b)] with an average molecular weight near 2 million, which corresponds to over 15,000 repeating units. Here, it is sufficient to note that poly(dich1orophosphaene)

15.1. Introduction 15.1.2. Differences Between Small and Large Molecules. 15.1.2.1. Special Characteristics of Small Cyclic and Linear Molecules.

17

is important as an intermediate for preparing poly(organophosphazenes) by substitutive techniques. The related halides (NPF,), and (NPBr,), also polymerize when heated. The ring-opening polymerization of octamethylcyclotetrasiloxane [Eq. (c)] is a wellknown process used industrially for the manufacture of silicone rubber. The polymerization is catalyzed by traces of acids or base and the products have molecular weights as high as 2 million. Both siloxane and chlorophosphazene polymerizations can be reversed at very high temperatures to reform cyclic oligomers. Cyclosilazanes do not yield high molecular weight polymers [Eq. (d)], presumably because side reactions interfere with chain propagation. However, the side reactions, particularly pyrolytic ring and chain coupling, provide access to silicon nitride ceramics, especially if the cyclosilazane bears a hydrogen atom attached to skeletal silicon3. As shown in Eq. (e), dodecamethylcyclohexasilane polymerizes when heated. However, polymerization is accompanied by a rearrangement to a poly(carbosilane), fSi(H)(Me)CH,+,, which is an intermediate in converting polysilanes to silicon nitride4. The solid-state polymerization of cyclodithiazyl [Eq. (f)] to polythiazyl is an extraordinary reaction. As discussed in 15.2.12, the polymerization process appears to be controlled by the crystal structure of the dimer. The polymer is a gold colored covalent metal. The polymerization of S, (rhombic sulfur) in the molten state is well-known. Viscosity increases indicate that the average chain length increases as the temperature is raised to N 18OoC, above which depolymerization to oligomers takes place. This polymerization and the polymerization of (SN), are free-radical processes. This is unusual in inorganic polymerization chemistry, where most of the reactions proceed by ionic mechanisms. It may not be possible to detect the end group in a very long chain polymer. Hence, it may be impossible to distinguish between open-chain high polymers and macrocyclic rings. Both show essentially the same physical properties. (H.R. ALLCOCK)

P. Wisian-Neilson, R. H. Neilson, J. Am. Chem. SOC.,102, 2848 (1980). R. H. Nielson, P. Wisian-Nielson, Chem. Rev., 88, 541 (1988). D. Seyferth, G. H. Wiseman, J. M. Schwark, Y.-F. Yu, C. A. Foutasse, in Inorganic and Organometallic Polymers, ACS Symposium Series 360, American Chemical Society, Washington, D.C. 1988, p. 142 R. West, J. Maxka, in Inorganic and Organometallic Polymers, ACS Symposium Series 360, American Chemical Society, Washington, D.C., 1988, p. 6.

15.1.2. Characteristic Differences Between Small and Large Molecules as Related to Their Synthesis. 15.1.2.1. Special Characteristics of Small Cyclic and Linear Molecules.

Small molecules have a high mobility in solution or molten states. Moreover, because they contain relatively few atoms and bonds, the number of accessible conformational states is limited, and few opportunities exist for one part of the molecule to shield another. These two characteristics ensure that small cyclic and linear molecules have a

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc. 15.1. Introduction 15.1.2. Differences Between Small and Large Molecules. 15.1.2.1. Special Characteristics of Small Cyclic and Linear Molecules.

17

is important as an intermediate for preparing poly(organophosphazenes) by substitutive techniques. The related halides (NPF,), and (NPBr,), also polymerize when heated. The ring-opening polymerization of octamethylcyclotetrasiloxane [Eq. (c)] is a wellknown process used industrially for the manufacture of silicone rubber. The polymerization is catalyzed by traces of acids or base and the products have molecular weights as high as 2 million. Both siloxane and chlorophosphazene polymerizations can be reversed at very high temperatures to reform cyclic oligomers. Cyclosilazanes do not yield high molecular weight polymers [Eq. (d)], presumably because side reactions interfere with chain propagation. However, the side reactions, particularly pyrolytic ring and chain coupling, provide access to silicon nitride ceramics, especially if the cyclosilazane bears a hydrogen atom attached to skeletal silicon3. As shown in Eq. (e), dodecamethylcyclohexasilane polymerizes when heated. However, polymerization is accompanied by a rearrangement to a poly(carbosilane), fSi(H)(Me)CH,+,, which is an intermediate in converting polysilanes to silicon nitride4. The solid-state polymerization of cyclodithiazyl [Eq. (f)] to polythiazyl is an extraordinary reaction. As discussed in 15.2.12, the polymerization process appears to be controlled by the crystal structure of the dimer. The polymer is a gold colored covalent metal. The polymerization of S, (rhombic sulfur) in the molten state is well-known. Viscosity increases indicate that the average chain length increases as the temperature is raised to N 18OoC, above which depolymerization to oligomers takes place. This polymerization and the polymerization of (SN), are free-radical processes. This is unusual in inorganic polymerization chemistry, where most of the reactions proceed by ionic mechanisms. It may not be possible to detect the end group in a very long chain polymer. Hence, it may be impossible to distinguish between open-chain high polymers and macrocyclic rings. Both show essentially the same physical properties. (H.R. ALLCOCK)

P. Wisian-Neilson, R. H. Neilson, J. Am. Chem. SOC.,102, 2848 (1980). R. H. Nielson, P. Wisian-Nielson, Chem. Rev., 88, 541 (1988). D. Seyferth, G. H. Wiseman, J. M. Schwark, Y.-F. Yu, C. A. Foutasse, in Inorganic and Organometallic Polymers, ACS Symposium Series 360, American Chemical Society, Washington, D.C. 1988, p. 142 R. West, J. Maxka, in Inorganic and Organometallic Polymers, ACS Symposium Series 360, American Chemical Society, Washington, D.C., 1988, p. 6.

15.1.2. Characteristic Differences Between Small and Large Molecules as Related to Their Synthesis. 15.1.2.1. Special Characteristics of Small Cyclic and Linear Molecules.

Small molecules have a high mobility in solution or molten states. Moreover, because they contain relatively few atoms and bonds, the number of accessible conformational states is limited, and few opportunities exist for one part of the molecule to shield another. These two characteristics ensure that small cyclic and linear molecules have a

18

15.1. Introduction 15.1.2 Differences Between Small and Large Molecules. 15.1.2.2.Special Characteristics of Long-Chain Molecules

high reactivity in substitution processes. The only serious exception to this rule is in the case of cage compounds, where substitutive attack from the back side of a skeletal atom may be blocked by other components of the cage. Small molecules possess a number of other characteristics that favor their separation and purification during synthetic procedures. They are generally volatile and can be separated by distillation, sublimation or vapor-phase chromatography. They pass through semipermeable membranes and are amenable to dialysis. Their small, generally rigid, molecular characteristics enable them to form single crystals and this, in turn, permits structural characterization by x-ray diffraction techniques. Small molecules are often more stable than large molecules at elevated temperatures for entropic reasons. (H R ALLCOCK)

15.1.2.2. Special Characteristics of Long-Chain Molecules. Long-chain polymers undergo molecular entanglement in solution, in the melt, and in the solid state. Hence, they yield viscous solution or molten liquids and are much tougher or elastomeric in the solid state than are small molecule systems. Because long chains can coil extensively, complications may exist when polymers are used as substrates for substitution reactions. Reagents have difficulty penetrating to the core of a random coil. Moreover, a polymer in solution generates its own environment and alters the polarity and solvation characteristics in the region of the reactive center. High polymers are nonvolatile. Hence, the usual preparative or separation techniques that require vaporization cannot be employed, except to remove low molecular weight impurities. Polymers cannot pass through semipermeable membranes; hence, they can be freed from small-molecule impurities by dialysis. Although vaporphase chromatography techniques cannot be employed for high polymers, gel-permeation chromatography techniques can be used. Gel-permeation chromatography is a method of separation in which large molecules in solution are eluted through a substrate that contains pores with diameters of, say, 100-2000 A. The larger the molecule, the less easily it penetrates the pores. Hence, large molecules are eluted faster than small molecules. This is a conventional method for the indirect estimation of molecular size and molecular weight. High polymers are generally so coiled and entangled in the solid state that insufficient order can be achieved to allow single crystals to form [an exception to this is provided by (SN),]. Hence, conventional x-ray crystallography cannot be employed as a structural tool. However, small microcrystalline domains may exist, and intramolecular order may be present in noncrystalline regions. Thus, helical transform xray analysis may permit structural information to be obtained. From a synthetic and chemical point of view, macromolecules are often more delicate than their oligorneric, small-molecule analogs. Their propensity to depolymerize or fragment at elevated temperatures has been mentioned earlier. In practice, the thermal stability of a macromolecule rarely comes close to that predicted on the basis of data derived from oligomers or bond-strength estimates. In substitution reactions, macromolecules must be handled with extreme care to avoid either crosslinking or chain cleavage reactions. Intramolecular coupling reactions present serious problems in polymer systems. An average of only 1.5 crosslinks per chain causes total insolubilization of a macromolecular system. In a macromolecule that contains, say, 15,000 repeating units,

15.1. Introduction 15.1.3. Chai n-C hai n , C hain-Ri ng and Ri ng-R i ng Equ i I i bria.

19

the chances of crosslinking occurring during a chemical reaction are uncomfortably high. Because the unique properties of polymers depend on the long chain length, any reactions that lead to chain cleavage (for example, as few as 10 cleavage reactions per 10,000 repeat unit chains) has unfortunate consequences with respect to these properties. Some of these problems do not apply to ladder polymers or rigid rod polymers. Generally, however, these species are of low solubility. Three reasons underlie the widespread interest in macromolecules, in spite of the synthetic difficulties that exist. First, the entanglement of long-chain macromolecules provides physical properties (strength, toughness, elasticity, fiber-forming properties, etc.) that cannot be obtained with small-molecule systems. Second, because polymers have a low volatility they can be used as engineering materials. Third, the odedimensional character of linear polymers is of considerable interest from the viewpoints of anisotropic physical properties, electrical phenomena, and information storage at the molecular level. (H.R. ALLCOCK)

15.1.3. Chain-Chain, Chain-Ring and Ring-Ring Equilibria. Because of the emphasis on carbon and its peculiarities (organic chemistry), where equilibration in standard laboratory reactions is rare, many preparative inorganic chemists view and interpret their data in terms of kinetically controlled reactions. However, in inorganic chemistry, full or partial equilibration dominates the reactionsincluding a considerable number that are usually discussed as if they were kinetically controlled. The reactions whereby chain and ring molecular structures interconvert by exchanging parts give such clear indications of equilibrium rather than kinetic control that this area of main-group chemistry is now generally considered from this viewpoint, even though some examples are kinetically controlled and are wiped out if the system is allowed to equilibrate fully. The ideas discussed in this section apply to preparative procedures for producing chain or cyclic molecules (including cage structures) in noncrystalline form, e.g., the condensation reixtions and that large group of reactions in which a small molecule is inserted into a macromolecule or a cyclic structure or either of the latter exchange distinguishable parts. The exchange of parts between oligomeric or polymeric chains has been analyzed’ for organic polymers in which the monomeric units are held together by labile linkages such as ester or amide groups. These systems behave as if they were inorganic, with the redistribution reactions involving scission of bonds to bridging atoms such as oxygen, nitrogen or sulfur rather than breaking of the carbon backbones of the monomeric building blocks. With respect to the resulting distribution function, a kinetic treatment gives the same results’ as a treatment assuming thermodynamic equilibrium because random probabilities are implicitly or explicitly assumed. However, a molecular sizedistribution function based on random sorting of the bonding sites of the constituent monomers does not agree with the size distributions of some well-studied inorganic systems, although it is appropriate for organic oligomers and macromolecules in which the monomers consist of a chain of two or more atoms separating the molecular sites where exchange takes place. This problem has been handled in a general manner by stochastic graph which accounts for the influence of surrounding atoms at any

20

15.1. Introduction 15.1.3. Chain-Chain, Chain-Ring and Ring-Ring Equilibria. 15.1.3.1. Bond Lability in Inorganic Systems.

distance within the molecule (not only the nearest neighbors) o n the thermodynamic properties (or reactivities) of the bonds involved in the exchange process. Although the theoretical basis of chain-chain equilibria is now sound and suitable for general use, the theory of ring-chain equilibria and interconversions between different sized rings requires detailed information concerning molecular structure-information of the kind that is embodied in experimental data-so that the theory is at this time necessarily semiempirical unless recourse is taken to including in the theoretical treatment complex and expensive a b initio quantum and statistical mechanics. Attempts have been to reduce the calculations to a simple form, but more needs to be done. (J.R. VAN WAZER)

1. P. J. Flory, J . Am. Chem. SOC.,64,27205 (1942); W. H. Stockmayer,J . Chem. Phys., 11,45 (1943); 12, 125 (1944); J . Polymer Sci., 9, 69 (1953). The Flory reference pertains to mono- and 2. 3. 4. 5. 6. 7.

difunctional units, whereas the papers by Stockmayer include higher functionalities. P. J. Flory, Principles of Polymer Chemistry, Cornell University Press, Ithaca, NY, 1953. D. W. Matula, L. C. D. Groenweghe, J. R. Van Wazer, J . Chem. Phys., 41, 3105 (1964). D. W. Matula, J. R. Van Wazer, J . Chem. Phys., 46, 3123 (1967). R. M. Levy, J. R. Van Wazer, J . Chem. Phys., 45, 1824 (1966). H. Jacobson, W. H. Stockmayer, J. Chem. Phys., 18, 1600 (1950). J. B. Carmichael, J. B. Kinsinger, Can. J . Chem., 42, 1996 (1964).

15.1.3.1. Bond Lability in Inorganic Systems.

Interest has centered on organic chemistry for nearly a hundred years, starting in the mid-19th century, because thermodynamically unstable organic compounds such as the hydrocarbons can remain virtually unchanged for centuries when protected from outside reactants at ambient temperatures. In classical organic chemistry, the emphasis on C-C bonds and the various homologous series based on them allowed the preparative chemist to focus attention on the functional groups without worrying about changes in the hydrocarbon tails. Now that chemists are learning to cope with quite labile structures, there has been a renaissance in inorganic chemistry and an explosion of knowledge in biochemistry. Biochemicals are labile owing to the atoms other than carbon that appear in the molecular backbones and that establish the molecular conformations which play important biochemical roles. The polyfunctional atoms that occupy central positions in inorganic molecules are classified into two broad categories: (a) those with low-lying vacant orbitals (e.g., the transition metals and the group-IIIB elements) and (b) the usual main-group elements. Even for the elements of groups IVB-VIIB and the noble gases, the heavier atoms readily exhibit coordination-number increase due to occupation of low-lying vacant orbitals so that their chemistries are quite similar to those of the transition-metal elements. Mainly because of the high coordination numbers (e.g., 6) associated with the transition metals and the other atoms having low-lying vacant orbitals, the gel point (see $15.1.3.3)occurs at stoichiometries for which the average chain length is small. Ring formation is also accentuated because of geometrical constraints from the high coordination numbers. Therefore these chemistries are dominated by relatively small polycyclics and little discussed amorphous macromolecular networks. For this reason, the ideas presented in this section are most readily applicable to the chemistries of the elements of the first four rows of groups IVB-VIB.

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc. 20

15.1. Introduction 15.1.3. Chain-Chain, Chain-Ring and Ring-Ring Equilibria. 15.1.3.1. Bond Lability in Inorganic Systems.

distance within the molecule (not only the nearest neighbors) o n the thermodynamic properties (or reactivities) of the bonds involved in the exchange process. Although the theoretical basis of chain-chain equilibria is now sound and suitable for general use, the theory of ring-chain equilibria and interconversions between different sized rings requires detailed information concerning molecular structure-information of the kind that is embodied in experimental data-so that the theory is at this time necessarily semiempirical unless recourse is taken to including in the theoretical treatment complex and expensive a b initio quantum and statistical mechanics. Attempts have been to reduce the calculations to a simple form, but more needs to be done. (J.R. VAN WAZER)

1. P. J. Flory, J . Am. Chem. SOC.,64,27205 (1942); W. H. Stockmayer,J . Chem. Phys., 11,45 (1943); 12, 125 (1944); J . Polymer Sci., 9, 69 (1953). The Flory reference pertains to mono- and 2. 3. 4. 5. 6. 7.

difunctional units, whereas the papers by Stockmayer include higher functionalities. P. J. Flory, Principles of Polymer Chemistry, Cornell University Press, Ithaca, NY, 1953. D. W. Matula, L. C. D. Groenweghe, J. R. Van Wazer, J . Chem. Phys., 41, 3105 (1964). D. W. Matula, J. R. Van Wazer, J . Chem. Phys., 46, 3123 (1967). R. M. Levy, J. R. Van Wazer, J . Chem. Phys., 45, 1824 (1966). H. Jacobson, W. H. Stockmayer, J. Chem. Phys., 18, 1600 (1950). J. B. Carmichael, J. B. Kinsinger, Can. J . Chem., 42, 1996 (1964).

15.1.3.1. Bond Lability in Inorganic Systems.

Interest has centered on organic chemistry for nearly a hundred years, starting in the mid-19th century, because thermodynamically unstable organic compounds such as the hydrocarbons can remain virtually unchanged for centuries when protected from outside reactants at ambient temperatures. In classical organic chemistry, the emphasis on C-C bonds and the various homologous series based on them allowed the preparative chemist to focus attention on the functional groups without worrying about changes in the hydrocarbon tails. Now that chemists are learning to cope with quite labile structures, there has been a renaissance in inorganic chemistry and an explosion of knowledge in biochemistry. Biochemicals are labile owing to the atoms other than carbon that appear in the molecular backbones and that establish the molecular conformations which play important biochemical roles. The polyfunctional atoms that occupy central positions in inorganic molecules are classified into two broad categories: (a) those with low-lying vacant orbitals (e.g., the transition metals and the group-IIIB elements) and (b) the usual main-group elements. Even for the elements of groups IVB-VIIB and the noble gases, the heavier atoms readily exhibit coordination-number increase due to occupation of low-lying vacant orbitals so that their chemistries are quite similar to those of the transition-metal elements. Mainly because of the high coordination numbers (e.g., 6) associated with the transition metals and the other atoms having low-lying vacant orbitals, the gel point (see $15.1.3.3)occurs at stoichiometries for which the average chain length is small. Ring formation is also accentuated because of geometrical constraints from the high coordination numbers. Therefore these chemistries are dominated by relatively small polycyclics and little discussed amorphous macromolecular networks. For this reason, the ideas presented in this section are most readily applicable to the chemistries of the elements of the first four rows of groups IVB-VIB.

15.1. Introduction 15.1.3. Chain-Chain, Chain-Ring and Ring-Ring Equilibria. 15.1.3.2. Some important Relationships Based on Theory.

21

When viewed broadly throughout the periodic table, the kinetic landscape (consisting of energy as the vertical coordinate and geometry in the horizontal) exhibits very high mountain ranges, with access from one valley to another by mountain passes for the lighter atoms having occupied orbitals. When progressing down a given group, e.g., group IVB or VB, the kinetic landscape becomes less mountainous and shades off into rolling hills so that the need for traversing mountain passes to go from one energy minimum to another becomes less. In other words, the details of the reaction mechanism in organic chemistry are paramount, while alternate kinetic pathways will be equivalent for tin. Indeed one can envisage situations in the synthetic chemistry of tin- or lead-based oligomeric molecules where the very concept of a reaction mechanism may be inapplicable. Organic substituents are usually the least labile of the monofunctional groups for undergoing redistribution and are therefore used as blocking groups. Methyl and phenyl groups are valuable in this respect, since they are thermally more stable than, for example, the ethyl group which in some systems produces ethylene at temperatures of several hundred degrees, leaving a hydrogen at its bonding site. Pyrolysis in organic chemistry consist of redistribution reactions, but similar processes occur at much lower temperatures in inorganic chemistry. Thus, as a rough rule of the thumb, reactions proceeding at red heat for carbon compounds occur at 200°C for the Si and at near RT for the Ge analogs, with the Sn compounds being in rapid dynamic equilibrium in a cooled solvent. The numerous exceptions to this rule usually occur when going from the row ending with neon to that ending with argon. The rate of equilibration of a distribution of a,w-dichloro oligomers based on oxygen-bridged, dimethyl-substituted group-IVB central atoms, starting from a mixture of the cyclic trimer and the chain compound exhibiting two central atoms, is an example. When the central atom is carbon', the system reaches equilibrium in ca. 2 d at 120"C, as compared to within 1 d at 200°C for Si2.However, in the case of Ge continuous rapid equilibration is found, corresponding to a lifetime of about 1 s at 35°C.

-

(J.R. VAN WAZER)

1. K. Moedritzer, J. R. Van Wazer, J . Org. Chem., 30, 3920 (1965). 2. K. Moedritzer, J. R. Van Wazer, J . Am. Chem. SOC.,86, 802 (1964). 3. K. Moedritzer, J. R. Van Wazer, Inorg. Chem., 4, 1753 (1965).

15.1-3.2.Some Important Relationships Based on Theory.

Consider a general system in which central atoms or moieties, each symbolized by Q, of functionality v are bonded to monofunctional substituents, T, and bridging atoms or groups (or even Q-Q bonds), symbolized by Z, with the Q-T and Q-Z connections being sufficientlylabile that an equilibrium distribution of the resulting molecules will be obtained by a scrambling process involving site exchanges of the T's and Z s with themselves and with each other. If there are v T's for each Q, only the simplest kind of molecule possible in this system, QT,, would result, whereas v/2 Z ' s per Q converts the system into the most condensed polymer possible or whatever cyclized molecules can be based on only Z-bridged Q units. We can thus refer to the overall system as the QT,-QZ,,, system (e.g., the SiCl,-SiO, system of the perchlorosiloxanes, for which Q = Si, T = C1, Z = 0 and v = 4). Furthermore, the stoichiometry of any chosen composition can be expressed in terms of the parameter R, defined as [v - ([T]/[Q])] or

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc. 15.1. Introduction 15.1.3. Chain-Chain, Chain-Ring and Ring-Ring Equilibria. 15.1.3.2. Some important Relationships Based on Theory.

21

When viewed broadly throughout the periodic table, the kinetic landscape (consisting of energy as the vertical coordinate and geometry in the horizontal) exhibits very high mountain ranges, with access from one valley to another by mountain passes for the lighter atoms having occupied orbitals. When progressing down a given group, e.g., group IVB or VB, the kinetic landscape becomes less mountainous and shades off into rolling hills so that the need for traversing mountain passes to go from one energy minimum to another becomes less. In other words, the details of the reaction mechanism in organic chemistry are paramount, while alternate kinetic pathways will be equivalent for tin. Indeed one can envisage situations in the synthetic chemistry of tin- or lead-based oligomeric molecules where the very concept of a reaction mechanism may be inapplicable. Organic substituents are usually the least labile of the monofunctional groups for undergoing redistribution and are therefore used as blocking groups. Methyl and phenyl groups are valuable in this respect, since they are thermally more stable than, for example, the ethyl group which in some systems produces ethylene at temperatures of several hundred degrees, leaving a hydrogen at its bonding site. Pyrolysis in organic chemistry consist of redistribution reactions, but similar processes occur at much lower temperatures in inorganic chemistry. Thus, as a rough rule of the thumb, reactions proceeding at red heat for carbon compounds occur at 200°C for the Si and at near RT for the Ge analogs, with the Sn compounds being in rapid dynamic equilibrium in a cooled solvent. The numerous exceptions to this rule usually occur when going from the row ending with neon to that ending with argon. The rate of equilibration of a distribution of a,w-dichloro oligomers based on oxygen-bridged, dimethyl-substituted group-IVB central atoms, starting from a mixture of the cyclic trimer and the chain compound exhibiting two central atoms, is an example. When the central atom is carbon', the system reaches equilibrium in ca. 2 d at 120"C, as compared to within 1 d at 200°C for Si2.However, in the case of Ge continuous rapid equilibration is found, corresponding to a lifetime of about 1 s at 35°C.

-

(J.R. VAN WAZER)

1. K. Moedritzer, J. R. Van Wazer, J . Org. Chem., 30, 3920 (1965). 2. K. Moedritzer, J. R. Van Wazer, J . Am. Chem. SOC.,86, 802 (1964). 3. K. Moedritzer, J. R. Van Wazer, Inorg. Chem., 4, 1753 (1965).

15.1-3.2.Some Important Relationships Based on Theory.

Consider a general system in which central atoms or moieties, each symbolized by Q, of functionality v are bonded to monofunctional substituents, T, and bridging atoms or groups (or even Q-Q bonds), symbolized by Z, with the Q-T and Q-Z connections being sufficientlylabile that an equilibrium distribution of the resulting molecules will be obtained by a scrambling process involving site exchanges of the T's and Z s with themselves and with each other. If there are v T's for each Q, only the simplest kind of molecule possible in this system, QT,, would result, whereas v/2 Z ' s per Q converts the system into the most condensed polymer possible or whatever cyclized molecules can be based on only Z-bridged Q units. We can thus refer to the overall system as the QT,-QZ,,, system (e.g., the SiCl,-SiO, system of the perchlorosiloxanes, for which Q = Si, T = C1, Z = 0 and v = 4). Furthermore, the stoichiometry of any chosen composition can be expressed in terms of the parameter R, defined as [v - ([T]/[Q])] or

22

15.1. Introduction 15.1.3. Chain-Chain, Chain-Ring and Ring-Ring Equilibria. 15.1.3.2. Some Important Relationships Based on Theory.

2([ZJ/[QJ), so that R ranges from 0 to v as the molecules making up the system become more condensed. In systems in which finite chain and ring structures coexist, it is advisable to calculate an Rchainsvalue for the chain population alone by subtracting the concentrations [TI, [ZJ and [Q] associated with cyclic molecules from the total concentrations of the T's, Z's and Q s respectively. (J.R. VAN WAZER)

15.1 3.2.1. The Chain Population.

If in a ring-free system the exchange of a T on any Q for a Z on any other is associated with zero enthalpy (as would be expected for truly additive bond energies), the overall scrambling process becomes entropy controlled and random sorting of the molecular parts is to be expected. The mathematics of such random redistribution processes' in the case considered here is equivalent to assembling completed noncyclic molecules by randomly selecting the connection sites on the T (one site), Z (two sites) and Q (v sites) units from a random mixture having the desired proportions of Z, T and Q. However, when the change in enthalpy for the scrambling is not zero (as is commonly the case in inorganic systems), stochastic graph theory' shows that it is necessary to carry out the random scrambling with molecular segments (preset assemblages of Q, T, and Z units) that may or may not be overlapping. In other words, neighboring atoms within the molecules contribute to the observed enthalpy of exchanging a T for a Z, and all of the atoms that contribute appreciably to the enthalpy must be included in the segments to be sorted before the sorting process can be treated by random statistics. The size of a segment centered on a particular atom or bond in the molecule is called its order of environment and is measured in terms of the number Q of units (usually atoms) to be included when going out along every atomic chain leading from that site. The number of chain units (value of Q)in each chain path within the smallest segment that can be randomly sorted with other segments of the same 0 is called the reorganizational heat order, p. The size-distribution function of the chain portion of many real systems can be interpreted quantitatively by assuming that p = 1, so that the enthalpy contributions are wholly ascribed to the particular mixture of T's and Z s on each of the pair of Q's between which the T-for-Z interchange takes place. This means that for p = 1 the molecular segments of the form QZiT,-i can be randomly sorted into noncyclic molecules (although Z's and T's cannot), but that the processes whereby they are intertransformed must be handled by a method that is amenable to thermodynamic treatment. The QZiT, -, segments are called building units and may be made stoichiometrically additive by splitting the Z units shared between them. We may now formulate the additive segments as Q(Z1,JiT,_, by assigning the symbol ZljZ to half of a bridging atom, bridging group or Q-Q bond. The segment for which i = 0 represents the smallest whole molecule in the system and is symbolized by n for the neso (or island) structure, whereas the one for which i = 1 is monofunctional, serves as an end group in oligomers and macromolecules and is denoted by e. The middle group, corresponding to i = 2 in Q(ZliZ),Tv - i, may be symbolized by m; and the branching groups of functionality i = 3,4, etc., may be represented by b3, b,, etc., respectively. All of these building units of functionality equal to or greater than unity (i.e., e, m, b, and b4) are the pieces from which the various molecules larger than n E QT, are assembled randomly.

15.1.3. Chain-Chain, Chain-Ring and Ring-Ring Equilibria. 15.1.3.2. Some Important Relationships Based on Theory. 15.1.3.2.1. The Chain Population.

23

The usual method for relating chemical reactions to thermodynamics is to set up chemical equations involving reactions between molecules-reactions that can be treated in terms of equilibrium constants. Obviously, if consideration were to be given to all of the possible reactions between the molecules present in a oligomeric or macromolecular mixture of the kind discussed here, there would be far too many equilibrium constants to be handled. However, according to the theory2, the only equilibrium constants needed are those dealing with the T-for-Z exchanges between those molecular segments for which 0 = p. For the common case of p = 1, the required (v - 2) equilibria are: i-QZ,

+ (v - i)QT,

vQZ,TV-,

(a)

where zi stands for the mole for which the equilibrium constant is (K')i = (zi)v/(zv)i(zo)v~i, fraction of the total Q in the Q(Z1,Z)iTv-isegments. Alternatively, the (v - 2) equations may represent the interaction of a pair of like segments undergoing a T-for-Z exchange: 2QZiTv-, e Q z i + l T v - i - l

+ QZi-iTv-,+l

(b)

for which Ki = (zi+l)(zi - J(zi)'. The enthalpy of each of these (v - 2) reactions involving molecular segments may be estimated from the equilibrium constants as follows: AGO = -RT In K = AHo - TAS" AGCand= -RTln K = AHR,,

-

TASg,,

(c) (4

If

AS" = AS:and

(e)

AHo % 6AG" = -RT ln(K/Krand)

(f)

then The random values, Krand,of the equilibrium constants can be calculated from the appropriate binomial coefficients. For example, for the ring-free part of a system for which v = 4, the five building units Q(Z1,Z)iTv-ifor i = 0, 1, 2, 3 and 4 exhibit (at a [Z]/[T] ratio of unity) concentrations proportional to the appropriate binomial coefficients, 1:4:6:4:1, so that for the compound for which i = 3 the value of Krandfor the reaction of Eq. (a) equals (4)4/(1)3(1)1= 256 and for the reaction of Eq. (b) it equals (1) (6)/(4)' = 0.3750. Throughout the remainder of $15.1.3, AHo of Eq. (f) is denoted as AHicramb. Once the appropriate set of equilibrium constants is known for the molecular segments for which 0 = p, the mole fractions (or concentrations) of these segments can be calculated for any particular composition (i.e., a chosen value of Rchains) from (i) the K s , (ii) the Rchainsitself and (iii) the normalization of the segments in chains zF (owing to the use of mole fractions of chains), according to which cy=o(zp) = 1, while remembering that Rchains= 2[Z]/[Q] for the chain fraction = iz;. As indicated, the building units, zp for i > 1, associated with cyclic structures are omitted from the calculation of Rchalns.In this way, (v + 1) simultaneous equations are obtained to be solved for an equal number of unknowns, zp with i = 0, 1, 2, . . . , v. Once the abundance of the molecular segments corresponding to 0 = p has been obtained, the molecular size distribution of the ring-free fraction of the molecules results from random sorting of these segments.

cy,o

24

15.1. Introduction 15.1.3. Chain-Chain, Chain-Ring and Ring-Ring Equilibria. 15.1.3.2. Some Important Relationships Based on Theory.

There are a number of mathematical relationship^^,^ of value for gaining understanding of the molecular distributions in systems for which p = 1. For example, it can be shown that the concentration maximum of a particular building unit in a ring-free distribution (or one in which the rings have been corrected for) occurs at the overall composition corresponding to its own composition. Likewise for such a system, the value of the (v - 1) equilibrium constants corresponding to Eq. (b) can be determined from the concentration of each building unit measured at the composition where it maximizes. In the case of equilibria between unbranched-chain molecules in a system for which p > 1, the value of p and the pth-order equilibrium constants may be determined from the concentrations of the two smallest observable molecules3. (J.R. VAN WAZER)

1. J. R. Van Wazer, Ann. N.Y. Acad. Sci., 159, 5 (1969). 2. D. W. Matula, L. C. D. Groenweghe, J. R. Van Wazer, J . Chem. Phys., 41, 3105 (1964) 3. D. W. Matula, J. R. Van Wazer, J . Chem. Phys., 46, 3123 (1967). 15.1.3.2.2. The Ring Population and Network Macromolecules.

A common situation involving ring-chain equilibria consists of one size of simple ring (composed of only Q(Zl,2)2T,- segments) appearing in conjunction with a distribution of chain molecules. Even in the case of v > 2, where chain branching can occur, the concentration of the simple-ring compound may be related to the chain-distribution function through a reaction whereby the ring is produced from a straight chain: e(m)je + (m)r -e(m)j+re

(a)

where e represents a monofunctional segment, QZ1,,T,- and m a bifunctional segment, Q(Zl,2)2Tv-2,with r being the number of such segments in the ring. Unlike any reaction involving exchange of parts between chain molecules, reactions relating ring molecules to chains are concentration dependent according to thermodynamic principles, so that the equilibrium constant must include the volume of the system. A general expression' for the constant of Eq. (a) for p = 1 is:

K," =

[middles in the r-sized ring] rV

[ends]

+ 2 [chain middles]

2 [chain middles]

1

where r is the number of middle groups in one size of ring and V is the average molar volume of the building units in the particular composition under study. When there are simple rings of several sizes, each size of ring may be related to the chain distribution by an equilibrium constant of the form of Eq. (b). Likewise, the rings may be related to each other by a ring-ring equilibrium constant, K,"aorb, of the usual form employed for molecules, a constant that is also volume dependent. Since in most of the cases that are amenable to experimental investigation there is not a large number of differently sized cyclic molecules, it is the usual practice to determine the proportion of each size of ring and to relate the rings to each other by K:& constants and the predominant ring to the chain population by a K: constant. Alternatively, the value of this constant as defined in Eq. (b) may be estimated* from approximate partition functions. Unfortunately for most inorganic systems there are insufficient thermodyn-

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc. 24

15.1. Introduction 15.1.3. Chain-Chain, Chain-Ring and Ring-Ring Equilibria. 15.1.3.2. Some Important Relationships Based on Theory.

There are a number of mathematical relationship^^,^ of value for gaining understanding of the molecular distributions in systems for which p = 1. For example, it can be shown that the concentration maximum of a particular building unit in a ring-free distribution (or one in which the rings have been corrected for) occurs at the overall composition corresponding to its own composition. Likewise for such a system, the value of the (v - 1) equilibrium constants corresponding to Eq. (b) can be determined from the concentration of each building unit measured at the composition where it maximizes. In the case of equilibria between unbranched-chain molecules in a system for which p > 1, the value of p and the pth-order equilibrium constants may be determined from the concentrations of the two smallest observable molecules3. (J.R. VAN WAZER)

1. J. R. Van Wazer, Ann. N.Y. Acad. Sci., 159, 5 (1969). 2. D. W. Matula, L. C. D. Groenweghe, J. R. Van Wazer, J . Chem. Phys., 41, 3105 (1964) 3. D. W. Matula, J. R. Van Wazer, J . Chem. Phys., 46, 3123 (1967). 15.1.3.2.2. The Ring Population and Network Macromolecules.

A common situation involving ring-chain equilibria consists of one size of simple ring (composed of only Q(Zl,2)2T,- segments) appearing in conjunction with a distribution of chain molecules. Even in the case of v > 2, where chain branching can occur, the concentration of the simple-ring compound may be related to the chain-distribution function through a reaction whereby the ring is produced from a straight chain: e(m)je + (m)r -e(m)j+re

(a)

where e represents a monofunctional segment, QZ1,,T,- and m a bifunctional segment, Q(Zl,2)2Tv-2,with r being the number of such segments in the ring. Unlike any reaction involving exchange of parts between chain molecules, reactions relating ring molecules to chains are concentration dependent according to thermodynamic principles, so that the equilibrium constant must include the volume of the system. A general expression' for the constant of Eq. (a) for p = 1 is:

K," =

[middles in the r-sized ring] rV

[ends]

+ 2 [chain middles]

2 [chain middles]

1

where r is the number of middle groups in one size of ring and V is the average molar volume of the building units in the particular composition under study. When there are simple rings of several sizes, each size of ring may be related to the chain distribution by an equilibrium constant of the form of Eq. (b). Likewise, the rings may be related to each other by a ring-ring equilibrium constant, K,"aorb, of the usual form employed for molecules, a constant that is also volume dependent. Since in most of the cases that are amenable to experimental investigation there is not a large number of differently sized cyclic molecules, it is the usual practice to determine the proportion of each size of ring and to relate the rings to each other by K:& constants and the predominant ring to the chain population by a K: constant. Alternatively, the value of this constant as defined in Eq. (b) may be estimated* from approximate partition functions. Unfortunately for most inorganic systems there are insufficient thermodyn-

15.1.3. Chain-Chain, Chain-Ring and Ring-Ring Equilibria. 15.1.3.2. Some Important Relationships Based on Theory. 15.1.3.2.2. The Ring Population and Network Macromolecules.

25

amic data or extrapolatory routines for estimating meaningful approximate partition functions. Although the mathematical framework is available', useful equations have not been developed to treat equilibria between a chain distribution and polycyclic or cage molecules. Now let us turn to the problem of the formation of network polymers in chaincontaining systems at equilibrium with respect to exchange of parts between molecules. When the central moiety Q has a functionality greater than two (v > 2), it is possible for infinite-network macromolecules to occur in the system owing to excessive chain branching. In other words, when there are insufficient end groups, QZ,,,T,- for v > 2, to cap off highly branched chain structures, these structures will be so large that they must extend from wall to wall of the containing vessel. The point at which there are barely sufficient end-groups to cap off all of the molecules and hence to keep them to a finite size is the gel point. Solution of the general equation for the gel point', for a system for which p = 1 and there is no tendency toward simple-ring formation gives:

1 i(i - l)zi = O Y

i=O

Expansion of this equation for v = 6 (a coordination shell of six filled with exchangeable Z and T substituents), letting the mole fraction of Q(Zl,2)1TV-ibe symbolized by [n] for i = 0, [el for i = 1, [m] for i = 2, [b3] for i = 3, [b4] for i = 4, etc., gives: [el

=

3[b3]

+ 8[b41 + 15[b,] + 24[b,]

Note in Eq. (d) that the [n] cancels since it is not a chain-building unit and that this is also true for the [m] units, which do not contribute to the branching or capping off of molecules. Obviously, for a system with v = 3, only the first term on the right side of Eq. (d) will be produced. This equation explains why systems based on atoms with coordination numbers greater than 4 in which there is scrambling between bridging and monofunctional substituents [e.g., alkaline iron(II1) solutions] exhibit mixtures of small molecules with amorphous gels or precipitates rather than oligomeric structures. In a system in which there is no tendency to form small cyclized molecules (whether simple rings or cages), ring formation still occurs in the network macromolecules existing beyond the gel point, owing to intolerable crowding in the highly branched trees of building units. The minimum number of ring closures C per building unit needed to make the system physically realizable can be calculated'. This same relationship can also be used to estimate the gel point in a system in which there are small rings, as is reported for p = 1 in Eq. (e), which is a more general form of Eq. (d): V i2zi C i(i - 2)zi = 2C ___ i=O izi

[

]

In the mathematically tractable ring-containing systems in which all cyclic molecules appearing below the gel point are composed of difunctional building units, there is one ring closure for each cyclic molecule, so that C = ~~2,,,i,(zi),,where o is the size of each ring in units. (J R. VAN WAZER)

1. D. W. Matula, L. C . D. Groenweghe, J. R. Van Wazer, J . Chem. Phys., 41, 3105 (1964). 2. R. M. Levy, J. R. Van Wazer, J . Chem. Phys., 45, 1824 (1966).

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc. 26

15.1. Introduction 15.1.3. Chain-Chain, Chain-Ring and Ring-Ring Equilibria. 15.1.3.3. Scrambling in Inorganic Systems.

15.1.3.3. Scrambling In Inorganic Systems.

There are three important categories in the practical chemistry of scrambling equilibria: (i) all-neso model systems, (ii) scrambling of bridging groups with each other and with monofunctional substituents and (iii) the scrambling of these between different central atoms. In the experimental study of molecular distributions during scrambling or in scrambled systems, it is necessary to determine' individual molecular species in a complicated mixture of closely related structures. Fractional distillation gives nine fractions when applied to a mixture of the 15 compounds resulting from scrambling methyl, ethyl and propyl groups in tetraalkyllead compounds'. A better method for separating those reorganization mixtures that can be volatilized without decomposition is gas chromatography, which has been applied to a study3 of the exchange of methyl, ethyl, propyl and butyl groups on silicon. Column chromatography has been used to investigate the molecular distribution in equilibrated mixtures of the condensed oxyacids of phosphorus4; and two-dimensional paper or thin-layer chromatography5 has proved unique in separating individual ring from chain species in phosphate mixtures. Perhaps the most useful of the procedures for determining6 the molecular distribution resulting from scrambling reactions is 'H, 31Pand I9Fnuclear magnetic resonance (NMR), which can be applied to samples that have been sealed in NMR tubes in an oxygen- and moisture-free atmosphere before being equilibrated. This method, which is applicable to an extremely wide range of equilibration rates and can be used to assay a large number of systems, is sensitive to the intramolecular environment of the magnetically active atoms so that molecular segments of the type discussed in $15.1.3.2 are precisely the items being assayed. This means that the concepts of graph theory can be applied in interpreting the NMR data'. (J.R. VAN WAZER)

J. R. Van Wazer, Ann. N.Y. Acad. Sci., 159, 5 (1969). G. H. Calingaert, H. A. Beatty, H. Soroos, J . Am. Chem. SOC.,62, 1099 (1940). F. H. Pollard, G. Nickless, P. C. Uden, J . Chrornatogr., 19, 28 (1965). S. Ohashi, Pure Appl. Chem., 44, 415 (1975). E. Karl-Kroupa, Anal. Chem., 28, 1091 (1956); W. J. van Ooij, J. P. W. Houtman, 2. Anal. Chem., 244, 38 (1969). The first reference is to paper and the second to thin-layer chromatography. 6. J. R. Van Wazer, L. C. D. Groenweghe, in Nuclear Magnetic Resonance in Chemistry, B. Pesce, ed., Academic Press, New York, 1965, pp. 283-298. 7. D. W. Matula, L. C. D. Groenweghe, J. R. Van Wazer, J . Chem. Phys., 41, 3105 (1964).

1. 2. 3. 4. 5.

15.1.3.3.1. All-Neso Systems.

Studies of scrambling reactions in systems in which two monofunctional substituents, T and Z , are scrambled on a central atom or moiety of functionality v led to the theory of random scrambling'. The scrambling of substituents that are chemically similar is close to random. Examples are found in the scrambling of methoxyl vs. ethoxyl groups, chlorine vs. bromine and, on the heavier elements where the exchange is sufficiently rapid, one alkyl group for another. However, the scrambling of some pairs of substituents on a given central moiety is far from random. For example, as shown in Figure 1, when tetramethyl silicate is equilibrated with tetraethyl silicate at 15OoC, the system is essentially random, whereas the equilibration of a mixture of tetrakisdimethylaminosilane with silicon tetrachloride is nonrandom, the mixed compounds being produced in

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc. 26

15.1. Introduction 15.1.3. Chain-Chain, Chain-Ring and Ring-Ring Equilibria. 15.1.3.3. Scrambling in Inorganic Systems.

15.1.3.3. Scrambling In Inorganic Systems.

There are three important categories in the practical chemistry of scrambling equilibria: (i) all-neso model systems, (ii) scrambling of bridging groups with each other and with monofunctional substituents and (iii) the scrambling of these between different central atoms. In the experimental study of molecular distributions during scrambling or in scrambled systems, it is necessary to determine' individual molecular species in a complicated mixture of closely related structures. Fractional distillation gives nine fractions when applied to a mixture of the 15 compounds resulting from scrambling methyl, ethyl and propyl groups in tetraalkyllead compounds'. A better method for separating those reorganization mixtures that can be volatilized without decomposition is gas chromatography, which has been applied to a study3 of the exchange of methyl, ethyl, propyl and butyl groups on silicon. Column chromatography has been used to investigate the molecular distribution in equilibrated mixtures of the condensed oxyacids of phosphorus4; and two-dimensional paper or thin-layer chromatography5 has proved unique in separating individual ring from chain species in phosphate mixtures. Perhaps the most useful of the procedures for determining6 the molecular distribution resulting from scrambling reactions is 'H, 31Pand I9Fnuclear magnetic resonance (NMR), which can be applied to samples that have been sealed in NMR tubes in an oxygen- and moisture-free atmosphere before being equilibrated. This method, which is applicable to an extremely wide range of equilibration rates and can be used to assay a large number of systems, is sensitive to the intramolecular environment of the magnetically active atoms so that molecular segments of the type discussed in $15.1.3.2 are precisely the items being assayed. This means that the concepts of graph theory can be applied in interpreting the NMR data'. (J.R. VAN WAZER)

J. R. Van Wazer, Ann. N.Y. Acad. Sci., 159, 5 (1969). G. H. Calingaert, H. A. Beatty, H. Soroos, J . Am. Chem. SOC.,62, 1099 (1940). F. H. Pollard, G. Nickless, P. C. Uden, J . Chrornatogr., 19, 28 (1965). S. Ohashi, Pure Appl. Chem., 44, 415 (1975). E. Karl-Kroupa, Anal. Chem., 28, 1091 (1956); W. J. van Ooij, J. P. W. Houtman, 2. Anal. Chem., 244, 38 (1969). The first reference is to paper and the second to thin-layer chromatography. 6. J. R. Van Wazer, L. C. D. Groenweghe, in Nuclear Magnetic Resonance in Chemistry, B. Pesce, ed., Academic Press, New York, 1965, pp. 283-298. 7. D. W. Matula, L. C. D. Groenweghe, J. R. Van Wazer, J . Chem. Phys., 41, 3105 (1964).

1. 2. 3. 4. 5.

15.1.3.3.1. All-Neso Systems.

Studies of scrambling reactions in systems in which two monofunctional substituents, T and Z , are scrambled on a central atom or moiety of functionality v led to the theory of random scrambling'. The scrambling of substituents that are chemically similar is close to random. Examples are found in the scrambling of methoxyl vs. ethoxyl groups, chlorine vs. bromine and, on the heavier elements where the exchange is sufficiently rapid, one alkyl group for another. However, the scrambling of some pairs of substituents on a given central moiety is far from random. For example, as shown in Figure 1, when tetramethyl silicate is equilibrated with tetraethyl silicate at 15OoC, the system is essentially random, whereas the equilibration of a mixture of tetrakisdimethylaminosilane with silicon tetrachloride is nonrandom, the mixed compounds being produced in

27

R = CI/Si

R = Cl/Si

= 4) of the following groups: A methoxyl vs. ethoxyl at 150"C, B methoxyl VS. chloro at 120"C, and C dimethylamino vs. chloro at 25°C. The dotted lines correspond to the random situation, and each curve cxhiblts a pcak value at the value of the parameter R corresponding to the composition of the compound it represents.

Figure 1. Equilibrium diagrams' for the exchange on silicon (v

R = C2HSO/Si

15.1. Introduction 15.1.3. Chain-Chain, Chain-Ring and Ring-Ring Equilibria. 15.1.3.3.Scrambling in Inorganic Systems.

28

TABLE1. DEVIATIONS FROM RANDOMNESS AH' in kJ mol-I of product"

Q

Z

Si P Si Ge P As Si Ge P

OC2H,

As

C1

CI

T

v

OCH,

4 3 4 4 3 3

OCH,

N(CH3)2 4 3 3

MZ,T,-, O.Ob

+0.4 - 14b - 12b - 13 --8(-14)" - 33 -28(-34)' - 36 -23(-28)'

MZ,T,-, 0.0

+ 0.4 - 16b - 15b

- 13 -12(-16)" - 38 -37(-38)' - 38 -23(-28)"

MZ,T,-,

+ 0.4 -

-1lb -lob -

- 27 -26(-23)' -

Ref. 2 3 2 4 5

6 2 4 5 6

Enthalpies of formation of the mixed building unit from the end-member units, according to Eq. (a) in 515.1 3.2.1. Corrections made for clerical or arithmetical errors in the original papers. ' Values in parentheses were obtained calorimetrically.

> 90 % yield. The equilibration of a tetraalkyl silicate with silicon tetrachloride is intermediate between these two other SiZ,-SiT, systems. Various equilibrated systems of this type may be intercompared by calculating the value of AH&mb according to Eq. (f) in $15.1.3.2.1for the reactions defined in Eq. (a) in that section, and the results of some such calculations are presented in Table 1. Note that a large negative value of the AH:crambcorresponds to an exothermic formation of the mixed compound and hence to its production at a higher level than expected from a randomly reorganizing mixture. If AH&,mbbecomes more negative than ca. -40 kJ per mole of mixed compound, its yield is quantitative and the equilibrium-composition graph looks like graph C of Figure 1 except that the peaks for the intermediate compounds rise to 100 %. On the other hand, if the value of AH&mb becomes as large as about f 4 0 kJ mole-', no mixed compounds are observed and the casual observer will conclude that no reaction has occurred even though, for some labile systems, rapid dynamic equilibration is proceeding at RT. The data in Table 1 indicate that for main-group central atoms Q, the deviations from randomness as measured by the AH&,ambare mainly affected by the choice of the pair of substituents being scrambled rather than the particular central atom. However, the value of AH' for a given Q-Z pair may be greater for elements having low-lying empty orbitals (e.g., the group-I11 elements and the transition metals). The deviations from randomness in substituent scrambling on a single kind of central atom are owing to poorly understood electronic effects at about the same level of complexity as the effects postulated in reaction-mechanism speculations. Some success has been achieved by may be the easiest way relating the values of AHZcramb to the omfunction. Indeed, AHtcramb to obtain ominformation for some substituents that cannot be handled otherwise. For exchange of different pairs of substituents on the dimethylsilicon moiety, the values of AHtcramb have been plotted7 so that all of the curves for a given T increase or decrease smoothly when going across a series of Z substituents, as shown in Figure 2. According to this plot and similar data, the various substituents have been placed in an correspond to the more widely separated pairs: order in which larger values of AHZGramb CN > Br > C1> N, > NCS(?) > F > SCH, > H(?) > OCH, > N(CH,),.

15.1.3. Chain-Chain, Chain-Ring and Ring-Ring Equilibria. 15.1.3.3. Scrambling in Inorganic Systems. 15.1.3.3.1. All-Neso Systems.

29

-10

d

m

E

5 LE

0

r

0

c a,

I

5

Figure 2. Curves7 showing the deviations from randomness, AH'&,,b, for the scrambling of various pairs of T and Z substituents on dimethylsilicon (v = 2). The sequence of Z substituents was arbitrarily chosen to give reasonably smooth curves. Any number of different substituents (Z, T, L, X, . . .) may be scrambled on a given central moiety (Q, with v > 1) and all possible combinations of the substituents must be considered. In such multisubstituent systems*, a nonredundant set of equilibrium constants must be so constituted that each one of the constants describes the relationship of a different mixed compound to other compounds in the overall system. As a result, the total number of independent constants equals the sum of the different kinds of molecular species other than those exhibiting a single kind of substituent. In addition to the equilibrium expressions of the form of either Eq. (a) or Eq. (b) in 515.1.3.2.1, there must be an equivalent relationship for those compounds based on more than two kinds of substituents (e.g., QZTL, QZT,L or QZTLX). For example, the compound QZT,L can be related to the overall system by an equation such as: QZ,

-

+ 2 QT4 + QL4 -4

or 2 QZT,L

QZ,TL

QZT,L

+ QT,L

(a)

(b)

Experimental examples* of systems of this type include the scrambling of methoxyl, dimethylamino and chloro groups on silicon (v = 4) and of chloro, bromo, iodo and phenoxyl groups on methylgermanium (v = 3), using NMR for the molecular assays. For a single kind of central moiety, the heat of a scrambling reaction is due chiefly to the nonadditivities of bond energies and so does not deviate greatly from the zero value for a wholly random process. However, when there is redistribution of substituents between two different central atoms, QZ, and MT,, the exchange of a Z for a T results in

30

15.1. Introduction 15.1.3. Chain-Chain, Chain-Ring and Ring-Ring Equilibria. 15.1.3.3. Scrambling in Inorganic Systems.

kJ/bond

- - o VS.

CI

0 vs. Br

0 VS. I

-5 vr. HI.

80 %l

69

3

c 0

80 40 El ICI) I ArlSi lBrl BlSi lBrl 1 BISI mi

20

0

lCli BIAS IBrI

Figure 3. Values9 of the entropy or free energy per bond for exchange of oxygen or sulfur with halo groups on pairs of different central moieties. These pairs (e.g., B/Sn) are shown with the moiety preferentially bonded to oxygen being given first.

the abolition of a Q-Z and an M-T bond with the concometant formation of a Q-T and an M-Z bond. This means that AH:crambusually exhibits a large value for scrambling processes involving exchange of substituents between different central moieties, so that no exchange (for positive values of AH&,,b) or complete exchange (for negative values of AH:cramb)will be observed for: pQZ,

+ vMT,

pQT,

+ vMZ,

(c>

The equilibrium constant for this intersystem reaction is designated herein as K,. When Q is an atom, an estimate of the size of AH&rambcan be made from thermodynamic tables even if the particular compound is not listed. T o the approximation that p = 1, tabulated values of AH' or AGO for an oxide (such as SiO,) can be used to estimate AHc for a scrambling reaction in which an ester, such as %(OR),, is involved. A graphg of the enthalpy or free energy per bond for a variety of pairs of central atoms is given in Figure 3 for the Z vs. T pairs of oxygen or sulfur vs. various halogens. In the scrambling of monofunctional substituents on more than one kind of central moiety, all the equilibria substituent-pair exchange on each central moiety must be considered as well as some reaction, such as that shown in Eq. (m), to account for the redistribution of substituents between each pair of central moieties. The exchange between dimethylsilicon and dimethylgermanium moietiesi illustrates the methodology and may be related to other work1',12 on exchange between methylsilicon and

15.1. Introduction 15.1.3. Chain-Chain, Chain-Ring and Ring-Ring Equilibria. 15.1.3.3. Scrambling in Inorganic Systems.

31

methylgermanium. The same formal treatment may be accorded to a given central atom with different numbers of blocking groups [e.g., CH,Si and (CH,),Si]. (J.R VAN WAZER)

G. Calingaert, H. A. Beatty, J . Am. Chem. SOC.,61, 2748 (1939). K. Moedritzer, J. R. Van Wazer, Inorg. Chem., 3, 268 (1964). K. Moedritzer, G. M. Burch, J. R. Van Wazer, H. K. Hofmeister, Inorg. Chem., 2, 1152 (1963). G. M. Burch, J. R. Van Wazer, J . Chem. SOC.,A, 586 (1966). E. Fluck, J. R. Van Wazer, Z . Anorg. Allg. Chem., 307,, 113 (1961). K. Moedritzer, J. R. Van Wazer, Inorg. Chem., 3, 139 (1964). H. Muller, J. R. Van Wazer, J . Organomet. Chem., 23, 395 (1970); K. Moedritzer, J. R. Van Wazer, Inorg. Chem., 7, 2105 (1968). 8. J. R. Van Wazer, S.Norval, Inorg. Chem., 4, 1294 (1965); K. Moedritzer, J. R. Van Wazer, R. E. Miller, 7, 1638 (1968). Examples of the scrambling of three substituents on silicon and of four substituents on methylsilicon. 9. J. R. Van Wazer, K. Moedritzer, J . Am. Chem. SOC.,90,47 (1968). 10. J. R. Van Wazer, K. Moedritzer, L. C. D. Groenweghe, J . Organomet. Chem., 5,420 (1966); K. Moedritzer, J. R. Van Wazer, J . Inorg. Nucl. Chem., 28, 957 (1966); K. Moedritzer, L. C. D. Groenweghe, J . Phys. Chem., 72,4380 (1968). 11. K. Moedritzer, J. R. Van Wazer, J . Inorg. Chem., 5, 547 (1966). 12. K. Moedritzer, J. R. Van Wazer, J . Inorg. Chim. Acta, I , 407 (1967). 1. 2. 3. 4. 5. 6. 7.

15.1.3.3.2. Oligomers and Polymers in Q2,-QT, Systems with 2 Bridging.

To the approximation that p = 1 in these systems, all-neso scrambling (in which the Z substituent is an atom acting as a bridge in the chain-ring family of compounds but with that atom blocked off by a nonexchangeable group in the all-neso case) can be considered as a model system for the interpretation of the scrambling equilibria involving chains and rings. For example, the exchange of a CH,O monofunctional group with a halogen gives information concerning the exchange of a bridging oxygen with the same halogen. Arsenic(II1) oxide dissolves in arsenic trihalides to give solutions that are increasingly viscous at the higher oxide concentrations but that can be separated into their original constituents by distillation'. Fluorine-19 NMR of the As,O,-AsF, mixtures discloses' 10 resonances: one for the neso molecule, AsF,, and three for end-group environments [e,, emand eb, where the boldface e refers to the F,As(Oliz) giving rise to the particular N M R peak and the e, m or b to the nearest-neighbor end, middle nearestneighbor or branch group that modulates the observed I9F shift]. Likewise, there are six middle-group, F A S ( O ~ , , ) ~resonances , (eme, emm, emb, mmm, mmb, bmb). The equilibrium constants for p = 1 have the values: K, and : K,

=

=

[ASF,][ASF(O,~~)~]/[ASF~(O~,~)]~ = 0.33 0.06

C ~ ~ ~ ~ , , z ~ , l C ~ ~ ~ z= ~0.31 ~ k, 0.02 , z ~ l / ~ ~ ~ ~ ~ ~ ~

with Krand= 0.3333. Few if any cyclic molecules appear below the gel point, so this system may be treated in terms of random sorting of the constituent atoms, without need to consider high-order segments or cyclization. The gel point for the As,O,-AsF, system is calculated from K, and K, to be at R 3 [F]/[As] = 1.485, a value close to the composition of the tetra-chain molecule, F, AsOAs(F)OAs(F)AsF,

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc.

15.1. Introduction 15.1.3. Chain-Chain, Chain-Ring and Ring-Ring Equilibria. 15.1.3.3. Scrambling in Inorganic Systems.

31

methylgermanium. The same formal treatment may be accorded to a given central atom with different numbers of blocking groups [e.g., CH,Si and (CH,),Si]. (J.R VAN WAZER)

G. Calingaert, H. A. Beatty, J . Am. Chem. SOC.,61, 2748 (1939). K. Moedritzer, J. R. Van Wazer, Inorg. Chem., 3, 268 (1964). K. Moedritzer, G. M. Burch, J. R. Van Wazer, H. K. Hofmeister, Inorg. Chem., 2, 1152 (1963). G. M. Burch, J. R. Van Wazer, J . Chem. SOC.,A, 586 (1966). E. Fluck, J. R. Van Wazer, Z . Anorg. Allg. Chem., 307,, 113 (1961). K. Moedritzer, J. R. Van Wazer, Inorg. Chem., 3, 139 (1964). H. Muller, J. R. Van Wazer, J . Organomet. Chem., 23, 395 (1970); K. Moedritzer, J. R. Van Wazer, Inorg. Chem., 7, 2105 (1968). 8. J. R. Van Wazer, S.Norval, Inorg. Chem., 4, 1294 (1965); K. Moedritzer, J. R. Van Wazer, R. E. Miller, 7, 1638 (1968). Examples of the scrambling of three substituents on silicon and of four substituents on methylsilicon. 9. J. R. Van Wazer, K. Moedritzer, J . Am. Chem. SOC.,90,47 (1968). 10. J. R. Van Wazer, K. Moedritzer, L. C. D. Groenweghe, J . Organomet. Chem., 5,420 (1966); K. Moedritzer, J. R. Van Wazer, J . Inorg. Nucl. Chem., 28, 957 (1966); K. Moedritzer, L. C. D. Groenweghe, J . Phys. Chem., 72,4380 (1968). 11. K. Moedritzer, J. R. Van Wazer, J . Inorg. Chem., 5, 547 (1966). 12. K. Moedritzer, J. R. Van Wazer, J . Inorg. Chim. Acta, I , 407 (1967). 1. 2. 3. 4. 5. 6. 7.

15.1.3.3.2. Oligomers and Polymers in Q2,-QT, Systems with 2 Bridging.

To the approximation that p = 1 in these systems, all-neso scrambling (in which the Z substituent is an atom acting as a bridge in the chain-ring family of compounds but with that atom blocked off by a nonexchangeable group in the all-neso case) can be considered as a model system for the interpretation of the scrambling equilibria involving chains and rings. For example, the exchange of a CH,O monofunctional group with a halogen gives information concerning the exchange of a bridging oxygen with the same halogen. Arsenic(II1) oxide dissolves in arsenic trihalides to give solutions that are increasingly viscous at the higher oxide concentrations but that can be separated into their original constituents by distillation'. Fluorine-19 NMR of the As,O,-AsF, mixtures discloses' 10 resonances: one for the neso molecule, AsF,, and three for end-group environments [e,, emand eb, where the boldface e refers to the F,As(Oliz) giving rise to the particular N M R peak and the e, m or b to the nearest-neighbor end, middle nearestneighbor or branch group that modulates the observed I9F shift]. Likewise, there are six middle-group, F A S ( O ~ , , ) ~resonances , (eme, emm, emb, mmm, mmb, bmb). The equilibrium constants for p = 1 have the values: K, and : K,

=

=

[ASF,][ASF(O,~~)~]/[ASF~(O~,~)]~ = 0.33 0.06

C ~ ~ ~ ~ , , z ~ , l C ~ ~ ~ z= ~0.31 ~ k, 0.02 , z ~ l / ~ ~ ~ ~ ~ ~ ~

with Krand= 0.3333. Few if any cyclic molecules appear below the gel point, so this system may be treated in terms of random sorting of the constituent atoms, without need to consider high-order segments or cyclization. The gel point for the As,O,-AsF, system is calculated from K, and K, to be at R 3 [F]/[As] = 1.485, a value close to the composition of the tetra-chain molecule, F, AsOAs(F)OAs(F)AsF,

32

15.1. Introduction 15.1.3. C hain-C hai n, C hai n-Ri ng and Ri ng-Ri ng Equ i Iibria. 15.1.3.3. Scrambling in Inorganic Systems.

for which R = 1.500. At 25°C the viscosity at the gel point is only about 1 poise and it increases rapidly at lower values of R, rising to lo5 poises at R = 0.88 and ca. 5 x lo7 at R = 0.75. Although the viscosity data suggest that the gel point lies at an R value 0.70, the reason for flow being observed beyond the gel point at R < 1.485 is the rapid rate of reorganization of the molecules making up this system. Indeed, an average lifetime of 0.01 s at the R = 1.00 composition and 0.4 s at R = 1.25 (where the viscosity is ca. 100 poises) owing to the exchange of F and 0 atoms between the building units was calculated from NMR line broadening. For the corresponding all-neso system3, AsF,-As(OCH,), , the average lifetime for F-vs.-(OCH,) scrambling (extrapolated to the solvent-free compositions) is ca. 0.002 sec. This value is much smaller than those measured on the neat polymer system because of the inhibition of the atomic exchange processes by the viscosity and the necessity for proper chain orientation to take place before the atoms can move from one molecular site to another. For the all-neso system, the substituent-exchange equilibrium constants (K, = 0.0138, K, = 0.080) are smaller than the corresponding constants for AsF3-As,03, suggesting that the As-0 bond is thermodynamically affected to an appreciable amount by the substitution of an As by a CH, group on the oxygen. system is The a,w-dibromopoly(dimethylsi1anes) in the (CH,),GeBr,-(CH,),GeS an example in which a ring molecule is in equilibrium with a distribution of chains. Under conditions (120°C for 12 h) in which the methyl groups do not rearrange4, the 'H NMR pattern of the methyl groups shows 11 resonances, i.e., that for the neso compound, (CH,),GeBr,, and the trimeric ring, [(CH,),GeS],, along with those of end- and middle-group overlapping segments involving two building units on either side of the particular one giving the resonance (e.g. the emmmm arrangement). For the chain population in this system, K = zo x Z ; / ( Z ~ ) ~ = 0.06, whereas for the equilibrium between the trimeric ring and the chain distribution [according to Eq. (b) in $15.1.3.2.21,K i = 50. For the all-neso system, (CH,),GeBr,-(CH,),Ge(SCH,),,the value of K is 0.025; and for the two related systems in which C1 or I is employed in place of the Br, the K for the all-neso system is also smaller than that for the comparable chain distribution. However, as would be expected if p were unity so that a constant of the form of Eq. (b) in $15.1.3.2.2 might properly be used, the value of K i is always found to equal the same number (50), regardless of the halogen involved. In the dimethylp~lysulfates~ (a distribution of straight chains in equilibrium with rings) for which Q = SO, with v = 2, T = CH,O, and Z = OljZ,the 'H NMR spectrum at 60 MHz exhibits a separate resonance for each of the molecules having from two to nine sulfur atoms in their structure, with the dimethyl sulfate molecule exchanging too slowly to be involved in the equilibria at <: 100°C. Since the NMR data correspond to the terminal atoms of straight chains, the area of each resonance is proportional to the concentration of its particular chain, while SO, units not associated with the chains are easily calculated from the starting composition and the observed chain population. For random scrambling of chain segments (i.e., when 0 = p), the equilibrium constant for the exchange of parts between chains: N

KT' = cj

+

x cj- J(cj)'

must equal unity. From testing 15 different compositions (ie., R values), the weighted average value of KY' is 0.020 for j = 2, 0.153 for j = 3, 0.70 for j = 4, 0.99 for j = 5 and numbers close to unity but with increasingly larger standard deviations for j > 5. These findings indicate that p is about 4. A more abstract treatment based on stochastic graph

15.1. Introduction 15.1.3. Chain-Chain, Chain-Ring and Ring-Ring Equilibria. 15.1.3.3. Scrambling in Inorganic Systems.

33

theory6 indicates that p I4 for this system. For p = 4 with v = 2, a set of four independent equilibrium constants must be employed to describe the equilibria between the chain molecules. The constants Kf""'for j = 1,2, 3 and 4 make up an appropriate set. Liquid sulfur trioxide consists of monomeric and trimeric SO,, with the proportions of the two varying with temperature. At 25"C, the equilibrium constant relating the monomer and trimer is KY?, = [SO,l3/[S3O,] = 3.65 x (mol 1-l)'. From the reaction shown in Eq. (a) in $15.1.3.2.2 for j > 4, an equilibrium constant K:, relating either the monomeric or trimeric SO, to a chain having more than four SO, groups, can be established. Thus, the six independent equilibrium constants needed to describe the overall scrambling equilibria in this system for which p = 4 can be KY' for j = 1,2,3 and 4, ICY:,, and one constant of the form of K:. The families of a,o-dimethoxyl poly(methy1phosph0nates)~for which p c 2, alkyl polyphosphates' for which p = 2, sodium polyphosphates6 for which p = 3 and aliphatic hydrocarbons6 for which p c 2 are examples of systems for which more than (v - 1) equilibrium constants are needed to represent the scrambling equilibria involving the chain population (since p > 1). For the aliphatic hydrocarbons, which make up the system CH,-[C], , the AH:cramb values are all positive so that the QZiT,-i building units for i = 1, 2 and 3 (i.e., the methyl, methylene and tertiary-carbon groups) are present in small amounts at equilibrium. As a result, the equilibrium mixtures consist of methane and amorphous, intractable chars containing some hydrogen as methyl, methylene and tertiary-carbon groups. The gel point lies not far from the methane (R = 4) composition. Similar behavior is observed for the compounds of the group-IV and -V elements which are held together by catenation, i.e., bonds between like elements. Examples are the PX,-[PI and SiX,-[Si], systems, in which X = H, C1 or Br. More of these scrambling systems exhibiting sufficiently large positive values of AH&mb to make equilibration in the system uninteresting to preparative chemists should be identified. (J.R VAN WAZER)

1. 2. 3. 4. 5. 6. 7. 8.

W. Wallace, Phil. Mag., 4 84, 358 (1858); E. Thilo, P. Flogel, Angew. Chem., 69, 754 (1957). J. R. Van Wazer, K. Moedritzer, D. W. Matula, J . Am. Chem. Soc., 86, 807 (1964). K. Moedritzer, J. R. Van Wazer, Inorg. Chem.,3, 139 (1964). K. Moedritzer, J. R. Van Wazer, J. Am. Chem. Soc., 87,2360 (1965). J. R. Van Wazer, D. Grant, C. H. Dungan, J. Am. Chem. Soc., 87, 3333 (1965). D. W. Matula, L. C . D. Groenweghe, J. R. Van Wazer, J. Chem. Phys., 41, 3105 (1964). D. Grant, J. R. Van Wazer, C. H. Dungan, J. Polymer Sci., A-I 5, 57 (1967). J. R. Van Wazer, S . Norval, J. Am. Chem. Soc., 88,4415 (1966).

15.1.3.3.3. Exchange of Segments Between Cyclic Molecules.

In systems exhibiting ring chain equilibria that favor cyclic molecules, compositions in which terminal groups are absent equilibrate between various cyclic species in which two or more different kinds of bridging Z groups connect the Q moieties. For example, when [(CH,),SiS],, symbolized by (SSS), is equilibrated' with [(CH,),SiN(CH,)],, symbolized by (NNN), the two mixed-ring species (SNN) and (SSN) are formed, along with the dimeric ring (SS). The three equilibrium constants needed to describe the system are K& = [(SSS)] [(SNN)]/[(SSN)]2, K&N = [(SSN)] [(NNN)]/[(SNN)IZ and K&ss = [(SSS)lZ/[(SS)l3. For the first two of these constants (which are formally equivalent to the pair of constants in the all-neso system QZ,-QT,), the random value

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc. 15.1. Introduction 15.1.3. Chain-Chain, Chain-Ring and Ring-Ring Equilibria. 15.1.3.3. Scrambling in Inorganic Systems.

33

theory6 indicates that p I4 for this system. For p = 4 with v = 2, a set of four independent equilibrium constants must be employed to describe the equilibria between the chain molecules. The constants Kf""'for j = 1,2, 3 and 4 make up an appropriate set. Liquid sulfur trioxide consists of monomeric and trimeric SO,, with the proportions of the two varying with temperature. At 25"C, the equilibrium constant relating the monomer and trimer is KY?, = [SO,l3/[S3O,] = 3.65 x (mol 1-l)'. From the reaction shown in Eq. (a) in $15.1.3.2.2 for j > 4, an equilibrium constant K:, relating either the monomeric or trimeric SO, to a chain having more than four SO, groups, can be established. Thus, the six independent equilibrium constants needed to describe the overall scrambling equilibria in this system for which p = 4 can be KY' for j = 1,2,3 and 4, ICY:,, and one constant of the form of K:. The families of a,o-dimethoxyl poly(methy1phosph0nates)~for which p c 2, alkyl polyphosphates' for which p = 2, sodium polyphosphates6 for which p = 3 and aliphatic hydrocarbons6 for which p c 2 are examples of systems for which more than (v - 1) equilibrium constants are needed to represent the scrambling equilibria involving the chain population (since p > 1). For the aliphatic hydrocarbons, which make up the system CH,-[C], , the AH:cramb values are all positive so that the QZiT,-i building units for i = 1, 2 and 3 (i.e., the methyl, methylene and tertiary-carbon groups) are present in small amounts at equilibrium. As a result, the equilibrium mixtures consist of methane and amorphous, intractable chars containing some hydrogen as methyl, methylene and tertiary-carbon groups. The gel point lies not far from the methane (R = 4) composition. Similar behavior is observed for the compounds of the group-IV and -V elements which are held together by catenation, i.e., bonds between like elements. Examples are the PX,-[PI and SiX,-[Si], systems, in which X = H, C1 or Br. More of these scrambling systems exhibiting sufficiently large positive values of AH&mb to make equilibration in the system uninteresting to preparative chemists should be identified. (J.R VAN WAZER)

1. 2. 3. 4. 5. 6. 7. 8.

W. Wallace, Phil. Mag., 4 84, 358 (1858); E. Thilo, P. Flogel, Angew. Chem., 69, 754 (1957). J. R. Van Wazer, K. Moedritzer, D. W. Matula, J . Am. Chem. Soc., 86, 807 (1964). K. Moedritzer, J. R. Van Wazer, Inorg. Chem.,3, 139 (1964). K. Moedritzer, J. R. Van Wazer, J. Am. Chem. Soc., 87,2360 (1965). J. R. Van Wazer, D. Grant, C. H. Dungan, J. Am. Chem. Soc., 87, 3333 (1965). D. W. Matula, L. C . D. Groenweghe, J. R. Van Wazer, J. Chem. Phys., 41, 3105 (1964). D. Grant, J. R. Van Wazer, C. H. Dungan, J. Polymer Sci., A-I 5, 57 (1967). J. R. Van Wazer, S . Norval, J. Am. Chem. Soc., 88,4415 (1966).

15.1.3.3.3. Exchange of Segments Between Cyclic Molecules.

In systems exhibiting ring chain equilibria that favor cyclic molecules, compositions in which terminal groups are absent equilibrate between various cyclic species in which two or more different kinds of bridging Z groups connect the Q moieties. For example, when [(CH,),SiS],, symbolized by (SSS), is equilibrated' with [(CH,),SiN(CH,)],, symbolized by (NNN), the two mixed-ring species (SNN) and (SSN) are formed, along with the dimeric ring (SS). The three equilibrium constants needed to describe the system are K& = [(SSS)] [(SNN)]/[(SSN)]2, K&N = [(SSN)] [(NNN)]/[(SNN)IZ and K&ss = [(SSS)lZ/[(SS)l3. For the first two of these constants (which are formally equivalent to the pair of constants in the all-neso system QZ,-QT,), the random value

34

15.1. Introduction 15.1.3. Chain-Chain, Chain-Ring and Ring-Ring Equilibria. 15.1.3.3. Scrambling in Inorganic Systems.

would be 0.3333. Since the observed values of KiiN = 0.32 k 0.02 at 120°C and 0.34 i-0.02, with KiiN = 0.98 i- 0.05 at 120°C and 0.90 5 0.04 at 25°C are quite close to the random one (with AH&,,b differing from zero by & 1.4 kJ), interchanging a bridging sulfur for a methylimide bridge does not greatly affect the basic structure of the ring2. The value of K&ss = 130 k 20 at 120°C and 3600 f 700 at 25°C. The large temperature dependence of the constant relating two different sizes of rings as compared to the tiny change with temperature for those relating the same sized ones is typical and suggests that ring size has more effect on the scrambling thermodynamics than the choice of the atoms making up the ring. For the exchange' of a bridging oxygen for a bridging sulfur in trimeric rings based on Q = (CH,),Ge moieties, Ki:o = 0.87 at 120°C and 1.07 at 25"C, with K;g0 = 0.59 at 120°C and 0.72 at 25°C. In this system, there is also a ring of another size, (0000), for = [(0000)]3/[(000)]4 = 0.24 at 120°C and 5.3 at 25°C. which K~ooo,ooo In a system having more than one kind of central moiety involved in ring structures, mixed rings can result from an equilibration process. For example', a mixture of [(CH,),SiS],, symbolized by (SiSiSi), and [(CH,),GeS],, symbolized by (GeGeGe), when equilibrated at 140°C forms the mixed rings (SiSiGe) and (SiGeGe), as well as the dimeric ring (SiSi). Again the equilibrium constants interrelating the trimeric rings are not far from the random values, whereas the constant relating the (SiSi) to the (SiSiSi) ring is large. (J.R. VAN WAZER)

1. K. Moedritzer, J. R. Van Wazer, J. Phys. Chem., 70, 2030 (1966); J. Am. Chem. SOC.,90, 1708 (1968); Inorg. Chim.Acta, I, 152 (1967). 2. R. M. Levy, J. R. Van Wazer, J. Chem. Phys., 45, 1824 (1966). 15.1.3.3.4. Chains and Rings from Scrambling Between Two Kinds of Central Moieties.

As was pointed out for all-neso systems in 415.1.3.3.1, the scrambling of two substituents T and Z between two kinds of central moieties Q and M usually involves the segregation of one substituent on one central moiety and the other on the other, owing to the large free-energy change usually involved in the interchange of different substituents between two different atoms, as shown in Figure 3 in that section. For the situation discussed below, in which Z is bridging, the usual sets of equilibrium constants that deal separately with the QZ,-QT, and the MT,-MZ, systems are required along with the appropriate ring-chain constants. In addition, there is an intersystem equilibrium constant such as that of Eq. (c) in $15.1.3.3.1 and a linkage constant to deal with the arrangement of Q and M moieties around each bridging Z . This latter constant is similar in form to that describing the exchange of substituents on a central atom having a functionality of two and has the same random value (Krand= 0.250). If this linkage constant shows the M-Z-M and Q-Z-Q arrangements to be preferred to the M-Z-Q, the system exhibits purely Q and purely M chain oligomers and, for macromolecules, block polymers in which there are long runs of either Q- or M-based building units. On the other hand, a linkage constant showing the Q-Z-M arrangement to be preferred corresponds to chains that are dominated by alternating Q and M moieties. Stochastic graph theory has been applied' to systems in which chains and rings are formed from scrambling between two kinds of central moieties so that a theoretical framework is available to aid in interpretation of the experimental data.

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc. 34

15.1. Introduction 15.1.3. Chain-Chain, Chain-Ring and Ring-Ring Equilibria. 15.1.3.3. Scrambling in Inorganic Systems.

would be 0.3333. Since the observed values of KiiN = 0.32 k 0.02 at 120°C and 0.34 i-0.02, with KiiN = 0.98 i- 0.05 at 120°C and 0.90 5 0.04 at 25°C are quite close to the random one (with AH&,,b differing from zero by & 1.4 kJ), interchanging a bridging sulfur for a methylimide bridge does not greatly affect the basic structure of the ring2. The value of K&ss = 130 k 20 at 120°C and 3600 f 700 at 25°C. The large temperature dependence of the constant relating two different sizes of rings as compared to the tiny change with temperature for those relating the same sized ones is typical and suggests that ring size has more effect on the scrambling thermodynamics than the choice of the atoms making up the ring. For the exchange' of a bridging oxygen for a bridging sulfur in trimeric rings based on Q = (CH,),Ge moieties, Ki:o = 0.87 at 120°C and 1.07 at 25"C, with K;g0 = 0.59 at 120°C and 0.72 at 25°C. In this system, there is also a ring of another size, (0000), for = [(0000)]3/[(000)]4 = 0.24 at 120°C and 5.3 at 25°C. which K~ooo,ooo In a system having more than one kind of central moiety involved in ring structures, mixed rings can result from an equilibration process. For example', a mixture of [(CH,),SiS],, symbolized by (SiSiSi), and [(CH,),GeS],, symbolized by (GeGeGe), when equilibrated at 140°C forms the mixed rings (SiSiGe) and (SiGeGe), as well as the dimeric ring (SiSi). Again the equilibrium constants interrelating the trimeric rings are not far from the random values, whereas the constant relating the (SiSi) to the (SiSiSi) ring is large. (J.R. VAN WAZER)

1. K. Moedritzer, J. R. Van Wazer, J. Phys. Chem., 70, 2030 (1966); J. Am. Chem. SOC.,90, 1708 (1968); Inorg. Chim.Acta, I, 152 (1967). 2. R. M. Levy, J. R. Van Wazer, J. Chem. Phys., 45, 1824 (1966). 15.1.3.3.4. Chains and Rings from Scrambling Between Two Kinds of Central Moieties.

As was pointed out for all-neso systems in 415.1.3.3.1, the scrambling of two substituents T and Z between two kinds of central moieties Q and M usually involves the segregation of one substituent on one central moiety and the other on the other, owing to the large free-energy change usually involved in the interchange of different substituents between two different atoms, as shown in Figure 3 in that section. For the situation discussed below, in which Z is bridging, the usual sets of equilibrium constants that deal separately with the QZ,-QT, and the MT,-MZ, systems are required along with the appropriate ring-chain constants. In addition, there is an intersystem equilibrium constant such as that of Eq. (c) in $15.1.3.3.1 and a linkage constant to deal with the arrangement of Q and M moieties around each bridging Z . This latter constant is similar in form to that describing the exchange of substituents on a central atom having a functionality of two and has the same random value (Krand= 0.250). If this linkage constant shows the M-Z-M and Q-Z-Q arrangements to be preferred to the M-Z-Q, the system exhibits purely Q and purely M chain oligomers and, for macromolecules, block polymers in which there are long runs of either Q- or M-based building units. On the other hand, a linkage constant showing the Q-Z-M arrangement to be preferred corresponds to chains that are dominated by alternating Q and M moieties. Stochastic graph theory has been applied' to systems in which chains and rings are formed from scrambling between two kinds of central moieties so that a theoretical framework is available to aid in interpretation of the experimental data.

15.1.3. Chain-Chain, Chain-Ring and Ring-Ring Equilibria. 15.1.3.3. Scrambling in Inorganic Systems. 15.1.3.3.4. Chains and Rings from Scrambling Between Central Moieties.

35

In the equilibrated system2:

heat is evolved and the 'H NMR pattern gives evidence of thorough reaction upon mixing dimethyldibromosilane with dimethylgermanium oxide, whereas mixtures of dimethylbromogermane with dimethylsilicon oxide show no evidence of reaction even after several weeks at 150°C. This behavior is in accord with the large value of 1 x 10'' for the intersystem constant, K,. The linkage equilibrium constant is ~ 0 . 3 5somewhat , larger than the random value of 0.250, and corresponds to a greater prevalence of Si-0-Si and Ge-0-Ge arrangements over that expected by chance. The constant for Eq. (b) in 515.1.3.2.1 for v = 2 and i = 1 is 0.150 for the silicon end-groups and 0.043 for the Ge end groups. This set of equilibrium constants for p = 1 corresponds to the molecular distribution indicated in Figure 1, in which it can be seen that the diagonal connecting the [(CH,),SiO], composition to the (CH,),GeBr, composition consists of a simple mixture of these two and divides the system into two characteristically different sectors. In the sector bounded by the (CH,),SiBr,, (CH,),GeBr, and [(CH,),SiO,], compositions, the molecules consist only of germanium-free siloxanes dissolved in dimethylgermanium dibromide, which is a chemically interacting solvent that masquerades as an inert one because of the large value of the intersystem equilibrium constant. In the other

Figure 1. A schematic diagramZindicating the molecular composition within the system (CH,),SiBr,-(CH,),GeBr,-[(CH,),SiO],-[(CH,),GeO], in which there is scrambling of bridging oxygens with bromo groups.

36

15.1. Introduction 15.1.3. Chain-Chain, Chain-Ring and Ring-Ring Equilibria. 15.1.3.3. Scrambling in Inorganic Systems.

sector bounded by [(CH,),SiO],, [(CH,),GeO],, and (CH,),GeBr,, mixed germanosiloxanes (all terminated with germanium-based end groups) are found along with dimethylgermanium dibromide but there is no dimethylsilicon dibromide-again in conformity with the large value of the intersystem equilibrium constant. Of course, close to the [(CH,),SiO],-[(CH,),GeO],, side of the diagram of Figure 1, there are ring compounds, but on the whole, cyclic molecules do not seem to play a significant role in this system. In the related sulfur-bridged system', for which Q = (CH,),Si, M = (CH,),Ge, Z = S, and T = C1 (interpretable on the basis of p = l), the intersystem constant is 2 x lo3 by contrast, a value not extremely far from the random one (K, = 1.000). As a result, the diagram of the system can not be divided into two sectors. The linkage equilibrium constant exhibits its random value, and the two constants describing Eq. (b) in $15.1.3.2.1 are set equal to the values found for the respective unmixed systems. Cyclic structures are the chief characteristic of sulfur-bridged systems, and this mixed system is no exception, exhibiting dimeric and trimeric silicon-based rings, a trimeric germaniumbased ring and the two mixed trimeric rings, (SiGeGe) and (SiSiGe). The two ring-chain and three ring-ring equilibrium constants evaluated for this system are found to agree with the values obtained on the related simple systems. In accord with the observed values of the pertinent equilibrium constants, this system is dominated by the cyclic and neso molecules and includes mixed Si-Ge ring and chain structures. In conclusion, the mathematical framework of stochastic graph theory can explain and quantify the data resulting from equilibria involving the exchange of parts between molecules, no matter how complicated the system and how many the kinds of exchangeable atoms and molecular segments involved. As indicated throughout $15.1.3, the usual advantage of a thermodynamic approach in interrelating different systems is multiplied considerably by the use of equilibria between molecular segments (since the segments will appear in a much wider variety of systems than could a like number of molecules), while simultaneously reducing drastically the number of equilibrium constants required to give a quantitative picture of a particular system. Just as for solubility products, ionization constants and standard electrode potentials, which have clarified complicated areas of chemistry, so the extensive areas of chemistry touched upon in $15.1.3and its subsections are now ready for formal organization. Systems with p > unity need more careful attention, and additional data should be obtained for the exchange of more pairs of substituents on various central moieties and for ring-chain and ring-ring equilibrium constants in a greater variety of systems. (J.R. VAN WAZER)

1. K. Moedritzer, J. R. Van Wazer, D. W. Matula, Inorg. Chim. Acta, 3, 559 (1969). 2. J. R. Van Wazer, K. Moedritzer, J. Am. Chem. Soc., 90,47 (1968).

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc.

15.2. Ring-Ring and Ring-Polymer lnterconversions 15.2.1, Introduction. When sulfur is heated from the melting point, the viscosity decreases slowly, rises sharply at 159"C, attains a maximum value at 187"C, and decreases Impurities can influence the viscosity considerably; however, when pure sulfur is used, the results are independent of the heating rate. Other physical properties, such as the specific heat and the density, also exhibit anomalous variations near 159°C. This unusual temperature dependency is explained if liquid sulfur is considered a mixture of two modifications, S, and S, ., Sulfur molecules in the rhombic, monoclinic, and liquid state, not too far above the melting point, are eight-atom rings, referred to as S, '. The other modification, S,, is composed of chain polymers of different lengths. The rise in viscosity at 159°C is explained by the partial conversion of S, molecules into polymers. This transition temperature is called the floor temperature. The difference between S, and S, can be demonstrated by heating liquid sulfur above 159°C and quenching it rapidly to ca. -2O"C, when an elastic substance is obtained. Part of this substance, i.e., S,, dissolves in CS, as S, molecules. The insoluble part is called S,. When kept at RT, the elastic sulfur hardens and rhombic crystals are observed. Elastic sulfur is a mixture of S, rings and chain molecules6. The fp of liquid sulfur does not always coincide with the mp of monoclinic sulfur7. The fp depression is dependent on the equilibration temperature and time'. When the liquid is kept for a longer time or at a higher temperature before the fp is measured, the depression is larger. The maximal depression occurs after a wait of several hours, the time necessary to reach equilibrium. This fp depression is attributed to a third modification of liquid sulfur, called S, '. The existence of this modification can also be demonstrated by quenching liquid sulfur rapidly to RT. The substance is extracted with CS, and the solution cooled to -80°C. Most of the dissolved S, precipitates and is removed. The remaining solution has a higher sulfur content than the solubility of S, indicates, proving the existence of an extra modification, S,, in the solution. The structure of S, is very complicated. A chain has been ruled out by ESR" and magnetic susceptibility measurement^'^*^^, which indicate the absence of unpaired electrons in low-T liquid sulfur. Cryoscopic measurements on CS, fractions, obtained by countercurrent distribution identify s,, s,, s,, and Sea, 23-33 components". Vibrational a mixture of ring molecules, of which s6, s,, s,,, s p e c t r o ~ c o p y ' shows ~ ~ ~ ~that ~ ' S,~ is ~~ ~ S,, and S, with n 2 20 are demonstrated. Melting rhombic and monoclinic sulfur gives mainly S, rings at first. However, when the temperature is held for some time above the mp, other rings are formed. Below 159°C the proportion of chain molecules is low, but at 159°C there is a sudden increase. On heating above 159°C the length of the chains increases, and at above ca. 200°C it decreases. Near the bp S,, S, and S, molecules are presentzorz1. The stable crystalline forms of solid selenium are trigonal or hexagonal, both composed of helical chains. Trigonal selenium melts at 493.7 K. Two monoclinic 37

38

15.2. Ring-Ring and Ring-Polymer Interconversions 15.2.1. Introduction.

allotropes are known, a- and P-selenium. The molecules constituting both allotropes are Se, When molten selenium is cooled not too slowly a glassy modification is formed. Glassy selenium can be compared with elastic sulfur; i.e., it is a mixture of Se, rings and chain molecules. Melting trigonal-, a- or P-selenium yields a mixture of chains and rings. Liquid selenium can therefore be compared with liquid sulfur above its floor temperature. In contrast to liquid sulfur, liquid selenium has no temperature range in which it is composed mainly of rings. Thus the floor temperature must lie below the melting point. exist, including those Many reviews of s u l f ~ r ’ ~ and - ~ ~selenium23,25~26,27~33-35 covering the properties of liquid s ~ l f ~ r ~ ~ ~ ~ ~ ~ ~ ~ (J.A. POULIS, J.-P. FRANCOIS, C.H. MASSEN, L.C. VAN POUCKE)

1. L. Z. Rotinjanz, Physik. Chem., 62, 609 (1908). 2. C. Farr, D. B. Macleod, Proc. Roy. SOC.(London), 97, 80 (1920). 3. R. F. Bacon, R. Fanelli, J. Am. Chem. Sac., 65, 639 (1943). 4. A. Smith, W. B. Holmes, Z. Phys. Chem., 42, 469 (1903) 54, 257 (1906). 5. B. E. Warren, J. T. Burwell, J. Chem. Phys., 3, 6 (1935). 6. J. J. Trillat, J. Forestier, Bull. SOC.Chim., 51, 248 (1932). 7. M. D. Gernez, C.R. Sdances Hebd. Acad. Sci., 82, 115 (1876). 8. T. K. Wiewiorowski, A. Parthasarathy, B. L. Slaten, J. Phys. Chem., 72, 1890 (1968). 9. A. H. W. Aten, Z. Phys. Chem., 86, 1 (1914); 88, 321 (1914). 10. P. W. Schenk, U. Thiimmler, Elektrochem., Ber. Bunsenges. Physik. Chem., 63, 1002 (1959). 11. P. W. Schenk, U. Thiimmler, Z. Anorg. Allg. Chem., 315, 271 (1962). 12. D. M. Gardner, G. K. Fraenkel, J. Am. Chem. Soc., 76, 5891 (1954); 78, 3279 (1956). 13. J. A. Poulis, C. H. Massen, P. van der Leeden, Trans. Faraday SOC.,58,474 (1962). 14. J. A. Poulis, W. Derbyshire, Trans. Faraday Soc., 59, 559 (1963). 15. K. Krebs, H. Beine, Z . Anorg. Allg. Chem., 355, 113 (1967). 16. R. Steudel, H.-J. Mausle, Angew. Chem., Int. Ed., 16, 112 (1977). 17. R. Steudel, H.4. Mausle, Angew. Chem., Int. Ed., 17, 56 (1978). 18. R. Steudel, H.4. Mausle, Z. Naturforsch., Teil A , 33, 951 (1978). 19. R. Steudel, H.4. Mausle, Angew. Chem., Int. Ed., 18, 152 (1979). 20. B. Meyer, T. Stroyer-Hansen, D. Jensen, T. V. Oommen, J. Am. Chem. Soc., 93, 1034 (1971). 21. B. Meyer, T. V. Oommen, D. Jensen, J. Phys. Chem. 75, 912 (1971). 22. K. E. Murphy, M. B. Altman, B. Wunderlich, J. Appl. Phys., 48, 4122 (1977). 23. P. Ungar, P. Cherin, in The Physics of Selenium and Tellurium, W. C. Cooper, ed. Pergamon, Oxford, 1969. 24. Gmelin Handbuch der Anorganischen Chemie,System-Nr.9, TeilA , Springer Verlag, Berlin, 1969, 1974. 25. P. Pascal, Nouveau Trait6 de Chimie MinPrale, Tome XIII, ed. Masson & Cie. Paris, 1960. 26. J. C. Bailar, H. J. Emelhus, R. Nyholm, A. F. Trotman-Dickenson, eds., Comprehensive Inorganic Chemistry, Vol. 2, Pergamon Press, Oxford, 1973. 27. M. Schmidt, W. Siebert, K. W. Bagnall, in Comprehensive Inorganic Chemistry, Vol. 15, J. C. Bailar, H. J. Emeltus, R. Nyholm, A. F. Trotman-Dickenson eds., Pergamon Press, Oxford, 1975. 28. B. Meyer, Chem. Rev., 64,429 (1964). 29. B. Meyer, Chem. Rev., 76, 367 (1976). 30. R. E. Harris, J. Phys. Chem., 74, 3102 (1970). 31. B. Meyer, ed., Elemental Sulfur, Chemistry and Physics, Interscience, New York, 1965. 32. J. A. Poulis, C. H. Massen, in Elemental Sulfur, Chemistry and Physics, B. Meyer, ed., Interscience, New York, 1965. 33. Gmelin Handbuch der Anorganischen Chemie, System-Nr. 10, Teil A, Springer-Verlag, Berlin, 1970, 1974. 34. E. Gerlach, P. Gross, eds., Selenium and Tellurium, Springer-Verlag, Berlin, 1979. 35. R. A. Zingaro, W. C. Cooper, Selenium, Van Nostrand, New York, 1974.

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc.

15.2. Ring-Ring and Ring-Polymer Interconversions 15.2.2. lnterconversion of Sulfur or Selenium Rings and Chains 15.2.2.1. Theory.

39

15.2.2. lnterconversion of Sulfur or Selenium Rings and Chains 15.2.2.1. Theory.

The properties of liquid sulfur are described by assuming that below 159°C the liquid is mainly composed of S, rings, while above that temperature there exists an equilibrium between rings and linear polymers, ST The interaction between rings and chains is characterized by: SF-,

+ s, t--s*

(4

where asterisks indicate the chain molecules. The equilibrium constant is assumed to be independent of the value of i. Another model, useful to extend the validity of the theory to chain lengths that are not multiples of eight and to rings S, with 1 # 8 between the melting point and 159"C, gives a reaction ~ c h e m e ~ . ~ :

s, t---ST The equilibrium constants of these reactions are defined by:

Three other quantities, N, W and P, whose values can be measured directly, can also be considered, where N is the total concentration of the polymer species expressed in mol kg- ' and given by: N=

W

(SF)

Substituting Eqs. (e) and (f) into Eq. (h), gives:

From Eq. (i), it can be seen that N is finite only when the condition (j) is fulfilled.

< When Eq. (j) holds, Eq. (i) can be replaced by: K3(S8)

(3

40

15.2. Ring-Ring and Ring-Polymer Interconversions 15.2.2. Interconversion of Sulfur or Selenium Rings and Chains 15.2.2.1. Theory.

The total concentration of S atoms involved in polymer formation, W, is given by: m

C j(S;)

W=

j=1

Using Eqs. (e), (f) and (g), if condition (j) is fulfilled, Eq. (1) can be replaced by: K'(S6)

= [K3(S,)]7/8

(1 - [K3(S,)]'/8}2

The average chain length, P, is then given by: W p=N Using Eqs. (k) and (m), P can also be calculated: 1

P=

1-[K~(sdl''~

Mass balance, applied to the S atoms, leads to:

Co = 8(S8)

+

lK1(S,)'18 I(+,)

+W

where C, is the number of moles of S atoms per kg sulfur. Using Eq. (m), the mass balance (p) can be transformed:

For simplicity the existence of other than eight-membered rings is ignored. This means that K, = 0, so that the second term of the right-hand side of Eq. (4) disappears. The relation: -RT In K

= AH' - TAS"

(4 introduces T into the theoretical development. When values for AH" and ASo are known, the following procedure can be used: 1. Calculate the values of K, and K' for the desired temperature range. 2. Substitute these values into Eq. (4) and solve this equation for (S,). 3. Substitute (S,) into Eqs. (k), (m) and (0).

The values found from this theoretical treatment can then be compared with experimental results, but an oversimplification of this outline is misleading. Step 2 of the procedure especially asks for a careful mathematical treatment. This treatment is followed here because it shows that apparently different theories',' are in fact similar. For the calculation of the quantities N, W and P, the values for A H " , AS'",AH: and AS: must be known. For sulfur the following values' are used: AH'" = 137,100J mol-' AH;

=

13,250 J mol-I

AS'" = 96 J K-' mol-' AS:

=

19.35 J K - ' mol-'

C , = 31.192 mol kg-'

The results of the computations for various temperatures are listed in Table 1.

130 150 160 161 165 170 180 200 240 300 350 400

403 423 433 434 438 443 453 473 513 573 623 673

0.196 0.236 0.258 0.260 0.269 0.281 0.304 0.352 0.458 0.635 0.794 0.960

1.74 x 1.21 x 2.98 x 3.26 x 4.58 x 7.04 x 1.60 x 7.45 x 1.13 x 3.29 x 3.31 x 2.38 x lo-*

lo-" lo-'' lo-''

lo-''

10-l'

5.102 4.237 3.876 3.846 3.717 3.559 3.289 2.841 2.183 1.575 1.259 1.042

Table 1. THEPOLYMERIZATION OF LIQUID SULFUR

x

260.5 x 502 x 1.676 2.697 x 4.681 x 8.35 x 1.61 x 4.25 x 1.85 x 9.83 x 2.97 x 7.54 x 10-3 10-3

10-4

10-4

10-5

lo-" lo-" 2.53 x 10-l' 1.59 x 0.0077 0.015 0.041 0.089 0.158 0.27 0.44 0.60 0.68 0.73

30 99 1.44 x 105 1.79 x 105 2.72 105 3.33 x 105 3.07 105 2.00 x 1 0 5 7.45 x 104 1.90 x 104 7.13 103 3.04 x 103

15.2. Ring-Ring and Ring-Polymer Interconversions 15.2.2. Interconversion of Sulfur or Selenium Rings and Chains 15.2.2.1. Theorv. 41

42

15.2. Ring-Ring and Ring-Polymer Interconversions 15.2.2. lnterconversion of Sulfur or Selenium Rings and Chains 15.2.2.1. Theory.

Table 2. THEPOLYMERIZATION OF LIQUID SELENIUM

CSe*l,

N,

Q,

P,

T("C)

TW)

K3

K

Eq. (9)

Eq. (k)

Eq. (ak)

eq. (0)

227 327 427 527 627 827 1027

500 600 700 800 900 1100 1300

1.596 2.337 3.068 3.762 4.411 5.455 6.398

1.27 x 8.36 x lo-' 1.68 x 1.576 x lo-' 9.07 x 1.15 6.70

0.624955 0.421235 0.30739 0.22714 0.16134 0.06397 0.0156

2.47 x 1.797 x lo-' 0.745 x lo-' 2.11 x lo-' 5.69 x lo-' 1.50 3.13

0.6052 0.734 0.806 0.857 0.898 0.96 0.99

3.10 x lo3 517 137 51.4 20.0 8.11 4.01

For selenium, the computations are performed with the following values4:

AH'" = 104,500 J mol-'

AS'" = 96 J K - ' mol-'

AH:

AS: = 22.86 J K-' mol-'

C,

= 9490 J mol-' =

12.665 mol kg-'

The results of the calculations for several temperatures are given in Table 2. As can be seen from Table 1, the K values are small. With this fact in mind Eq. (9) is transformed:

K'=AxB

(s)

A and B are given by: A

B=

= C, - 8y8

0)

K;'*(l - Kk'8y)Z Y

where y is (S8)'". As K is so small, Eq. (t) can be approximated for most purposes by A = 0 or B = 0. The following considers when this approximation is allowed, and which of the two factors A or B has to be taken as zero. When A = 0:

yA =

(y8

When B = 0: yB -- K;i/s

The solution depends on the temperature. The choice is determined by two conditions: Eq. (x), which follows from Eq. (j): y < K;'"

and Eq. (y), which follows from Eq. (4):

(4

15.2. Ring-Ring and Ring-Polymer Interconversions 15.2.2. Interconversion of Sulfur or Selenium Rings and Chains 15.2.2.1. Theory.

43

Since K, is temperature dependent, there must be a temperature for which yA = yB,or for which:

K, = -

8

co

This temperature is the polymerization temperature, T,. Consider first temperatures, T, above the polymerization temperature, i.e., T > T,. From Eq. (r) and from the relevant data it follows that K, increases with temperature. Therefore, for the high-temperature region: (aa) This can be seen in Fig. 1, where A and B are schematically plotted as a function of y for T > T,. From Eqs. (w) and (aa) it follows that yA does not satisfy condition (x), leaving for consideration only the B solution. Formally stated, this B solution, yB = (1/K3)’18, does not satisfy condition (x). However, the relation A x B = 0 is only an approximation of Eq. (s). This means that a higher approximation finds the intersection of curve B not with the y axis, but with a

\ \

Figure 1. Factors A and B as a function of y for T > T, or K, > 32 do not satisfy conditions (x) and (y).

x

S/lOOO. The dashed lines

44

15.2. Ring-Ring and Ring-Polymer Interconversions 15.2.2. Interconversion of Sulfur or Selenium Rings and Chains 15.2.2.1. Theory.

horizontal line drawn just above the horizontal axis. This leads to two intersections because curve B is parabolic. These two intersections have y values that are both close to yB. The distance is so small that when the solutions of Eq. (s) are computed, care must be exercised to avoid computational difficulties. Figure 1 shows that the smallest of these two roots satisfies condition (x). The difference between this root and yB is so small that it can be neglected throughout. The situation for T < T, schematized in Fig. 2. Here condition (y) leads to the conclusion that only solution yA is acceptable. This means that in this approximation there is no chainlike sulfur in the liquid for T < T,, although estimation of the amount of polymeric sulfur existing at T < T,, may be of interest. To get this the approximation y = yA + Ay is used: A(yA f Ay)B(yA) = K’

(ab)

The solution is given in first approximation by:

YA

Using the relevant data in Table 1 it can be seen that Ay/yA is ca. - 10- l o at 150°C. How wide is the interval about T, where the approximations used for T > T, and for T < T, are not applicable? It is a merit of the complete treatment2 that this temperature

Figure 2. Factors A and B as a function of y for T < T, or K, < 32 do not satisfy conditions (x) and (y).

x S/lOOO. The dashed lines

15.2. Ring-Ring and Ring-Polymer Interconversions 15.2.2. lnterconversion of Sulfur or Selenium Rings and Chains 15.2.2.1. Theory.

45

trajectory can be estimated without solving Eq. (s) numerically. In this estimation the question of determining the temperature trajectory, is stated as the determination of a trajectory of K, values. The value of K, for which yA = yB is called K30. In order to obtain the K, trajectory, AK,, the condition whereby A and B must be equally small is replaced by the solution of Eq. (ad) for AK,: B(yA) = K

A(yB)

(ad)

Substitution of Eqs. (v) and (w) into Eqs. (t) and (u) gives, in a first-order approximation: AK3 = 2

(9 -

(K’)”,

If the relevant value for K’ from Table 1 is substituted into Eq. (ae), AK, = 7 x is obtained, which corresponds to AT = O.Ol”C, which is indeed negligible. Summarizing: 1

for T > T,

(S,) = y; = K3 ,--

(S,) = Y i = 7 L O

for T < T,

(ad

To carry out comparisons with experimental data, these solutions are substituted into Eqs. (k), (m) and ( 0 ) . However, this leads to complications for when T > T,, the approximations used in deriving Eq. (af) cause the denominators of Eqs. (k) and (m) to become zero. This difficulty can be bypassed by combining Eqs. (k), (m) and (p) and taking Eq. (af) into account, since T > T,. The results are: N

=

{K’(S,)[C, - 8(S,)]}”2 W = Co - 8(S,)

(ah) (ail

Another useful quantity is the fraction @ of the S atoms that form a polymer; @ can be calculated with: @=-

W CO

The quantities K,, K‘, y:, N, @ and P are given in Table 1 for several temperatures. The equations used are indicated in the table. Table 1 serves for comparisons with experimental results. In Table 2 the quantities K,, K’, (Se,), N, @ and P are listed for liquid selenium at several temperatures. Since the K‘ values are not as small as those for liquid sulfur, it is not necessary to use approximations. The quantities [Se,], N and Po can be calculated directly with Eqs. (q), (k) and (o), respectively. As for sulfur, the second term of the right-hand side of Eq. (4) is omitted because of the lack of data. The small value of the average chain length for liquid selenium at high temperatures (> 1000 K) makes application of the theory doubtful as long as the possible existence of such molecules as Se, is not accounted for in the liquid state. (J.A POULIS, J -P. FRANFOIS, C H MASSEN, L.C. VAN POUCKE)

46

1. 2. 3. 4.

15.2. Ring-Ring and Ring-Polymer Interconversions 15.2.2. lnterconversion of Sulfur or Selenium Rings and Chains 15.2.2.2. Experimental Techniques

G. Gee, Trans. Favaday Soc., 48, 515 (1952). A. V. Tobolsky, A. Eisenberg, J. Am. Chem. SOC.,81, 780 (1959). J. A. Poulis, C. H. Massen, A. Bsenberg, A. V. Tobolsky, J. Am. Chem., Soc. 87,413 (1965). A. Eisenberg, A. V. Tobolsky, J. Polym. Sci., 46, 19 (1960).

15.2.2.2. Experimental Techniques

Three different kinds of experiments can be used to study the polymerization of liquid sulfur and selenium. First, those which lead to direct results without any parameter adaptation are represented in $15.2.2.3 and $15.2.2.5, Figs. 1 and 2. Second are those where the measured quantity is linearly dependent on the ring-chain composition of the liquid, but where the value of the quantity for a pure polymer is not known. Here the difference of this quantity between eight-membered rings and chains must be adapted. This means that for a quantity A,, measured as a function of T, we use:

'

Usually Aringis known either from the solid-state measurements or from the liquid sulfur measurements below 159°C. The parameter Achain- Aringcan be adapted by comparing of the measured A,, with the value of QT1, obtained from other experiments at a particular temperature TI. Once the value of Achain- Arlngis established, DT can be calculated from A,. Because of this procedure and the necessity of comparing with other results, the outcomes of the experiments of this category are less than those of the first category. The third category comprises experiments where more than one parameter has to be adapted. It is obvious that results obtained from the first category are to be preferred over the others. 15.2.2.2.1, Viscosity Measurements.

As explained in 515.2.1, the physical property most affected by the polymerization of sulfur is the viscosity','. The viscosity increases four decades at 159°C owing to the quickly increasing amount of polymers. At still higher temperatures the viscosity decreases as the polymers shorten. It is difficult to quantify this qualitative picture. The only exception is the steep rise of the viscosity at 159"C,which makes a precise value of T, possible, leading to a stringent relation between the values of the theoretical parameters. Calculating the dependency of the viscosity of liquid sulfur on the chain length and the polymer fraction must be done by analogy with experimental data concerning other polymeric material. This comparison involves adapting of more than one parameter and brings the viscosity measurements into the third category of techniques (see 515.2.2.2). The change of the viscosity as a function of temperature is less p r o n o ~ n c e d ~for -~ liquid selenium because the total liquid phase is above the floor temperature. (J.A POULIS, J.-P. FRANGOIS, C.H MASSEN, L C.VAN POUCKE)

1. R. F. Bacon, R. Fanelli, J. Am. Chem. Soc., 65, 639 (1943). 2. F. J. Touro, T. K. Wiewiorowski, J. Phys. Chem., 70, 239 (1966). 3. S. Dobinski, J. Wesolowski, Bull. Acad. Polon. A , 9 (1937) (cited in Gmelin Handbuch der Anorganischen Chemie, System-Nr 10).

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc.

46

1. 2. 3. 4.

15.2. Ring-Ring and Ring-Polymer Interconversions 15.2.2. lnterconversion of Sulfur or Selenium Rings and Chains 15.2.2.2. Experimental Techniques

G. Gee, Trans. Favaday Soc., 48, 515 (1952). A. V. Tobolsky, A. Eisenberg, J. Am. Chem. SOC.,81, 780 (1959). J. A. Poulis, C. H. Massen, A. Bsenberg, A. V. Tobolsky, J. Am. Chem., Soc. 87,413 (1965). A. Eisenberg, A. V. Tobolsky, J. Polym. Sci., 46, 19 (1960).

15.2.2.2. Experimental Techniques

Three different kinds of experiments can be used to study the polymerization of liquid sulfur and selenium. First, those which lead to direct results without any parameter adaptation are represented in $15.2.2.3 and $15.2.2.5, Figs. 1 and 2. Second are those where the measured quantity is linearly dependent on the ring-chain composition of the liquid, but where the value of the quantity for a pure polymer is not known. Here the difference of this quantity between eight-membered rings and chains must be adapted. This means that for a quantity A,, measured as a function of T, we use:

'

Usually Aringis known either from the solid-state measurements or from the liquid sulfur measurements below 159°C. The parameter Achain- Aringcan be adapted by comparing of the measured A,, with the value of QT1, obtained from other experiments at a particular temperature TI. Once the value of Achain- Arlngis established, DT can be calculated from A,. Because of this procedure and the necessity of comparing with other results, the outcomes of the experiments of this category are less than those of the first category. The third category comprises experiments where more than one parameter has to be adapted. It is obvious that results obtained from the first category are to be preferred over the others. 15.2.2.2.1, Viscosity Measurements.

As explained in 515.2.1, the physical property most affected by the polymerization of sulfur is the viscosity','. The viscosity increases four decades at 159°C owing to the quickly increasing amount of polymers. At still higher temperatures the viscosity decreases as the polymers shorten. It is difficult to quantify this qualitative picture. The only exception is the steep rise of the viscosity at 159"C,which makes a precise value of T, possible, leading to a stringent relation between the values of the theoretical parameters. Calculating the dependency of the viscosity of liquid sulfur on the chain length and the polymer fraction must be done by analogy with experimental data concerning other polymeric material. This comparison involves adapting of more than one parameter and brings the viscosity measurements into the third category of techniques (see 515.2.2.2). The change of the viscosity as a function of temperature is less p r o n o ~ n c e d ~for -~ liquid selenium because the total liquid phase is above the floor temperature. (J.A POULIS, J.-P. FRANGOIS, C.H MASSEN, L C.VAN POUCKE)

1. R. F. Bacon, R. Fanelli, J. Am. Chem. Soc., 65, 639 (1943). 2. F. J. Touro, T. K. Wiewiorowski, J. Phys. Chem., 70, 239 (1966). 3. S. Dobinski, J. Wesolowski, Bull. Acad. Polon. A , 9 (1937) (cited in Gmelin Handbuch der Anorganischen Chemie, System-Nr 10).

15.2.2. lnterconversion of Sulfur or Selenium Rings and Chains 15.2.2.2.Experimental Techniques 15.2.2.2.3. Quenching Experiments. 4. 5. 6. 7.

47

H. Krebs, Z . Anorg. Allg. Chem., 263, 305 (1950). H. Krebs, Z . Anorg. Allg. Chem., 265, 156 (1951). D. E. Harrison, J. Chem. Phys., 41, 844 (1964). R. C. Keezer, M. W. Bailey, Mat. Res. Bull., 2, 185 (1967).

15.2.2.2.2. Ultrasonic Absorption in Liquid Selenium.

The dependency on temperature of the ultrasonic absorption' can be understood by assuming that only the polymers undergo relaxation in the frequency interval used (30-70 MHz), probably owing to the break up of the polymer chain. Further, it is assumed that the rings contribute as a non-relaxing absorption floor. This term decreases with increasing temperature. The relaxation term has a maximum value at 500°C. The sum of the two terms exhibits both a minimum and a maximum and fits with the experimental data. It would be interesting to carry out a combined ultrasonic and ESR experiment in order to investigate whether ultrasonic absorption would influence the ESR spectrum. (J.A. POULIS, J.-P. FRANFOIS, C.H MASSEN, L.C VAN POUCKE)

1. E. M. Ring, R. T. Beyer, J. Acoust. Soc. Am., 62, 582 (1977).

15.2.2.2.3. Quenching Experiments.

The weight fraction of the polymers in liquid sulfur are determined by rapidly chilling droplets of the liquid's'. After hardening the sample is extracted with solvent. The insoluble fraction is the polymer. The experimental data are shown in Fig. 1. At high temperatures there is an uncertainty about the rate of cooling in the first part of the

Figure 1. The weight fraction Q, of polymer as a function of temperature for liquid sulfur. The line is calculated from theory with parameter values as noted in the text. M, data from quenching data from quenching experiment^^,^. +, some of the data from vibrational experiments1. 0 , spectroscopy5.

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc. 15.2.2. lnterconversion of Sulfur or Selenium Rings and Chains 15.2.2.2.Experimental Techniques 15.2.2.2.3. Quenching Experiments. 4. 5. 6. 7.

47

H. Krebs, Z . Anorg. Allg. Chem., 263, 305 (1950). H. Krebs, Z . Anorg. Allg. Chem., 265, 156 (1951). D. E. Harrison, J. Chem. Phys., 41, 844 (1964). R. C. Keezer, M. W. Bailey, Mat. Res. Bull., 2, 185 (1967).

15.2.2.2.2. Ultrasonic Absorption in Liquid Selenium.

The dependency on temperature of the ultrasonic absorption' can be understood by assuming that only the polymers undergo relaxation in the frequency interval used (30-70 MHz), probably owing to the break up of the polymer chain. Further, it is assumed that the rings contribute as a non-relaxing absorption floor. This term decreases with increasing temperature. The relaxation term has a maximum value at 500°C. The sum of the two terms exhibits both a minimum and a maximum and fits with the experimental data. It would be interesting to carry out a combined ultrasonic and ESR experiment in order to investigate whether ultrasonic absorption would influence the ESR spectrum. (J.A. POULIS, J.-P. FRANFOIS, C.H MASSEN, L.C VAN POUCKE)

1. E. M. Ring, R. T. Beyer, J. Acoust. Soc. Am., 62, 582 (1977).

15.2.2.2.3. Quenching Experiments.

The weight fraction of the polymers in liquid sulfur are determined by rapidly chilling droplets of the liquid's'. After hardening the sample is extracted with solvent. The insoluble fraction is the polymer. The experimental data are shown in Fig. 1. At high temperatures there is an uncertainty about the rate of cooling in the first part of the

Figure 1. The weight fraction Q, of polymer as a function of temperature for liquid sulfur. The line is calculated from theory with parameter values as noted in the text. M, data from quenching data from quenching experiment^^,^. +, some of the data from vibrational experiments1. 0 , spectroscopy5.

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc. 15.2.2. lnterconversion of Sulfur or Selenium Rings and Chains 15.2.2.2.Experimental Techniques 15.2.2.2.3. Quenching Experiments. 4. 5. 6. 7.

47

H. Krebs, Z . Anorg. Allg. Chem., 263, 305 (1950). H. Krebs, Z . Anorg. Allg. Chem., 265, 156 (1951). D. E. Harrison, J. Chem. Phys., 41, 844 (1964). R. C. Keezer, M. W. Bailey, Mat. Res. Bull., 2, 185 (1967).

15.2.2.2.2. Ultrasonic Absorption in Liquid Selenium.

The dependency on temperature of the ultrasonic absorption' can be understood by assuming that only the polymers undergo relaxation in the frequency interval used (30-70 MHz), probably owing to the break up of the polymer chain. Further, it is assumed that the rings contribute as a non-relaxing absorption floor. This term decreases with increasing temperature. The relaxation term has a maximum value at 500°C. The sum of the two terms exhibits both a minimum and a maximum and fits with the experimental data. It would be interesting to carry out a combined ultrasonic and ESR experiment in order to investigate whether ultrasonic absorption would influence the ESR spectrum. (J.A. POULIS, J.-P. FRANFOIS, C.H MASSEN, L.C VAN POUCKE)

1. E. M. Ring, R. T. Beyer, J. Acoust. Soc. Am., 62, 582 (1977).

15.2.2.2.3. Quenching Experiments.

The weight fraction of the polymers in liquid sulfur are determined by rapidly chilling droplets of the liquid's'. After hardening the sample is extracted with solvent. The insoluble fraction is the polymer. The experimental data are shown in Fig. 1. At high temperatures there is an uncertainty about the rate of cooling in the first part of the

Figure 1. The weight fraction Q, of polymer as a function of temperature for liquid sulfur. The line is calculated from theory with parameter values as noted in the text. M, data from quenching data from quenching experiment^^,^. +, some of the data from vibrational experiments1. 0 , spectroscopy5.

15.2. Ring-Ring and Ring-Polymer lnterconversions 15.2.2. lnterconversion of Sulfur or Selenium Rings and Chains 15.2.2.2. Experimental Techniques

48

$0

200

300

400

500

600

7b0

-

800 T ( T

Figure 2. The weight fraction @ of polymer as a function of temperature for liquid selenium. The line is calculated from theory with parameter values as noted in the text. D, data from quenching experiments3.

quenching process. The quenching experiments for selenium' are shown together with theoretical results in Fig. 2. (J.A. POULIS, J.-P. FRANGOIS, C.H MASSEN, L.C VAN POUCKE)

1. D. L. Hammick, W. Cousins, E. Langford, J. Chem. SOC.,797 (1928). 2. A. H. W. Aten, Z. Phys. Chem., 86, l(1914). 3. G. Briegleb, 2. Phys. Chem., A144 321 (1929). 4. R. E. Harris, J . Phys. Chem., 74 3102 (1970). 5. A. T. Ward, M. B. Meyers, J . Phys. Chern., 73, 1374 (1969). 15.2.2.2.4. Diamagnetic Susceptibility Measurements.

A physical quantity that can be calculated for a mixture by supposing linearity with composition is the diamagnetic s~sceptibilityl-~, which therefore belongs to the second category ($15.2.2.2).After adaption of the value of the parameter xchain- xring,the results fit with theoretical predictions between 159 and 300°C. (J A. POULIS, J.-P. FRANCOIS, C H MASSEN, L C. VAN POUCKE)

1. J. A. Poulis, W. Derbyshire, Trans. Faraday SOC.,59, 559 (1963). 2. C . H. Massen, A. G. L. M. Weyts, J. A. Poulis, Trans. Faraday SOC.,60, 317 (1964). 3. J. A. Gardner, M. Cutler, Phys. Rev., B 20, 529 (1979).

15.2.2.2.5. Paramagnetic Susceptibility Measurements.

Above 300°C an increase in susceptibility occurs, which is caused by the free-radical ends of the polymer chains. The paramagnetic contribution to these centers is given by: @N,4 S(S + l)p2 32P3kT

(a)

where N, is Avogadro's number, p the Bohr magneton, k Boltzmann's constant and S the spin quantum number of a free radical. q!~ and P are the weight fraction and the mean

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc. 15.2. Ring-Ring and Ring-Polymer lnterconversions 15.2.2. lnterconversion of Sulfur or Selenium Rings and Chains 15.2.2.2. Experimental Techniques

48

$0

200

300

400

500

600

7b0

-

800 T ( T

Figure 2. The weight fraction @ of polymer as a function of temperature for liquid selenium. The line is calculated from theory with parameter values as noted in the text. D, data from quenching experiments3.

quenching process. The quenching experiments for selenium' are shown together with theoretical results in Fig. 2. (J.A. POULIS, J.-P. FRANGOIS, C.H MASSEN, L.C VAN POUCKE)

1. D. L. Hammick, W. Cousins, E. Langford, J. Chem. SOC.,797 (1928). 2. A. H. W. Aten, Z. Phys. Chem., 86, l(1914). 3. G. Briegleb, 2. Phys. Chem., A144 321 (1929). 4. R. E. Harris, J . Phys. Chem., 74 3102 (1970). 5. A. T. Ward, M. B. Meyers, J . Phys. Chern., 73, 1374 (1969). 15.2.2.2.4. Diamagnetic Susceptibility Measurements.

A physical quantity that can be calculated for a mixture by supposing linearity with composition is the diamagnetic s~sceptibilityl-~, which therefore belongs to the second category ($15.2.2.2).After adaption of the value of the parameter xchain- xring,the results fit with theoretical predictions between 159 and 300°C. (J A. POULIS, J.-P. FRANCOIS, C H MASSEN, L C. VAN POUCKE)

1. J. A. Poulis, W. Derbyshire, Trans. Faraday SOC.,59, 559 (1963). 2. C . H. Massen, A. G. L. M. Weyts, J. A. Poulis, Trans. Faraday SOC.,60, 317 (1964). 3. J. A. Gardner, M. Cutler, Phys. Rev., B 20, 529 (1979).

15.2.2.2.5. Paramagnetic Susceptibility Measurements.

Above 300°C an increase in susceptibility occurs, which is caused by the free-radical ends of the polymer chains. The paramagnetic contribution to these centers is given by: @N,4 S(S + l)p2 32P3kT

(a)

where N, is Avogadro's number, p the Bohr magneton, k Boltzmann's constant and S the spin quantum number of a free radical. q!~ and P are the weight fraction and the mean

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc. 15.2. Ring-Ring and Ring-Polymer lnterconversions 15.2.2. lnterconversion of Sulfur or Selenium Rings and Chains 15.2.2.2. Experimental Techniques

48

$0

200

300

400

500

600

7b0

-

800 T ( T

Figure 2. The weight fraction @ of polymer as a function of temperature for liquid selenium. The line is calculated from theory with parameter values as noted in the text. D, data from quenching experiments3.

quenching process. The quenching experiments for selenium' are shown together with theoretical results in Fig. 2. (J.A. POULIS, J.-P. FRANGOIS, C.H MASSEN, L.C VAN POUCKE)

1. D. L. Hammick, W. Cousins, E. Langford, J. Chem. SOC.,797 (1928). 2. A. H. W. Aten, Z. Phys. Chem., 86, l(1914). 3. G. Briegleb, 2. Phys. Chem., A144 321 (1929). 4. R. E. Harris, J . Phys. Chem., 74 3102 (1970). 5. A. T. Ward, M. B. Meyers, J . Phys. Chern., 73, 1374 (1969). 15.2.2.2.4. Diamagnetic Susceptibility Measurements.

A physical quantity that can be calculated for a mixture by supposing linearity with composition is the diamagnetic s~sceptibilityl-~, which therefore belongs to the second category ($15.2.2.2).After adaption of the value of the parameter xchain- xring,the results fit with theoretical predictions between 159 and 300°C. (J A. POULIS, J.-P. FRANCOIS, C H MASSEN, L C. VAN POUCKE)

1. J. A. Poulis, W. Derbyshire, Trans. Faraday SOC.,59, 559 (1963). 2. C . H. Massen, A. G. L. M. Weyts, J. A. Poulis, Trans. Faraday SOC.,60, 317 (1964). 3. J. A. Gardner, M. Cutler, Phys. Rev., B 20, 529 (1979).

15.2.2.2.5. Paramagnetic Susceptibility Measurements.

Above 300°C an increase in susceptibility occurs, which is caused by the free-radical ends of the polymer chains. The paramagnetic contribution to these centers is given by: @N,4 S(S + l)p2 32P3kT

(a)

where N, is Avogadro's number, p the Bohr magneton, k Boltzmann's constant and S the spin quantum number of a free radical. q!~ and P are the weight fraction and the mean

15.2.2. Interconversion of Sulfur or Selenium Rings and Chains 15.2.2.2. Experimental Techniques 15.2.2.2.5. Paramagnetic Susceptibility Measurements.

150

200

300

250

450

400

350

500

550

600

650

49

700T(OC)

Figure 1. The number N of moles polymer per 1000 g liquid sulfur. The line is calculated from theory with parameter values as noted in the text. 0, data from paramagnetic susceptibility measurements5. 0 , data from ESR measurements6. +, data from ESR measurements6.

chain length of the polymers, respectively, which are defined in 415.2.2.1. The paramagnetic contribution can be used as a direct measurement of the quantity N, the total concentration of polymer'chains, in mol kg-'. Since no parameters have to be adapted, paramagnetic susceptibility measurements belong to the first category in $15.2.2.2. Some results for liquid sulfur',' are shown in Fig. 1. The same procedure has been employed for liquid selenium3 (Fig. 2). (J.A POULIS, J -P. FRANGOIS, C.H. MASSEN, L C. VAN POUCKE)

Nl t

I 10-51 100

0 I

200

I

300

I

400

I

500

I

600

I

700

I

800

I

900

I

1000 llOOT(oC)

Figure 2. The number N of moles polymer per 1000 g liquid selenium. The line is calculated from theory with parameter values as noted in the text. f,data from paramagnetic susceptibility measurements3. l 7 , data from paramagnetic susceptibility measurements4. 0, data from ESR measurements6.

50

1. 2. 3. 4. 5. 6. 7.

15.2. Ring-Ring and Ring-Polymer Interconversions 15.2.2. Interconversion of Sulfur or Selenium Rings and Chains 15.2.2.2. Experimental Techniques

J. A. Poulis, C. H. Massen, P. van der Leeden, Trans. Faruday SOC.,58, 474 (1962). J. A. Poulis, W. Derbyshire, Trans. Furaday Soc., 59, 559 (1963). C. H. Massen, A. G. L. M. Weyts, J. A. Poulis, Trans. Faraday SOC.,60, 317 (1964). W. Freyland, M. Cutler, J. Chem. SOC.Furaday II, 76, 756 (1980). J. A. Poulis, C. H. Massen, P. van den Leeden, Trans. Faraday SOC., 58,474 (1962). D. C. Konigsberger, Ph.D. Thesis, University of Technology, Eindoven. D. M. Gardner, G. K. Fraenkel, J . Am. Chem. SOC.,76, 5891 (1954); 78, 3279 (1956).

15.2.2.2.6. Electron Spin Resonance Absorption (ESR).

An ESR e ~ p e r i m e n t l -is~a direct measurement of the quantity N. Just as in the case of the paramagnetic susceptibility measurements in 815.2.2.2.5,no parameters have to be adapted. The number of spins (twice the number of chain molecules) can be calculated from the surface area of the resonance peak. The measurements call for a careful approach insofar as the quotient of this surface area and the number of spins has to be calibrated using a reference sample with a known number of spins. Apart from this, the ESR measured linewidth could give information about the reaction rate of

Information concerning reaction rates can be of interest in explaining the observed time needed to attain equilibrium in the T range between the mp and T,. The results of the magnetic susceptibility and the ESR measurements are compared to theory in Figs. 1 and 2 in 815.2.2.2.5for sulfur and selenium, respectively. (J.A POULIS, J.-P FRANFOIS, C H. MASSEN, L.C. VAN

POUCKE)

1. D. M. Gardner, G. K. Fraenkel, J. Am. Chem. SOC., 76, 5891 (1954); 78, 3279 (1956). 2. D. C. Koningsberger, T. De Neef, Chem. Phys. Lett., 4, 615 (1970). 3. D. C. Koningsberger, J. H. M. C. Van Wolput, P. C. U. Rieter, Chem. Phys. Lett. 8, 145 (1971). 4. D. C . Koningsberger, Ph. D. Thesis, University of Technology, Eindhoven, 1971.

15.2.2.2.7. Optical Absorption.

It has been supposed in the past that the shift of the absorption edge with temperature is not influenced by the polymerization above 159°C'. The spurious shift is attributed to a weak absorption band in the energy range between 1 and 1.5 eV. The overlapping of the edge and this band is deconvoluted and the surface area under the band taken as proportional to the molar concentration, N, of the chains. The results fit well with the theoretical values for N after a parameter is adapted. In the immediate vicinity of T,, the optical absorbance is partly caused by light scattering'. This scattering, comparable with critical point opalescence, is due to composition fluctuations generated at the onset of the ring-chain transitions, which are accompanied by dielectric constant fluctuations. (J.A POULIS, J.-P. FRANGOIS, C H MASSEN, L.C. VAN POUCKE)

1. G. Weser, F. Hensel, W. W. Warren Jr., Ber. Bunsenges. Phys. Chem., 82, 588 (1978).

2. M. Zanini, J. Taw, Appl. Opt., 15, 3149 (1976).

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc. 50

1. 2. 3. 4. 5. 6. 7.

15.2. Ring-Ring and Ring-Polymer Interconversions 15.2.2. Interconversion of Sulfur or Selenium Rings and Chains 15.2.2.2. Experimental Techniques

J. A. Poulis, C. H. Massen, P. van der Leeden, Trans. Faruday SOC.,58, 474 (1962). J. A. Poulis, W. Derbyshire, Trans. Furaday Soc., 59, 559 (1963). C. H. Massen, A. G. L. M. Weyts, J. A. Poulis, Trans. Faraday SOC.,60, 317 (1964). W. Freyland, M. Cutler, J. Chem. SOC.Furaday II, 76, 756 (1980). J. A. Poulis, C. H. Massen, P. van den Leeden, Trans. Faraday SOC., 58,474 (1962). D. C. Konigsberger, Ph.D. Thesis, University of Technology, Eindoven. D. M. Gardner, G. K. Fraenkel, J . Am. Chem. SOC.,76, 5891 (1954); 78, 3279 (1956).

15.2.2.2.6. Electron Spin Resonance Absorption (ESR).

An ESR e ~ p e r i m e n t l -is~a direct measurement of the quantity N. Just as in the case of the paramagnetic susceptibility measurements in 815.2.2.2.5,no parameters have to be adapted. The number of spins (twice the number of chain molecules) can be calculated from the surface area of the resonance peak. The measurements call for a careful approach insofar as the quotient of this surface area and the number of spins has to be calibrated using a reference sample with a known number of spins. Apart from this, the ESR measured linewidth could give information about the reaction rate of

Information concerning reaction rates can be of interest in explaining the observed time needed to attain equilibrium in the T range between the mp and T,. The results of the magnetic susceptibility and the ESR measurements are compared to theory in Figs. 1 and 2 in 815.2.2.2.5for sulfur and selenium, respectively. (J.A POULIS, J.-P FRANFOIS, C H. MASSEN, L.C. VAN

POUCKE)

1. D. M. Gardner, G. K. Fraenkel, J. Am. Chem. SOC., 76, 5891 (1954); 78, 3279 (1956). 2. D. C. Koningsberger, T. De Neef, Chem. Phys. Lett., 4, 615 (1970). 3. D. C. Koningsberger, J. H. M. C. Van Wolput, P. C. U. Rieter, Chem. Phys. Lett. 8, 145 (1971). 4. D. C . Koningsberger, Ph. D. Thesis, University of Technology, Eindhoven, 1971.

15.2.2.2.7. Optical Absorption.

It has been supposed in the past that the shift of the absorption edge with temperature is not influenced by the polymerization above 159°C'. The spurious shift is attributed to a weak absorption band in the energy range between 1 and 1.5 eV. The overlapping of the edge and this band is deconvoluted and the surface area under the band taken as proportional to the molar concentration, N, of the chains. The results fit well with the theoretical values for N after a parameter is adapted. In the immediate vicinity of T,, the optical absorbance is partly caused by light scattering'. This scattering, comparable with critical point opalescence, is due to composition fluctuations generated at the onset of the ring-chain transitions, which are accompanied by dielectric constant fluctuations. (J.A POULIS, J.-P. FRANGOIS, C H MASSEN, L.C. VAN POUCKE)

1. G. Weser, F. Hensel, W. W. Warren Jr., Ber. Bunsenges. Phys. Chem., 82, 588 (1978).

2. M. Zanini, J. Taw, Appl. Opt., 15, 3149 (1976).

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc. 50

1. 2. 3. 4. 5. 6. 7.

15.2. Ring-Ring and Ring-Polymer Interconversions 15.2.2. Interconversion of Sulfur or Selenium Rings and Chains 15.2.2.2. Experimental Techniques

J. A. Poulis, C. H. Massen, P. van der Leeden, Trans. Faruday SOC.,58, 474 (1962). J. A. Poulis, W. Derbyshire, Trans. Furaday Soc., 59, 559 (1963). C. H. Massen, A. G. L. M. Weyts, J. A. Poulis, Trans. Faraday SOC.,60, 317 (1964). W. Freyland, M. Cutler, J. Chem. SOC.Furaday II, 76, 756 (1980). J. A. Poulis, C. H. Massen, P. van den Leeden, Trans. Faraday SOC., 58,474 (1962). D. C. Konigsberger, Ph.D. Thesis, University of Technology, Eindoven. D. M. Gardner, G. K. Fraenkel, J . Am. Chem. SOC.,76, 5891 (1954); 78, 3279 (1956).

15.2.2.2.6. Electron Spin Resonance Absorption (ESR).

An ESR e ~ p e r i m e n t l -is~a direct measurement of the quantity N. Just as in the case of the paramagnetic susceptibility measurements in 815.2.2.2.5,no parameters have to be adapted. The number of spins (twice the number of chain molecules) can be calculated from the surface area of the resonance peak. The measurements call for a careful approach insofar as the quotient of this surface area and the number of spins has to be calibrated using a reference sample with a known number of spins. Apart from this, the ESR measured linewidth could give information about the reaction rate of

Information concerning reaction rates can be of interest in explaining the observed time needed to attain equilibrium in the T range between the mp and T,. The results of the magnetic susceptibility and the ESR measurements are compared to theory in Figs. 1 and 2 in 815.2.2.2.5for sulfur and selenium, respectively. (J.A POULIS, J.-P FRANFOIS, C H. MASSEN, L.C. VAN

POUCKE)

1. D. M. Gardner, G. K. Fraenkel, J. Am. Chem. SOC., 76, 5891 (1954); 78, 3279 (1956). 2. D. C. Koningsberger, T. De Neef, Chem. Phys. Lett., 4, 615 (1970). 3. D. C. Koningsberger, J. H. M. C. Van Wolput, P. C. U. Rieter, Chem. Phys. Lett. 8, 145 (1971). 4. D. C . Koningsberger, Ph. D. Thesis, University of Technology, Eindhoven, 1971.

15.2.2.2.7. Optical Absorption.

It has been supposed in the past that the shift of the absorption edge with temperature is not influenced by the polymerization above 159°C'. The spurious shift is attributed to a weak absorption band in the energy range between 1 and 1.5 eV. The overlapping of the edge and this band is deconvoluted and the surface area under the band taken as proportional to the molar concentration, N, of the chains. The results fit well with the theoretical values for N after a parameter is adapted. In the immediate vicinity of T,, the optical absorbance is partly caused by light scattering'. This scattering, comparable with critical point opalescence, is due to composition fluctuations generated at the onset of the ring-chain transitions, which are accompanied by dielectric constant fluctuations. (J.A POULIS, J.-P. FRANGOIS, C H MASSEN, L.C. VAN POUCKE)

1. G. Weser, F. Hensel, W. W. Warren Jr., Ber. Bunsenges. Phys. Chem., 82, 588 (1978).

2. M. Zanini, J. Taw, Appl. Opt., 15, 3149 (1976).

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc. 15.2.2. lnterconversion of Sulfur or Selenium Rings and Chains 15.2.2.2. Experimental Techniques 15.2.2.2.10. Dielectric Constant Measurements of Sulfur.

51

15.2.2.2.8. Vibrational Spectroscopy of Sulfur.

The extent of polymerization can be measured by comparing the relative intensities of the contribution to the vibrational spectrum of rings and chains'. Assuming the vibrational scattering efficiencies to be equal for rings and chains and both to be independent of temperature, the spectrum allows the direct determination of the quantity @, without adapting a parameter. So far, the uncertainty, A@, is not better than 0.1. The fact that the vibrational spectroscopy results are greater than those from quenching experiments may indicate inefficient quenching'. (J.A. POULIS, J.-P. FRANGOIS, C H. MASSEN, L.C. VAN POUCKE)

1. A. T. Ward, M. B. Myers, J . Phys. Chem., 73, 1374 (1969). 2. G. Gee, Trans. Faraday SOC.,48, 515 (1952).

15.2.2.2.9. Nuclear Magnetic Resonance (NMR) Measurements on Selenium.

For NMR measurements 77Se,which has a natural abundance of 7.5 at %; can be used. Working with material enriched to 94.38 at %, part of the measured shift can be attributed to a difference in chemical shift between ring and chain molecules'. The remaining part is temperature dependent and attributed to the magnetic influence of the paramagnetic chain ends on the nuclei. In the relation between this shift and the number of electron spins causing it, the hyperfine coupling constant appears as a proportional constant. This constant can be determined by comparing the NMR results with paramagnetic susceptibility data (see 515.2.2.2.5).In conclusion, the NMR measurements give a determination of the quantity N after adaption of a constant. Once this adaption is realized, the method can be used to study of the pressure dependence of the equilibrium. (J A POULIS, J.-P. FRANCOIS, C.H. MASSEN, L C. VAN POUCKE)

1. W. W. Warren Jr., R. Dupree, Phys. Rev., B 22, 2257 (1980).

15.2.2.2.10. Dielectric Constant Measurements of Sulfur.

Unlike the magnetic susceptibility, the dielectric constant of a mixture cannot be calculated directly by taking a linear combination of the dielectric constants of the components. Instead, the relationship given in Eq. (a) indicates how to work with the molar polarization, which is a linear combination of the polarizabilities of the components. For a mixture of species i:

where P is the molar polarization, E the dielectric constant, d the density, N, Avogadro's number, Mi the concentration of species i and aiand pi the polarizability and permanent dipole moment, respectively of species i. Below T,, the molar polarization is temperature dependent', because of a second, chain form of the S, rings; the common form of the S, rings is the crown form. This explanation is based on the fact that the reaction rates between the different sulfur species may be smaller below T, than above, since polymers may facilitate ring conversions by acting as intermediate products. When heating is not performed slowly during the E measurements, the composition of the rings below T,

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc. 15.2.2. lnterconversion of Sulfur or Selenium Rings and Chains 15.2.2.2. Experimental Techniques 15.2.2.2.10. Dielectric Constant Measurements of Sulfur.

51

15.2.2.2.8. Vibrational Spectroscopy of Sulfur.

The extent of polymerization can be measured by comparing the relative intensities of the contribution to the vibrational spectrum of rings and chains'. Assuming the vibrational scattering efficiencies to be equal for rings and chains and both to be independent of temperature, the spectrum allows the direct determination of the quantity @, without adapting a parameter. So far, the uncertainty, A@, is not better than 0.1. The fact that the vibrational spectroscopy results are greater than those from quenching experiments may indicate inefficient quenching'. (J.A. POULIS, J.-P. FRANGOIS, C H. MASSEN, L.C. VAN POUCKE)

1. A. T. Ward, M. B. Myers, J . Phys. Chem., 73, 1374 (1969). 2. G. Gee, Trans. Faraday SOC.,48, 515 (1952).

15.2.2.2.9. Nuclear Magnetic Resonance (NMR) Measurements on Selenium.

For NMR measurements 77Se,which has a natural abundance of 7.5 at %; can be used. Working with material enriched to 94.38 at %, part of the measured shift can be attributed to a difference in chemical shift between ring and chain molecules'. The remaining part is temperature dependent and attributed to the magnetic influence of the paramagnetic chain ends on the nuclei. In the relation between this shift and the number of electron spins causing it, the hyperfine coupling constant appears as a proportional constant. This constant can be determined by comparing the NMR results with paramagnetic susceptibility data (see 515.2.2.2.5).In conclusion, the NMR measurements give a determination of the quantity N after adaption of a constant. Once this adaption is realized, the method can be used to study of the pressure dependence of the equilibrium. (J A POULIS, J.-P. FRANCOIS, C.H. MASSEN, L C. VAN POUCKE)

1. W. W. Warren Jr., R. Dupree, Phys. Rev., B 22, 2257 (1980).

15.2.2.2.10. Dielectric Constant Measurements of Sulfur.

Unlike the magnetic susceptibility, the dielectric constant of a mixture cannot be calculated directly by taking a linear combination of the dielectric constants of the components. Instead, the relationship given in Eq. (a) indicates how to work with the molar polarization, which is a linear combination of the polarizabilities of the components. For a mixture of species i:

where P is the molar polarization, E the dielectric constant, d the density, N, Avogadro's number, Mi the concentration of species i and aiand pi the polarizability and permanent dipole moment, respectively of species i. Below T,, the molar polarization is temperature dependent', because of a second, chain form of the S, rings; the common form of the S, rings is the crown form. This explanation is based on the fact that the reaction rates between the different sulfur species may be smaller below T, than above, since polymers may facilitate ring conversions by acting as intermediate products. When heating is not performed slowly during the E measurements, the composition of the rings below T,

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc. 15.2.2. lnterconversion of Sulfur or Selenium Rings and Chains 15.2.2.2. Experimental Techniques 15.2.2.2.10. Dielectric Constant Measurements of Sulfur.

51

15.2.2.2.8. Vibrational Spectroscopy of Sulfur.

The extent of polymerization can be measured by comparing the relative intensities of the contribution to the vibrational spectrum of rings and chains'. Assuming the vibrational scattering efficiencies to be equal for rings and chains and both to be independent of temperature, the spectrum allows the direct determination of the quantity @, without adapting a parameter. So far, the uncertainty, A@, is not better than 0.1. The fact that the vibrational spectroscopy results are greater than those from quenching experiments may indicate inefficient quenching'. (J.A. POULIS, J.-P. FRANGOIS, C H. MASSEN, L.C. VAN POUCKE)

1. A. T. Ward, M. B. Myers, J . Phys. Chem., 73, 1374 (1969). 2. G. Gee, Trans. Faraday SOC.,48, 515 (1952).

15.2.2.2.9. Nuclear Magnetic Resonance (NMR) Measurements on Selenium.

For NMR measurements 77Se,which has a natural abundance of 7.5 at %; can be used. Working with material enriched to 94.38 at %, part of the measured shift can be attributed to a difference in chemical shift between ring and chain molecules'. The remaining part is temperature dependent and attributed to the magnetic influence of the paramagnetic chain ends on the nuclei. In the relation between this shift and the number of electron spins causing it, the hyperfine coupling constant appears as a proportional constant. This constant can be determined by comparing the NMR results with paramagnetic susceptibility data (see 515.2.2.2.5).In conclusion, the NMR measurements give a determination of the quantity N after adaption of a constant. Once this adaption is realized, the method can be used to study of the pressure dependence of the equilibrium. (J A POULIS, J.-P. FRANCOIS, C.H. MASSEN, L C. VAN POUCKE)

1. W. W. Warren Jr., R. Dupree, Phys. Rev., B 22, 2257 (1980).

15.2.2.2.10. Dielectric Constant Measurements of Sulfur.

Unlike the magnetic susceptibility, the dielectric constant of a mixture cannot be calculated directly by taking a linear combination of the dielectric constants of the components. Instead, the relationship given in Eq. (a) indicates how to work with the molar polarization, which is a linear combination of the polarizabilities of the components. For a mixture of species i:

where P is the molar polarization, E the dielectric constant, d the density, N, Avogadro's number, Mi the concentration of species i and aiand pi the polarizability and permanent dipole moment, respectively of species i. Below T,, the molar polarization is temperature dependent', because of a second, chain form of the S, rings; the common form of the S, rings is the crown form. This explanation is based on the fact that the reaction rates between the different sulfur species may be smaller below T, than above, since polymers may facilitate ring conversions by acting as intermediate products. When heating is not performed slowly during the E measurements, the composition of the rings below T,

52

15.2. Ring-Ring and Ring-Polymer Interconversions 15.2.2.Interconversion of Sulfur or Selenium Rings and Chains 15.2 2.2. Experimental Techniques

differ, from the equilibrium composition. At or above T,, the composition of the different rings suddenly attains equilibrium value and, therefore, produces a spurious discontinuity in E. When adapting a parameter, the dielectric constant measurements could, above T,, be considered direct measurements of @. As the value of the polarizability of the polymer does not follow from any other experiment, the method is considered to belong to the second category in 515.2.2.2. (J.A. POULIS, J.-P. FRANGOIS, C.H. MASSEN, L.C. VAN POUCKE)

1. M. E. Bauer, D. A. Horsma, J. Phys. Chem., 78, 1670 (1974). 15.2.2.2.11. Specific Heat Measurements on Sulfur and Selenium.

The position of the S, - S, transition in a differential thermal analysis (DTA) experiment with liquid sulfur' may be used to estimate reaction rates after corrections for delay caused by thermoconductivity. At and above T,, DTA measurements show a peak. This has been used to determine the dependence of T, on pressure. The temperature of the DTA peak is measured at each pressure, but the results are not in agreement with theoretical predictions'. This discrepancy is ascribed to a difference in the molar volumes of rings and chains. The difference becomes much smaller with increasing pressure. The DTA curves3 show no evidence for a T, mechanism in the case of liquid selenium, The specific heat at constant pressure, C,, vs. T curve of liquid selenium shows a minimum near 500°C4 s 5 . The increase after this minimum is ascribed to depolymerization, which is endothermic. However, care must be exercized in determining the polymer-ring equilibrium mechanism from the heat capacity measurements. The following contribute to the specific heat6: 1. The variation of @ with temperature 2. The variation of P with temperature 3. The vibrational heat capacity, which rises linearly with T and plays, therefore, an important part in the position of the minimum of the measured c, values 4. The normal free volume increase When these four contribution combine, a discrepancy with the experimental values of ca. 10% is observed that arises from an error in the commonly accepted values of the ring-chain equilibrium. A more careful estimate of the uncertainties involved when using data in the theoretical expressions would be useful. (J.A. POULIS, J.-P. FRANGOIS, C.H. MASSEN, L.C. VAN POUCKE)

1. 2. 3. 4. 5.

M. Kuballa, G. M. Schneider, Ber. Bunsenges. Phys. Chem., 75, 513 (1971). A. Eisenberg, J. Chem. Phys., 39, 1852 (1963). A. Datta, V. Krishnan, J. Thermal Anal., 17, 31 (1979). S. S. Chang, A. B. Bestul, J. Chem. Thermodyn.,6, 325 (1974). S. Hua-Cheng, U. Gaur, B. Wunderlich, J. Polym. Sci., Polym. Phys. Ed., 18, 449 (1980).

15.2.2.2.12. Density Measurements.

The law of corresponding states can be applied to the measured values of the density of liquid sulfur and selenium as a function of temperature and pressure in the vicinity of

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc.

52

15.2. Ring-Ring and Ring-Polymer Interconversions 15.2.2.Interconversion of Sulfur or Selenium Rings and Chains 15.2 2.2. Experimental Techniques

differ, from the equilibrium composition. At or above T,, the composition of the different rings suddenly attains equilibrium value and, therefore, produces a spurious discontinuity in E. When adapting a parameter, the dielectric constant measurements could, above T,, be considered direct measurements of @. As the value of the polarizability of the polymer does not follow from any other experiment, the method is considered to belong to the second category in 515.2.2.2. (J.A. POULIS, J.-P. FRANGOIS, C.H. MASSEN, L.C. VAN POUCKE)

1. M. E. Bauer, D. A. Horsma, J. Phys. Chem., 78, 1670 (1974). 15.2.2.2.11. Specific Heat Measurements on Sulfur and Selenium.

The position of the S, - S, transition in a differential thermal analysis (DTA) experiment with liquid sulfur' may be used to estimate reaction rates after corrections for delay caused by thermoconductivity. At and above T,, DTA measurements show a peak. This has been used to determine the dependence of T, on pressure. The temperature of the DTA peak is measured at each pressure, but the results are not in agreement with theoretical predictions'. This discrepancy is ascribed to a difference in the molar volumes of rings and chains. The difference becomes much smaller with increasing pressure. The DTA curves3 show no evidence for a T, mechanism in the case of liquid selenium, The specific heat at constant pressure, C,, vs. T curve of liquid selenium shows a minimum near 500°C4 s 5 . The increase after this minimum is ascribed to depolymerization, which is endothermic. However, care must be exercized in determining the polymer-ring equilibrium mechanism from the heat capacity measurements. The following contribute to the specific heat6: 1. The variation of @ with temperature 2. The variation of P with temperature 3. The vibrational heat capacity, which rises linearly with T and plays, therefore, an important part in the position of the minimum of the measured c, values 4. The normal free volume increase When these four contribution combine, a discrepancy with the experimental values of ca. 10% is observed that arises from an error in the commonly accepted values of the ring-chain equilibrium. A more careful estimate of the uncertainties involved when using data in the theoretical expressions would be useful. (J.A. POULIS, J.-P. FRANGOIS, C.H. MASSEN, L.C. VAN POUCKE)

1. 2. 3. 4. 5.

M. Kuballa, G. M. Schneider, Ber. Bunsenges. Phys. Chem., 75, 513 (1971). A. Eisenberg, J. Chem. Phys., 39, 1852 (1963). A. Datta, V. Krishnan, J. Thermal Anal., 17, 31 (1979). S. S. Chang, A. B. Bestul, J. Chem. Thermodyn.,6, 325 (1974). S. Hua-Cheng, U. Gaur, B. Wunderlich, J. Polym. Sci., Polym. Phys. Ed., 18, 449 (1980).

15.2.2.2.12. Density Measurements.

The law of corresponding states can be applied to the measured values of the density of liquid sulfur and selenium as a function of temperature and pressure in the vicinity of

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc.

52

15.2. Ring-Ring and Ring-Polymer Interconversions 15.2.2.Interconversion of Sulfur or Selenium Rings and Chains 15.2 2.2. Experimental Techniques

differ, from the equilibrium composition. At or above T,, the composition of the different rings suddenly attains equilibrium value and, therefore, produces a spurious discontinuity in E. When adapting a parameter, the dielectric constant measurements could, above T,, be considered direct measurements of @. As the value of the polarizability of the polymer does not follow from any other experiment, the method is considered to belong to the second category in 515.2.2.2. (J.A. POULIS, J.-P. FRANGOIS, C.H. MASSEN, L.C. VAN POUCKE)

1. M. E. Bauer, D. A. Horsma, J. Phys. Chem., 78, 1670 (1974). 15.2.2.2.11. Specific Heat Measurements on Sulfur and Selenium.

The position of the S, - S, transition in a differential thermal analysis (DTA) experiment with liquid sulfur' may be used to estimate reaction rates after corrections for delay caused by thermoconductivity. At and above T,, DTA measurements show a peak. This has been used to determine the dependence of T, on pressure. The temperature of the DTA peak is measured at each pressure, but the results are not in agreement with theoretical predictions'. This discrepancy is ascribed to a difference in the molar volumes of rings and chains. The difference becomes much smaller with increasing pressure. The DTA curves3 show no evidence for a T, mechanism in the case of liquid selenium, The specific heat at constant pressure, C,, vs. T curve of liquid selenium shows a minimum near 500°C4 s 5 . The increase after this minimum is ascribed to depolymerization, which is endothermic. However, care must be exercized in determining the polymer-ring equilibrium mechanism from the heat capacity measurements. The following contribute to the specific heat6: 1. The variation of @ with temperature 2. The variation of P with temperature 3. The vibrational heat capacity, which rises linearly with T and plays, therefore, an important part in the position of the minimum of the measured c, values 4. The normal free volume increase When these four contribution combine, a discrepancy with the experimental values of ca. 10% is observed that arises from an error in the commonly accepted values of the ring-chain equilibrium. A more careful estimate of the uncertainties involved when using data in the theoretical expressions would be useful. (J.A. POULIS, J.-P. FRANGOIS, C.H. MASSEN, L.C. VAN POUCKE)

1. 2. 3. 4. 5.

M. Kuballa, G. M. Schneider, Ber. Bunsenges. Phys. Chem., 75, 513 (1971). A. Eisenberg, J. Chem. Phys., 39, 1852 (1963). A. Datta, V. Krishnan, J. Thermal Anal., 17, 31 (1979). S. S. Chang, A. B. Bestul, J. Chem. Thermodyn.,6, 325 (1974). S. Hua-Cheng, U. Gaur, B. Wunderlich, J. Polym. Sci., Polym. Phys. Ed., 18, 449 (1980).

15.2.2.2.12. Density Measurements.

The law of corresponding states can be applied to the measured values of the density of liquid sulfur and selenium as a function of temperature and pressure in the vicinity of

15.2. Ring-Ring and Ring-Polymer Interconversions 15.2.3. In Oligomeric Catenates of P, As, Sb and Bi 15.2.3.1. General Trends.

53

the critical point'. From this treatment it follows that for sulfur the number of atoms per molecule is between 2 and 3. For selenium the isochores are more complicated. They show a kink at ca. 1000°C, which indicates a transition to a more dense liquid. This explanation is supported by the fact that the expansion coefficient shows a minimum. (J.A. POULIS, J.-P. FRANFOIS, C.H. MASSEN, L.C. VAN POUCKE)

1. R. Fischer, R. W. Schmutzler, F. Hensel, J. Non-Cryst. Solids, 35 & 36, 1295 (1980). 15.2.2.2.13. Thermal Conductivity in Liquid Selenium.

The rapid increase in the thermal conductivity of liquid selenium' with temperature can be attributed to the photon component of the thermal conductivity. For liquids with a small absorption coefficient, this radiation term should rise as a third power of the absolute temperature. From the results of the thermal conductivity data we can, therefore, get information about two optical parameters, i.e., the optical absorption coefficient, CI,and the refractive index, n, in the form a/n2. (J.A. POULIS, J.-P. FRANFOIS, C.H. MASSEN, L.C. VAN POUCKE)

1. A. H. Abou El Ela, H. H. A. Labib, K. A. H. Sharaf, Acta Phys. Acud. Sci. Hung., 47, 353 (1979). 15.2.2.2.14. Electrical Conductivity of Sulfur.

The electrical conductivity of sulfur may be attributed to the free chain ends, which give rise to carriers. Unfortunately, impurities also contribute to the number of carriers', especially monovalent impurities, which can bind the chain ends. As these impurities have higher activation energies than the free ends of the chains, the free-end electrical contribution dominate above a certain floor temperature. For a monovalent impurity of 1 ppm this bottom temperature is ca. 550°C. For an impurity concentration of 10 ppm, the bottom temperature will be above 800°C. The high-T part of the measurements shows an activation energy of 1.05 eV, which, after correction for the temperature dependency of N* (the number of chain ends), leads to an acceptor energy of 1.3 eV. This is in good agreement with the results of optical absorption measurements2. (J.A. POULIS, J.-P. FRANGOIS, C.H. MASSEN, L.C. VAN POUCKE)

1. M. Edeling, R. W. Schmutzler, F. Hensel, Philos, Mug., B 39, 547 (1979). 2. G . Weser, F. Hensel, W. W. Warren Jr., Ber. Bunsenges. Phys. Chem., 82, 588 (1978).

15.2.3. In Oligomeric Catenates of P, As, Sb and Bi 15.2.3.1. General Trends.

Ring-ring or ring-chain interconversions of poly-(RE), formulations (R = organic group, E = P, As or Sb)' resemble in their complexity the dynamics of the chalcogen allotropes, but with the added dimension of differing behavior with R-group modification. Further resemblance of group-VB and -VIB behavior is found in the difficulty in separating purely thermal phenomena from those influenced by trace-level impurities.

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc. 15.2. Ring-Ring and Ring-Polymer Interconversions 15.2.3. In Oligomeric Catenates of P, As, Sb and Bi 15.2.3.1. General Trends.

53

the critical point'. From this treatment it follows that for sulfur the number of atoms per molecule is between 2 and 3. For selenium the isochores are more complicated. They show a kink at ca. 1000°C, which indicates a transition to a more dense liquid. This explanation is supported by the fact that the expansion coefficient shows a minimum. (J.A. POULIS, J.-P. FRANFOIS, C.H. MASSEN, L.C. VAN POUCKE)

1. R. Fischer, R. W. Schmutzler, F. Hensel, J. Non-Cryst. Solids, 35 & 36, 1295 (1980). 15.2.2.2.13. Thermal Conductivity in Liquid Selenium.

The rapid increase in the thermal conductivity of liquid selenium' with temperature can be attributed to the photon component of the thermal conductivity. For liquids with a small absorption coefficient, this radiation term should rise as a third power of the absolute temperature. From the results of the thermal conductivity data we can, therefore, get information about two optical parameters, i.e., the optical absorption coefficient, CI,and the refractive index, n, in the form a/n2. (J.A. POULIS, J.-P. FRANFOIS, C.H. MASSEN, L.C. VAN POUCKE)

1. A. H. Abou El Ela, H. H. A. Labib, K. A. H. Sharaf, Acta Phys. Acud. Sci. Hung., 47, 353 (1979). 15.2.2.2.14. Electrical Conductivity of Sulfur.

The electrical conductivity of sulfur may be attributed to the free chain ends, which give rise to carriers. Unfortunately, impurities also contribute to the number of carriers', especially monovalent impurities, which can bind the chain ends. As these impurities have higher activation energies than the free ends of the chains, the free-end electrical contribution dominate above a certain floor temperature. For a monovalent impurity of 1 ppm this bottom temperature is ca. 550°C. For an impurity concentration of 10 ppm, the bottom temperature will be above 800°C. The high-T part of the measurements shows an activation energy of 1.05 eV, which, after correction for the temperature dependency of N* (the number of chain ends), leads to an acceptor energy of 1.3 eV. This is in good agreement with the results of optical absorption measurements2. (J.A. POULIS, J.-P. FRANGOIS, C.H. MASSEN, L.C. VAN POUCKE)

1. M. Edeling, R. W. Schmutzler, F. Hensel, Philos, Mug., B 39, 547 (1979). 2. G . Weser, F. Hensel, W. W. Warren Jr., Ber. Bunsenges. Phys. Chem., 82, 588 (1978).

15.2.3. In Oligomeric Catenates of P, As, Sb and Bi 15.2.3.1. General Trends.

Ring-ring or ring-chain interconversions of poly-(RE), formulations (R = organic group, E = P, As or Sb)' resemble in their complexity the dynamics of the chalcogen allotropes, but with the added dimension of differing behavior with R-group modification. Further resemblance of group-VB and -VIB behavior is found in the difficulty in separating purely thermal phenomena from those influenced by trace-level impurities.

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc. 15.2. Ring-Ring and Ring-Polymer Interconversions 15.2.3. In Oligomeric Catenates of P, As, Sb and Bi 15.2.3.1. General Trends.

53

the critical point'. From this treatment it follows that for sulfur the number of atoms per molecule is between 2 and 3. For selenium the isochores are more complicated. They show a kink at ca. 1000°C, which indicates a transition to a more dense liquid. This explanation is supported by the fact that the expansion coefficient shows a minimum. (J.A. POULIS, J.-P. FRANFOIS, C.H. MASSEN, L.C. VAN POUCKE)

1. R. Fischer, R. W. Schmutzler, F. Hensel, J. Non-Cryst. Solids, 35 & 36, 1295 (1980). 15.2.2.2.13. Thermal Conductivity in Liquid Selenium.

The rapid increase in the thermal conductivity of liquid selenium' with temperature can be attributed to the photon component of the thermal conductivity. For liquids with a small absorption coefficient, this radiation term should rise as a third power of the absolute temperature. From the results of the thermal conductivity data we can, therefore, get information about two optical parameters, i.e., the optical absorption coefficient, CI,and the refractive index, n, in the form a/n2. (J.A. POULIS, J.-P. FRANFOIS, C.H. MASSEN, L.C. VAN POUCKE)

1. A. H. Abou El Ela, H. H. A. Labib, K. A. H. Sharaf, Acta Phys. Acud. Sci. Hung., 47, 353 (1979). 15.2.2.2.14. Electrical Conductivity of Sulfur.

The electrical conductivity of sulfur may be attributed to the free chain ends, which give rise to carriers. Unfortunately, impurities also contribute to the number of carriers', especially monovalent impurities, which can bind the chain ends. As these impurities have higher activation energies than the free ends of the chains, the free-end electrical contribution dominate above a certain floor temperature. For a monovalent impurity of 1 ppm this bottom temperature is ca. 550°C. For an impurity concentration of 10 ppm, the bottom temperature will be above 800°C. The high-T part of the measurements shows an activation energy of 1.05 eV, which, after correction for the temperature dependency of N* (the number of chain ends), leads to an acceptor energy of 1.3 eV. This is in good agreement with the results of optical absorption measurements2. (J.A. POULIS, J.-P. FRANGOIS, C.H. MASSEN, L.C. VAN POUCKE)

1. M. Edeling, R. W. Schmutzler, F. Hensel, Philos, Mug., B 39, 547 (1979). 2. G . Weser, F. Hensel, W. W. Warren Jr., Ber. Bunsenges. Phys. Chem., 82, 588 (1978).

15.2.3. In Oligomeric Catenates of P, As, Sb and Bi 15.2.3.1. General Trends.

Ring-ring or ring-chain interconversions of poly-(RE), formulations (R = organic group, E = P, As or Sb)' resemble in their complexity the dynamics of the chalcogen allotropes, but with the added dimension of differing behavior with R-group modification. Further resemblance of group-VB and -VIB behavior is found in the difficulty in separating purely thermal phenomena from those influenced by trace-level impurities.

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc. 15.2. Ring-Ring and Ring-Polymer Interconversions 15.2.3. In Oligomeric Catenates of P, As, Sb and Bi 15.2.3.1. General Trends.

53

the critical point'. From this treatment it follows that for sulfur the number of atoms per molecule is between 2 and 3. For selenium the isochores are more complicated. They show a kink at ca. 1000°C, which indicates a transition to a more dense liquid. This explanation is supported by the fact that the expansion coefficient shows a minimum. (J.A. POULIS, J.-P. FRANFOIS, C.H. MASSEN, L.C. VAN POUCKE)

1. R. Fischer, R. W. Schmutzler, F. Hensel, J. Non-Cryst. Solids, 35 & 36, 1295 (1980). 15.2.2.2.13. Thermal Conductivity in Liquid Selenium.

The rapid increase in the thermal conductivity of liquid selenium' with temperature can be attributed to the photon component of the thermal conductivity. For liquids with a small absorption coefficient, this radiation term should rise as a third power of the absolute temperature. From the results of the thermal conductivity data we can, therefore, get information about two optical parameters, i.e., the optical absorption coefficient, CI,and the refractive index, n, in the form a/n2. (J.A. POULIS, J.-P. FRANFOIS, C.H. MASSEN, L.C. VAN POUCKE)

1. A. H. Abou El Ela, H. H. A. Labib, K. A. H. Sharaf, Acta Phys. Acud. Sci. Hung., 47, 353 (1979). 15.2.2.2.14. Electrical Conductivity of Sulfur.

The electrical conductivity of sulfur may be attributed to the free chain ends, which give rise to carriers. Unfortunately, impurities also contribute to the number of carriers', especially monovalent impurities, which can bind the chain ends. As these impurities have higher activation energies than the free ends of the chains, the free-end electrical contribution dominate above a certain floor temperature. For a monovalent impurity of 1 ppm this bottom temperature is ca. 550°C. For an impurity concentration of 10 ppm, the bottom temperature will be above 800°C. The high-T part of the measurements shows an activation energy of 1.05 eV, which, after correction for the temperature dependency of N* (the number of chain ends), leads to an acceptor energy of 1.3 eV. This is in good agreement with the results of optical absorption measurements2. (J.A. POULIS, J.-P. FRANGOIS, C.H. MASSEN, L.C. VAN POUCKE)

1. M. Edeling, R. W. Schmutzler, F. Hensel, Philos, Mug., B 39, 547 (1979). 2. G . Weser, F. Hensel, W. W. Warren Jr., Ber. Bunsenges. Phys. Chem., 82, 588 (1978).

15.2.3. In Oligomeric Catenates of P, As, Sb and Bi 15.2.3.1. General Trends.

Ring-ring or ring-chain interconversions of poly-(RE), formulations (R = organic group, E = P, As or Sb)' resemble in their complexity the dynamics of the chalcogen allotropes, but with the added dimension of differing behavior with R-group modification. Further resemblance of group-VB and -VIB behavior is found in the difficulty in separating purely thermal phenomena from those influenced by trace-level impurities.

54

15.2. Ring-Ring and Ring-Polymer Interconversions 15.2.3. In Oligomeric Catenates of P, As, Sb and Bi 15.2.3.2. Phosphorus.

Nonetheless, some trends based upon variation of R and E emerge. As the bulk of the alkyl group increases (particularly in the a positive relative to E), there is a greater tendency to form small rings, cycl~-(RE),,,~~, and less tendency to form catena structures. With aromatic R groups the trends are not nearly as clear, reflecting the better ability of a planar substituent to adopt a nondemanding orientation. As group VB is descended, the tendency to form catena in favor of cyclo modifications increases. Thus, few stable catena-(RP), but many cyclo-(RP), structures are known; both catena-(RAs), and cyclo-(RAs), structures are known and, with but a few unusual examples, only catena-(RSb), structures appear to form’. The decreasing E-E bond energies on descending the group leads to an increasing ease of rearrangement and an increasing sensitivity of systems to the effect of impurities. (A.L. RHEINGOLD)

1. Homoatomic bonding in Bi is not well developed and is known to occur in only a few not very stable dibismuth species. Tetraphenyldibismuth:F. Calderazzo, A. Morrillo, G. Pelizzi, R. Poli, J. Chem. SOC.,Chem. Commun., 507 (1983); tetraalkyldibismuth:H. J. Breunig, D. Muller, Angew. Chem., Int. Ed. Engl., 21, 439 (1982); H. J. Bruenig, D. Muller, Z . Naturforsch., Ted B, 38, 125 (1983); A. J. Ashe, 111, E. G. Ludwig, Jr., J. Oleksyszyn, Organometallics, 2, 1859 (1983): A. J. Ashe, I11 F. J. Drone, Organometallics, 3, 495 (1984); Dibismuth and tetrabismuth metal complexes: A. M. Arif, A. H. Cowley, N. C. Norman, M. Pakulski, J. Am. Chem. SOC.,107, 1062 (1985); K. H. Whitmire, K. S. Raghuveer, M. R. Churchill, J. C. Fettinger, R. F. See, J. Am. Chem. Soc., 108, 2778 (1986); K. H. Whitmire, T. A. Albright, S.-K. Kang, M. R. Churchill, J. C. Fettinger, Inorg. Chem., 25, 2799 (1986). 2. A. L. Rheingold, in Homoatomic Rings, Chains and Macromolecules of the Main-Group Elements, A. L. Rheingold, ed. Elsevier, Amsterdam, 1977, pp. 385-408.

15.2.3.2. Phosphorus.

Phosphorus is second only to1,’ carbon in the structural diversity of its homoatomic bonded forms’s’. In addition to monocyclic ring sizes from three to six, many different polycyclic structures have been identifiedi3. Phenyl-substituted cyclopolyphosphines, cyclo-(PhP),, are known for n = 3, 4, 5 and 6. There is increasing evidence that cyclo-(PhP), is the thermodynamically most stable form. Both cyclo-(PhP), and cyclo-(PhP), spontaneously rearrange to cyclo-(PhP), by a mechanism that may involve four-centered intermediate^,,^. For the cyclo-(PhP), to cyclo-(PhP), conversion the proposed scheme is: 2 P,

--P,

p6

+

p4

p3

P,

P,

2

PI0

P5

+ P5

(a) (b)

whereas for the cyclo-(PhP), to cyclo-(PhP), conversion:

P6

-P4

PI,

2 P,

( 4

The latter conversion occurs faster in THF than in aromatic hydrocarbons, suggesting that the proposed four-centered intermediate involves charge-separated stages of development stabilized by the more polar solvent6.

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc.

54

15.2. Ring-Ring and Ring-Polymer Interconversions 15.2.3. In Oligomeric Catenates of P, As, Sb and Bi 15.2.3.2. Phosphorus.

Nonetheless, some trends based upon variation of R and E emerge. As the bulk of the alkyl group increases (particularly in the a positive relative to E), there is a greater tendency to form small rings, cycl~-(RE),,,~~, and less tendency to form catena structures. With aromatic R groups the trends are not nearly as clear, reflecting the better ability of a planar substituent to adopt a nondemanding orientation. As group VB is descended, the tendency to form catena in favor of cyclo modifications increases. Thus, few stable catena-(RP), but many cyclo-(RP), structures are known; both catena-(RAs), and cyclo-(RAs), structures are known and, with but a few unusual examples, only catena-(RSb), structures appear to form’. The decreasing E-E bond energies on descending the group leads to an increasing ease of rearrangement and an increasing sensitivity of systems to the effect of impurities. (A.L. RHEINGOLD)

1. Homoatomic bonding in Bi is not well developed and is known to occur in only a few not very stable dibismuth species. Tetraphenyldibismuth:F. Calderazzo, A. Morrillo, G. Pelizzi, R. Poli, J. Chem. SOC.,Chem. Commun., 507 (1983); tetraalkyldibismuth:H. J. Breunig, D. Muller, Angew. Chem., Int. Ed. Engl., 21, 439 (1982); H. J. Bruenig, D. Muller, Z . Naturforsch., Ted B, 38, 125 (1983); A. J. Ashe, 111, E. G. Ludwig, Jr., J. Oleksyszyn, Organometallics, 2, 1859 (1983): A. J. Ashe, I11 F. J. Drone, Organometallics, 3, 495 (1984); Dibismuth and tetrabismuth metal complexes: A. M. Arif, A. H. Cowley, N. C. Norman, M. Pakulski, J. Am. Chem. SOC.,107, 1062 (1985); K. H. Whitmire, K. S. Raghuveer, M. R. Churchill, J. C. Fettinger, R. F. See, J. Am. Chem. Soc., 108, 2778 (1986); K. H. Whitmire, T. A. Albright, S.-K. Kang, M. R. Churchill, J. C. Fettinger, Inorg. Chem., 25, 2799 (1986). 2. A. L. Rheingold, in Homoatomic Rings, Chains and Macromolecules of the Main-Group Elements, A. L. Rheingold, ed. Elsevier, Amsterdam, 1977, pp. 385-408.

15.2.3.2. Phosphorus.

Phosphorus is second only to1,’ carbon in the structural diversity of its homoatomic bonded forms’s’. In addition to monocyclic ring sizes from three to six, many different polycyclic structures have been identifiedi3. Phenyl-substituted cyclopolyphosphines, cyclo-(PhP),, are known for n = 3, 4, 5 and 6. There is increasing evidence that cyclo-(PhP), is the thermodynamically most stable form. Both cyclo-(PhP), and cyclo-(PhP), spontaneously rearrange to cyclo-(PhP), by a mechanism that may involve four-centered intermediate^,,^. For the cyclo-(PhP), to cyclo-(PhP), conversion the proposed scheme is: 2 P,

--P,

p6

+

p4

p3

P,

P,

2

PI0

P5

+ P5

(a) (b)

whereas for the cyclo-(PhP), to cyclo-(PhP), conversion:

P6

-P4

PI,

2 P,

( 4

The latter conversion occurs faster in THF than in aromatic hydrocarbons, suggesting that the proposed four-centered intermediate involves charge-separated stages of development stabilized by the more polar solvent6.

55

15.2. Ring-Ring and Ring-Polymer lnterconversions 15.2.3. In Oligomeric Catenates of P, As, Sb and Bi 15.2.3.3. Arsenic.

As an illustration of the effect sample history can have upon chemical and physical properties, the rate of thermal decomposition of Ph(H)PP(H)Ph, which forms PhPH,, catena-H,(PhP), and cyclo-(PhP),, is strongly influenced by light, phosphides and traces of strong acids or bases7. Cyclopolyphosphines of mixed substitution are prepared by the redistribution of parts in ring-scrambling reactions. Thus, when heated with cyclo-(MeP), or cyclo-(PhP), cyclo-(EtP), forms all of the possible mixed-substituent cyclopentamers'. The general instability of catena-(PR), relative to cyclo-(PR), can be illustrated by the products of decomposition of several catena tri- (and tetra-) phosphines, e.g.:

4 (CF,),PP(CF,)P(CF,),

-

cyclo-(CF,P),

+ 4 (CF,),PP(CF,,

+ 5 EtzPPEt, 2 (Ph,P),P205~yclo-(PhP)' + Ph,P + PhZPPPh,

5 Et,PPEtPEt,-

cyclo-(EtP),

(e)' (f)

O

(8)"

The first reactiop is catalyzed by stopcock grease, mercury and bases. The reverse cyclo + catena conversion has also been observed": cyClo-(CF,P),

HzO

(h)

CF3(H)PPCF3P(H)CF3

The combination of cyclo-(CF,P), and MePH, leads to the quantitative conversion of CF,-substituted cyclopentamer to the Me-substituted cy~lopentamer'~.

$ cyclo-(CF,P),

+ MePH,

CF,PH,

+

cyclo-(CH,P),

(9

(A.L. RHEINGOLD)

M. Baudler, Angew. Chem., Znt. Ed. Engl., 26,419 (1987). H.-G. von Schnering, W. Honle, Chem. Rev., 88, 243 (1988). M. Baudler, Angew. Chem., Znt. Ed. Engl., 21,492 (1982). M. Baudler, B. Carlsohn, B. Kloth, B. Koch, 2. Anorg. Allg. Chem., 432, 67 (1977). M. Baudler, Pure Appl. Chem., 52,155 (1980). A. H. Cowley, D. S. Dierdorf, J. Am. Chem. SOC.,91,6609 (1969). M. Baudler, B. Carlsohn, D. Koch, P. Medda, Chem. Ber., 111, 1210 (1978). U. Schmidt, R. Schroer, H. Achenbach, Angew. Chem., Znt. Ed. Engl., 5, 316 (1966), see also M. Baudler, B. Carlsohn, Chem. Ber., 110, 2404 (1977). 9. A. B. Berg, J. F. Nixon, J. Am. Chem. SOC.,86, 356 (1964). 10. E. Wilberg, M. van Ghemen, G. Muller-Schiedmayer, Angew. Chem.,Int. Ed. Engl., 2,646 (1963). 11. H. Schumann, A. Roth, 0. Stelzer, Angew. Chem., Int. Ed. Engl., 7,218 (1968). 12. W. Mahler, A. B. Burg, J. Am. Chem. SOC.,80, 6161 (1958). 13. A. H. Cowley, J. Am. Chem. SOC., 89,5990 (1967). 1. 2. 3. 4. 5. 6. 7. 8.

15.2.3.3. Arsenic.

The cyclic form of MeAs appears to be exclusively a pentamer. Proton-NMR studies of cyclo-(MeAs), have led to speculation about the dynamics of ring behavi~rl-~. Above 170"C, the 2:2: 1, three-line pattern coalescences to a single line, which is clearly resolved

above 200°C. The importance of speculation about symmetry-generating, ring-opening-ring-closing and inversional equilibria is dampened by the observation that the temperature of onset of coalescence is sample dependent-hist~ry'~~. In fact, when special effort is made to minimize factors that may degrade the cyclopentamer, no coalescence at all is observed'. The specific incorporation of traces of iodine or nonoxidizing acids can

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc.

55

15.2. Ring-Ring and Ring-Polymer lnterconversions 15.2.3. In Oligomeric Catenates of P, As, Sb and Bi 15.2.3.3. Arsenic.

As an illustration of the effect sample history can have upon chemical and physical properties, the rate of thermal decomposition of Ph(H)PP(H)Ph, which forms PhPH,, catena-H,(PhP), and cyclo-(PhP),, is strongly influenced by light, phosphides and traces of strong acids or bases7. Cyclopolyphosphines of mixed substitution are prepared by the redistribution of parts in ring-scrambling reactions. Thus, when heated with cyclo-(MeP), or cyclo-(PhP), cyclo-(EtP), forms all of the possible mixed-substituent cyclopentamers'. The general instability of catena-(PR), relative to cyclo-(PR), can be illustrated by the products of decomposition of several catena tri- (and tetra-) phosphines, e.g.:

4 (CF,),PP(CF,)P(CF,),

-

cyclo-(CF,P),

+ 4 (CF,),PP(CF,,

+ 5 EtzPPEt, 2 (Ph,P),P205~yclo-(PhP)' + Ph,P + PhZPPPh,

5 Et,PPEtPEt,-

cyclo-(EtP),

(e)' (f)

O

(8)"

The first reactiop is catalyzed by stopcock grease, mercury and bases. The reverse cyclo + catena conversion has also been observed": cyClo-(CF,P),

HzO

(h)

CF3(H)PPCF3P(H)CF3

The combination of cyclo-(CF,P), and MePH, leads to the quantitative conversion of CF,-substituted cyclopentamer to the Me-substituted cy~lopentamer'~.

$ cyclo-(CF,P),

+ MePH,

CF,PH,

+

cyclo-(CH,P),

(9

(A.L. RHEINGOLD)

M. Baudler, Angew. Chem., Znt. Ed. Engl., 26,419 (1987). H.-G. von Schnering, W. Honle, Chem. Rev., 88, 243 (1988). M. Baudler, Angew. Chem., Znt. Ed. Engl., 21,492 (1982). M. Baudler, B. Carlsohn, B. Kloth, B. Koch, 2. Anorg. Allg. Chem., 432, 67 (1977). M. Baudler, Pure Appl. Chem., 52,155 (1980). A. H. Cowley, D. S. Dierdorf, J. Am. Chem. SOC.,91,6609 (1969). M. Baudler, B. Carlsohn, D. Koch, P. Medda, Chem. Ber., 111, 1210 (1978). U. Schmidt, R. Schroer, H. Achenbach, Angew. Chem., Znt. Ed. Engl., 5, 316 (1966), see also M. Baudler, B. Carlsohn, Chem. Ber., 110, 2404 (1977). 9. A. B. Berg, J. F. Nixon, J. Am. Chem. SOC.,86, 356 (1964). 10. E. Wilberg, M. van Ghemen, G. Muller-Schiedmayer, Angew. Chem.,Int. Ed. Engl., 2,646 (1963). 11. H. Schumann, A. Roth, 0. Stelzer, Angew. Chem., Int. Ed. Engl., 7,218 (1968). 12. W. Mahler, A. B. Burg, J. Am. Chem. SOC.,80, 6161 (1958). 13. A. H. Cowley, J. Am. Chem. SOC., 89,5990 (1967). 1. 2. 3. 4. 5. 6. 7. 8.

15.2.3.3. Arsenic.

The cyclic form of MeAs appears to be exclusively a pentamer. Proton-NMR studies of cyclo-(MeAs), have led to speculation about the dynamics of ring behavi~rl-~. Above 170"C, the 2:2: 1, three-line pattern coalescences to a single line, which is clearly resolved

above 200°C. The importance of speculation about symmetry-generating, ring-opening-ring-closing and inversional equilibria is dampened by the observation that the temperature of onset of coalescence is sample dependent-hist~ry'~~. In fact, when special effort is made to minimize factors that may degrade the cyclopentamer, no coalescence at all is observed'. The specific incorporation of traces of iodine or nonoxidizing acids can

56

15.2. Ring-Ring and Ring-Polymer Interconversions 15.2.3. In Oligomeric Catenates of P, As, Sb and Bi 15.2.3.3. Arsenic.

-

be used to “tune” the coalescence temperature in the range from 150 to 205°C. A sample of cyclo-(MeAs) prepared from methylarsine and dibenzyl mercury: MeAsH,

+ Bz,Hg

i(MeAs),

+ 2 BzH + Hg

(4

showed no tendency to coalesce when first heated to 210”C, but did show steadily decreasing coalescence temperatures for subsequent reheatings,. The viscosity of freshly prepared and purified cyclo-(MeAs), that has not been previously exposed to light or temperatures above 5OCC,shows no deviations from linearity in log viscosity vs. 1/T plots between 30 and 200°C. Repeated temperature cycles, however, show increasing deviations from linearity above 140°C in a direction indicative of an increase in high molecular weight components4. Tetramethyldiarsine, a thermal decomposition product of cyclo-(MeAs), ,causes the collapse of the three-line NMR pattern for the cyclopentamer to a single line at 150°CI . At a Me:As ratio of about 1.5, an exchange lifetime of 0.3 s is determined along with an activational energy of 50 & 12 kJ mol-I. It was proposed that in Me,As-terminated catena forms, the middle MeAs units exchanged two or three times faster than did either starting material. Specifically, an equilibrium is established, which as drawn below lies predominately to the right: Me(AsMe),Me i-mer chain

+ (AsMe),

(b)

+ x-mer ring

(c)

Me(AsMe),-.Me (i - x)-mer chain

When cyclo-(MeAs) is treated with iodine in benzene, rapid iodine decolarization occurs, As- As ring bonds are cleaved and solid products form6. End-group analysis has shown that molecular weights between 2700 and 4300 are obtained, depending more upon the initial concentration of cyclopentamer than upon the As:I ratio’. The solid products, however, were washed with ether prior to analysis, a procedure accompanied by the removal of I, from the polymers. The assignments of mol wts based upon their end-group analyses therefore appears questionable. As a result consistent with the principles of scrambling reactions*, the action of iodine on arsenomethane more likely creates initially an oligomer mixture of iodineterminated, benzene-soluble polymethylpolyarsines, which eventually disproportionates to a mixture of high and low molecular weight products. The mixture can readily attain its most stable arrangement of parts since the relevant bond energies in the redistribution of parts, D(As-As) and D(As-I), are both weak and similar, giving no significant enthalpic advantage to any particular composition. The formation of the high molecular weight product, (MeAs),, which is insoluble in benzene, appears to be the “driving force” in this equilibrium-controlled process: n(MeAs), I(MeAs),,I

-

+ I,

benzene

I(MeAs),,I

MeAsI,

+ (MeAs),

(e)

The formation and disproportionation of the oligomer mixture is a complex process that has been studied by dynamic NMR spectroscopyg. When, in Eqs. (d) and (e), the initial value of n is varied from 0.2 to 20, and the NMR spectrum followed up to the initiation of precipitation of (MeAs),, a single, broad spectral line is seen at 35°C whose variation in chemical shift depends upon n. As the value of n decreases, the chemical shift

15.2. Ring-Ring and Ring-Polymer Interconversions 15.2.3. In Oligomeric Catenates of P, As, Sb and Bi 15.2.3.3.Arsenic.

57

approaches a value identical to that of MeAsI, (62.49). For values of n between 20 and 50, the single line breaks up into a number of peaks resembling 2:1:2 multiplet (61.72) of cyclo-(MeAs), . The transition from a broad single line to the characteristic three-line cyclo-(MeAs), spectrum is not smooth; additional lines, possibly due to transient fouror six-membered ring species, demonstrate the complex nature of these equilibria. For n > 50, the spectrum in all details is indistinguishable from pure cyclo-(MeAs),. The initiation of polymer precipitation at room temperature in dilute solutions varies from seconds to days depending upon n and the sample history of cyclo-(MeAs), . Curiously, brief exposure to long-wavelength UV radiation completely inhibits all solid product formation9. An interesting feature of the preparation of cyclo-(MeAs), by the reduction of MeAsO(ONa), is the frequent presence of varying quantities of red-brown solids contaminating the desired yellow oil. Traces of nonoxidizing acids or halogen can lead to virtually complete conversion to the solid product. Isolation and analysis of the amorphous red-brown solid indicates the approximate empirical composition MeAs, presumably a high molecular weight catena polymer; heating the solid to 205-210°C causes substantial reversion to cyclo-(MeAs), , but since this is also the temperature at which Me,AsAsMe, is formed in detectable quantities, the solid -+ liquid conversion can be considered catalytic in nature9. A crystalline, purple-black product of the same empirical formula, MeAs, is formed on the long-term combination of MeAsH, and MeAsI, after 400-600 days at 60°C in benzenelo. This solid possesses a unique, infinitely extended ladder array of As-Asbonded MeAs units'':

Under identical conditions, the ethyl- and higher alkyl-substituted arsines lead only to cyclooligomeric productsg. Additionally, attempts to form mixed methyl-ethyl polymers lead only to substitutionally pure products: MeAsH, MeAsC1,

+ EtAsC1, + EtAsH,

>--(MeAs), + (EtAs), ladder

ring

Further, if a pair of alkyl groups, neither of which is methyl, is chosen for a mixedsubstitution reaction, all possible mixed ring products are obtained (as shown by mass spectral molecular-ion identifications): RASH,

+ R'AsCl,

-

all possible mixed rings

+ HCl

(h)

where R # R # Me. Thus, Me is the only alkyl substitution that can lead to stable high molecular weight polyarsenic products. Solutions of repeatedly recrystallized cyclo-(PhAs), are stable to atmospheric oxidation, but the addition of traces of acid, iodine and many other reagents causes an immediate uptake of oxygen to form a variety of oligomeric structures". At high levels of

58

15.2. Ring-Ring and Ring-Polymer lnterconversions 15.2.3. In Oligomeric Catenates of P, As, Sb and Bi 15.2.3.4. Antimony.

I,, no oxygen uptake is also observed as complete conversion to PhAsI, is the apparently preferred reaction. The combination of cyclo-(MeAs), with a primary dihydride, RAsH,, produces the equilibrium relationships: n (MeAs),

+ RAsH, + 5 (RAs), + 5n CH,AsH,

(0

The equilibrium may be driven to the right by the removal of the more volatile CH,AsH,. The exchange of organyl groups is q~antitative'~. This reaction may prove to be important in the preparation of new cyclopolyarsines. (A.L. RHEINGOLD)

F. Knoll, H. C. Marsmann, J. R. Van Wazer, J. Am. Chem. SOC.,91,4986 (1969). P. S. Elmes, S. Middleton, B. 0. West, Aust. J. Chem., 23, 1559 (1970). E. J. Wells, R. C. Fergusson, J. C. Halett, L. K. Peterson, Can. J. Chem., 46, 2733 (1968). A. L. Rheingold, J. M. Bellama, J. Organomet. Chem., 102, 445 (1975). A. L. Rheingold, P. Choudhury, J. Organomet. Chem., 128, 155 (1977). A. L. Rheingold, J. M. Bellama, J. Chem. Soc., Chem. Commun., 1058 (1969). M. Ya. Kraft, V. V. Katyshkina, Dokl. Akad. Nauk SSSR, 66,207 (1949). J. R. Van Wazer, in Homoatomic Rings, Chains and Macromolecules of the Main-Group Elements, A. L. Rheingold, ed. Elsevier, Amsterdam, 1977, pp. 1-24. 9. A. L. Rheinhold, in Homoatomic Rings, Chains and Macromolecules of the Main-Group Elements, A. L. Rheingold, ed. Elsevier, Aysterdam, 1977, pp. 385-408. 10. A. L. Rheingold, J. M. Bellama, J. E. Lewis, Inorg. Chem., 12, 2845 (1973). 11. J. J. Daly, F. Sanz, Helv. Chim. Acta, 53, 1879 (1970). 12. F. F. Blicke, F. D. Smith, J. Am. Chem. Soc., 52,2946 (1930). 13. V. K. Gupta, L. K. Krannick, C. L. Wilkins, Inorg. Chem., 26, 1638 (1987). 1. 2. 3. 4. 5. 6. I. 8.

15.2.3.4. Antimony.

Only an extremely limited literature on Sb-Sb-bonded rings and chains is available. Unlike phosphorus and arsenic, primary organoantimony reagents are reluctant to form stable ring structures. Rather, intractable, often antimony-rich solids of very approximate composition RSb are obtained in reactions useful in the preparation of P or As rings. For example, PhSbC1, and PhSbH, in ether at -25°C produces a black solid, C6H5,06Sbl,04Clo,14, which suggests a material of average composition Cl-(PhSb),,-Cll. Tetraalkyl distibines, R,SbSbR,, are thermally unstable. The decomposition of Me,SbSbMe, is accelerated by stopcock grease and results in the formation of a black solid of approximate composition SbMe,,,,; (n-Bu),SbSb(Bu-n), decomposes similarly'. In contrast, Ph,SbSbPh, is found t o be considerably more stable. The clear preference for polymeric rather than cyclooligomeric primary organoantimony derivatives suggests that the polymeric state is the more thermodynamically stable. Given the relatively weak Sb-Sb bond strength, lowered steric restrictions and a likely more weakly stabilizing influence of transannular interactions, it is reasonable that ringchain equilibria strongly favor catena antimony structures3. Several cyclopolystibines have been crystallographically characterized: cyclo-(PhSb), (as the d i ~ x a n a t e cy~lo-(t-BuSb),~ )~ and ~yclo-(mesitylSb)4C,H,~. (A.L. RHEINGOLD)

1. F. Klages, W. Rapp, Chem. Ber., 88, 384 (1955). 2. A. B. Burg, L. R. Grant, J. Am. Chem. SOC.,89, 1 (1959).

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc.

58

15.2. Ring-Ring and Ring-Polymer lnterconversions 15.2.3. In Oligomeric Catenates of P, As, Sb and Bi 15.2.3.4. Antimony.

I,, no oxygen uptake is also observed as complete conversion to PhAsI, is the apparently preferred reaction. The combination of cyclo-(MeAs), with a primary dihydride, RAsH,, produces the equilibrium relationships: n (MeAs),

+ RAsH, + 5 (RAs), + 5n CH,AsH,

(0

The equilibrium may be driven to the right by the removal of the more volatile CH,AsH,. The exchange of organyl groups is q~antitative'~. This reaction may prove to be important in the preparation of new cyclopolyarsines. (A.L. RHEINGOLD)

F. Knoll, H. C. Marsmann, J. R. Van Wazer, J. Am. Chem. SOC.,91,4986 (1969). P. S. Elmes, S. Middleton, B. 0. West, Aust. J. Chem., 23, 1559 (1970). E. J. Wells, R. C. Fergusson, J. C. Halett, L. K. Peterson, Can. J. Chem., 46, 2733 (1968). A. L. Rheingold, J. M. Bellama, J. Organomet. Chem., 102, 445 (1975). A. L. Rheingold, P. Choudhury, J. Organomet. Chem., 128, 155 (1977). A. L. Rheingold, J. M. Bellama, J. Chem. Soc., Chem. Commun., 1058 (1969). M. Ya. Kraft, V. V. Katyshkina, Dokl. Akad. Nauk SSSR, 66,207 (1949). J. R. Van Wazer, in Homoatomic Rings, Chains and Macromolecules of the Main-Group Elements, A. L. Rheingold, ed. Elsevier, Amsterdam, 1977, pp. 1-24. 9. A. L. Rheinhold, in Homoatomic Rings, Chains and Macromolecules of the Main-Group Elements, A. L. Rheingold, ed. Elsevier, Aysterdam, 1977, pp. 385-408. 10. A. L. Rheingold, J. M. Bellama, J. E. Lewis, Inorg. Chem., 12, 2845 (1973). 11. J. J. Daly, F. Sanz, Helv. Chim. Acta, 53, 1879 (1970). 12. F. F. Blicke, F. D. Smith, J. Am. Chem. Soc., 52,2946 (1930). 13. V. K. Gupta, L. K. Krannick, C. L. Wilkins, Inorg. Chem., 26, 1638 (1987). 1. 2. 3. 4. 5. 6. I. 8.

15.2.3.4. Antimony.

Only an extremely limited literature on Sb-Sb-bonded rings and chains is available. Unlike phosphorus and arsenic, primary organoantimony reagents are reluctant to form stable ring structures. Rather, intractable, often antimony-rich solids of very approximate composition RSb are obtained in reactions useful in the preparation of P or As rings. For example, PhSbC1, and PhSbH, in ether at -25°C produces a black solid, C6H5,06Sbl,04Clo,14, which suggests a material of average composition Cl-(PhSb),,-Cll. Tetraalkyl distibines, R,SbSbR,, are thermally unstable. The decomposition of Me,SbSbMe, is accelerated by stopcock grease and results in the formation of a black solid of approximate composition SbMe,,,,; (n-Bu),SbSb(Bu-n), decomposes similarly'. In contrast, Ph,SbSbPh, is found t o be considerably more stable. The clear preference for polymeric rather than cyclooligomeric primary organoantimony derivatives suggests that the polymeric state is the more thermodynamically stable. Given the relatively weak Sb-Sb bond strength, lowered steric restrictions and a likely more weakly stabilizing influence of transannular interactions, it is reasonable that ringchain equilibria strongly favor catena antimony structures3. Several cyclopolystibines have been crystallographically characterized: cyclo-(PhSb), (as the d i ~ x a n a t e cy~lo-(t-BuSb),~ )~ and ~yclo-(mesitylSb)4C,H,~. (A.L. RHEINGOLD)

1. F. Klages, W. Rapp, Chem. Ber., 88, 384 (1955). 2. A. B. Burg, L. R. Grant, J. Am. Chem. SOC.,89, 1 (1959).

15.2. Ring-Ring and Ring-Polymer Interconversions 15.2.3. In Oligomeric Catenates of P, As, Sb and Bi 15.2.3.5. Stabilized Rings and Chains.

59

3. P.Choudhury, M. F. El-Shazly, C. Spring A. L. Rheingold, Znorg. Chem., 18, 543 (1979). 4. H. J. Bruenig, K. Haberle, M. Drager, T. Severengiz, Angew. Chem., Znt. Ed. Engl., 24, 72 (1985). 5. 0. Mundt, G. Becker, M. J. Wessely, H. J. Bruenig, H. Kischkel, 2.Anorg. Alloy. Chem., 486, 70 (1982).

6. M. Ates, H. J. Breunig, S . Giilec, W. Offermann, K. Habeile, M. Drager, Chem. Eer., 122, 473 (1989).

15.2.3.5. Stabilized Rings and Chalns.

Thermodynamic factors determine the most stable sizes of homoatomic rings and chains the P, As and Sb. Exactly the same set of factors is at work in the allotropic forms of S; all allotropes other than cyclo-S, eventually revert to this most stable form. This high degree of specificity derives from the interplay of several factors, of which the most important is likely a balance between optimization of transannular stabilization and minimization of steric forces, and the apparent ease with which thermodynamically less stable forms can rearrange to more stable forms, preventing the isolation of forms that may be kinetically favored. Yet, like so many examples from other areas of chemistry, coordination to a transition-metal center alters the importance of factors prevailing in the uncoordinated state. The effect results from the addition of metal atomic orbitals to the ligand molecular orbital set; the stabilization of cyclobutadiene by coordination to the Fe tricarbonyl group provides a now classical example of this phenomenon'. Reactions of the cyclopolyarsines (and to a lesser extent the cyclopolyphosphines) with metal carbonyls and cyclopentadienyl metal carbonyls has produced an array of sizes for rings and chains of RAs units unknown in the uncoordinated state'. In some cases under relatively mild reaction conditions, the R groups are lost and complexes containing catenates of naked As atoms are formed; single products may contain both naked and substitued members in the same chain. Several examples of transition-metal complexes containing three or more linked group-VB atoms are given below. A useful concept in sorting out the various products of these reactions is the application of isolobal relationships3. A group-VB atom and an RE unit are related, respectively, to the common organic fragments CH and CH,. These fragments are easily generated from the elements themselves (e.g., white phosphorus, P,, or yellow arsenic, As,) and from ring-opening reactions of the cyclopolyphosphines or arsines. This is one example of complexes containing rings of As atoms being formed from As,& Thus, the fragment RAsAsAsR is related to the ally1 radical, CH,CHCH,. Complexes of RAsAsAsR (R = Me, Ph), formed by the fragmentation and partial loss of organic substitution from cyclopolyarsines, are known that are structurally identical to '. The metallacyclobutane z-ally1complexes, e.g., [(q5-C5Me,)Mo(CO),(q3-RAsAsAsR)] analog complex (q5-C,H,),Ti(AsMe), may be formed6 from cyclo-(MeAs), and (q5-C,H5),TiCl,. Dinuclear complexes containing chains of four RE units may be formed directly from metal carbonyls and the cyclic precursors; Fe(CO), and cyclo-(MeE), form7 [Fe(CO),],[catena-q2, p-(MeE),] for E = P and As, and [(qs-C,H,)Mo(CO),]2 and cyclo-(MeAs), form' a five-membered catenate complex, [(q5-C,H,)Mo(CO),],]catena-$, p-(MeAs),]. The six-membered ring system in (CO),Mo[$-(AsPh),] is derived from cyclo-(PhAs), but has undergone a chair to twist-boat conformational change to

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc. 15.2. Ring-Ring and Ring-Polymer Interconversions 15.2.3. In Oligomeric Catenates of P, As, Sb and Bi 15.2.3.5. Stabilized Rings and Chains.

59

3. P.Choudhury, M. F. El-Shazly, C. Spring A. L. Rheingold, Znorg. Chem., 18, 543 (1979). 4. H. J. Bruenig, K. Haberle, M. Drager, T. Severengiz, Angew. Chem., Znt. Ed. Engl., 24, 72 (1985). 5. 0. Mundt, G. Becker, M. J. Wessely, H. J. Bruenig, H. Kischkel, 2.Anorg. Alloy. Chem., 486, 70 (1982).

6. M. Ates, H. J. Breunig, S . Giilec, W. Offermann, K. Habeile, M. Drager, Chem. Eer., 122, 473 (1989).

15.2.3.5. Stabilized Rings and Chalns.

Thermodynamic factors determine the most stable sizes of homoatomic rings and chains the P, As and Sb. Exactly the same set of factors is at work in the allotropic forms of S; all allotropes other than cyclo-S, eventually revert to this most stable form. This high degree of specificity derives from the interplay of several factors, of which the most important is likely a balance between optimization of transannular stabilization and minimization of steric forces, and the apparent ease with which thermodynamically less stable forms can rearrange to more stable forms, preventing the isolation of forms that may be kinetically favored. Yet, like so many examples from other areas of chemistry, coordination to a transition-metal center alters the importance of factors prevailing in the uncoordinated state. The effect results from the addition of metal atomic orbitals to the ligand molecular orbital set; the stabilization of cyclobutadiene by coordination to the Fe tricarbonyl group provides a now classical example of this phenomenon'. Reactions of the cyclopolyarsines (and to a lesser extent the cyclopolyphosphines) with metal carbonyls and cyclopentadienyl metal carbonyls has produced an array of sizes for rings and chains of RAs units unknown in the uncoordinated state'. In some cases under relatively mild reaction conditions, the R groups are lost and complexes containing catenates of naked As atoms are formed; single products may contain both naked and substitued members in the same chain. Several examples of transition-metal complexes containing three or more linked group-VB atoms are given below. A useful concept in sorting out the various products of these reactions is the application of isolobal relationships3. A group-VB atom and an RE unit are related, respectively, to the common organic fragments CH and CH,. These fragments are easily generated from the elements themselves (e.g., white phosphorus, P,, or yellow arsenic, As,) and from ring-opening reactions of the cyclopolyphosphines or arsines. This is one example of complexes containing rings of As atoms being formed from As,& Thus, the fragment RAsAsAsR is related to the ally1 radical, CH,CHCH,. Complexes of RAsAsAsR (R = Me, Ph), formed by the fragmentation and partial loss of organic substitution from cyclopolyarsines, are known that are structurally identical to '. The metallacyclobutane z-ally1complexes, e.g., [(q5-C5Me,)Mo(CO),(q3-RAsAsAsR)] analog complex (q5-C,H,),Ti(AsMe), may be formed6 from cyclo-(MeAs), and (q5-C,H5),TiCl,. Dinuclear complexes containing chains of four RE units may be formed directly from metal carbonyls and the cyclic precursors; Fe(CO), and cyclo-(MeE), form7 [Fe(CO),],[catena-q2, p-(MeE),] for E = P and As, and [(qs-C,H,)Mo(CO),]2 and cyclo-(MeAs), form' a five-membered catenate complex, [(q5-C,H,)Mo(CO),],]catena-$, p-(MeAs),]. The six-membered ring system in (CO),Mo[$-(AsPh),] is derived from cyclo-(PhAs), but has undergone a chair to twist-boat conformational change to

60

15.2. Ring-Ring and Ring-Polymer Interconversions 15.2.3. In Oligomeric Catenates of P, As, Sb and Bi 15.2.3.5. Stabilized Rings and Chains.

facilitate coordinationg. However, this reaction with Mo(CO), is likely more complex; the analogous product is also formed from cyclo-(MeP),, which had to undergo a ring expansion to form a six-membered ring”. Ring expansion is a common result of the reactions of cyclopolyphosphines and arsines with metal carbonyls: e.g., Mo(CO), combines with either cyclo-(MeP), or whose heavy-atom struccyclo-(PhAs), to form [Mo(C0),l2[q6, p-cyclo-(RE),] 1131’, ture is that of trishomoc~bane’~, C,,H,,, the same structure as found in PlI3-l4 and Asl13-15. The nine-membered ring formed has a cyclononane conformation; six of the nine ring members are coordinated, three each, to two Mo(CO), groups. The same overall structure is maintained when Mn,(CO),, or Re,(CO),, replaces Mo(CO),, but owing to the increase by one in electron count for Mn or Re compared to Mo, one of the RAs units in the ring looses an organic group to convert it from a two-electron to a oneelectron donor: /

As

R - A s ~M ‘As

-

~

A

As / s ~M

‘As

the cyclononane ring also contains an isoelectronic synthesized “arsenic atom” in the form of an Re(CO), group’,. In this manner, {[cyc1o-(MeAs),(As)Re(C0),]Re,(C0),} is able to retain an isolobal relationship to the other trishomocubane structures. This result is a demonstration of the very strong tendency for main-group-rich metal complexes to adopt structures with analogs with main-group precedent, i.e., adopting the more open structural extremes associated with main-group clusters, in contrast to metalrich clusters, which more often adopt structures at or near the closo limit. If [(q5-C,H,)Mo(CO),], and cyclo-(MeAs), are slowly heated at a rate that allows both demethylation and decarbonylation to occur simultaneously, then both processes can be complete and [q5-C,H,Mo]2[q4, p-cyclo-(As),] is formed; it may be described as a 27-electron “triple-decker sandwich” containing an allyl-ene distorted ring of five As atoms as an isolobal analog of a cyclopentadienyl ringI7. The related all-phosphorus in the phosphorus structure the structure may be obtained from white central ring is near regular in the P-P bond distances. Molecular orbital calculations confirm the differences between the P and As systems21.22.Additionally, the reaction of white phosphorus with [(~5-C,Me,)Fe(CO),], produces the ferrocene-like s t r u c t ~ r e ’[(q5-(C,Me,)Fe(P,-q5). ~~~~ (A.L. RHEINGOLD)

1. 2. 3. 4. 5. 6. 7. 8. 9.

10.

G. F. Emerson, L. Watts, R. Pettit, J. Am. Chem. SOC.,87, 131 (1965). A.-J. DiMaio, A. L. Rheingold, Chem. Rev., 90, 0000 (1990). R. Hoffmann, Angew. Chem., Int. Ed. Engl., 21, 711 (1982). I. Bernal, H. Brunner, W. Meier, H. Pfisterer, J. Wachter, M. L. Zeigler, Angew. Chem., Int. Ed. Engl., 23, 438 (1984). J. R. Harper, M. E. Fountain, A. L. Rheingold, Orgunometallzcs, 8, 0000 (1989). P. Mercando, A.-J. DiMaio, A. L. Rheingold, Angew. Chem., Int. Ed. Engl., 26, 244 (1987). P. S. Elmes, B. 0. West, in J . Organomet. Chern., 32, 365 (1971). (a) A. L. Rheingold, M. R. Churchill, J. Organomet. Chem., 243, 165 (1983), (b) A. L. Rheingold, M. J. Foley, P. J. Sullivan, J. Am. Chem. SOC.,104, 4727 (1982). A. L. Rheingold, M. E. Fountain, Organometallics, 5, 2410 (1986). P. S. Elmes, B. M. Gatehouse, B. 0. West, J. Organomet. Chem., 82, 235 (1974).

15.2.4. In SiliconSilicon Systems 15.2.4.1. Formation of Cyclic Silicon-Silicon Systems 15.2.4.1.1. Perarylcyclosilanes.

61

11. P. S. Elmes, B. M. Gatehouse, D. J. Lloyd, B. 0. West, J. Chem. SOC.,Chem. Commun., 953 (1974). 12. A. L. Rheingold, M. L. Fountain, A.-J. DiMaio, J. Am. Chem. Soc., 109, 141 (1987). 13. G. R. Underwood, B. Ramamoorthy, Tetrahedron Lett., 4125 (1970). This structure is sometimes 14. 15. 16. 17. 18. 19.

also called UFOsane. H.-G. von Schnering, Angew. Chem., Znt. Ed. Engl., 20, 33 (1981). H.-G. von Schnering, W. Honle, Chem. Rev., 88, 243 (1988). A.-J. DiMaio, A. L. Rheingold, Organometallics 6, 1138 (1987). A. L. Rheingold, M. J. Foley, P. J. Sullivan, J. Am. Chem. SOC.,104, 4727 (1982). 0. J. Scherer, H. Sitzmann, G. Womershauser, Angew. Chem., Int. Ed. Engl., 24, 351 (1985). 0. J. Scherer, J. Schwalb, G. Wolmershauser, W. Kaim, R. Gross, Angew. Chem., Int. Ed. Engl.,

25, 363 (1986). 20. 0.J. Scherer, J. Schwalb, H. Swarowsky, G. Wolmerhauser, W. Kaim, R. Gross, Chem. Ber., 121, 443 (1988). 21. E. D. Jemmis, A. C . Reddy, Organometallics 7, 1561 (1988). 22. W. Tremmel, R. Hoffman, M. Kertesz, J. Am. Chem. SOC.,111, 2030 (1989). 23. 0. J. Scherer, T. Briick, Angew. Chem., Int. Ed. Engl., 26, 59 (1987). 24. 0. J. Scherer, T. Briick, G. Wolmerhauser, Chem. Ber., 121, 935 (1988).

15.2.4. In Silicon-Silicon Systems 15.2.4.1, Formation of Cyclic SiliconSilicon Systems

.

15.2.4.1 .l Perarylcyclosilanes.

The perphenylsilacycles, (Ph,Si), where n = 4,5 and 6, can be made from diphenyldichlorosilane with Na metal, but are best obtained by condensing diphenyldichlorosilane with Li metal'-3. With just 2 equiv Li in THF, the kinetic product (Ph,Si),, obtained in yields3 up to 75%:

5 Ph,SiCl,

+ 100 Li

THF

(Ph,Si),

+ 10 LiCl

(a)

Coupling Ph,SiCl, with Mg also produces (Ph,Si), in yields4 up to 60%. Electrolytic reduction of Ph,SiCl, also gives (Ph,Si), exclusively5.When larger amounts of Li and longer times are used, conversion to the thermodynamically favored product, (Ph,Si), takes place (see 815.2.4.2). The preferred method for synthesis of the five-membered ring involves the use both of xs Li and of a catalyst for ring-ring interconversion, such as (Ph,Si), or Ph,SiSiMe, 6 3 7 . These compounds produce Ph,SiLi, the active catalyst for the reaction:

5 Ph,SiCl,

+ xs Li

THF, 65°C

(Ph,Si),

+ 10 LiCl

The six-membered ring is obtained only as a minor byproduct in these syntheses, and no specific synthesis for (Ph,Si), is available. A 3.1% yield is reported from the Ph,SiCl,-Li reaction of Eq. (a)',*. The perphenylcyclosilanes are colorless crystals with high mp. In separating these compounds, use is made of the greater solubility of (Ph,Si), in benzene or toluene than either the four- or six-membered ring; the latter two therefore precipitate from the reaction mixture, leaving (Ph,Si), in solution. One other arylcyclosilane, (p-tolyl,Si),, is

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc. 15.2.4. In SiliconSilicon Systems 15.2.4.1. Formation of Cyclic Silicon-Silicon Systems 15.2.4.1.1. Perarylcyclosilanes.

61

11. P. S. Elmes, B. M. Gatehouse, D. J. Lloyd, B. 0. West, J. Chem. SOC.,Chem. Commun., 953 (1974). 12. A. L. Rheingold, M. L. Fountain, A.-J. DiMaio, J. Am. Chem. Soc., 109, 141 (1987). 13. G. R. Underwood, B. Ramamoorthy, Tetrahedron Lett., 4125 (1970). This structure is sometimes 14. 15. 16. 17. 18. 19.

also called UFOsane. H.-G. von Schnering, Angew. Chem., Znt. Ed. Engl., 20, 33 (1981). H.-G. von Schnering, W. Honle, Chem. Rev., 88, 243 (1988). A.-J. DiMaio, A. L. Rheingold, Organometallics 6, 1138 (1987). A. L. Rheingold, M. J. Foley, P. J. Sullivan, J. Am. Chem. SOC.,104, 4727 (1982). 0. J. Scherer, H. Sitzmann, G. Womershauser, Angew. Chem., Int. Ed. Engl., 24, 351 (1985). 0. J. Scherer, J. Schwalb, G. Wolmershauser, W. Kaim, R. Gross, Angew. Chem., Int. Ed. Engl.,

25, 363 (1986). 20. 0.J. Scherer, J. Schwalb, H. Swarowsky, G. Wolmerhauser, W. Kaim, R. Gross, Chem. Ber., 121, 443 (1988). 21. E. D. Jemmis, A. C . Reddy, Organometallics 7, 1561 (1988). 22. W. Tremmel, R. Hoffman, M. Kertesz, J. Am. Chem. SOC.,111, 2030 (1989). 23. 0. J. Scherer, T. Briick, Angew. Chem., Int. Ed. Engl., 26, 59 (1987). 24. 0. J. Scherer, T. Briick, G. Wolmerhauser, Chem. Ber., 121, 935 (1988).

15.2.4. In Silicon-Silicon Systems 15.2.4.1, Formation of Cyclic SiliconSilicon Systems

.

15.2.4.1 .l Perarylcyclosilanes.

The perphenylsilacycles, (Ph,Si), where n = 4,5 and 6, can be made from diphenyldichlorosilane with Na metal, but are best obtained by condensing diphenyldichlorosilane with Li metal'-3. With just 2 equiv Li in THF, the kinetic product (Ph,Si),, obtained in yields3 up to 75%:

5 Ph,SiCl,

+ 100 Li

THF

(Ph,Si),

+ 10 LiCl

(a)

Coupling Ph,SiCl, with Mg also produces (Ph,Si), in yields4 up to 60%. Electrolytic reduction of Ph,SiCl, also gives (Ph,Si), exclusively5.When larger amounts of Li and longer times are used, conversion to the thermodynamically favored product, (Ph,Si), takes place (see 815.2.4.2). The preferred method for synthesis of the five-membered ring involves the use both of xs Li and of a catalyst for ring-ring interconversion, such as (Ph,Si), or Ph,SiSiMe, 6 3 7 . These compounds produce Ph,SiLi, the active catalyst for the reaction:

5 Ph,SiCl,

+ xs Li

THF, 65°C

(Ph,Si),

+ 10 LiCl

The six-membered ring is obtained only as a minor byproduct in these syntheses, and no specific synthesis for (Ph,Si), is available. A 3.1% yield is reported from the Ph,SiCl,-Li reaction of Eq. (a)',*. The perphenylcyclosilanes are colorless crystals with high mp. In separating these compounds, use is made of the greater solubility of (Ph,Si), in benzene or toluene than either the four- or six-membered ring; the latter two therefore precipitate from the reaction mixture, leaving (Ph,Si), in solution. One other arylcyclosilane, (p-tolyl,Si),, is

62

15.2. Ring-Ring and Ring-Polymer Interconversions 15.2.4. In Silicon-Silicon Systems 15.2.4.1. Formation of Cyclic Silicon-Silicon Systems

,

made by a condensation reaction, from p-Tol,SiCl, and Li g. Perarylcyclosilanes can also be made by ring closure from open-chain compounds (see $15.2.4.2). Several sterically hindered arylcyclotrisilanes have been synthesized by condensing the corresponding dichlorosilanes with Li naphthalide'O,ll:

where Ar = Mes; 2,6-xylyl, 2,6-diethylphenyL (R. WEST)

H. Gilman, G. L. Schwebke, J. Am. Chem. Soc., 86,2613 (1964). H. Gilman, G. L. Schwebke, J. Organomet. Chem., 3, 382 (1965). H. Gilman, G. L. Schwebke, Adv. Organomet. Chem., 1, 132 (1964). H. Gilman, D. J. Peterson, R. A. Tomasi, A. L. Harrell, J. Organomet. Chem., 4, 167 (1965). 5. E. Hengge, J. Litscher, Monatsh. Chem., 109, 1217 (1978). 6. E. Hengge, H. Marketz, Monatsh. Chem., 100, 890 (1969). 7. S . M. Chen, A. Katti, T. Blinka, R. West, Synthesis, 684 (1985). 8. E. Hengge, D. Kovar, 2. Anorg. Allg. Chem., 459, 123 (1979). 9. M. F. Lemanski, E. P. Schramm, Znorg. Chem., 15, 1489, 2515 (1976). 10. S . Masamune, S . Murakami, J. T. Snow, H. Tobita, D. J. Williams, Organometallics, 3, 3330 1. 2. 3. 4.

(1984).

11. S . Masamune, Y. Hanyawa, S . Murakami, T. Bally, J. Blount, J. Am. Chem. 'Soc., 104, 1150

(1982).

15.2.4.1.2 Permethylcyclosilanes.

The six-membered ring (Me,%),, originally obtained only in low yield', is the first of

> 100 alkylcyclosilanes synthesized. Among cyclic alkylsilanes the compounds in the

permethyl series (Me,Si), are best known. (i) Alkali-Metal Condensations. Of the several reactions for the synthesis of methylcyclosilanes, the most important and general involve condensation of Me,SiCl, with alkali metals. The product distribution depends on the metal and the conditions (see Table 1) e.g., dimethyldichlorosilane in T H F is added to xs Na-K alloy in the THF2*3at 65°C: Me,SiCl,

T (Me,Si), a+ (Me,Si), + (Me,%), xs Na-K

90

x

9%

(a)

1%

The use of Na-K alloy is hazardous because this material inflames spontaneously in air. Precautions should be taken to prevent exposure of Na-K to O , , and a dry chemical extinguisher suitable for alkali-metal fires should be available. This condensation takes place in two stages. In the first the product is soluble linear polymer, along with small amounts of (Me,%),, n = 5-7. The polymer undergoes depolymerization in the second stage, producing the equilibrium mixture of (Me,Si), rings (see $15.2.4.3). From this (Me#), can be isolated in 85% yield by recrystallization from ethanol. Smaller amounts of (Me,%), and (Me,Si), can be obtained from the mixture, either by fractional sublimation [for (Me,Si),] or by preparative high-pressure liquid chromatography (HPLC). These and other peralkylcyclopolysilanes are colorless crystals. The same equilibrium is reached when xs K is used as the condensing agent, and the hazard is reduced here because the metal can be added as a solid and removed in the

-

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc. 62

15.2. Ring-Ring and Ring-Polymer Interconversions 15.2.4. In Silicon-Silicon Systems 15.2.4.1. Formation of Cyclic Silicon-Silicon Systems

,

made by a condensation reaction, from p-Tol,SiCl, and Li g. Perarylcyclosilanes can also be made by ring closure from open-chain compounds (see $15.2.4.2). Several sterically hindered arylcyclotrisilanes have been synthesized by condensing the corresponding dichlorosilanes with Li naphthalide'O,ll:

where Ar = Mes; 2,6-xylyl, 2,6-diethylphenyL (R. WEST)

H. Gilman, G. L. Schwebke, J. Am. Chem. Soc., 86,2613 (1964). H. Gilman, G. L. Schwebke, J. Organomet. Chem., 3, 382 (1965). H. Gilman, G. L. Schwebke, Adv. Organomet. Chem., 1, 132 (1964). H. Gilman, D. J. Peterson, R. A. Tomasi, A. L. Harrell, J. Organomet. Chem., 4, 167 (1965). 5. E. Hengge, J. Litscher, Monatsh. Chem., 109, 1217 (1978). 6. E. Hengge, H. Marketz, Monatsh. Chem., 100, 890 (1969). 7. S . M. Chen, A. Katti, T. Blinka, R. West, Synthesis, 684 (1985). 8. E. Hengge, D. Kovar, 2. Anorg. Allg. Chem., 459, 123 (1979). 9. M. F. Lemanski, E. P. Schramm, Znorg. Chem., 15, 1489, 2515 (1976). 10. S . Masamune, S . Murakami, J. T. Snow, H. Tobita, D. J. Williams, Organometallics, 3, 3330 1. 2. 3. 4.

(1984).

11. S . Masamune, Y. Hanyawa, S . Murakami, T. Bally, J. Blount, J. Am. Chem. 'Soc., 104, 1150

(1982).

15.2.4.1.2 Permethylcyclosilanes.

The six-membered ring (Me,%),, originally obtained only in low yield', is the first of

> 100 alkylcyclosilanes synthesized. Among cyclic alkylsilanes the compounds in the

permethyl series (Me,Si), are best known. (i) Alkali-Metal Condensations. Of the several reactions for the synthesis of methylcyclosilanes, the most important and general involve condensation of Me,SiCl, with alkali metals. The product distribution depends on the metal and the conditions (see Table 1) e.g., dimethyldichlorosilane in T H F is added to xs Na-K alloy in the THF2*3at 65°C: Me,SiCl,

T (Me,Si), a+ (Me,Si), + (Me,%), xs Na-K

90

x

9%

(a)

1%

The use of Na-K alloy is hazardous because this material inflames spontaneously in air. Precautions should be taken to prevent exposure of Na-K to O , , and a dry chemical extinguisher suitable for alkali-metal fires should be available. This condensation takes place in two stages. In the first the product is soluble linear polymer, along with small amounts of (Me,%),, n = 5-7. The polymer undergoes depolymerization in the second stage, producing the equilibrium mixture of (Me,Si), rings (see $15.2.4.3). From this (Me#), can be isolated in 85% yield by recrystallization from ethanol. Smaller amounts of (Me,%), and (Me,Si), can be obtained from the mixture, either by fractional sublimation [for (Me,Si),] or by preparative high-pressure liquid chromatography (HPLC). These and other peralkylcyclopolysilanes are colorless crystals. The same equilibrium is reached when xs K is used as the condensing agent, and the hazard is reduced here because the metal can be added as a solid and removed in the

-

15.2.4. In SiliconSilicon Systems 15.2.4.1. Formation of Cyclic Silicon-Silicon Systems 15.2.4.1.2. Permethylcyclosilanes.

TABLE1. YIELD OF CYCLOSILANES, (Me,Si),, (wt%) Metal xsNa-K or K xs Na xs Li xs Li 2.0 Li xsNa-K 2.0 K

Solvent THF Toluene THF THF THF THF THF

CONDENSATION OF Me,SiC1, WITH ALKALI METALS

n =5

6

7

9

90

1

0"

9 5

650b 65"

17 31

<1 60 90 17 52 64

T("C)

-

FROM

65" 100"

0" 0""

63

8

9

>9

High polymer

2,3,8

> 99 1 7 2.6 1.8

Refs.

16 1.3 0.4

3 0.8 0.3

N

2

>5

8 7 5 7 8 4

With Ph,SiLi catalyst. Slow addition.

same form from the cooled mixture after reaction4. The equilibrium mixture of (Me&), rings is also attained with xs Li, if (Ph,Si), or Ph,SiSiMe, is added as a redistribution catalystSs6.Using Li in the absence of such a catalyst the approach to equilibrium is slow and much polymer is obtained. If an exactly equivalent amount of Li, or a deficiency, is used to condense Me,SiCl,, a product distribution is obtained in which the six- and eight-membered rings dominate among the oligomeric products (see Table 1)'. This provides the best method for synthesis of (Me,Si), . Higher cyclopolysilanes are also formed in this Li condensation. Nonequilibrium product distributions are also obtained when condensation is carried out with an equivalent amount of K ', or when the dimethyldichlorosilane is added slowly (over 8-10 h) to Na-K alloy, to prevent polymer formationg. The amounts of (Me,%), and (Me,%), are increased relative to (Me,%),, and higher permethylcyclosilanes are also present, including all (Me,%), rings up to n = 35. After removal of most of the (Me,Si), and (Me,Si), by sublimation, these mixtures are best separated by reverse-phase HPLC (see Fig. 1). The rings with n = 5-19, and 24 are known. These

Figure 1. High-pressure liquid chromatography (HPLC) of mixture of (Me,Si), rings obtained in reaction of Me,SiCl, with Na-K (after ref. 8). Each peak represents a separate oligomer, from n = 5 to n = 35.

64

15.2. Ring-Ring and Ring-Polymer Interconversions 15.2.4. In Silicon-Silicon Systems 15.2.4.1. Formation of Cyclic Silicon-Silicon Systems

I 4

I.0

0.S’

Figure 2. Yields of (Me,Si), rings from reaction of Me,SiCI, with 2.0 equiv of K4. cyclopolysilanes form the largest homologous series known to chemistry, excepting only the cycloalkanes, and their isolation provides an opportunity for the study of ring-size effects in Si chemistry. Yields of rings decline through about n = 10 and then level off (for Na-K) or increase (with K) as shown4~* in Fig. 2. A medium-ring effect, of the type known in cycloalkane chemistry, is found also in the cyclosilane condensations. Because of the increasing entropic requirement for ring closure as n increases, yields should decrease with increasing n. For (Me& rings with n = 10-16, the increased entropy must be more than compensated by decreased nonclassical strain as the ring size increases. (ii) Other Syntheses of Permethylcyclosilanes. Other reactions produce (Me&), , and some may be useful synthetically. The methoxide-ion-catalyzed disproportionation of (Me,SiOMe), leads to (Me,Si), lo. The starting material may be made from the residue from the direct process for making methylchlorosilanes (see 55.2.7.2.1): 6 (Me,SiOMe),

[OMel-

43 %

6 Me,Si(OMe),

+ (Me,Si),

The synthesis of (Me,Si), also proceeds’ from dimethylchlorosilane and Li: 6 Me,SiHCl

+ 6 Li

6 LiCl

+ (Me,Si), + 3 H,

In this process only 1 equiv of Li is used, so H, must be lost at some stage.

(c)

15.2.4. In Silicon-Silicon Systems 15.2.4.1. Formation of Cyclic Silicon-Silicon Systems 15.2.4.1.2. Permethylcyclosilanes.

65

Triphenylsilyllithium, which catalyzes equilibration of methylcyclosilanes, also converts certain linear polysilanes to (Me,Si), and (Me,Si), 12: Ph,Si(Me,Si),SiPh,

PhaSiLi

(Me$),

+ Ph,SiSiPh,

(4

where m = 1-5; n = 5, 6. This redistribution must be kinetically controlled, because the reaction of Ph,SiMe,SiSiPh, with Ph,SiLi leads initially to (Me,Si), in larger quantity than (Me,Si), . Depolymerization of linear polysilanes to cyclic oligomers13 is described in $15.2.4.3. The photolysis of cyclosilanes with 254-nm UV light leads to elimination of divalent Si intermediates (silylenes), and reclosure of the ring. In addition to being a source of silylenes, this reaction is useful in the synthesis of cyclosilanes'4. In the permethyl series, photolysis of (Me,Si), leads successively to (Me,Si), and (Me,Si),: (Me,Si), (Me,Si),

s(Me,%), hv hv

(Me,%),

+ [Me,Si:]

+ [Me,Si:]

Further photolysis produces the ring-opened product, H(SiMe,),H. Finally, the permethylcyclosilanes with n = 4-6 can be made by alkylating the perhalocyclosilanes, which can in turn be synthesized from the perphenyl rings' ,,16: (Ph,Si), where n

+ 2n HCl

= 4-6;

+

(C12Si)n n Me,Zn

AICI~

(C12Si), + 2n PhH

(Me$),

+ ZnC1,

(h)

where n = 4-6. Dialkylzincs, rather than organolithium or organomagnesium halide reagents, must be used for the alkylation; otherwise ring interconversion takes place. Although lengthy, this route may be the best one for making large quantities of the reactive compound (Me,Si),. (R. WEST)

1. C. Burkhard, J. Am. Chem. SOC.,71, 63 (1949). 2. R. West, L. F. Brough, W. Wojnowski, Inorg. Synth., 19, 265 (1979). 3. L. F. Brough, R. West, W. Wojnowski, unpublished work; L. F. Brough, Ph.D. Thesis, Univ. Wisconsin, Madison, WI, 1980. 4. A. Katti, C. W. Carlson, R. West, J. Organomet. Chem., 271, 353 (1984). 5. S.-M. Chen, A. Katti, T. Blinka, R. West, Synthesis, 684 (1985). 6. H. Gilman, R. A. Tomasi, J . Org. Chem., 128, 1651 (1963). 7. K. Matsumura, L. F. Brough, R. West, J. Chem. Soc., Chem. Commun., 1092 (1978). 8. L. F. Brough, R. West, J. Am. Chem. Soc., 103, 3049 (1981). 9. L. F. Brough, K. Matsumura, R. West, Angew. Chem., Int. Ed. Engl., 18, 955 (1979). 10. H. Watanabe, K. Higuchi, M. Kobayashi,T. Kitahara, Y. Nagai, J. Chem. Soc., Chem. Commun., 704 (1977). 11. W. Peterson, B. Arkles, US. Pat. 4,276,424 (1981); Chem. Abstr., 95, 113,145 (1981). 12. M. Kumada, M. Ishikawa, S. Sakamoto, S. Maeda, J. Organomet. Chem., 17,223 (1969). 91, 5440 (1969). 13. E. Carberry, R. West, J. Am. Chem. SOC., 14. M. Ishikawa, M. Kumada, J. Chem. Soc., Chem. Commun.,612 (1970); J. Organornet. Chem.,42, 375 (1972).

66

15.2. Ring-Ring and Ring-Polymer Interconversions 15.2.4. In Silicon-Silicon Systems 15.2.4.1. Formation of Cyclic Silicon-Silicon Systems

15. E. Hengge, J. Organomet. Chem. Libr.,9, 267 (1979). 16. E. Hengge, D. Kovar, J. Organomet. Chem., 125, C29 (1977); 2. Anorg. Allg. Chem., 459, 123 (1979).

15.2.4.1.3. Other Peralkylcyclosilanes.

The reaction of dichlorosilanes with alkali metals, especially Li, synthesizes various dialkylcyclosilanes. Typically 10-20% xs Li is used in THF a t 0°C to RT’. 1,2Dichlorodisilanes can also be used’ : RR‘SiC1,

- xs Li, THF

0°C

xs Li, THF

(RRSi),

RRSiCI-SiClRR’

0°C

The ring size of the principal product depends on the steric bulk of the substituents, as shown in Table 13.

TABLE1. CYCLOSILANE PRODUCTS FROM RR’SiCI, + Li R

R‘

n

Ref.

Me Me

Me

6 5 5 5 5 4 4 4 4 4 3

4 5 6 1 1 1 1 1 7 1 1

Et Et

Et

n-Pr i-Pr Me,CCH, i-Pr i-Bu

n-Pr Me Me i-Pr i-Bu

Me Me,SiCH, Me,CCH,

t-Bu

Me,SiCH, Me,CCH,

The highly hindered cyclotrisilane (t-Bu,Si), is obtained by lithium naphthalide condensationa: t-Bu,SiI, or (t-Bu,SiBr),

L i + C10H8 THF, -78°C

(t-Bu,Si),

(b)

Sodium condensation of the corresponding dibromide produces ([Et3Si),Si],g, and two hexaalkylcyclotrisilanes are synthesized by reductive chlorination of 1,3-dichlorotrisilanes: l o Cl(SiR,),Cl

Li CloHe +

(R,Si),

where R = i-Pr, Et,CH. A photochemical method is used to generate (i-PrzSi)4, apparently through the intermediacy of the disilene i-Pr,Si = SiPr2-i:l l .

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc.

66

15.2. Ring-Ring and Ring-Polymer Interconversions 15.2.4. In Silicon-Silicon Systems 15.2.4.1. Formation of Cyclic Silicon-Silicon Systems

15. E. Hengge, J. Organomet. Chem. Libr.,9, 267 (1979). 16. E. Hengge, D. Kovar, J. Organomet. Chem., 125, C29 (1977); 2. Anorg. Allg. Chem., 459, 123 (1979).

15.2.4.1.3. Other Peralkylcyclosilanes.

The reaction of dichlorosilanes with alkali metals, especially Li, synthesizes various dialkylcyclosilanes. Typically 10-20% xs Li is used in THF a t 0°C to RT’. 1,2Dichlorodisilanes can also be used’ : RR‘SiC1,

- xs Li, THF

0°C

xs Li, THF

(RRSi),

RRSiCI-SiClRR’

0°C

The ring size of the principal product depends on the steric bulk of the substituents, as shown in Table 13.

TABLE1. CYCLOSILANE PRODUCTS FROM RR’SiCI, + Li R

R‘

n

Ref.

Me Me

Me

6 5 5 5 5 4 4 4 4 4 3

4 5 6 1 1 1 1 1 7 1 1

Et Et

Et

n-Pr i-Pr Me,CCH, i-Pr i-Bu

n-Pr Me Me i-Pr i-Bu

Me Me,SiCH, Me,CCH,

t-Bu

Me,SiCH, Me,CCH,

The highly hindered cyclotrisilane (t-Bu,Si), is obtained by lithium naphthalide condensationa: t-Bu,SiI, or (t-Bu,SiBr),

L i + C10H8 THF, -78°C

(t-Bu,Si),

(b)

Sodium condensation of the corresponding dibromide produces ([Et3Si),Si],g, and two hexaalkylcyclotrisilanes are synthesized by reductive chlorination of 1,3-dichlorotrisilanes: l o Cl(SiR,),Cl

Li CloHe +

(R,Si),

where R = i-Pr, Et,CH. A photochemical method is used to generate (i-PrzSi)4, apparently through the intermediacy of the disilene i-Pr,Si = SiPr2-i:l l .

15.2.4. In SiliconSilicon Systems 15.2.4.1. Formation of Cyclic Silicon-Silicon Systems 15.2.4.1.3. Other Peralkylcyclosilanes.

%

67

SiPr,-i

i-PrJi

hv

C,,H,,

+ i-Pr,Si=Si(Pr-i),

x2

(i-Pr,Si),

(d)

Another cyclotetrasilane prepared by a route involving probable disilene dimerization is [(Me,Si),Si],”: (Me,Si),SiOMe

a

[(Me,Si),Si]

x2

(Me,Si),Si=Si(SiMe,),

IX

C(Me,Si),Sil, Alkali-metal condensation is also used to synthesize organosilicon rotanes from the cyclopolymethylenesilanes (CH,),SiCl, and (CH,),SiCl, l 1 : n (CH,),SiC1,

+2K

n (CH,),SiCl,

+ 2 Li

where n = 5, 6;

THF

[(CH,),Si],

THF O T

[(CH,),Si],

+ 2n KC1 + 2n LiCl

(f)

(8)

where n = 5-10. The products have symmetrical structures, e.g., for [(CH,),Si], (A). In xs Li, (CH,),SiCI, reacts to give the expected rotanes with n = 5 and 6, along with a rearranged (C,H,Si), isomer (B).

A

B

Marked dependence of product distribution on conditions is observed for the diethyl as seen in Table 2. Condensations with an equivalent amount of Li at 0°C produces a mixture consisting mainly of the seven- and five-membered rings (cf. the analogous reaction in the permethyl series where the six- and eight-membered rings dominate; Table 1, 515.2.4.1.2). Potassium or Na-K alloy give the equilibrium mixture, which is mostly (Et,Si), . Sodium condensation produces mainly the four-membered ring (Et,Si), . Thus by judicious choice of conditions alkali-metal condensation may be used to synthesize in major amount any of the rings (Et2Si),, where n = 4,5 or 7.

68

15.2. Ring-Ring and Ring-Polymer Interconversions 15.2.4. In Silicon-Silicon Systems 15.2.4.1. Formation of Cyclic Silicon-Silicon Systems

TABLE2. PRODUCTS FROM EtzSiC1, CONDENSATION WITH ALKALI METALS'S~ ~~

(EtZSi),

Metal

T(T)

2.0 Li 2.0 Li 2.2 Li 2.2 K 2.2 Na

0 -40 0 65 110

Solvent THF THF THF THF C,H,

n

=4

0 0 0 0 56

(7%

5

6

7

8

14 7 29 91 9

4 15 4 Trace 0

41 33 22 0 0

8 2 2 0 0

Large rings may be formed in nonequilibrium syntheses; e.g.; all rings from n = 4 to n = 38 are found in the mixture of (EtMeSi), isomer produced from EtMeSiCl, and 2.0 equiv K 5 . Large rings are also formed in the synthesis of organosilicon rotamers", (R. WEST)

1. H. Watanabe, T. Muraoka, M. Kageyama, K. Yoshizumi, Y. Nagai, Organometallics, 3, 141 (1984) and refs. therein. 2. H. Watanabe, J. Inose, K. Fukushima, Y. Kougo, Y. Nagai, Chem. Lett., 1711 (1983). 3. Y. Nagai, H. Watanabe, M. Matsumoto, in Silicon Chemistry, J. Y. Corey, E. R. Corey, P. P. Gaspar, eds., Ellis Horwood, 1988, p. 247. 4. S. M. Chen, R. West, Organomet. Synth., 4, 506, (1988). 5. A. Katti, C. W. Carlson, R. West, J. Organomet. Chem., 271, 353 (1984). 6. C. W. Carlson, R. West, Organometallics, 2, 1792 (1983). 7. B. J. Helmer, R. West, Organometallics, I , 1458 (1982). 8. A. Schafer, M. Weidenbruch, K. Peters, H. G. von Schnering, Angew. Chem., Int. Ed. Engl., 23, 302 (1984). 9. H. Matsumoto, A. Sakamoto, Y. Nagai, J. Chem. SOC.,Chem. Commun., 1468 (1986). 10. S. Masamune, H. Tobita, S. Murakami, J. Am. Chem. SOC.,105,6524 (1983). 11. H. Watanabe, Y. Kougo, Y. Nagai, J. Chem. SOC.,Chem. Commun., 66 (1984). 12. Y . 4 . Chen, P. P. Gaspar, Organometallics, 1, 1410 (1987). 13. C. W. Carlson, Z.-H. Zhang, R. West, Organometallics, 2, 453 (1983). 15.2.4.1.4. Alkylarylcyclosilanes.

Condensation of PhMeSiCl, with Li gives of the all-trans isomer of (PhMeSi), in 10% yield'. The reaction of PhMeSiC1, with Li and Ph,SiSiMe, catalyst produces high yields of (PhMeSi), and (PhMeSi), as a complex mixture of isomers'. Cocondensation of different dichlorosilanes is also used to synthesize substituted cyclics; e.g., treatment of a 5 : 1 mixture of Me,SiCl, and PhMeSiC1, with Li produces : mainly phenyl-undeca-methyl-cyclo-hexasilane3 5 Me,SiCl,

+ PhMeSiC1,

-BPh

(Here and elsewhere in these sections, dots in structural formulas indicate Si atoms bound to a sufficient number of alkyl groups to bring the valence to four.) (R. WEST)

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc.

68

15.2. Ring-Ring and Ring-Polymer Interconversions 15.2.4. In Silicon-Silicon Systems 15.2.4.1. Formation of Cyclic Silicon-Silicon Systems

TABLE2. PRODUCTS FROM EtzSiC1, CONDENSATION WITH ALKALI METALS'S~ ~~

(EtZSi),

Metal

T(T)

2.0 Li 2.0 Li 2.2 Li 2.2 K 2.2 Na

0 -40 0 65 110

Solvent THF THF THF THF C,H,

n

=4

0 0 0 0 56

(7%

5

6

7

8

14 7 29 91 9

4 15 4 Trace 0

41 33 22 0 0

8 2 2 0 0

Large rings may be formed in nonequilibrium syntheses; e.g.; all rings from n = 4 to n = 38 are found in the mixture of (EtMeSi), isomer produced from EtMeSiCl, and 2.0 equiv K 5 . Large rings are also formed in the synthesis of organosilicon rotamers", (R. WEST)

1. H. Watanabe, T. Muraoka, M. Kageyama, K. Yoshizumi, Y. Nagai, Organometallics, 3, 141 (1984) and refs. therein. 2. H. Watanabe, J. Inose, K. Fukushima, Y. Kougo, Y. Nagai, Chem. Lett., 1711 (1983). 3. Y. Nagai, H. Watanabe, M. Matsumoto, in Silicon Chemistry, J. Y. Corey, E. R. Corey, P. P. Gaspar, eds., Ellis Horwood, 1988, p. 247. 4. S. M. Chen, R. West, Organomet. Synth., 4, 506, (1988). 5. A. Katti, C. W. Carlson, R. West, J. Organomet. Chem., 271, 353 (1984). 6. C. W. Carlson, R. West, Organometallics, 2, 1792 (1983). 7. B. J. Helmer, R. West, Organometallics, I , 1458 (1982). 8. A. Schafer, M. Weidenbruch, K. Peters, H. G. von Schnering, Angew. Chem., Int. Ed. Engl., 23, 302 (1984). 9. H. Matsumoto, A. Sakamoto, Y. Nagai, J. Chem. SOC.,Chem. Commun., 1468 (1986). 10. S. Masamune, H. Tobita, S. Murakami, J. Am. Chem. SOC.,105,6524 (1983). 11. H. Watanabe, Y. Kougo, Y. Nagai, J. Chem. SOC.,Chem. Commun., 66 (1984). 12. Y . 4 . Chen, P. P. Gaspar, Organometallics, 1, 1410 (1987). 13. C. W. Carlson, Z.-H. Zhang, R. West, Organometallics, 2, 453 (1983). 15.2.4.1.4. Alkylarylcyclosilanes.

Condensation of PhMeSiCl, with Li gives of the all-trans isomer of (PhMeSi), in 10% yield'. The reaction of PhMeSiC1, with Li and Ph,SiSiMe, catalyst produces high yields of (PhMeSi), and (PhMeSi), as a complex mixture of isomers'. Cocondensation of different dichlorosilanes is also used to synthesize substituted cyclics; e.g., treatment of a 5 : 1 mixture of Me,SiCl, and PhMeSiC1, with Li produces : mainly phenyl-undeca-methyl-cyclo-hexasilane3 5 Me,SiCl,

+ PhMeSiC1,

-BPh

(Here and elsewhere in these sections, dots in structural formulas indicate Si atoms bound to a sufficient number of alkyl groups to bring the valence to four.) (R. WEST)

15.2. Ri ng-Ri ng and Ri ng-Pol y mer Interconversions 15.2.4. In Silicon-Silicon Systems 15.2.4.2. Reactions of Cyclosilanes.

69

1. E. Hengge, E. Lanzen, Synth. Inorg. Metal-Org. Chem., 3, 93 (1972). 2. S. M. Chen, L. D. David, K. J. Haller, C. L. Wadsworth, R. West, Organometallics, 2,409 (1983). 3. B. J. Helmer, R. West, J. Organomet. Chem., 236, 21 (1982).

15.2.4.2. Reactions of Cyclosilanes. 15.2.4.2.1. Substitutional Reactions.

The well-known cleavage of phenyl groups from Si under acidic conditions yields functionalized cyclosilanes'*'. Treatment of (Ph,Si), with HI yields the corresponding pentaiodo compound3, in which one I is on each Si atom. The latter can be used to synthesize other substituted rings:

A similar reaction is carried out with (Ph,Si), and HBr, forming (PhSiBr), which may also be carried on to other cyclic compounds4. Upon treatment with HCl-AlC1, or HBr-AlBr,, all three perphenyl rings are transformed to their perhalo derivatives, which may be converted to the cyclic hydrosilanes with LiAlH, l q 5 : (Ph,Si),

- - HX

AlX3

LiAIH4

(Six,),

(H,Si),

The mixed arylalkylcyclosilane (PhMeSi), undergoes similar cleavage of the phenyl rings6: (PhMeSi),

HI

(MeSiI),

MeLi

(Me,Si),

(4

Although a more difficult reaction, substitution of methyl groups on (Me,Si), can also be accomplished. The reagent of choice is HSiCl, in the presence of a PtCl, catalyst' : HSiCl3 Me11Si6C1

benzene

Under forcing conditions additional methyl groups are replaced, leading utimately' to (MeSiCl),. Replacement of methyl by C1 in (Me$), also takes place with HCl-AlCl,, but rearranged products are also obtainedg. Photochemical coupling of the hydride of the six-membered ring, using t-butoxy radical as a hydrogen-abstracting reagent, produces a bi-ring compoundlOsll:

Substituted cyclosilanes are known. The chloro-substituted six-membered ring undergoes the usual transformations of chlorosilanes (Scheme 1).

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc. 15.2. Ri ng-Ri ng and Ri ng-Pol y mer Interconversions 15.2.4. In Silicon-Silicon Systems 15.2.4.2. Reactions of Cyclosilanes.

69

1. E. Hengge, E. Lanzen, Synth. Inorg. Metal-Org. Chem., 3, 93 (1972). 2. S. M. Chen, L. D. David, K. J. Haller, C. L. Wadsworth, R. West, Organometallics, 2,409 (1983). 3. B. J. Helmer, R. West, J. Organomet. Chem., 236, 21 (1982).

15.2.4.2. Reactions of Cyclosilanes. 15.2.4.2.1. Substitutional Reactions.

The well-known cleavage of phenyl groups from Si under acidic conditions yields functionalized cyclosilanes'*'. Treatment of (Ph,Si), with HI yields the corresponding pentaiodo compound3, in which one I is on each Si atom. The latter can be used to synthesize other substituted rings:

A similar reaction is carried out with (Ph,Si), and HBr, forming (PhSiBr), which may also be carried on to other cyclic compounds4. Upon treatment with HCl-AlC1, or HBr-AlBr,, all three perphenyl rings are transformed to their perhalo derivatives, which may be converted to the cyclic hydrosilanes with LiAlH, l q 5 : (Ph,Si),

- - HX

AlX3

LiAIH4

(Six,),

(H,Si),

The mixed arylalkylcyclosilane (PhMeSi), undergoes similar cleavage of the phenyl rings6: (PhMeSi),

HI

(MeSiI),

MeLi

(Me,Si),

(4

Although a more difficult reaction, substitution of methyl groups on (Me,Si), can also be accomplished. The reagent of choice is HSiCl, in the presence of a PtCl, catalyst' : HSiCl3 Me11Si6C1

benzene

Under forcing conditions additional methyl groups are replaced, leading utimately' to (MeSiCl),. Replacement of methyl by C1 in (Me$), also takes place with HCl-AlCl,, but rearranged products are also obtainedg. Photochemical coupling of the hydride of the six-membered ring, using t-butoxy radical as a hydrogen-abstracting reagent, produces a bi-ring compoundlOsll:

Substituted cyclosilanes are known. The chloro-substituted six-membered ring undergoes the usual transformations of chlorosilanes (Scheme 1).

70

15.2. Ring-Ring and Ring-Polymer Interconversions 15.2.4. In Silicon-Silicon Systems 15.2.4.2. Reactions of Cyclosilanes. ~

~~

Scheme 1

Polycyclic silanes are also formed by treating undecamethylcyclohexasilanylpotassium with dihalosilicon compounds":

e-K + (ClSiMe,),

(f)

The five-membered ring (Me2Si), may also be chlorodemethylated with HSiC1,-PtCl,, but an alternative synthesis of the C1 derivative starting from (Me$), may be more convenient. When (Me&), is treated with AlCl, in the presence of Me,SiCl, it undergoes both ring contraction and regiospecific chlorodemethylation on the exocyclic Si: (Me,Si),

+ Me,SiCl

*''I3

e S i M e 2 C I + Me,Si

Chlorine in the product can be replaced by phenyl by treatment with phenylmagnesium halide reagent. The resulting compound undergoes base cleavage of the exocyclic Si-Si bond, so that it it can be converted to the ethoxy and eventually to the C1 derivati~e~~'~:

(R. WEST)

1. E. Hengge, J. Organomet. Chem. Libr., 9, 267 (1979). 2. E. Hengge, in Homoatomic Rings, Chains and Macromolecules of Main-Group Elements, A. L. Rheingold, ed., Elsevier, Amsterdam, 1977, p. 235. 3. E. Hengge, D. Kovar, H. P. Sollradl, Monatsh. Chem., 110, 805 (1979). 4. E. Hengge, H. G. Schuster, W. Peter, J. Organomet. Chem., 186, C45 (1980). 5. E. Hengge, D. Kovar, J . Organomet. Chem., 125, C29 (1977); Z . Anorg. Allg. Chem., 459, 123 (1979). 6. E. Hengge, E. Lanzen, Synth. Inorg. Metal-Org. Chem., 3,. 93 (1972). 7. M. Ishikawa, K. Kumada, J. Chem. Soc., Chem. Commun., 567 (1969); Synth. Inorg. Metal-Org. Chem., 1, 191, 229 (1971).

15.2. Ring-Ring and Ring-Polymer lnterconversions 15.2.4. In Silicon-Silicon Systems 15.2.4.2. Reactions of Cyclosilanes. 8. 9. 10. 11. 12.

71

E. Hengge, W. Peter, J. Organomet. Chem., 148, C22 (1978). M. Kumada, J. Organomet. Chem., 100, 127 (1975). M. Kumada, Intra-Sci. Chem. Rep., 7, 121 (1973). M. Ishikawa, M. Kumada, Adu. Organornet. Cheh., 19, 51 (1981). F. K. Mitter, E. Hengge, J. Organomet. Chem., 332, 47 (1982).

15.2.4.2.2. Ring-Cleavage Reactions.

The perphenylcyclosilanes, (Ph,Si), and (Ph,Si), , can be cleaved with halogenating agents, hydrogen halides or alkali metals to give linear silanes' :

- - 2 Li

Li(Ph,Si),Li

I2

(Ph,Si),

I(Ph,Si),I

(a)

Br(Ph,Si),Bt

(b)

1

H(Ph,Si),X,

where X = C1, Br, I; Li(Ph,Si),Li

- -2Li

Br2

(Ph,Si),

These cleavage reactions are more rapid for (Ph,Si), than for (Ph,Si),, because of ring strain in the four-membered ring. The six-membered ring (Ph,Si), is resistant to cleavage, producing only fragments. The action of alkali metals on a,w-dihaloperphenylpolysilanes converts them back to cyclic compounds. In a reaction series that must be complex, K,PtCl, converts Li(Ph,Si),Li to the five-membered ring, (Ph,Si), . This reaction is used in the p-tolyl series, to synthesize the otherwise inaccessible (p-Tol,Si), ': (p-Tol,Si),

Li

THF

Li(p-Tol,Si),Li

Kz[PtCkl

(p-Tol,Si),

The dilithio and dihalo derivatives of both the tetrasilane and pentasilane are intermediates for the synthesis of other catenated silanes, both linear and cyclic, e.g.lS3:

Li(Ph,Si),Li

-

-

Li(Ph,Si),Li

PhzSiHCl

(Me0)sPO

-

Me(Ph,Si),Me

H(Ph,Si),H

PCI 5

Cl(Ph,Si),CI

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc. 15.2. Ring-Ring and Ring-Polymer lnterconversions 15.2.4. In Silicon-Silicon Systems 15.2.4.2. Reactions of Cyclosilanes. 8. 9. 10. 11. 12.

71

E. Hengge, W. Peter, J. Organomet. Chem., 148, C22 (1978). M. Kumada, J. Organomet. Chem., 100, 127 (1975). M. Kumada, Intra-Sci. Chem. Rep., 7, 121 (1973). M. Ishikawa, M. Kumada, Adu. Organornet. Cheh., 19, 51 (1981). F. K. Mitter, E. Hengge, J. Organomet. Chem., 332, 47 (1982).

15.2.4.2.2. Ring-Cleavage Reactions.

The perphenylcyclosilanes, (Ph,Si), and (Ph,Si), , can be cleaved with halogenating agents, hydrogen halides or alkali metals to give linear silanes' :

- - 2 Li

Li(Ph,Si),Li

I2

(Ph,Si),

I(Ph,Si),I

(a)

Br(Ph,Si),Bt

(b)

1

H(Ph,Si),X,

where X = C1, Br, I; Li(Ph,Si),Li

- -2Li

Br2

(Ph,Si),

These cleavage reactions are more rapid for (Ph,Si), than for (Ph,Si),, because of ring strain in the four-membered ring. The six-membered ring (Ph,Si), is resistant to cleavage, producing only fragments. The action of alkali metals on a,w-dihaloperphenylpolysilanes converts them back to cyclic compounds. In a reaction series that must be complex, K,PtCl, converts Li(Ph,Si),Li to the five-membered ring, (Ph,Si), . This reaction is used in the p-tolyl series, to synthesize the otherwise inaccessible (p-Tol,Si), ': (p-Tol,Si),

Li

THF

Li(p-Tol,Si),Li

Kz[PtCkl

(p-Tol,Si),

The dilithio and dihalo derivatives of both the tetrasilane and pentasilane are intermediates for the synthesis of other catenated silanes, both linear and cyclic, e.g.lS3:

Li(Ph,Si),Li

-

-

Li(Ph,Si),Li

PhzSiHCl

(Me0)sPO

-

Me(Ph,Si),Me

H(Ph,Si),H

PCI 5

Cl(Ph,Si),CI

72

15.2. Ring-Ring and Ring-Polymer lnterconversions 15.2.4. In Silicon-Silicon Systems 15.2.4.2. Reactions of Cyclosilanes.

By similar reactions, heterocyclic arylpolysilanes are synthesized, including compounds containing TiZ,C, N, S, P, Ge, Sn and B atoms in the ring'94.A related ring-closure gives a heterocycle with a Pt atom in the ring2:

The cleavage of (Me,Si), to form a,w-dihalopermethylpolysilanesuses as halogenating agentss-" Cl,, Br,, I,, PCl, and SO,Cl,. The initial breaking of the (Me,%), ring to form the 1,6-dihalo compound is followed by cleavage of Si-Si bonds in the chain, leading to smaller fragments. This reaction sequence is followed for the cleavages by C1, lo and Br, 'I. After the initial ring-opening, further breaking of Si-Si bonds is random, except that cleavage of the terminal disilane link is slow. Therefore, Cl(SiMe,),Cl reacts to give ca. equal amounts of Cl(Me,Si),Cl, Cl(Me,Si),Cl and Cl(Me,Si),CI (dotted lines indicate positions of bond-breaking): (Me,Si),

c12

-C 1

I

,

.

.

I

,

c

1

2Cl./-./m

+

C l A C 1

+

Cl*C1

6)

Further reaction converts the entire mixture to CI(Me,Si),Cl in high yield. The appearance and disappearance of the several compounds can be followed by gas chromatography, as shown in Fig. 1. If the reaction is stopped at the appropriate time, a maximum yield of 35% Cl(Me,Si),Cl is obtained". Similar cleavage reactions produce CI(Me,Si),Cl from (Me,%),, Cl(Me,Si),Cl from (Me,Si),, etc. For the latter a high (>go%) yield of Cl(Me,Si),Cl can be obtained because the cleavage of the fivemembered ring is more rapid than subsequent chlorination. The dihalo intermediates, especially Cl(Me,Si),Cl, are useful in synthesizing longchain permethylsilanes. Treatment of Cl(Me,Si),Cl with 1 equiv CH3MgI gives 90% of the monochloro compound CH,(Me,Si),Cl 1 2 , which can be coupled with itself or with the dichloro compound. Linear polysilanes containing up to 24 Si atoms are prepared in this way',: CH,(Me,Si),Cl

+ Cl(Me,Si),Cl

-

CH,(Me,Si),,CH,

(j)

CH,(Me,Si),&H, CH,(Me,Si),,CH,

Obtained from Et,SiCl, and Na metal, (Et,Si), is quite reactive toward Si-Si cleavage because of ring strain14. Several ring-cleavage reactions of this compound are summarized in Scheme 1. Cyclotrisilanes generally are highly susceptible to cleavage reactions. The reactions of (t-Bu,Si), are reviewed' 5 .

15.2.4. In Silicon-Silicon Systems 15.2.4.2. Reactions of Cyclosilanes. 15.2.4.2.2. Ring-Cleavage Reactions.

1.0(Si6Me12 0

#\ \ \

0.5

.

Figure 1. Yields of Cl(Me,Si),CI compounds from chlorination of (Me,Si), as a function of amount of C1, added, from ref. 10.

H(Et,Si),X

H(Et,Si),H

(Et,Si),O

HOAc

RO(Et,Si),H

,

X( Et Si),X (X = C1, Br, I)

Scheme 1

CIEt,SiO(EtpSi),C1

73

15.2. Ring-Ring and Ring-Polymer lnterconversions 15.2.4. In Silicon-Silicon Systems 15.2.4.2. Reactions of Cyclosilanes.

74

Several ring-cleavage reactions of (i-PrzSi)4have been studied16: HCl-c&

(i-PrzSi)4--

H(i-Pr,Si),Cl 93 %

(k)

-

(i-PrzSi)4

12-CsH6

I(i-Pr,Si)J 95 %

(i-PrzSi)4 where n = 1-4.

rn-ClCsH4C03H

(i-PrzSi)40n

(R. WEST)

1. E. Hengge, Top. Curr. Chem., 51, 1 (1974); J. Organomet. Chem. Libr., 9, 267 (1979). 2. M. F. Lemanski, E. P. Schramm, Inorg. Chem., 15, 1489,2515 (1976). 3. E. Hengge, D. Wofler, J. Organomet. Chem., 66,413 (1974). 4. E. Hengge, R. Sommer, Monatsh. Chem., 108, 1413 (1977). 5. E. Hengge, G. Kollman, Monatsh. Chem., 109, 477 (1978). 6. E. Hengge, H. Stiiger, Monatsh. Chem., 111, 1043 (1980). 7. M. S. Holtman, E. P. Schramm, J. Organomet. Chem., 187, 197 (1980). 8. H. Gilman, S. Inoue, J. Organomet. Chem., 15,237 (1968). 9. D. K. Sen, D. Ballard, H. Gilman, J. Organomet. Chem., 29, 3418 (1964). 10. W. Wojnowski, C. J. Hurt, R. West, J. Organomet. Chem., 124,271 (1977). 11. C. H. Middlecamp, W. Wojnowski, R. West, J. Organomet. Chem., 140, 133 (1977). 12. W. G. Boberski, A. L. Allred, J. Am. Chem. Soc., 96, 1244 (1974). 13. W. G. Boberski, A. L. Allred, J. Organomet. Chem., 71, C27 (1974). 14. C. W. Carlson, R. West, Organometallics, 2, 1798, 1801 (1983). 15. M. Weidenbruch, Comments Inorg, Chem., 5, 247 (1986). 16. H. Watanabe, T. Muraoka, M. Kageyame, Y. Nogai, J. Organomet. Chem., 216, C45 (1981). 15.2.4.2.3. Photolysis.

Irradiation of cyclosilanes with UV light (usually 254 nm) leads to ring contraction and elimination of silylene fragments. This process provides a valuable method both for generating silylenes and for synthesizing smaller cyclosilane rings from larger ones; e.g., photolysis of (Et,Si), provides the best route' to (Et,Si),, and photolysis of (Me,Si), gives (Me,Si), (see $15.2.4.1.2). Linear polysilanes also produce silylenes upon irradiation; this photolysis has been used for the matrix isolation of unstable silylenes' : Mes

\ /

Si(SiMe,),

hv, 3-MP

(Me,Si),

+ [Mes-Si-El

(a)

E

where Ex = CH,, H, C1, OMe, Ph, t-Bu, i-Pr, and 3-MP = 3-methylpentane, as well as for the synthesis of stable disilenes. Cyclotetrasilanes such as (Me,Si), undergo ring opening to 1,Csilyl diradicals upon

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc. 15.2. Ring-Ring and Ring-Polymer lnterconversions 15.2.4. In Silicon-Silicon Systems 15.2.4.2. Reactions of Cyclosilanes.

74

Several ring-cleavage reactions of (i-PrzSi)4have been studied16: HCl-c&

(i-PrzSi)4--

H(i-Pr,Si),Cl 93 %

(k)

-

(i-PrzSi)4

12-CsH6

I(i-Pr,Si)J 95 %

(i-PrzSi)4 where n = 1-4.

rn-ClCsH4C03H

(i-PrzSi)40n

(R. WEST)

1. E. Hengge, Top. Curr. Chem., 51, 1 (1974); J. Organomet. Chem. Libr., 9, 267 (1979). 2. M. F. Lemanski, E. P. Schramm, Inorg. Chem., 15, 1489,2515 (1976). 3. E. Hengge, D. Wofler, J. Organomet. Chem., 66,413 (1974). 4. E. Hengge, R. Sommer, Monatsh. Chem., 108, 1413 (1977). 5. E. Hengge, G. Kollman, Monatsh. Chem., 109, 477 (1978). 6. E. Hengge, H. Stiiger, Monatsh. Chem., 111, 1043 (1980). 7. M. S. Holtman, E. P. Schramm, J. Organomet. Chem., 187, 197 (1980). 8. H. Gilman, S. Inoue, J. Organomet. Chem., 15,237 (1968). 9. D. K. Sen, D. Ballard, H. Gilman, J. Organomet. Chem., 29, 3418 (1964). 10. W. Wojnowski, C. J. Hurt, R. West, J. Organomet. Chem., 124,271 (1977). 11. C. H. Middlecamp, W. Wojnowski, R. West, J. Organomet. Chem., 140, 133 (1977). 12. W. G. Boberski, A. L. Allred, J. Am. Chem. Soc., 96, 1244 (1974). 13. W. G. Boberski, A. L. Allred, J. Organomet. Chem., 71, C27 (1974). 14. C. W. Carlson, R. West, Organometallics, 2, 1798, 1801 (1983). 15. M. Weidenbruch, Comments Inorg, Chem., 5, 247 (1986). 16. H. Watanabe, T. Muraoka, M. Kageyame, Y. Nogai, J. Organomet. Chem., 216, C45 (1981). 15.2.4.2.3. Photolysis.

Irradiation of cyclosilanes with UV light (usually 254 nm) leads to ring contraction and elimination of silylene fragments. This process provides a valuable method both for generating silylenes and for synthesizing smaller cyclosilane rings from larger ones; e.g., photolysis of (Et,Si), provides the best route' to (Et,Si),, and photolysis of (Me,Si), gives (Me,Si), (see $15.2.4.1.2). Linear polysilanes also produce silylenes upon irradiation; this photolysis has been used for the matrix isolation of unstable silylenes' : Mes

\ /

Si(SiMe,),

hv, 3-MP

(Me,Si),

+ [Mes-Si-El

(a)

E

where Ex = CH,, H, C1, OMe, Ph, t-Bu, i-Pr, and 3-MP = 3-methylpentane, as well as for the synthesis of stable disilenes. Cyclotetrasilanes such as (Me,Si), undergo ring opening to 1,Csilyl diradicals upon

15.2. Ring-Ring and Ring-Polymer Interconversions 15.2.4. In Silicon-Silicon Systems 15.2.4.3. Ring-Ring and Chain-Ring Interconversion.

75

irradiation. Certain sterically hindered cyclotetrasilanes, however, photolyze to produce first the cyclotrisilane and then the disilene, as silylene fragments are eliminated4. (R,Si),

hv

254nm [R,Si]

+ (R,Si),

hv

[R,Si]

+ [R,Si=SiR,]

(b)

where R = i-Pr, i-Bu; CH,SiMe,, CHMeEt,. The silylenes and disilenes are shown to be present by trapping reactions. Photolysis of the stable peralkylcyclotrisilanes [(Me,CCH,),Si],5, ( ~ - B u , S ~and ) ~ ~[(Et,CH),Si], also produce the disilene and are unstable : (R,Si),

hv

[R,Si]

+ [R,Si=SiR,]

(c)

where R = Me3CCH,, t-Bu, Et,CH. Photolysis of hindered arylcyclotrisilanes provides an important pathway to stable disilenes: (RRSi),

hv

RRSi=SiRR

(4

where R = R = mes, 2,6-dimethylphenyl, 2,6-diethylphenyl. An interesting example is the photolysis of the two stereoisomers of [t-Bu(Mesityl)Si],, which takes place with retention of configuration to give the Z and E stereoisomers of the disilene'. (R. WEST)

1. C. W. Carlson, R. West, Organometallics, 2, 1792 (1983). 2. M. J. Michalczyk, M. J. Fink, D. J. De Young, C. W. Carlson, K. M. Welsh, R. West, J. Michl, Silcion, Germanium, Tin Lead Compds., 9, 75 (1986). 3. R. West, Angew. Chem., Int. Ed. Engl., 26, 1201 (1987) 4. H. Watanabe, Y. Kougo, M. Kato, M. Kuwabara, T. Okawa, Y. Nagai, Bull. Chem. SOC.Jpn., 57, 3019 (1984). 5. H. Watanabe, T. Okawa, M. Kato, Y. Nagai, J. Chem. SOC.,Chem. Commun., 781 (1983) 6. A. Schafer, M. Weidenbruch, S . Pohl, J. Organomet. Chem., 282, 305 (1985). 7. S. Masamune, H. Tobita, S. Murakami, J. Am. Chem. SOC.,105,6524 (1983). 8. S. Murakami, S. Collins, S. Masamune, Tetrahedron Lett., 2131 (1984).

15.2.4.3. Ring-Ring and Chaln-Ring Interconversion.

Cyclic polysilanes are more stable than chain polymers. Attempts to produce high polymers of polysilanes by ring opening of cyclic compounds are unsuccessful, but conversions from linear polymers or chain oligomers to cyclic compounds are feasible. Most of the limited work in this area is done with permethylpolysilanes. As mentioned in $15.2.4.1.2, the preparation of methylcyclosilanes from Me,SiCl, and alkali metals proceeds through polymer formation followed by depolymerization, at least in some cases Polymer obtained by Na condensation of Me,SiCl, can also be equilibrated with strong bases (Na-K, Ph,SiLi) or cracked thermally, producing mainly the five- and six-membered ring oligomers3: '3'.

A , 300°C

where n

65T

=

5,6. Some rearrangement products are also produced in the thermal cracking.

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc. 15.2. Ring-Ring and Ring-Polymer Interconversions 15.2.4. In Silicon-Silicon Systems 15.2.4.3. Ring-Ring and Chain-Ring Interconversion.

75

irradiation. Certain sterically hindered cyclotetrasilanes, however, photolyze to produce first the cyclotrisilane and then the disilene, as silylene fragments are eliminated4. (R,Si),

hv

254nm [R,Si]

+ (R,Si),

hv

[R,Si]

+ [R,Si=SiR,]

(b)

where R = i-Pr, i-Bu; CH,SiMe,, CHMeEt,. The silylenes and disilenes are shown to be present by trapping reactions. Photolysis of the stable peralkylcyclotrisilanes [(Me,CCH,),Si],5, ( ~ - B u , S ~and ) ~ ~[(Et,CH),Si], also produce the disilene and are unstable : (R,Si),

hv

[R,Si]

+ [R,Si=SiR,]

(c)

where R = Me3CCH,, t-Bu, Et,CH. Photolysis of hindered arylcyclotrisilanes provides an important pathway to stable disilenes: (RRSi),

hv

RRSi=SiRR

(4

where R = R = mes, 2,6-dimethylphenyl, 2,6-diethylphenyl. An interesting example is the photolysis of the two stereoisomers of [t-Bu(Mesityl)Si],, which takes place with retention of configuration to give the Z and E stereoisomers of the disilene'. (R. WEST)

1. C. W. Carlson, R. West, Organometallics, 2, 1792 (1983). 2. M. J. Michalczyk, M. J. Fink, D. J. De Young, C. W. Carlson, K. M. Welsh, R. West, J. Michl, Silcion, Germanium, Tin Lead Compds., 9, 75 (1986). 3. R. West, Angew. Chem., Int. Ed. Engl., 26, 1201 (1987) 4. H. Watanabe, Y. Kougo, M. Kato, M. Kuwabara, T. Okawa, Y. Nagai, Bull. Chem. SOC.Jpn., 57, 3019 (1984). 5. H. Watanabe, T. Okawa, M. Kato, Y. Nagai, J. Chem. SOC.,Chem. Commun., 781 (1983) 6. A. Schafer, M. Weidenbruch, S . Pohl, J. Organomet. Chem., 282, 305 (1985). 7. S. Masamune, H. Tobita, S. Murakami, J. Am. Chem. SOC.,105,6524 (1983). 8. S. Murakami, S. Collins, S. Masamune, Tetrahedron Lett., 2131 (1984).

15.2.4.3. Ring-Ring and Chaln-Ring Interconversion.

Cyclic polysilanes are more stable than chain polymers. Attempts to produce high polymers of polysilanes by ring opening of cyclic compounds are unsuccessful, but conversions from linear polymers or chain oligomers to cyclic compounds are feasible. Most of the limited work in this area is done with permethylpolysilanes. As mentioned in $15.2.4.1.2, the preparation of methylcyclosilanes from Me,SiCl, and alkali metals proceeds through polymer formation followed by depolymerization, at least in some cases Polymer obtained by Na condensation of Me,SiCl, can also be equilibrated with strong bases (Na-K, Ph,SiLi) or cracked thermally, producing mainly the five- and six-membered ring oligomers3: '3'.

A , 300°C

where n

65T

=

5,6. Some rearrangement products are also produced in the thermal cracking.

76

15.2. Ring-Ring and Ring-Polymer Interconversions 15.2.4. In Silicon-Silicon Systems 15.2.4.3. Ring-Ring and Chain-Ring Interconversion. TABLE1. TION FOR

ENTHALPIESAND ENTROPIESOF CYCLOPOLYSILANES AND CYCLOALKANES

Redistribution

5 (Me&), 7 (Me&), 7 (Me,Si),

6 (Me,Si), 6 (Me,%), 5 (Me,Si), 6 C,H,O 6 C,H,4 5 C,H,4

a a

C6H12

7 C6Hl2

AH (kJ mol-')

AS (J K-mol-')

- 74 - 13 i77 - 140 - 140 50

- 84 138 f213 -205 -29 f218 f

+

C5H10

REDISTRIBU-

In THF solution.

* In liquid phase.

Ring-ring interconversions are important in polysilane chemistry. The enthalpy and entropy changes for ring interconversion in the equilibrium between the permethyl (Me,Si), rings in THF solution catalyzed by Na-K4 are listed in Table 1 for ring interconversion are listed in Table 1, together with similar data for the cycloalkanes (as pure liquids). Thus (Me,Si), is the most stable of the rings, although the difference between n = 6 and 7 is slight. The destabilization of n = 5 and 7 relative to n = 6 is due to methyl-methyl interactions, combined with some angle strain in n = 5. Similar reasons are used to explain why cyclohexane is more stable than cycloheptane or cyclopentane5. Comparing the methylcyclosilane and cycloalkane families, the six-membered ring is the most stable in each series, but the differences between compounds are less among the cyclosilanes because methyl-methyl steric interactions in permethylsilanes are actually lower than H-H repulsions in alkanes6. Ring-ring equilibration among other cyclosilanes is known only semiquantitatively, with results shown in Table 2. As the steric requirement of the substituent groups on Si is increased, the thermodynamically preferred ring size becomes smaller. Thus for the (Et,Si), and (Ph,Si), series the most stable ring is n = 5, while for (t-BuMeSi),, n = 4 is favored, and for still larger substituents only the three-membered ring is formed, e.g.,

TABLE 2. PRODUCTS OF RING-RING EQUILIBRATION REACTIONS IN THF, 30°C

Wt % at equilibrium for System

Catalyst

(Me,Si), (MeEtSi), (Et2Si)" (PhMeSi), (t-BuMeSi), (Ph,Si),

K K K Li, Ph,SiLi Li, Ph,SiLi Li, Ph,SiLi

n

=4

5

6

7

Ref.

2

7 61 96 27. 1 70

92 38 2 67

11 Trace -

4 7

-

> 99 10

-

-

Trace

-

8 9 10 9

15.2. Ring-Ring and Ring-Polymer Interconversions 15.2.4. In Silicon-Silicon Systems 15.2.4.3. Ring-Ring and Chain-Ring Interconversion. TABLE3. RING-RING EQUILIBRATION:% (t-BuSiMe), ISOMERS'O

YIELDS

77

OF

~

Reaction t-BuMeSiCl, + Li,THF" Statistical distribution Equilibration, Ph,SiLi Photoequilibration

A

B

C

D

9 12.5 97 65

49 50 3 30

29 25 Trace 5

13 12.5 0

~~~

0

a 3% of (t-BuMeSi), IS also produced Figures for (t-BuMeSi), have been normalized to 100%;the actual yield of (t-BuMeSi), isomers is 85%.

(t-Bu,Si), and (Mes,Si),. As explained in $15.2.5.1, ring-ring equilibration may be useful in synthesizing the most stable cyclooligomer, when this is desired. With the methyl-butylcyclosilanes'O~' Li condensation produces a nearly statistical mixture of (t-BuMeSi), isomers (see Table 3), together with small amounts of the (t-BuMeSi), isomers. The four-ring isomers are identified by a combination of 'H NMR spectroscopy and x-ray crystallography. Compounds B and C can be isolated from this mixture by fractional crystallization. Equilibration with Ph,SiLi converts the mixture almost entirely to the most stable all-trans isomer, A, which can then be purified by crystallization. This isomer is less crowded than the others because it can adopt a folded conformation in which each tert-butyl group is in a pseudoequatorial position (see Fig. 1)lZ. Irradiation of the (t-BuMeSi), isomers with UV light at I. = 300 nm leads to ring opening to the 1P-diradical and then to reclosing of the ring lead to a photostationary state". The composition of the resulting (pseudo) photoequilibrium mixture is also shown in Table 3.

',

Figure 1. ORTEP diagram of molecular structure of all-trans isomer of (t-BuMeSi),".

78

15.2. Ring-Ring and Ring-Polymer lnterconversions 15.2.4. In Silicon-Silicon Systems 15.2.4.3. Ring-Ring and Chain-Ring Interconversion.

Another kind of equilibration, catalyzed by AlCl,, interconverts ring compounds of different structure. This reaction is effective only for permethylcyclosilanes13~14. Some examples are: (Me,Si), (Me,Si),

(Me,Si),

(Me,Si),

-0-a -J3( --13(

The thermodynamic properties for chain branching in methylpolysilanes applies also to acyclic compound^'^:

x ew-

(h)

(Me,Si),Si

t

(R. WEST)

1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15.

R. West, L. F. Brough, W. Wojnowski, Inorg. Synth., 19, 265 (1979). M. Kumada, K. Sakamoto, M. Ishikawa, J. Organomet. Chem., 17, 231 (1969). E. Carberry, R West, J. Am. Chem. SOC.,91, 5440 (1969). L. F. Brough, R. West, J. Organomet. Chem., 194, 134 (1980). E. L. Eliel, Elements of Stereochemistry, John Wiley and Sons, New York, 1966. D. A. Stanislawski, A. C. Buchanan, 111, R. West, J. Am. Chem. SOC.,100,7791 (1978) and refs. therein. A. Katti, C. W. Carlson, R. West, J. Organomet. Chem., 271, 353 (1984). C. W. Carlson, R. West, Organometallics, 2, 1798 (1983). S. M. Chen, L. D. David, K. J. Haller, C. J. Wadsworth, R. West, Organometallics, 2,409 (1983). B. Helmer, R. West, Organometallics, I , 1458 (1982). M. Bierbaum, R. West, J. Organomet. Chem., 131, 179, 189 (1977). C. J. Hurt, J. C. Calabrese, R. West, J. Organomet. Chem., 91, 273 (1975). M. Kumada, J. Organomet. Chem., 100, 127 (1975). T. A. Blinka, R. West, Organometallics, 5, 128 (1986). M. Ishikawa, J. Kyoda, H. Ikeda, H. Kotake, T. Hashimoto, M. Kumada, J. Am. Chem. Soc., 103,4845 (1981).

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc. 15.2. Ring-Ring and Ring-Polymer Interconversions 15.2.4. In Silicon-Silicon Systems 15.2.4.4. Cage Polysilanes.

79

15.2.4.4. Cage Polysilanes.

Little is yet known of cage polysilanes, although many such structures should be possible. In the permethyl series when MezSiC1, and MeSiC1, are cocondensed with Na-K alloy in THF in the presence of naphthalene, six different polycyclic compounds are produced, all in small yield, along with (Me$), and polymer': MeJiC1,

+ MeSiC1,

-Na-K, THF

Me,,Si, Me,,$, Me,,Silo Me18Silo MelBSil1 MezZSil,

(a)

5% 2 0.5 4

0.5 0.1

The products are separated by a combination of fractional crystallization and preparative gas chromatography. Structures are known with certainty' only for the first two; in addition the cis- or trans-decalin stucture is probable for Me,,Si,,:

Me,,Si,

Me, ,Si9

Me18%

0

An adamantane structure is consistent with the formula Me,,Silo, but the NMR spectrum of the compound is too complex for this formulation. Cage polysilanes undergo only two chemical reactions. Photolysis of Me,,Si, leads to silylene extrusion and the formation of the 2.2.1-heptasilabicyclo-heptane':

and, assuming that the decalin structure is correct for Mel,Silo, treatment of this substance with AlCl, gives spirosilane products3:

Condensation of i-PrSiC1, and i-Pr,SiCl, gives two polycyclic products in small yieldsz4: R

I

RSiCl,

-

I

R,Si-Si-SiR,

l

l

+

R,Si-SSi-SiR,

I

R

I

R,Si-Si-Si-SiR, R,Si-Si-Si-SiR,

I

80

15.2. Ring-Ring and Ring-Polymer Interconversions 15 2.4. In Silicon-Silicon Systems 15.2.4.5. Silane High Polymers.

where R = i-OPr. A highly strained bicyclo compound is synthesized according to5s6: Ar,Si-Si(Bu-t)Cl

I

I

Ar,Si-SiAr

Li, c

Ar

Bu-t Ar

lo~*

THF

t-Bu

and a remarkable octasilacubane is obtained from7: Na, toluene

t-BuMe,SiBr3w (R. WEST)

1. 2. 3. 4. 5. 6. 7.

R. West, A. Indriksons, J. Am. Chem. SOC.,94, 6110 (1972). W. Stallings, J. Donahue, Inorg. Chem., 15, 524 (1976). M. Kumada, J. Organomet. Chem., 100, 127 (1975). H. Matsumoto, H. Miyamoto, N. Kojima, N. Nagai, J. Chem. SOC.,Chem. Commun., 1316 (1987). S. Masamune, Y. Kabe, S. Collins, D. J. Williams, R. Jones, J. Am. Chem. SOC., 107, 5552 (1988). R. Jones, D. J. Williams, Y. Kabe, S . Masamune, Angew. Chem., Int. Ed. Engl., 25, 173 (1986). H. Matsumoto, K. Higuchi, Y. Hochino, H. Koike, Y. Naoi, Y. Nagai, J. Chem. SOC., Chem. Commun., 1083 (1988).

15.2.4.5. Silane High Polymers.

The chemistry of polysilane high polymers is undergoing rapid development because of their scientific interest and technological applications'". The usual synthesis of polysilanes is by the condensation of diorganodichlorosilanes with finely divided sodium metal in an inert diluent at T > 100°C. Either homopolymers or copolymers can be made: R

RR'SiC1,

R

R"R"'SiC1,

R'

R

R"

R"'

The organic groups can be varied widely as long as they withstand treatment with Na metal. Sterically bulky substituents, however, give only cyclic and linear oligomers rather than high polymers. The properties of the resulting polymers depend on the nature of the substituent groups. The dimethyl, diethyl and diphenyl polymers are exceptional in being highly crystalline, insoluble, intractable materials. Lowering the symmetry or increasing the length of the alkyl groups decreases the crystallinity and leads to the thermoplastic properties typical of most polysilanes. n-Alkylmethyl polymers such as (n-C,H,,SiMe), have glass transitions below RT and so behave as elastomers; dialkyl polymers such as [(n-C,H,),Si], are crystalline but soluble and meltable; arylmethyl polymers such as (PhMeSi), are partially crystalline and some copolymers, such as the phenylmethyldimethyl polymer (PhMeSi),(Me,Si),(m/n x l), are amorphous.

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc. 80

15.2. Ring-Ring and Ring-Polymer Interconversions 15 2.4. In Silicon-Silicon Systems 15.2.4.5. Silane High Polymers.

where R = i-OPr. A highly strained bicyclo compound is synthesized according to5s6: Ar,Si-Si(Bu-t)Cl

I

I

Ar,Si-SiAr

Li, c

Ar

Bu-t Ar

lo~*

THF

t-Bu

and a remarkable octasilacubane is obtained from7: Na, toluene

t-BuMe,SiBr3w (R. WEST)

1. 2. 3. 4. 5. 6. 7.

R. West, A. Indriksons, J. Am. Chem. SOC.,94, 6110 (1972). W. Stallings, J. Donahue, Inorg. Chem., 15, 524 (1976). M. Kumada, J. Organomet. Chem., 100, 127 (1975). H. Matsumoto, H. Miyamoto, N. Kojima, N. Nagai, J. Chem. SOC.,Chem. Commun., 1316 (1987). S. Masamune, Y. Kabe, S. Collins, D. J. Williams, R. Jones, J. Am. Chem. SOC., 107, 5552 (1988). R. Jones, D. J. Williams, Y. Kabe, S . Masamune, Angew. Chem., Int. Ed. Engl., 25, 173 (1986). H. Matsumoto, K. Higuchi, Y. Hochino, H. Koike, Y. Naoi, Y. Nagai, J. Chem. SOC., Chem. Commun., 1083 (1988).

15.2.4.5. Silane High Polymers.

The chemistry of polysilane high polymers is undergoing rapid development because of their scientific interest and technological applications'". The usual synthesis of polysilanes is by the condensation of diorganodichlorosilanes with finely divided sodium metal in an inert diluent at T > 100°C. Either homopolymers or copolymers can be made: R

RR'SiC1,

R

R"R"'SiC1,

R'

R

R"

R"'

The organic groups can be varied widely as long as they withstand treatment with Na metal. Sterically bulky substituents, however, give only cyclic and linear oligomers rather than high polymers. The properties of the resulting polymers depend on the nature of the substituent groups. The dimethyl, diethyl and diphenyl polymers are exceptional in being highly crystalline, insoluble, intractable materials. Lowering the symmetry or increasing the length of the alkyl groups decreases the crystallinity and leads to the thermoplastic properties typical of most polysilanes. n-Alkylmethyl polymers such as (n-C,H,,SiMe), have glass transitions below RT and so behave as elastomers; dialkyl polymers such as [(n-C,H,),Si], are crystalline but soluble and meltable; arylmethyl polymers such as (PhMeSi), are partially crystalline and some copolymers, such as the phenylmethyldimethyl polymer (PhMeSi),(Me,Si),(m/n x l), are amorphous.

15.2. Ring-Ring and Ring-Polymer Interconversions 15.2.4. In Silicon-Silicon Systems 15.2.4.5. Silane High Polymers.

81

The synthesis of polymers according to Eqs. (a) and (b) is always accompanied by formation of cyclic and linear oligomers, and yields of high polymer are often low. The sodium coupling reaction is difficult to control, and results are not fully reproducible. Polymer molecular weight distributions, as determined by gel permeation chromatography, are typically bimodal, suggesting that at least two mechanisms for chain extension may occur. The yield and molecular weight are influenced by bulk solvent effects, by the nature of the sodium dispersion, and by the rate and order of the addition of reagents3p4. Monomodal molecular weight distributions are obtained when the condensation is carried out near RT with ultrasound activation'. Polysilanes are produced by polymerization of silylenes, but this method is not generally used for polymer synthesis. A promising alternative method is transition-metalcatalyzed condensation of diorganosilanes, RRSiH, 6.

-

PhSiH,

-

(q5-Cp)zTiMez

(PhSiH),

where n 10. To date, however, only oligomers are obtained. Polysilanes are produced from thermolysis of polymeric mercury compounds, obtained from diorganosilanes and t-Bu,Hg : RR'SiH,

+

t-BuzHg

fRR'Si-Hg-f,

7 (RRSi),

This route has been used to prepare, e.g., (PhMeSi),. Crosslinked polysilane polymers useful as precursors for silicon carbide are obtained by redistribution from chl~rodisilanes~ : (MeSiCl,),

+ MeSiCl,SiClMe,

[~-Bu~P]CI

[Me,Si(MeSiCl), ,(MeSi),],

+ MeSiC1, + Me,SiCl,

(e)

Soluble crosslinked polysilane polymers have been made by reduction of n-hexyltrichlorosilane with liq Na-K alloy promoted by ultrasound [cf. Eq. (f)]". Polysilane polymers show reactivity similar to that of other liner polysilanes. The polymers are inert to air and moisture but in solution react with bases to cleave Si-Si bonds and release hydrogen. Peroxide oxidation of polysilanes converts Si-Si into siloxane, Si-0-Si, linkages. A few reactions of side-chain groups have been investigated. Homopolymers or copolymers containing cyclohexenylethyl groups add hydrogen halides under electronpain acceptor catalysis to give haloalkyl polymers" : Br

+Si+6Si+m ?ye

I

Me

I

Me

0

' HBr ~ ~ +Si+dSiX r a ye I

Me

1

Me

HCl

I I

f s i H sMe i t m

I

Me

Me

(f)

82

15.2. Ring-Ring and Ring-Polymer Interconversions 15.2.4.In Silicon-Silicon Systems 15.2.4.5. Silane High Polymers.

Polymers that contain Si-H groups show the usual reactivity of the silane hydrogen, e.g., addition of olefins with H,PtCl, catalysis12. Chloromethylation of phenylmethylpolysilane takes place’, :

Me

Me

Polysilanes’ containing phenolic -OH groups are synthesized by deprotonating trimethylsiloxyaryl substituents with methanolI4; the resulting polymers are soluble in base. Finally, aryl groups can be cleaved from the polysilane chain by trifluoroacetic acid, giving polymers with highly reactive OCOCF, groups attached to silicon”. The uses of polysilanes as photoresist materials16, and as photoinitiators’* both depend on the photoscission of the polymers by ultraviolet light. Initial steps in this process are chain cleavage to radicals and elimination of silylene, RR‘Si 19: R

-Si-Si-SiI l R

R

R

I 1 e R R ’ S i : + --Si-SiI I

R’ R‘

l

R

R’

-Si.

I I

R

R

I

I

I

+ eSi-Si-I

but complex subsequent reactions also take place”. Polysilanes are important as thermal precursors to high-purity silicon carbide”. In the best known process poly(dimethy1silylene) is heated under Ar at 320°C for several hours. Among the chemical changes is insertion of methylene groups into Si-Si bonds, producing a carbosilane polymer. Volatile components are removed by heating under vacuum, and the residual polymer is fractionated and drawn into fibers. After surface oxidation to prevent melting, the fibers are heated slowly to 1300”C, producing fibers of silicon carbide:

(Me,Si),

320T

H

I

fSi-CH,-);

I CH,

1300°C

S i c + CH,

+ H,

The copolymer (PhMeSi),(Me,Si), is also used as a silicon carbide precursor, especially for monolithic formed objects. (R. WEST)

1. R. West, J. Organomet. Chem., 300, 327 (1986). 2. M. Zeldin, K. J. Wynne, H. R. Allcock, eds., Inorganic and Organometallic Polymers, ACS Symp. Ser. 360, American Chemical Society, Washington, D.C., 1988. 3. J. M. Zeigler, Polym. Preprints, 27, 109 (1986); Macromolecules, in press.

15.2.5. in Boron-Nitrogen Systems. 15.2.5.1. Synthesis of Ring Compounds 15.2.5.1.1. Borazines and Related Compounds.

83

4. D. J. Worsfold, Ch. 8 in ref. 2. 5. H. K. Kim, K. Matyjaszewski, J. Am. Chem. SOC.,110, 3321 (1988). 6. C. T. Aitken, J. F. Harrod, E. Samuel, J. Organomet. Chem., 279, C11 (1985); J. Am. Chem. SOC., 108, 4059 (1986); Can. J. Chem., 64, 1677 (1986). 7. E. Hengge, F. K. Mitter, Monatsh. Chem., 117, 721 (1986). 8. J. Maxka, PhD. Thesis, University of Wisconsin, Madison, WI, 1988. 9. R. H. Baney, J. H. Gaul, Jr., T. K. Hilty, Orgunometallics, 2, 859 (1983). 10. P. A. Bianconi, T. W. Weidman, J. Am. Chem. Soc., 110, 2342 (1988) 11. H. Stuger, R. West, Macromolecules, 18, 2349 (1985). 12. X. H. Zhang, R. West, Macromolecules, in press. 13. H. Ban, K. Sukegawa, S. Tagawa, Macromolecules, 20, 1175 (1987). 14. R. Horiguchi, Y. Onishi, S. Hayase, Macromolecules, 21,304 (1988). 15. K. Matyjaszewski, Y-L. Chen, H. K. Kim, Ch. 6 in ref. 2. 16. R. D. Miller, D. Hofer, D. R. McKean, C. G. Willson, R. West, R. Trefonas, in “Materialsfor Microlithogruphy, ACS Syrnp. Ser. 266, L. Thompson, C. G. Willson, J. M. J. Frechet, eds., American Chemical Society, Washington, D.C., 1984., pp. 293-310. 17. R. D. Miller, J. F. Rabolt, R. Sooriyakumaran, W. Fleming, G. N. Fickes, B. L. Farmer, H. Kuzmany, Ch. 4 in ref. 2. 18. A. R. Wolff, R. West, Appl. Organomet. Chem., 1, (1987). 19. P. Trefonas, R. Miller, R. West, J. Am. Chem. Soc., 107, 2737 (1985). 20. J. Michl, J. W. Downing, T. Karatsu, A. J. McKinley, G. Poggi, G . M. Wallraf, R. Sooriyakumaran, R. D. Miller, Pure Appl. Chem., 60,959 (1988). 21. S. Yajima, K. Okamuro, J. Hayashi, M. Omori, J. Am. Ceram. Soc., 59,324 (1976).

15.2.5. in Boron-Nitrogen Systems. 15.2.5.1. Synthesis of Ring Compounds 15.2.5.1 .l.Borazines and Related Compounds.

The most widely studied and best characterized of the boron-nitrogen ring com-

I

I

pounds are the borazines, of generic formula (-N-B-)3. The six-membered ring is readily formed to yield borazine or substituted borazines directly via elimination reactions. Substituted amines or amine hydrochlorides react with boron trihalides or tetrahydroborate salts to yield borazines. Thus 1,3,5-trimethylborazine and 1,3,5trimethyl-2,4,6-trichloroborazine are produced in essentially quantitative yields by 1:

3 [MeNH3]C1

+ 3 Li[BH,]

3 [MeNH,]Cl

EtzO

+ 3 BC1,

PhCl

(MeNBH),

+ 3 LiCl + 9 H,

(MeNBCl),

-

(4

+ 9 HCl

The parent compound borazine, (HNBH), , can be prepared similarly’:

3 [NH,]Cl

+ 3 Na[BH,]

(HNBH),

+ 9 H,

(c>

However, the method is inefficient, and borazine itself is better prepared3 by NaBH, bond in the preformed ring compound (HNBCI),. The reduction of the B-Cl 2,4,6-trichloroborazine can be prepared by reaction of NH,Cl with BC1, or the and other B-trichloroborazines acetonitrile adduct of BCl, in c h l ~ r o b e n z e n e ~This .

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc. 15.2.5. in Boron-Nitrogen Systems. 15.2.5.1. Synthesis of Ring Compounds 15.2.5.1.1. Borazines and Related Compounds.

83

4. D. J. Worsfold, Ch. 8 in ref. 2. 5. H. K. Kim, K. Matyjaszewski, J. Am. Chem. SOC.,110, 3321 (1988). 6. C. T. Aitken, J. F. Harrod, E. Samuel, J. Organomet. Chem., 279, C11 (1985); J. Am. Chem. SOC., 108, 4059 (1986); Can. J. Chem., 64, 1677 (1986). 7. E. Hengge, F. K. Mitter, Monatsh. Chem., 117, 721 (1986). 8. J. Maxka, PhD. Thesis, University of Wisconsin, Madison, WI, 1988. 9. R. H. Baney, J. H. Gaul, Jr., T. K. Hilty, Orgunometallics, 2, 859 (1983). 10. P. A. Bianconi, T. W. Weidman, J. Am. Chem. Soc., 110, 2342 (1988) 11. H. Stuger, R. West, Macromolecules, 18, 2349 (1985). 12. X. H. Zhang, R. West, Macromolecules, in press. 13. H. Ban, K. Sukegawa, S. Tagawa, Macromolecules, 20, 1175 (1987). 14. R. Horiguchi, Y. Onishi, S. Hayase, Macromolecules, 21,304 (1988). 15. K. Matyjaszewski, Y-L. Chen, H. K. Kim, Ch. 6 in ref. 2. 16. R. D. Miller, D. Hofer, D. R. McKean, C. G. Willson, R. West, R. Trefonas, in “Materialsfor Microlithogruphy, ACS Syrnp. Ser. 266, L. Thompson, C. G. Willson, J. M. J. Frechet, eds., American Chemical Society, Washington, D.C., 1984., pp. 293-310. 17. R. D. Miller, J. F. Rabolt, R. Sooriyakumaran, W. Fleming, G. N. Fickes, B. L. Farmer, H. Kuzmany, Ch. 4 in ref. 2. 18. A. R. Wolff, R. West, Appl. Organomet. Chem., 1, (1987). 19. P. Trefonas, R. Miller, R. West, J. Am. Chem. Soc., 107, 2737 (1985). 20. J. Michl, J. W. Downing, T. Karatsu, A. J. McKinley, G. Poggi, G . M. Wallraf, R. Sooriyakumaran, R. D. Miller, Pure Appl. Chem., 60,959 (1988). 21. S. Yajima, K. Okamuro, J. Hayashi, M. Omori, J. Am. Ceram. Soc., 59,324 (1976).

15.2.5. in Boron-Nitrogen Systems. 15.2.5.1. Synthesis of Ring Compounds 15.2.5.1 .l.Borazines and Related Compounds.

The most widely studied and best characterized of the boron-nitrogen ring com-

I

I

pounds are the borazines, of generic formula (-N-B-)3. The six-membered ring is readily formed to yield borazine or substituted borazines directly via elimination reactions. Substituted amines or amine hydrochlorides react with boron trihalides or tetrahydroborate salts to yield borazines. Thus 1,3,5-trimethylborazine and 1,3,5trimethyl-2,4,6-trichloroborazine are produced in essentially quantitative yields by 1:

3 [MeNH3]C1

+ 3 Li[BH,]

3 [MeNH,]Cl

EtzO

+ 3 BC1,

PhCl

(MeNBH),

+ 3 LiCl + 9 H,

(MeNBCl),

-

(4

+ 9 HCl

The parent compound borazine, (HNBH), , can be prepared similarly’:

3 [NH,]Cl

+ 3 Na[BH,]

(HNBH),

+ 9 H,

(c>

However, the method is inefficient, and borazine itself is better prepared3 by NaBH, bond in the preformed ring compound (HNBCI),. The reduction of the B-Cl 2,4,6-trichloroborazine can be prepared by reaction of NH,Cl with BC1, or the and other B-trichloroborazines acetonitrile adduct of BCl, in c h l ~ r o b e n z e n e ~This .

15.2. Ring-Ring and Ring-Polymer Interconversions 15.2.5.in Boron-Nitrogen Systems. 15.2.5.1. Synthesis of Ring Compounds

84

prepared by reactions analogous to that of Eq. (b) are important starting materials for preparing many substituted borazines that are not readily accessible by direct formation. An improved procedure for the direct synthesis of borazine via Eq. (c) has been reported5. The generalized preparative reaction for borazines shown in Eq. (d) has been widely utilized: R,N

+ BX,

- -

3 (RNBX), + 2 RX

R,NBX,

(4

where R typically can be combinations of H, alkyl, aryl or silyl, and X can be combinations of H, alkyl, aryl, halo or amino. These are preparative procedures of long standing, and they are reviewed extensively1s6-s. The ease of elimination of the RX molecule of Eq. (d) depends on steric considerations and on the facility of RN and BX bond cleavage. The choice of R combinations and X combinations in the starting materials can direct the syntheses to the desired products. Reactions of primary amines with BC1, have been most widely exploited owing to the ease of elimination of HCl. The addition of tertiary amine to these reactions facilitates sterically hindered eliminations. B-Fluoroborazines are less easily prepared. The reaction of a primary amine with a methylboron fluoride results in the elimination of methaneg rather than HF. However, H F elimination can be assisted by the addition of Me,NBF, to the reaction". The latter compound abstracts H F as the fluoroborate salt: RNH,

+ BF, + 2 Me,NBF,

-

3 (RNBF),

+ 2 [Me,NH][BF,]

(e)

B-Fluoroborazines have also been prepared by elimination of silyl fluorides. Facile cleavage of the Si-N bond renders silylamines very effective in the synthesis of these and other B-haloborazines, and they are extensively utilized. Reactions involving elimination of silyl halides have been found particularly useful in sterically hindered systems' '. Contrasting its broad versatility, the method of Eq. (d) is unsuited to the preparation of perfluoroalkyl or perfluoroaryl borazines, giving products in poor yield and of unsatisfactory purity. These borazines can be prepared in good yields, however, by a transamination reaction between the appropriate amine and a preformed amine-borane ~ o r n p l e x ' ~:~ ' Et,NBCl,

-

+ C,F,NH, + Et,N

3 (C,F,NBCl), + 2 [Et,NH]Cl

(f)

Because of the relative convenience of handling alkyl or aryl boranes as their amine complexes, transaminations have been utilizedl4,l for their reactions as well: Et,NBH,R

+ NH,

4 (HNBR), + 2 H, + Et,N

(g)

Another route to the formation of borazines is the reaction between borate esters, amines and A1 under H, pressure16: B(OPh),

+ A1 + RNH,

H2

3 (RNBH), + Al(OPh),

This reaction at 20,000 kPa and 120-150°C is inappropriate as a laboratory synthesis, but the method has considerable versatility. Hydroboration of nitriles has also been shown to be an effective means of synthesizing selected B-substituted borazines, in particular the N-haloalkylborazines". The propensity for cyclization into six-membered ring structures is clearly pronounced in these systems. Four- and eight-membered rings of analogous constitution are

85

15.2.5.in Boron-Nitrogen Systems.

15.2.5.1. Synthesis of Ring Compounds 15.2.5.1.1. Borazines and Related Compounds.

relatively uncommon and obtain only as a consequence of steric interactions of the substituent groups. Pyrolysis of the addition compound t-BuNH,BCI, gives (t-BuNBCI), with the elimination of HCI. Less hindered amines may give mixtures that include borazines, but t-alkyl primary amine adducts yield exclusively the tetramer”. Dimeric four-membered rings can also result from eliminations from sterically hindered systems. They are often obtained when bulky silyl or amino groups remain as ring substituents’. The novel transf~rmationl~ of the tin-nitrogen ring dimer of Eq. (i) is a method that provides good yields of the boron-nitrogen dimers under mild conditions. (t-BuNSnMe,),

+ RBCI,

-

(t-BuNBR),

+ 2 Me,SnCl,

(0

where R = Me or C1. Reactive monomeric species of RNBR’ composition can be isolated by trapping at low temperatures. The monomers are prepared by low-pressure thermal degradation of boranamines” or azidoboranes21. Cyclization reactions can occur on warming to RT. In the absence of steric hindrance, trimeric borazine rings invariably form, but with sufficiently bulky substituents (e.g., R = R = t-Bu), dimeric species result. However, even with nonbulky substituents many monomers can form cyclic dimers at low temperature if subjected to catalysis by certain transition-metal complexes or the isonitrile t-BuNC 22. With intermediate steric requirements, specifically if R = t-Bu and R’ = i-Pr or sec-Bu, thermal oligomerization forms trimers that are not typical borazines. These trimers are bicyclic rings with structures analogous to Dewar benzenez3. They are commonly referred to as Dewar borazines. The properties and reactions of RNBR’ species (boranimines or iminoboranes) have been reviewedz4. Caution: Synthetic procedures for borazines may present various hazards. Preparations should be attempted only by persons familiar with the character of the reagents used. (R. K. BUNTING)

1. R. A. Geanangel, S. G. Shore, Prep. Inorg. React., 3, 123 (1966). 2. D. T. Haworth, L. F. Hohnstedt, Chem. Ind. (London), 559 (1960). 3. G. H. Dahl, R. Schaeffer, J. Inorg. Nucl. Chem., 12, 380 (1960). 4. E. F. Rothgery, L. F. Hohnstedt, Inorg. Chem., 6, 1065 (1967). 5. V. S . Volkov, A. A. Pukhov, K. G. Myakishev, Izv. Sib.Otd. Akad. Nauk SSSR,Ser. Khim. Nauk, 3, 116 (1983); Chem. Abstr., 99, 205,078 (1983). 6. H. Steinberg, R. J. Brotherton, Organoboron Chemistry, Vol. 2, Interscience, New York, 1966, Ch. 8. 7. K. Niedenzu, J. W. Dawson, Boron-Nitrogen Compounds, Academic Press, New York, 1965, Ch. 3. 8. A. Meller, Fortschr. Chem. Forsch., 15, 146 (1970). 9. V. E. Wiberg, G. Harold, Z. Naturforsch., Teil B, 6, 338 (1951). 10. J. J. Harris, B. Rudner, Inorg. Chem., 8, 1258 (1969). 11. P.Paetzold, A. Richter, T. Thijssen, S. Wurtenberg, Chem. Ber., 112, 3811 (1979). 12. A. Meller, V. Gutmann, M. Wechsberg, Inorg. Nucl. Chem. Lett., 1, 79 (1965). 13. M. Wechsberg, V. Gutmann, Monatsh. Chem. 97, 619 (1966). 14. M. F. Hawthorne, J. Am. Chem. SOC.,83,833 (1961). 15. I. B. Atkinson, D. B. Clapp, C. A. Beck, B. R. Currell, J. Chem. SOC.,Dalton Trans., 182 (1972). 16. E. C . Ashby, R. A. Kovar, Inorg. Chem., 10, 1524 (1971). 17. A. J. Leffler, Inorg. Chem., 3, 145 (1964). 18. H. S. Turner, R. J. Warne, J . Chem. Soc., 6421 (1965). 19. W. Storch, W. Jacktiess, J. Noth, G. Winter, Angew. Chem., Znt. Ed. EngL, 16, 478 (1977). 20. P. Paetzold, E. Schroder, G. Schmid, R. Boese, Chem. Ber., 118, 3205 (1985) and refs. therein.

86

15.2. Ring-Ring and Ring-Polymer lnterconversions 15.2.5. in Boron-Nitrogen Systems. 15.2.5.1. Synthesis of Ring Compounds

21. 22. 23. 24.

H.-U. Meier, P. Paetzold, E. Schroder, Chern. Ber., 117, 1954 (1984). K. Delpy, H.-U. Meier, P. Paetzold, C. von Plotho, 2. Naturforsch. Teil B, 39, 1696 (1984). P. Paetzold, C. von Plotho, G. Schmid, R. Boese, Z . Naturforsch, Teil39, 1069 (1984). P. Paetzold, Adu. Inorg. Chem., 31,123 (1987).

~~

15.2.5.1.2. Coordinately Saturated 6-N

Rings.

I I

The six-membered boranamine ring compounds of composition (-N-B-),

I I

can

be prepared by addition reactions across the B-N bonds of preformed borazines', but there are also routes to the direct synthesis of these and other ring boranamines. Such compounds have been synthesized by reactions similar to those used for the preparation of borazines ($15.2.5.1.1).If the general reaction between compounds of the type R,N and BX, involves combinations of R groups and of X groups that facilitate elimination of only one RX molecule, the R,NBX, cyclic oligomers can result. Thus the pyrolysis of Me,NHBH, yields the cyclic boranamine dime? (Me,NBH,), , while pyrolysis of MeNH,BH, can yield3 the boranamine trimer (at 100OC) or the borazine (at 200°C): 3 MeNH,BH,

100°C

(MeNHBH,),

200°C

(MeNBH),

(a)

The additional substituent on each atom of the saturated boranamines provides increased steric inhibition to cyclic oligomerization. Cyclic four-membered rings are more prevalent than in the analogous unsaturated compounds. The boranamine Et,NBCl, is known only as the monomer, but Me,NBH,, Me,NBHMe, MeNHBMe, and H,NBPh, exist in varying degrees of monomer-dimer equilibria4. In the absence of steric strain the preference for the six-membered ring is indicated by formation of (MeNHBH,), as the sole cyclic product5 of the reaction: [MeNH,]Cl

--

+ Li[BH,]

3 (MeNHBH,), + LiCl + 2 H,

(b)

The unsubstituted cyclic boranamine polymers (H,NBH,), cannot be prepared by pyrolysis of NH,BH,. These compounds can be formed6 by preparation of the unsymmetrical cleavage product of diborane in liq NH3 and subsequent treatment with NaNH, : B2H6 [(NH,),BH,I+

+ 2 NH3

+ NaCNH,]

+ [BHJ(H,NBH,), + 2 NH, + N a +

[(NH,),BH,]+

(c) (d)

These reactions can produce the cyclic dimer, trimer, tetramer (trace) and pentamer, with the unusual 10-membered ring compound (H,NBH,), being the principal cyclic species (80 %>. Except for the pentamer this preparative procedure provides low yields of cyclic products. The dimer and trimer are better prepared by other routes, the former by an interconversion reaction ($15.2.5.4) and the latter by NaBH, reduction of the B-trichloro compound5. Caution: Diborane is extremely hazardous and can explode violently on contact with the atmosphere. Syntheses using B,H, should be attempted only by persons well versed in nonatmospheric techniques. (R. K. BUNTING)

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc.

86

15.2. Ring-Ring and Ring-Polymer lnterconversions 15.2.5. in Boron-Nitrogen Systems. 15.2.5.1. Synthesis of Ring Compounds

21. 22. 23. 24.

H.-U. Meier, P. Paetzold, E. Schroder, Chern. Ber., 117, 1954 (1984). K. Delpy, H.-U. Meier, P. Paetzold, C. von Plotho, 2. Naturforsch. Teil B, 39, 1696 (1984). P. Paetzold, C. von Plotho, G. Schmid, R. Boese, Z . Naturforsch, Teil39, 1069 (1984). P. Paetzold, Adu. Inorg. Chem., 31,123 (1987).

~~

15.2.5.1.2. Coordinately Saturated 6-N

Rings.

I I

The six-membered boranamine ring compounds of composition (-N-B-),

I I

can

be prepared by addition reactions across the B-N bonds of preformed borazines', but there are also routes to the direct synthesis of these and other ring boranamines. Such compounds have been synthesized by reactions similar to those used for the preparation of borazines ($15.2.5.1.1).If the general reaction between compounds of the type R,N and BX, involves combinations of R groups and of X groups that facilitate elimination of only one RX molecule, the R,NBX, cyclic oligomers can result. Thus the pyrolysis of Me,NHBH, yields the cyclic boranamine dime? (Me,NBH,), , while pyrolysis of MeNH,BH, can yield3 the boranamine trimer (at 100OC) or the borazine (at 200°C): 3 MeNH,BH,

100°C

(MeNHBH,),

200°C

(MeNBH),

(a)

The additional substituent on each atom of the saturated boranamines provides increased steric inhibition to cyclic oligomerization. Cyclic four-membered rings are more prevalent than in the analogous unsaturated compounds. The boranamine Et,NBCl, is known only as the monomer, but Me,NBH,, Me,NBHMe, MeNHBMe, and H,NBPh, exist in varying degrees of monomer-dimer equilibria4. In the absence of steric strain the preference for the six-membered ring is indicated by formation of (MeNHBH,), as the sole cyclic product5 of the reaction: [MeNH,]Cl

--

+ Li[BH,]

3 (MeNHBH,), + LiCl + 2 H,

(b)

The unsubstituted cyclic boranamine polymers (H,NBH,), cannot be prepared by pyrolysis of NH,BH,. These compounds can be formed6 by preparation of the unsymmetrical cleavage product of diborane in liq NH3 and subsequent treatment with NaNH, : B2H6 [(NH,),BH,I+

+ 2 NH3

+ NaCNH,]

+ [BHJ(H,NBH,), + 2 NH, + N a +

[(NH,),BH,]+

(c) (d)

These reactions can produce the cyclic dimer, trimer, tetramer (trace) and pentamer, with the unusual 10-membered ring compound (H,NBH,), being the principal cyclic species (80 %>. Except for the pentamer this preparative procedure provides low yields of cyclic products. The dimer and trimer are better prepared by other routes, the former by an interconversion reaction ($15.2.5.4) and the latter by NaBH, reduction of the B-trichloro compound5. Caution: Diborane is extremely hazardous and can explode violently on contact with the atmosphere. Syntheses using B,H, should be attempted only by persons well versed in nonatmospheric techniques. (R. K. BUNTING)

15.2. Ring-Ring and Ring-Polymer Interconversions 15.2.5.in Boron-Nitrogen Systems. 15.2.5.2.Factors Affecting Stability of Ring Species.

87

1. R. A. Geanangel, S . G. Shore, Prep. Znorg. React., 3, 123 (1966). 78,2061 (1956). 2. R. E. McCoy, S. H. Bauer, J. Am. Chem. SOC., 3. T. C. Bissot, R. W. Parry, J. Am. Chem. SOC.,73, 953 (1951). 4. K. Niedenzu, J. W. Dawson, in The Chemistry of Boron and Its Compounds,E. L. Muetterties, ed., Wiley, New York, 1967, p. 377. 5. D. F. Gaines, R. Schaeffer, J. Am. Chem. SOC.,85, 395 (1963). 6. K. W. Boddeker, S. G. Shore, R. K. Bunting, J. Am. Chem. SOC.,88,4396 (1966).

15.2.5.2. Factors Affecting Stability of Ring Species.

The thermodynamic bond strength of the B-N framework in borazines is, in part, a consequence of multiple-bond contributions from the unshared pair of electrons on nitrogen with the vacant p-orbital of boron, as represented by the resonance forms:

I

I

B -N ../ \..N-B

I

1 - 1

B-

\”/

N

I

B

-N B\\

\N-

I1

N

I

/B-

-

I

B -N / \N-B

I

I1

B-

I

The significance of B-N multiple bonding was overemphasized in the early conjecture on borazines’, but the contribution from some degree of multiple bonding is clearly evidenced. The B-N ring bond strengths are thus augmented by electron-donating substituents on nitrogen and electron-withdrawing substituents on boron’ and diminished by the reverse. However, the strength of ring bonding is of secondary importance to the thermal stability of borazines, as most thermal decompositions appear to be initiated by elimination of an exocyclic molecule followed by ring rupture. Thus the steric effects and the bond cleavage facility of the substituent groups dominate both the thermal and hydrolytic stability of these species. The coordinately saturated rings have weaker B-N bonding, as evidenced by the oligomerization equilibria ($15.2.5.1.2) and by interconversion reactions ($15.2.5.4). Inasmuch as the substituent groups dominate the nature of stability, however, saturated ring compounds may have greater thermal and/or hydrolytic stability than the borazines; (HNBH), decomposes slowly at RT3, whereas (H,NBH& is stable4 up to 150°C. B-Trihaloborazines are more thermally stable than (HNBH), , and they decompose only if heated above their melting points. They are hydrolyzed rapidly, however (the fluoroborazine least rapidly), owing to the typical hydrolytic instability of B-halogen bonds, whereas (HNBH), is hydrolyzed only slowly at room temperature. Alkyl or aryl substitution on either the boron or the nitrogen atoms tends to increase the thermal stability of borazines, and the fully substituted compounds have very high thermal stabilities. The hydrolytic stability of fully alkylated or arylated borazines is less impressive than their thermal stabilities, but hydrolysis is slow in the absence of acid. Aryl substitution on nitrogen is much more influential at retarding hydrolysis than is aryl substitution at boron5. The four and eight-membered ring analogs of the borazines have weaker ring bonding, but the extensive steric requirements necessary for their formation also renders them thermally and hydrolytically stable6. The compound (t-BuNBCl), can be readily

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc. 15.2. Ring-Ring and Ring-Polymer Interconversions 15.2.5.in Boron-Nitrogen Systems. 15.2.5.2.Factors Affecting Stability of Ring Species.

87

1. R. A. Geanangel, S . G. Shore, Prep. Znorg. React., 3, 123 (1966). 78,2061 (1956). 2. R. E. McCoy, S. H. Bauer, J. Am. Chem. SOC., 3. T. C. Bissot, R. W. Parry, J. Am. Chem. SOC.,73, 953 (1951). 4. K. Niedenzu, J. W. Dawson, in The Chemistry of Boron and Its Compounds,E. L. Muetterties, ed., Wiley, New York, 1967, p. 377. 5. D. F. Gaines, R. Schaeffer, J. Am. Chem. SOC.,85, 395 (1963). 6. K. W. Boddeker, S. G. Shore, R. K. Bunting, J. Am. Chem. SOC.,88,4396 (1966).

15.2.5.2. Factors Affecting Stability of Ring Species.

The thermodynamic bond strength of the B-N framework in borazines is, in part, a consequence of multiple-bond contributions from the unshared pair of electrons on nitrogen with the vacant p-orbital of boron, as represented by the resonance forms:

I

I

-N -B

../ B \..NI

1 - 1

B-

\”/

N

I

B

-N B\\

\N-

I1

N

I

/B-

-

-N -B

I

I1

/

B \N-

I

B-

I

The significance of B-N multiple bonding was overemphasized in the early conjecture on borazines’, but the contribution from some degree of multiple bonding is clearly evidenced. The B-N ring bond strengths are thus augmented by electron-donating substituents on nitrogen and electron-withdrawing substituents on boron’ and diminished by the reverse. However, the strength of ring bonding is of secondary importance to the thermal stability of borazines, as most thermal decompositions appear to be initiated by elimination of an exocyclic molecule followed by ring rupture. Thus the steric effects and the bond cleavage facility of the substituent groups dominate both the thermal and hydrolytic stability of these species. The coordinately saturated rings have weaker B-N bonding, as evidenced by the oligomerization equilibria ($15.2.5.1.2) and by interconversion reactions ($15.2.5.4). Inasmuch as the substituent groups dominate the nature of stability, however, saturated ring compounds may have greater thermal and/or hydrolytic stability than the borazines; (HNBH), decomposes slowly at RT3, whereas (H,NBH& is stable4 up to 150°C. B-Trihaloborazines are more thermally stable than (HNBH), , and they decompose only if heated above their melting points. They are hydrolyzed rapidly, however (the fluoroborazine least rapidly), owing to the typical hydrolytic instability of B-halogen bonds, whereas (HNBH), is hydrolyzed only slowly at room temperature. Alkyl or aryl substitution on either the boron or the nitrogen atoms tends to increase the thermal stability of borazines, and the fully substituted compounds have very high thermal stabilities. The hydrolytic stability of fully alkylated or arylated borazines is less impressive than their thermal stabilities, but hydrolysis is slow in the absence of acid. Aryl substitution on nitrogen is much more influential at retarding hydrolysis than is aryl substitution at boron5. The four and eight-membered ring analogs of the borazines have weaker ring bonding, but the extensive steric requirements necessary for their formation also renders them thermally and hydrolytically stable6. The compound (t-BuNBCl), can be readily

88

15 2. Ring-Ring and Ring-Polymer lnterconversions 15.2.5. in Boron-Nitrogen Systems. 15.2 5.3. Preparation of Linear Boron-Nitrogen Oligomers.

sublimed, and it is unaffected by boiling water-a stark contrast to the character of the six-membered B-haloborazines. Some exocyclic multiple-bond character may be influential at stabilizing the fourmembered unsaturated rings; the majority of known compounds contain a silicon substituent at the ring nitrogen atoms or a nitrogen substituent at the ring boron atoms. The coordinately saturated ring compounds owe their stabilities to both additional steric effects and the unavailability of sites for electrophilic or nucleophilic attack. A clear discrimination between the relative influence of these factors is often difficult, but the fact that (H,NBH,), is unaffected by boiling water while (HNBH), is rapidly hydrolyzed must be largely a consequence of the latter. (R K. BUNTING)

1. The similarity between benzene and borazine is superficial, and parallels are found only among

2. 3. 4. 5. 6.

the physical properties. L. Ya. Rikhter, L. M. Sverdlov, Russ. J. Phys. Chem., 49, 1598 (1975). C. J. Wolf, R. H. Toeniskoetter, J. Phys. Chem., 67, 88 (1963). G. H. Dahl, R. Schaeffer, J. Am. Chem. SOC.,83, 3032 (1961). W. Gerrard, E. F. Mooney, D. E. Pratt, J. Appl. Chem. (London), 13, 127 (1963). H. S. Turner, R. J. Warne, Adv. Chern. Ser., 42, 290 (1964).

15.2.5.3. Preparation of Linear Boron-Nitrogen Oligomers. Owing to the tendency to cyclization, there are few known linear oligomers containing alternating B-N bonds, and those that have been characterized lack the thermal and hydrolytic stability of comparable cyclic species. Linear compounds containing two B-N units and capped by amino groups can be prepared':

3 RNH,

+ 2 PhBCl,

-

RNHBPhNRBPhNHR

+ 4 HCl

(a)

where R = n-, i-Bu. These compounds are sterically inhibited from cyclizing, and on heating they undergo linear extension with elimination of amine. Analogous B-chloro products are obtained' from the reaction of 2,6- or 2,4,6-substituted arylamines with BCl,. Such oligomers are readily hydrolyzed in boiling water, in contrast to similarly substituted ring compounds. Believed to be short-chain oligomers, (H,NBH,), species have been synthesized: B,H,

+ LiCNH,]

EtzO

(H,NBH,),

+ LiBH,

The products are obtained as mixtures. They are amorphous and not well characterized, but mol wt determinations indicate average values of 3 < n < 4.5. Their linear structure is only inferred from their lack of the thermal and hydrolytic stability that characterize the cyclic boranamines. A route to the specific formation of boranamine dimers, end-capped by hydrogen atoms, is available via reaction of aminodiboranes with amines,: R,NB,H5

+ R,N

-

R,NBH,NR,BH,

(c)

The reaction is applicable to combinations of R as H or Me. Caution: Diborane and substituted diboranes are extremely hazardous materials and should be handled only by persons experienced in nonatmospheric techniques. (R. K. BUNTING)

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc.

88

15 2. Ring-Ring and Ring-Polymer lnterconversions 15.2.5. in Boron-Nitrogen Systems. 15.2 5.3. Preparation of Linear Boron-Nitrogen Oligomers.

sublimed, and it is unaffected by boiling water-a stark contrast to the character of the six-membered B-haloborazines. Some exocyclic multiple-bond character may be influential at stabilizing the fourmembered unsaturated rings; the majority of known compounds contain a silicon substituent at the ring nitrogen atoms or a nitrogen substituent at the ring boron atoms. The coordinately saturated ring compounds owe their stabilities to both additional steric effects and the unavailability of sites for electrophilic or nucleophilic attack. A clear discrimination between the relative influence of these factors is often difficult, but the fact that (H,NBH,), is unaffected by boiling water while (HNBH), is rapidly hydrolyzed must be largely a consequence of the latter. (R K. BUNTING)

1. The similarity between benzene and borazine is superficial, and parallels are found only among

2. 3. 4. 5. 6.

the physical properties. L. Ya. Rikhter, L. M. Sverdlov, Russ. J. Phys. Chem., 49, 1598 (1975). C. J. Wolf, R. H. Toeniskoetter, J. Phys. Chem., 67, 88 (1963). G. H. Dahl, R. Schaeffer, J. Am. Chem. SOC.,83, 3032 (1961). W. Gerrard, E. F. Mooney, D. E. Pratt, J. Appl. Chem. (London), 13, 127 (1963). H. S. Turner, R. J. Warne, Adv. Chern. Ser., 42, 290 (1964).

15.2.5.3. Preparation of Linear Boron-Nitrogen Oligomers. Owing to the tendency to cyclization, there are few known linear oligomers containing alternating B-N bonds, and those that have been characterized lack the thermal and hydrolytic stability of comparable cyclic species. Linear compounds containing two B-N units and capped by amino groups can be prepared':

3 RNH,

+ 2 PhBCl,

-

RNHBPhNRBPhNHR

+ 4 HCl

(a)

where R = n-, i-Bu. These compounds are sterically inhibited from cyclizing, and on heating they undergo linear extension with elimination of amine. Analogous B-chloro products are obtained' from the reaction of 2,6- or 2,4,6-substituted arylamines with BCl,. Such oligomers are readily hydrolyzed in boiling water, in contrast to similarly substituted ring compounds. Believed to be short-chain oligomers, (H,NBH,), species have been synthesized: B,H,

+ LiCNH,]

EtzO

(H,NBH,),

+ LiBH,

The products are obtained as mixtures. They are amorphous and not well characterized, but mol wt determinations indicate average values of 3 < n < 4.5. Their linear structure is only inferred from their lack of the thermal and hydrolytic stability that characterize the cyclic boranamines. A route to the specific formation of boranamine dimers, end-capped by hydrogen atoms, is available via reaction of aminodiboranes with amines,: R,NB,H5

+ R,N

-

R,NBH,NR,BH,

(c)

The reaction is applicable to combinations of R as H or Me. Caution: Diborane and substituted diboranes are extremely hazardous materials and should be handled only by persons experienced in nonatmospheric techniques. (R. K. BUNTING)

89

15.2. Ring-Ring and Ring-Polymer Interconversions 15.2.5. in Boron-Nitrogen Systems. 15.2.5.4. Interconversions in Boron-Nitrogen Systems.

1. W. Gerrard, High Temperature Resistance and Thermal Degradation of Polymers, Monograph 13,

Society of Chemical Industry, London, 1961, p. 328. 2. B. K. Bartlett, H. S . Turner, R. J. Warne, M. A. Young, I. J. Lawrenson, J . Chem. Soc., A , 479 (1966). 3. D. L. Denton, A. D. Johnson 11, C. W. Hickam Jr., R. K. Bunting, S. G. Shore, J. h o r g . Nucl. Chem., 37, 1037 (1975). 4. G. A. Hahn, R. Schaeffer, J. Am. Chern. Soc., 86, 1503 (1964).

15.2.5.4. lnterconversions in Boron-Nitrogen Systems.

There are several interesting interconversion reactions in which the ring size of a cyclic oligomer is expanded or contracted. None of the mechanisms of these has been established. Whereas the reaction of hydrogen halides with borazines typically results in addition across the B-N bond to form the coordinately saturated trihalo rings, it has been observed’ that treatment of (HNBCI), with acetic acid (or acetic anhydride) effects a similar addition, along with chlorine replacement, but with a concomitant reduction in ring size to the dimer: (HNBCI),

+ 6 HOAc

-

4 [H,NB(OAc),], + 3 HCI

(a)

-

A quantitative irreversible transformation from tetramer to trimer occurs2 when (BuNBPh), is heated to 250°C: 3 (BuNBPh),

4 (BuNBPh),

(b)

A wide variety of dimers (RNBR’), react quantitatively with additional monomer to expand into borazine?, and (t-BuNBPr-i), reacts with t-BuNBPr-i to form the Dewar borazine4 as expected ($15.2.5.1.1). Certain RNBR species that are sterically inhibited from forming trimeric borazines exhibit a dimer-tetramer equilibrium3. If R = R’ = i-Pr, or if R = t-Bu and R = Me only dimers exist at elevated temperatures, whereas tetramers are the stable form at 20°C. More sterically strained species can exist only in the dimer form, and less strain allows formation of the stable borazine trimer. Owing to the similar steric requirements for dimers and for Dewar borazines, the bicyclo (i-PrNBBu-t), transforms into dimer at 200°C The coordinately saturated cyclic dimer (H,NBH,), readily undergoes transformation to the trimer in solution at RT3.

’.

The reaction appears to be quantitative in NH,, MeOH or Et,O. Pyrolysis of the pentamer (H,NBH,), yields the dimeric species (H,NBH,), among other products6. The interconversion to dimer is by no means complete but it is best known method for acquiring (H,NBH,),. (R. K. BUNTING)

G. L. Brennan, G. H. Dahl, R. Schaeffer, J. Am. Chem. SOC.,82, 6248 (1960). B. R. Currell, W. Gerrard, M. Khodabocus, Chem. Cornmun., 77 (1966). K. Delpy, H.-U. Meier, P. Paetzold, C. von Plotho, Z. Naturforsch., Teil B,39, 1696 (1984). P. Paetzold, C. von Plotho, G. Schmid, R. Boese, Z. Naturforsch., Teil B, 39, 1069 (1984). 5. H. Steuer, A. Meller, G. Elter., J . Organornet. Chem., 295, l(1985). 6. K. W. Boddeker, S. G. Shore, R. K. Bunting, J. Am. Chem. SOC.,88,4396 (1966).

1. 2. 3. 4.

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc.

89

15.2. Ring-Ring and Ring-Polymer Interconversions 15.2.5. in Boron-Nitrogen Systems. 15.2.5.4. Interconversions in Boron-Nitrogen Systems.

1. W. Gerrard, High Temperature Resistance and Thermal Degradation of Polymers, Monograph 13,

Society of Chemical Industry, London, 1961, p. 328. 2. B. K. Bartlett, H. S . Turner, R. J. Warne, M. A. Young, I. J. Lawrenson, J . Chem. Soc., A , 479 (1966). 3. D. L. Denton, A. D. Johnson 11, C. W. Hickam Jr., R. K. Bunting, S. G. Shore, J. h o r g . Nucl. Chem., 37, 1037 (1975). 4. G. A. Hahn, R. Schaeffer, J. Am. Chern. Soc., 86, 1503 (1964).

15.2.5.4. lnterconversions in Boron-Nitrogen Systems.

There are several interesting interconversion reactions in which the ring size of a cyclic oligomer is expanded or contracted. None of the mechanisms of these has been established. Whereas the reaction of hydrogen halides with borazines typically results in addition across the B-N bond to form the coordinately saturated trihalo rings, it has been observed’ that treatment of (HNBCI), with acetic acid (or acetic anhydride) effects a similar addition, along with chlorine replacement, but with a concomitant reduction in ring size to the dimer: (HNBCI),

+ 6 HOAc

-

4 [H,NB(OAc),], + 3 HCI

(a)

-

A quantitative irreversible transformation from tetramer to trimer occurs2 when (BuNBPh), is heated to 250°C: 3 (BuNBPh),

4 (BuNBPh),

(b)

A wide variety of dimers (RNBR’), react quantitatively with additional monomer to expand into borazine?, and (t-BuNBPr-i), reacts with t-BuNBPr-i to form the Dewar borazine4 as expected ($15.2.5.1.1). Certain RNBR species that are sterically inhibited from forming trimeric borazines exhibit a dimer-tetramer equilibrium3. If R = R’ = i-Pr, or if R = t-Bu and R = Me only dimers exist at elevated temperatures, whereas tetramers are the stable form at 20°C. More sterically strained species can exist only in the dimer form, and less strain allows formation of the stable borazine trimer. Owing to the similar steric requirements for dimers and for Dewar borazines, the bicyclo (i-PrNBBu-t), transforms into dimer at 200°C The coordinately saturated cyclic dimer (H,NBH,), readily undergoes transformation to the trimer in solution at RT3.

’.

The reaction appears to be quantitative in NH,, MeOH or Et,O. Pyrolysis of the pentamer (H,NBH,), yields the dimeric species (H,NBH,), among other products6. The interconversion to dimer is by no means complete but it is best known method for acquiring (H,NBH,),. (R. K. BUNTING)

G. L. Brennan, G. H. Dahl, R. Schaeffer, J. Am. Chem. SOC.,82, 6248 (1960). B. R. Currell, W. Gerrard, M. Khodabocus, Chem. Cornmun., 77 (1966). K. Delpy, H.-U. Meier, P. Paetzold, C. von Plotho, Z. Naturforsch., Teil B,39, 1696 (1984). P. Paetzold, C. von Plotho, G. Schmid, R. Boese, Z. Naturforsch., Teil B, 39, 1069 (1984). 5. H. Steuer, A. Meller, G. Elter., J . Organornet. Chem., 295, l(1985). 6. K. W. Boddeker, S. G. Shore, R. K. Bunting, J. Am. Chem. SOC.,88,4396 (1966).

1. 2. 3. 4.

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc. 90

15.2. Ring-Ring and Ring-Polymer Interconversions 15.2.5. in Boron-Nitrogen Systems. 15.2.5.5. Preparation of Boron-Nitrogen Polymers.

15.2.5.5. Preparation of Boron-Nitrogen Polymers. 15.2.5.5.1. Linear Boron-Nitrogen Chains.

Linear polymers built upon chains of alternating B and N atoms typically have neither the thermal nor the hydrolytic stability of cyclic species similarly constituted, and none of the high polymers has been thoroughly characterized. Preparative procedures and the properties of many polymeric species are summarized’, and investigations on polymerization mechanisms are reviewed’. Most mechanisms remain largely speculative. The tendency toward small ring formation precludes the synthesis of linear polymers by elimination of H, or HCl from such compounds as RNH’BH, or RNH’BCl,. Even such compounds with highly hindered amines tend toward ring formation (four- or eightmembered rings) rather than polymer formation when pyrolyzed ($15.2.5.1.1). Reactions involving elimination of amines or chloroalkanes have met with some success. Linear polymers (RNBPh), are3 products from the thermal decomposition (distillation) of (RNH),BPh compounds when the amino substituents are sufficiently bulky (R = i-Bu, s-Bu, t-Bu or i-Pr). The degree of polymerization is inferred from viscosity measurements to be between 10 and 20. The products hydrolyze rapidly but are thermally stable to 350°C. Linear polymers are also obtained by elimination of n-BuC1 through pyrolysis of n-Bu3NBCl,, resulting in as many as nine n-BuNBCI units linearly assembled4. Chain termination was assumed to be via loss of BCl,, resulting in polymers capped with n-Bu and n-Bu,N moieties. The inclusion of NBN linkages in heterocyclic rings inhibits cyclization of BN unitss:

This compound is a promising means of generating linear condensation polymers, but studiess on the polymerization reactions are only preliminary. Coordinately saturated polymers of composition (H’NBH,), are identified6 but not characterized. They are amorphous materials and their polymerization degree is believed to be quite low ($15.2.5.3).A polymer of the same formula unit is prepared from pyrolysis of NH,BH, under a low pressure of nitrogen. This polymer lacks sufficient solubility for molecular weight determination but is crystalline7, in contrast to the materials of the same composition reported earlier, and its x-ray diffraction pattern differs from that of any known cyclic boranamine. Pyrolysis of H,NBF,, followed by low-T condensation and subsequent warming of the product, yields the polymer of composition (H’NBF,), . H,NBF3

- - HF

H,NBF,

4 (H,NBF,),

(a)

This product is soluble without reaction in water, and light scattering techniques

15.2.5. in Boron-Nitrogen Systems. 15.2.5.5. Preparation of Boron-Nitrogen Polymers. 15.2.5.5.2. Polycyclic Chains.

91

establish' its molecular weight as near 23,000. Thermal studies of its stability are inconclusive owing to rapid vaporization of the material above 200°C. (R. K. BUNTING)

1. 2. 3. 4. 5.

6. 7. 8.

I. B. Atkinson, B. R. Currell, Inorg. Macromol. Rev., 1, 203 (1970). S. U.Sheikh, J. Therm. Anal., 21, 343 (1981). J. E. Burch, W. Gerrard, E. F. Mooney, J. Chem. Soc., 2200 (1962). I. B. Atkinson, C. A. Beck, B. R. Currell, S. U. Sheikh, International Symposium on Inorganic Polymers, Northern Polytechnic, London, April 9-1 1, 1969. S. Y. Shaw, D. A. DuBois, R. H. Nielson, in Inorganic Organometallics Polymers, ACS Symp. Ser. 360, M. Zeldin, K. J. Wynne, H. R. Allcock, eds., American Chemical Society, Washington, D.C., 1987, p. 31. K. W. Boddeker, S. G. Shore, R. K. Bunting, J. Am. Chem. Soc., 88,4396 (1966). R. Komm, R. A. Geanangel, R. Liepins, Inorg. Chem., 22, 1684 (1983). S. Y. Pusatcioglu, H. A. McGee, Jr., A. L. Fricke, J. C. Hassler, J. Appl. Polym. Sci., 21, 1561 (1977).

15.2.5.5.2. Polycyclic Chains.

Polycyclic B-N chains have been formed by the direct attachment of borazine rings or by the attachment of boron atoms of successive rings to an intervening NR group. There are also examples of cyclic species joined by other atoms, but only B-N-linked polymers are considered here. The treatment of borazine in a silent electric discharge apparatus effects elimination of H, and coupling of ring units'. A diphenyl analog, N,B,H,,, and a naphthalene analog, N,B,H,, have been identified among the products. The pyrolysis of B-substituted dimethylaminoborazine rings at 250°C gives ringlinked polymers with the elimination of amine' :

(HNBNRZ),

-

1

-

n

n

but crosslinking results if higher T is used. Boron-substituted n-Bus borazines react similarly, the disubstituted derivative readily eliminating n-BUSH in an analogous reaction3. The borazine rings coupled at ring boron atoms by NH can be prepared" by a controlled pyrolysis of boranes such as B(NHR),; polymers thus linked can be formed by pyrolysis of borazines that have two B-amino substituents and all other ring atoms alkylated. An amine molecule is eliminated and NR links the rings: n-Bu B Me"

I

'NMe I

BNHR

RNHB Me

--

+ RNHZ

n Me

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc. 15.2.5. in Boron-Nitrogen Systems. 15.2.5.5. Preparation of Boron-Nitrogen Polymers. 15.2.5.5.2. Polycyclic Chains.

91

establish' its molecular weight as near 23,000. Thermal studies of its stability are inconclusive owing to rapid vaporization of the material above 200°C. (R. K. BUNTING)

1. 2. 3. 4. 5.

6. 7. 8.

I. B. Atkinson, B. R. Currell, Inorg. Macromol. Rev., 1, 203 (1970). S. U.Sheikh, J. Therm. Anal., 21, 343 (1981). J. E. Burch, W. Gerrard, E. F. Mooney, J. Chem. Soc., 2200 (1962). I. B. Atkinson, C. A. Beck, B. R. Currell, S. U. Sheikh, International Symposium on Inorganic Polymers, Northern Polytechnic, London, April 9-1 1, 1969. S. Y. Shaw, D. A. DuBois, R. H. Nielson, in Inorganic Organometallics Polymers, ACS Symp. Ser. 360, M. Zeldin, K. J. Wynne, H. R. Allcock, eds., American Chemical Society, Washington, D.C., 1987, p. 31. K. W. Boddeker, S. G. Shore, R. K. Bunting, J. Am. Chem. Soc., 88,4396 (1966). R. Komm, R. A. Geanangel, R. Liepins, Inorg. Chem., 22, 1684 (1983). S. Y. Pusatcioglu, H. A. McGee, Jr., A. L. Fricke, J. C. Hassler, J. Appl. Polym. Sci., 21, 1561 (1977).

15.2.5.5.2. Polycyclic Chains.

Polycyclic B-N chains have been formed by the direct attachment of borazine rings or by the attachment of boron atoms of successive rings to an intervening NR group. There are also examples of cyclic species joined by other atoms, but only B-N-linked polymers are considered here. The treatment of borazine in a silent electric discharge apparatus effects elimination of H, and coupling of ring units'. A diphenyl analog, N,B,H,,, and a naphthalene analog, N,B,H,, have been identified among the products. The pyrolysis of B-substituted dimethylaminoborazine rings at 250°C gives ringlinked polymers with the elimination of amine' :

(HNBNRZ),

-

1

-

n

n

but crosslinking results if higher T is used. Boron-substituted n-Bus borazines react similarly, the disubstituted derivative readily eliminating n-BUSH in an analogous reaction3. The borazine rings coupled at ring boron atoms by NH can be prepared" by a controlled pyrolysis of boranes such as B(NHR),; polymers thus linked can be formed by pyrolysis of borazines that have two B-amino substituents and all other ring atoms alkylated. An amine molecule is eliminated and NR links the rings: n-Bu B Me"

I

'NMe I

BNHR

RNHB Me

--

+ RNHZ

n Me

92

15.2. Ring-Ring and Ring-Polymer Interconversions 15.2.5. in Boron-Nitrogen Systems. 15.2.5.5. Preparation of Boron-Nitrogen Polymers.

1

where R = H, Et, Ph. A similar linkage of borazine rings by NH is achieved5 by the reaction of (Me,Si),NH with 2,4-dichloroborazines; however, in these instances the unusual macrocycles of Eq. (c) form instead of extended linear polymerization.

R

1 4 MeN/ 1 B \ NMe ClB BCl \N/ Me

+ 4 (Me,Si),NH

-

b/’\TiJM; \N/B-N Me

+ 8 Me,SiCl

H 4

(c> where R = Me, Et, i-Pr. However, crosslinked polycyclic chains of high polymerization degree result if 2,4,6-trichloroborazine is used6 in place of the 2,4-dichloroborazine. These polymers are not fully characterized but are produced as gels in CH,Cl, for subsequent thermal degradation to boron nitride. Copolymerization reactions also yield polycyclic borazines with directly linked rings. The reaction of N-lithiated borazines with B-chloroborazines gives polymers’ :

LiN

Me /B\

MeB\

NLi

I

I

/

N Me

BMe

+

ClB/

I

MeN

Me N

\ BCl

\B/ Me

I

NMe

Polymers containing up to 10 cyclic units are identified. A similar copolymerization results’ from heating an equimolar mixture of 1,3,5triphenyl- and 2,4,6-triphenylborazines to 420°C. Some benzene is eliminated as well as hydrogen, but all the borazine rings appear to be directly attached rather than linked by aromatic rings. (R. K. BUNTING)

A. W. Laubengayer, 0. T. Beachley, Adv. Chem. Ser., 42, 281 (1964). W. Gerrard, H. R. Hudson, E. F. Mooney, J. Chem. Soc., 113 (1962). B. M. Mikhailov, A. F. Galkin, Bull. Acad. Sci. USSR,Div. Chem. Sci., 539 (1969). V. Gutmann, A. Meller, R. Schlegel, Monatsh. Chem., 94, 1071 (1963). A. Meller, H. J. Fullgrabe, Z. Naturforsch., Ted B, 33, 156 (1978). C. K. Narula, R. T. Paire, R. Schaeffer, in Inorganic and Organometallzc Polymers, ACS Symp. Ser. 360, M. Zeldin, K. J. Wynne, H. R. Alcock, eds., American Chemical Society, Washington, D.C., 1987, p. 378. 7. R. I. Wagner, J. L. Bradford, Inorq. Chem., I , 99 (1962). 8. N. I. Beklsova, V. V. Korshak, L. G. Komarova, Vysokomol.Soedin. Ser. B, 10,101 (1968); Chem. Abstr., 68, 96, 235 (1968).

1. 2. 3. 4. 5. 6.

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc.

93

15.2.6. in Boron-Oxygen and Boron-Sulfur Systems 15.2.6.1. Anionic Boron-Oxygen Compounds.

15.2.6. in Boron-Oxygen and Boron-Sulfur Systems 15.2.6.1. Anionic Boron-Oxygen Compounds.

There is an extensive and complex structural chemistry of boron-oxygen anionic species (polyborates), in aqueous or nonaqueous solution and in the melt or solid state. Six-membered ring formation dominates, but the structural chemistry of the species is complicated since boron exists in either 3- or 4-coordinate environments, or various combinations of these. Thus most polymeric species consist of (B-0), rings joined by boron atoms linked to an intervening oxygen atom, or y rings sharing a common boron atom. Relatively few definitive structural have emerged. However, the number of 4-coordinate boron atoms, n, in an anionic unit is given by: n = C - 0, + 0,

(a)

where C is the anion charge, and 0, and 0, represent, respectively, the number of single and triply bonded oxygen atoms3. Thus in most hydrated mineral structures (0,= 0), the number of 4-coordinate boron atoms equals the anion charge. An important exception is the hexaborate structure, with free fused (B-0), rings sharing a common oxygen atom (0, = 1). There is an increase in the density of solids and in hydrolytic stability that clearly parallels an increase in the ratio of 4- to 3-coordinate boron atoms. Significant stability requires one or two tetracoordinate boron atoms per ring4. In aqueous solution polyborates exist in complex equilibria among species of varying degrees of oligomerization depending upon pH and concentration53637. At low pH molecular B(OH), predominates. An increase in pH effects an increase in the tetrahedral-total boron ratio to the upper limit of 1 in [B(OH),]-, exemplified by the intermediate species:

OH-

___f

HO,-

I

I

____)

o/ \o I

HO/ B\

0

B / \

o/

OH-

0, B -/ O

HO/

B

B

‘0’

,OH ‘0

OH -

I

B

\OH

OH

OH

I

HO-B\

I I JB-OHl o 1 OH

OH -

(b)

94

15.2. Ring-Ring and Ring-Polymer Interconversions 15.2.6. in Boron-Oxygen and Boron-Sulfur Systems 15.2.6.2. Molecular Boron-Oxygen Species.

High polymers are uncommon in protic solvents owing to the ease of bond cleavage at trigonal boron, but in aprotic solvents the polymerization can increase to the degree that gelation occurs'. Reviews on polyborate structures and reactions, both on hydrate^^*^-'^ and a n h y d r o ~ s ~ , 'materials, ~~' should be consulted for details on specific systems. (R.K. BUNTING) 1. C. L. Christ, J. R. Clark, Phys. Chem. Minerals, 2, 59 (1977). 2. C. L. Crist, J. Geol, Educ., 20, 235 (1972). 3. There are a few known minerals that contain oxygen with a formal 2 charge (quadruply bound); see R. 0. Gould, P. J. Nelmec, S.E. B. Gould, J. Phys. C Solid State Phys., 14,5259 (1981) and refs. therein. 4. B. P. Tarasevich, E. V. Kuznetsov, Russ. Chem. Revs., 56,203 (1987); Usp. Khirn. 56,353 (1987). 5. R. E. Mesmer, C. F. Baes Jr., F. H. Sweeton, Inorg. Chern., 11, 537 (1972). 6. L. Maya, Inorg. Chern., 15, 2179 (1976). 7 . B. P. Tarasevich, N. Z. Sirotkina, A. S. Il'in, Ye. V. Kuznetsov, Polymer Sci. USSR, 27, 503 (1985): Vysokomol.Soyed., A27, 451 (1985). 8. C. J. Brinker, B. C. Bunker, D. R. Tallant, K. J. Ward, R. J. Kirkpatrick, in Inorganic and Organometallic Polymers, ACS Symp. Ser. 360, M. Zeldin, K. J. Wynne, H. R. Allcock, eds., American Chemical Society, Washington, D.C., 1988, Ch. 26. 9. J. B. Farmer, Adv. Inorg. Chem. Radiochem, 25, 187 (1982). 10. G. Heller, Fortschr. Chem. Forsch., 131, 39 (1986). 11. N. I. Leonyuk, L. I. Leonyuk, Kristallokhimiya Bezvodnykh Boratov (Crystal Chemistry of Anhydrous Borates), Izd. Moskov. Gos. Univ., Moscow, 1983; Chem. Abstr., 99,185,471 (1983).

15.2.6.2. Molecular Boron-Oxygen Species.

The most prevalent boron-oxygen oligomeric systems are based on the sixmembered (B-0), ring, formally a derivative of boroxine, (HBO), . These compounds are synthesized directly, the tendency to ring formation usually precluding the isolation of any intermediate products. Boroxine itself is prepared by a high-T hydrolysis of elemental boron' or explosive The compound is unstable, however, and reactions of boron hydrides in decomposes rapidly under ordinary conditions to B 2 0 3and B2H6. Haloboroxines are similarly unstable but can be prepared by a high T reaction of B,O, with BX, (X = F, C1, Br), trapping the products at low T These compounds readily redistribute to the original reactants, although (ClBO), is stable at RT while isolated and kept in the gas phase6. Organo derivatives of boroxine are much more stable than boroxine or its halo derivatives. They are formally anydrides of boronic acids, and most are prepared nearly quantitatively by dehydration of the corresponding acid5:

'.

3 RB(OH),

-

(RBO),

+ 3 H,O

(4

This is conveniently accomplished by azeotropic distillation of H,O from a solvent such as toluene. Excellent reviews summarize the established preparative procedures for a great many o r g a n o b o r ~ x i n e s ~ ~Versatile '~~. preparative methods include heterofunctional condensation, which occurs readily at RT: 3 RB(0R)Cl-

(RBO),

+ 3 RCl

(b)

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc. 94

15.2. Ring-Ring and Ring-Polymer Interconversions 15.2.6. in Boron-Oxygen and Boron-Sulfur Systems 15.2.6.2. Molecular Boron-Oxygen Species.

High polymers are uncommon in protic solvents owing to the ease of bond cleavage at trigonal boron, but in aprotic solvents the polymerization can increase to the degree that gelation occurs'. Reviews on polyborate structures and reactions, both on hydrate^^*^-'^ and a n h y d r o ~ s ~ , 'materials, ~~' should be consulted for details on specific systems. (R.K. BUNTING) 1. C. L. Christ, J. R. Clark, Phys. Chem. Minerals, 2, 59 (1977). 2. C. L. Crist, J. Geol, Educ., 20, 235 (1972). 3. There are a few known minerals that contain oxygen with a formal 2 charge (quadruply bound); see R. 0. Gould, P. J. Nelmec, S.E. B. Gould, J. Phys. C Solid State Phys., 14,5259 (1981) and refs. therein. 4. B. P. Tarasevich, E. V. Kuznetsov, Russ. Chem. Revs., 56,203 (1987); Usp. Khirn. 56,353 (1987). 5. R. E. Mesmer, C. F. Baes Jr., F. H. Sweeton, Inorg. Chern., 11, 537 (1972). 6. L. Maya, Inorg. Chern., 15, 2179 (1976). 7 . B. P. Tarasevich, N. Z. Sirotkina, A. S. Il'in, Ye. V. Kuznetsov, Polymer Sci. USSR, 27, 503 (1985): Vysokomol.Soyed., A27, 451 (1985). 8. C. J. Brinker, B. C. Bunker, D. R. Tallant, K. J. Ward, R. J. Kirkpatrick, in Inorganic and Organometallic Polymers, ACS Symp. Ser. 360, M. Zeldin, K. J. Wynne, H. R. Allcock, eds., American Chemical Society, Washington, D.C., 1988, Ch. 26. 9. J. B. Farmer, Adv. Inorg. Chem. Radiochem, 25, 187 (1982). 10. G. Heller, Fortschr. Chem. Forsch., 131, 39 (1986). 11. N. I. Leonyuk, L. I. Leonyuk, Kristallokhimiya Bezvodnykh Boratov (Crystal Chemistry of Anhydrous Borates), Izd. Moskov. Gos. Univ., Moscow, 1983; Chem. Abstr., 99,185,471 (1983).

15.2.6.2. Molecular Boron-Oxygen Species.

The most prevalent boron-oxygen oligomeric systems are based on the sixmembered (B-0), ring, formally a derivative of boroxine, (HBO), . These compounds are synthesized directly, the tendency to ring formation usually precluding the isolation of any intermediate products. Boroxine itself is prepared by a high-T hydrolysis of elemental boron' or explosive The compound is unstable, however, and reactions of boron hydrides in decomposes rapidly under ordinary conditions to B 2 0 3and B2H6. Haloboroxines are similarly unstable but can be prepared by a high T reaction of B,O, with BX, (X = F, C1, Br), trapping the products at low T These compounds readily redistribute to the original reactants, although (ClBO), is stable at RT while isolated and kept in the gas phase6. Organo derivatives of boroxine are much more stable than boroxine or its halo derivatives. They are formally anydrides of boronic acids, and most are prepared nearly quantitatively by dehydration of the corresponding acid5:

'.

3 RB(OH),

-

(RBO),

+ 3 H,O

(4

This is conveniently accomplished by azeotropic distillation of H,O from a solvent such as toluene. Excellent reviews summarize the established preparative procedures for a great many o r g a n o b o r ~ x i n e s ~ ~Versatile '~~. preparative methods include heterofunctional condensation, which occurs readily at RT: 3 RB(0R)Cl-

(RBO),

+ 3 RCl

(b)

15.2. Ring-Ring and Ring-Polymer lnterconversions 15.2.6. in Boron-Oxygen and Boron-Sulfur Systems 15.2.6.3. Boron-Sulfur Compounds

-

95

if trace quantities of FeCI, are present, and the redistribution reaction: BR,

+ B203

(RBO),

(c)

The latter method is restricted to n-alkyl or cycloalkyl derivatives; a-branched alkyl derivatives isomerize under the reaction conditions. Alkoxy and aryloxy boroxines are5,’**.Formally esters of metaboric acid (HOBO),, which itself is prepared by thermal dehydration of boric acid (azeotropically in toluene). Dehydration of B(OH), in alcohols or phenols directly yields (ROBO), or (ArOBO),, respectively. The hydration interconversions among boric acid, metaboric acid and boric oxide:

correctly suggest the use of any of these as the starting material, but the highly crosslinked ring network in B,O, requires more severe conditions for its reactions7. Other synthetic routes are procedures analogous to those for the alkylboroxines. Thus (RO),BCl produces (ROBO), with trace quantities of electron-pair acceptor catalysts (cf. Eq. b), and (RO),B with B,O, redistributes to (ROBO), (cf. Eq. c ) ~ . Higher polymers of B-0 systems consist of interconnected (B-0), rings. Such materials are prepared by dehydration of mixtures of ROB(OH), and B(OH),, the RO/B ratio influencing the polymerization degree7. However, such systems are not well characterized. Hydrolysis occurs at trigonal boron, and these molecular materials are even more susceptible to nucleophilic attack than are the anionic polyborates (915.2.6.1.1). Caution: Many boron hydrides are hazardous and can explode violently on contact with the atmosphere. Preparations using B,H, or B,H, should be attempted only by persons experienced in nonatmospheric techniques. (R. K. BUNTING)

W. P. Sholette, R. F. Porter, J . Phys. Chem., 67, 177 (1963). G. H. Lee, S. H. Bauer, S. E. Wiberley, J. Phys. Chem., 67, 1742 (1963). L. Barton, G. A. Grimm, R. F. Porter, Inorg. Chem., 5,2076 (1966). B. S . Ault, J. Mol. Struct., 159, 291 (1987). I. Haiduc, The Chemistry of Inorganic Ring Systems, Part 1,Wiley-Interscience, London, 1970, p. 216. 6. D. J. Knowles, A. S. Buchanan, Inorg. Chem., 4, 1799 (1965). 7. H. Steinberg, Organoboron Chemistry, Vol. 1, Wiley-Interscience, New York, 1964, Ch. 9. 8. W. Gerrard, The Organic Chemistry ofBoron, Academic Press, New York, 1961.

1. 2. 3. 4. 5.

15.2.6.3. Boron-Sulfur Compounds.

Although B-0 oligomeric systems are based almost exclusively on trimeric sixmembered rings ($15.2.6.1), four- and eight-membered rings are found in the B-S system. Six-membered rings predominate, however. The reaction of BBr, with H,S yields (HSBS),, which on further heating converts’ irrevrsibly to (HSBS),. The latter, the thio analog of trimetaboric acid, is a convenient source for preparing other substituted (B-S), ring compounds: 2 (HSBS),

+ BE

-

2 (EBS),

+ B2S3 + 3 H,S

(4

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc. 15.2. Ring-Ring and Ring-Polymer lnterconversions 15.2.6. in Boron-Oxygen and Boron-Sulfur Systems 15.2.6.3. Boron-Sulfur Compounds

-

95

if trace quantities of FeCI, are present, and the redistribution reaction: BR,

+ B203

(RBO),

(c)

The latter method is restricted to n-alkyl or cycloalkyl derivatives; a-branched alkyl derivatives isomerize under the reaction conditions. Alkoxy and aryloxy boroxines are5,’**.Formally esters of metaboric acid (HOBO),, which itself is prepared by thermal dehydration of boric acid (azeotropically in toluene). Dehydration of B(OH), in alcohols or phenols directly yields (ROBO), or (ArOBO),, respectively. The hydration interconversions among boric acid, metaboric acid and boric oxide:

correctly suggest the use of any of these as the starting material, but the highly crosslinked ring network in B,O, requires more severe conditions for its reactions7. Other synthetic routes are procedures analogous to those for the alkylboroxines. Thus (RO),BCl produces (ROBO), with trace quantities of electron-pair acceptor catalysts (cf. Eq. b), and (RO),B with B,O, redistributes to (ROBO), (cf. Eq. c ) ~ . Higher polymers of B-0 systems consist of interconnected (B-0), rings. Such materials are prepared by dehydration of mixtures of ROB(OH), and B(OH),, the RO/B ratio influencing the polymerization degree7. However, such systems are not well characterized. Hydrolysis occurs at trigonal boron, and these molecular materials are even more susceptible to nucleophilic attack than are the anionic polyborates (915.2.6.1.1). Caution: Many boron hydrides are hazardous and can explode violently on contact with the atmosphere. Preparations using B,H, or B,H, should be attempted only by persons experienced in nonatmospheric techniques. (R. K. BUNTING)

W. P. Sholette, R. F. Porter, J . Phys. Chem., 67, 177 (1963). G. H. Lee, S. H. Bauer, S. E. Wiberley, J. Phys. Chem., 67, 1742 (1963). L. Barton, G. A. Grimm, R. F. Porter, Inorg. Chem., 5,2076 (1966). B. S . Ault, J. Mol. Struct., 159, 291 (1987). I. Haiduc, The Chemistry of Inorganic Ring Systems, Part 1,Wiley-Interscience, London, 1970, p. 216. 6. D. J. Knowles, A. S. Buchanan, Inorg. Chem., 4, 1799 (1965). 7. H. Steinberg, Organoboron Chemistry, Vol. 1, Wiley-Interscience, New York, 1964, Ch. 9. 8. W. Gerrard, The Organic Chemistry ofBoron, Academic Press, New York, 1961.

1. 2. 3. 4. 5.

15.2.6.3. Boron-Sulfur Compounds.

Although B-0 oligomeric systems are based almost exclusively on trimeric sixmembered rings ($15.2.6.1), four- and eight-membered rings are found in the B-S system. Six-membered rings predominate, however. The reaction of BBr, with H,S yields (HSBS),, which on further heating converts’ irrevrsibly to (HSBS),. The latter, the thio analog of trimetaboric acid, is a convenient source for preparing other substituted (B-S), ring compounds: 2 (HSBS),

+ BE

-

2 (EBS),

+ B2S3 + 3 H,S

(4

96

15.2. Ring-Ring and Ring-Polymer Interconversions 15.2.6. in Boron-Oxygen and BoronSulfur Systems 15.2.6.3. Boron-Sulfur Compounds. ~~

where E = Me, MeO, Me,N, C1, Br. Replacement of BE, with AlF, or HI in Eq. (a) forms (FBS), or (IBS),, respectively’. The halo-substituted (B-S), rings can disproportionate to BE, and B,S,, but they are more stable than their oxygen analogs (515.2.6.1.2), increasing in stability with E-group size’. Other versatile synthetic methods for ring (B-S), compounds are’:

3 RBCl, 3 ArBCl,

+ 3 H,S

+ (R,SiS),

-

+ 6 HCl (ArBS), + 3 R,SiCl, (RBS),

(b) (c)

Many mercaptoboranes exhibit oligomerization tendencies and form stable cyclic species,, in contrast to the alkoxyboranes, which are almost exclusively monomeric. The cyclic compounds are coordination oligomers, with bonding characteristics like those of the coordinately saturated BN oligomers (515.2.5.1.2). The reaction of B,H, with RSH produces intermediate high polymers, which rearrange on standing (R = Et, n-Pr, n-Bu) or in solution (R = Me, t-Bu) to give cyclic trimers,:

f B,H,

+ RSH

-

3 (H,BSR), + H,

(4

These cyclic trimers are typically stable toward oxidation, can be distilled without decomposition and are more resistant to hydrolysis than are the cyclic species that contain tervalent boron. Caution: B,H, is hazardous and can explode on contact with the atmosphere. (R K BUNTING)

1. F. Armitage, Inorganic Rings and Cages, Edward Arnold, London, 1972, Ch. 3. 2. I. Haiduc, The Chemistry of Inorganic Ring Systems Part I, Wiley-Interscience, London, 1970, Ch. 111. 3. B. M. Mikhailov, in Progress in Boron Chemistry, Vol. 3, R. B. Brotherton, H. Steinberg, eds., Pergamon Press, Oxford, 1970, Ch. 5 .

15.2.6.4. Ring-ring and Ring-chain interconversions. Trimeric B-0 ring formation so dominates this system that ring-ring or ring-chain interconversions are rare. Structural conversions result in ring fusion or the formation of bicyclic rings’.’ of these trimeric units. However, (CIBO), is the dominant cyclic unit formed from the high-T reaction of BC1, with B,O,, whereas the trimer is the preferred product at low T

’.

(CIBO),

-

(CIBO),

(a)

The conversion is estimated to be exothermic by about 30 kJ. A dimeric form of the thio analog of metaboric acid, (HSBS),, is isolated from the reaction of BBr, or B,S, with H,S. This undergoes an irreversible ring expansion4 to (HSBS), on heating to 90°C. The trimer (EtSBS),, however, converts to the dimer on attempted vacuum distillation, and (MeSBS), undergoes spontaneous ring expansion at ambient T to tetrameric or higher cyclic species4. Six-membered rings prevail in most syntheses, however, and other cyclic units owe their existence to a kinetic stability5. In solution, (Me’NBS), undergoes spontaneous

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc. 96

15.2. Ring-Ring and Ring-Polymer Interconversions 15.2.6. in Boron-Oxygen and BoronSulfur Systems 15.2.6.3. Boron-Sulfur Compounds. ~~

where E = Me, MeO, Me,N, C1, Br. Replacement of BE, with AlF, or HI in Eq. (a) forms (FBS), or (IBS),, respectively’. The halo-substituted (B-S), rings can disproportionate to BE, and B,S,, but they are more stable than their oxygen analogs (515.2.6.1.2), increasing in stability with E-group size’. Other versatile synthetic methods for ring (B-S), compounds are’:

3 RBCl, 3 ArBCl,

+ 3 H,S

+ (R,SiS),

-

+ 6 HCl (ArBS), + 3 R,SiCl, (RBS),

(b) (c)

Many mercaptoboranes exhibit oligomerization tendencies and form stable cyclic species,, in contrast to the alkoxyboranes, which are almost exclusively monomeric. The cyclic compounds are coordination oligomers, with bonding characteristics like those of the coordinately saturated BN oligomers (515.2.5.1.2). The reaction of B,H, with RSH produces intermediate high polymers, which rearrange on standing (R = Et, n-Pr, n-Bu) or in solution (R = Me, t-Bu) to give cyclic trimers,:

f B,H,

+ RSH

-

3 (H,BSR), + H,

(4

These cyclic trimers are typically stable toward oxidation, can be distilled without decomposition and are more resistant to hydrolysis than are the cyclic species that contain tervalent boron. Caution: B,H, is hazardous and can explode on contact with the atmosphere. (R K BUNTING)

1. F. Armitage, Inorganic Rings and Cages, Edward Arnold, London, 1972, Ch. 3. 2. I. Haiduc, The Chemistry of Inorganic Ring Systems Part I, Wiley-Interscience, London, 1970, Ch. 111. 3. B. M. Mikhailov, in Progress in Boron Chemistry, Vol. 3, R. B. Brotherton, H. Steinberg, eds., Pergamon Press, Oxford, 1970, Ch. 5 .

15.2.6.4. Ring-ring and Ring-chain interconversions. Trimeric B-0 ring formation so dominates this system that ring-ring or ring-chain interconversions are rare. Structural conversions result in ring fusion or the formation of bicyclic rings’.’ of these trimeric units. However, (CIBO), is the dominant cyclic unit formed from the high-T reaction of BC1, with B,O,, whereas the trimer is the preferred product at low T

’.

(CIBO),

-

(CIBO),

(a)

The conversion is estimated to be exothermic by about 30 kJ. A dimeric form of the thio analog of metaboric acid, (HSBS),, is isolated from the reaction of BBr, or B,S, with H,S. This undergoes an irreversible ring expansion4 to (HSBS), on heating to 90°C. The trimer (EtSBS),, however, converts to the dimer on attempted vacuum distillation, and (MeSBS), undergoes spontaneous ring expansion at ambient T to tetrameric or higher cyclic species4. Six-membered rings prevail in most syntheses, however, and other cyclic units owe their existence to a kinetic stability5. In solution, (Me’NBS), undergoes spontaneous

15.2. Ring-Ring and Ring-Polymer Interconversions 15.2.7. in Cage Polyboranes and Carboranes.

97

ring expansion to (Me,NBS),. Pure (PhBS), is synthesized from PhBC1, and (Me,Si),S under mild conditions, but a higher T synthesis yields5 mixtures of’(PhBS), and (PhBS), . Alkylthioboranes are isolated as high molecular weight coordination polymers. The solid (H,BSR), polymers (R = Me, t-Bu) are stable but convert to trimers in solution whereas liquid products (R = Et, n-Prm n-Bu) slowly convert to trimers on standing6. (R. K. BUNTING)

1. B. P. Tarasevich, E. V. Kuznetsov, (1987) Russ. Chem. Revs., 56, 203 Usp. Khzm.,56, 353 (1987). 2. J. B. Farmer, Adv. Inorg. Chem. Radiochem, 25, 187 (1982). 3. E. F. Porter, S. K. Gupta, J . Phys. Chem., 68, 280 (1964). 4. E. Wiberg, W. Sturm, Angew. Chem., 67, 483 (1955). 5. H. Noth, W. Rattay, J . Organometallic Chem., 312, 139 (1986). 6. B. M. Mikhailov, in Progress in Boron Chemistry, Vol. 3, R. J. Brotherton, H. Steinberg, eds., Pergamon Press, Oxford, 1970, p. 348.

15.2.7. in Cage Polyboranes and Carboranes. The major, nearly exclusive, emphasis in polyborane polymer synthesis centers about the polymers of closo-carborane derivatives, especially closo-carborane siloxanes. Two carboranes having high thermal, oxidative and hydrolytic stabilities are the pentagonal bipyramidal C,B,H, and the icosahedral C,B,,H,, (55.3.2.7); thus it is not surprising that nearly all of the polymers containing carborane units are comprised of one or the other of these two cage systems. Only one of the two known isomers of C,B,H,, with cage carbons located in the 2 and 4 positions of the pentagonal pyramidal framework, is utilized in this respect. All three isomers of C,B,,H,, are known but the bulk of the polymer studies on this cage system have concentrated on the 1,2 and the 1,7 isomers, and chiefly the latter cage carborane; not surprisingly polymers constructed from these two C,B,,H,, isomers are found to be less crystalline and more elastomeric than those constructed from the more symmetrical 1,12 Numerous polymers incorporating these carboranes are known but this section covers only those in which the carborane is incorporated as part of a largely “inorganic” polymeric unit. Many organic polymers incorporating cage carborane units have been

’.

(T ONAK)

1. D. J. Mangold, Appl. Polym. Symp., No. 11, 157 (1969). 2. V. V. Korshak, I. G. Sarishvili, M. V. Sobolevskii, Usp. Khim., 36,2068 (1967); Chem. Abstr., 68, 59,891 (1968). 3. V. V. Korshak, A. F. Zhigach, I. G. Sarishvili, M. V. Sobolevskii, Progr. Polim. Khim., 321 (1968); Chem Abstr, 72, 13,080 (1970). 4. H. A. Schroeder, Rubber Age, 101, 58 (1969). 5. H. A. Schroeder, Inorg. Macromol. Rev., I , 45 (1970). 6. T. L. Heying, in Progress zn Boron Chemistry, R. J. Brotherton, H. Steinberg, eds., Vol. 2, New York, 1970, p. 119. 7. R. E. Williams, Pure Appl. Chem., 29, 569 (1972); US. Natl. Tech. Inform. Serv., AD No. 732,031, 33 pp. (1971); Chem. Abstr., 76, 154,375 (1972). 8. G. Schroderheim, Plastvaerlden, 55, 57 (1973); Chem. Abstr., 79, No. 6424 (1973). 9. M. B. Roller, C. K. Schoff, J. K. Gillham, SOC.Plust. Eng. Tech. Pap., 20, 671 (1974); Chem. Abstr., 81, 106,450 (1974).

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc. 15.2. Ring-Ring and Ring-Polymer Interconversions 15.2.7. in Cage Polyboranes and Carboranes.

97

ring expansion to (Me,NBS),. Pure (PhBS), is synthesized from PhBC1, and (Me,Si),S under mild conditions, but a higher T synthesis yields5 mixtures of’(PhBS), and (PhBS), . Alkylthioboranes are isolated as high molecular weight coordination polymers. The solid (H,BSR), polymers (R = Me, t-Bu) are stable but convert to trimers in solution whereas liquid products (R = Et, n-Prm n-Bu) slowly convert to trimers on standing6. (R. K. BUNTING)

1. B. P. Tarasevich, E. V. Kuznetsov, (1987) Russ. Chem. Revs., 56, 203 Usp. Khzm.,56, 353 (1987). 2. J. B. Farmer, Adv. Inorg. Chem. Radiochem, 25, 187 (1982). 3. E. F. Porter, S. K. Gupta, J . Phys. Chem., 68, 280 (1964). 4. E. Wiberg, W. Sturm, Angew. Chem., 67, 483 (1955). 5. H. Noth, W. Rattay, J . Organometallic Chem., 312, 139 (1986). 6. B. M. Mikhailov, in Progress in Boron Chemistry, Vol. 3, R. J. Brotherton, H. Steinberg, eds., Pergamon Press, Oxford, 1970, p. 348.

15.2.7. in Cage Polyboranes and Carboranes. The major, nearly exclusive, emphasis in polyborane polymer synthesis centers about the polymers of closo-carborane derivatives, especially closo-carborane siloxanes. Two carboranes having high thermal, oxidative and hydrolytic stabilities are the pentagonal bipyramidal C,B,H, and the icosahedral C,B,,H,, (55.3.2.7); thus it is not surprising that nearly all of the polymers containing carborane units are comprised of one or the other of these two cage systems. Only one of the two known isomers of C,B,H,, with cage carbons located in the 2 and 4 positions of the pentagonal pyramidal framework, is utilized in this respect. All three isomers of C,B,,H,, are known but the bulk of the polymer studies on this cage system have concentrated on the 1,2 and the 1,7 isomers, and chiefly the latter cage carborane; not surprisingly polymers constructed from these two C,B,,H,, isomers are found to be less crystalline and more elastomeric than those constructed from the more symmetrical 1,12 Numerous polymers incorporating these carboranes are known but this section covers only those in which the carborane is incorporated as part of a largely “inorganic” polymeric unit. Many organic polymers incorporating cage carborane units have been

’.

(T ONAK)

1. D. J. Mangold, Appl. Polym. Symp., No. 11, 157 (1969). 2. V. V. Korshak, I. G. Sarishvili, M. V. Sobolevskii, Usp. Khim., 36,2068 (1967); Chem. Abstr., 68, 59,891 (1968). 3. V. V. Korshak, A. F. Zhigach, I. G. Sarishvili, M. V. Sobolevskii, Progr. Polim. Khim., 321 (1968); Chem Abstr, 72, 13,080 (1970). 4. H. A. Schroeder, Rubber Age, 101, 58 (1969). 5. H. A. Schroeder, Inorg. Macromol. Rev., I , 45 (1970). 6. T. L. Heying, in Progress zn Boron Chemistry, R. J. Brotherton, H. Steinberg, eds., Vol. 2, New York, 1970, p. 119. 7. R. E. Williams, Pure Appl. Chem., 29, 569 (1972); US. Natl. Tech. Inform. Serv., AD No. 732,031, 33 pp. (1971); Chem. Abstr., 76, 154,375 (1972). 8. G. Schroderheim, Plastvaerlden, 55, 57 (1973); Chem. Abstr., 79, No. 6424 (1973). 9. M. B. Roller, C. K. Schoff, J. K. Gillham, SOC.Plust. Eng. Tech. Pap., 20, 671 (1974); Chem. Abstr., 81, 106,450 (1974).

98

15.2. Ring-Ring and Ring-Polymer Interconversions 15.2.7. in Cage Polyboranes and Carboranes. 15.2.7.1. Polyborane Oligomers and Polymers; Syntheses.

10. H. A. Schroeder, Nuova Chim., 50, 65 (1974); Chem. Abstr., 80, 96,387 (1974). 11. J. F. Ditter, A. J. Gotcher, in The Synthesis of New Polymers: Modern Methods, N. Yoda, ed., New York, 1975; U.S. Clearinghouse Fed. Sci. Tech. Inform., AD No. 770625/2GA (1973). 12. M. B. Roller, J. K. Gillham, Polym. Eng. Sci., 14, 567 (1974). 13. J. F. Ditter, Gmelin Handbuch der Anorganischen Chemie,Borverbindungen 6, Vol. 27, SpringerVerlag, Berlin, 69 (1975). 14. W. Stumpf, Themen Chem. Bors., 121 (1976); Chem. Abstr., 88, 23,433 (1976). 15. N. I. Bekasova, Usp. Khim., 53, 107 (1984); 105, 79,482 (1986). 16. E. N. Peters, Ind. Eng. Chem. Prod. Res. Dev., 23, 28 (1984). 17. V. V. Korshak, N. I. Bekasova, L. I. Komarova, M. A. Surikova, Vysokomol.Soedin., B, 30, 116 (1988); Chem. Abstr. 108, 187,376 (1988). 18. 0. A. Mel’nik, A. A. Sakharova, T. M. Frunze, Usp. Khim., 57, 1529 (1988); Chem. Abstr. 109, 231,584 (1988).

15.2.7.1. Polyborane Ollgomers and Polymers; Syntheses.

There have been no definitive studies reporting the preparation and characterization of well-defined polymeric cyclic or cage boranes; however, many synthetic preparations of monomeric polyboranes give rise to poorly defined high molecular weight solid side products with ratios close to 1:1’. In all probability these solids contain oligomeric units of BH clusters. Some plausible structural suggestions have been advanced for polymeric boranes produced from the condensation of polyborane units (each condensation involves formal loss of B,H,, or two BH, units). This type of condensation leads to the arachno-B,H,+, + following generalizations: hypho-B,H,+, + (B,-zH,+2),, (B,-,HJx, nido-B,H,+, + (Bn-2Hn-2)x and closo-B,H,+, + (B,H,-,),. Thus, onlyfor the nido class is a 1:l ratio of B and H predicted; however, for large values of n the distinctions among the classes of polymers become smaller. A (B,H,), polymer is then

Figure 1. The proposed structure for a (B,H,), polymer2.

Figure 2. The proposed structure for a (Bl6H1& polymer2.

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc. 98

15.2. Ring-Ring and Ring-Polymer Interconversions 15.2.7. in Cage Polyboranes and Carboranes. 15.2.7.1. Polyborane Oligomers and Polymers; Syntheses.

10. H. A. Schroeder, Nuova Chim., 50, 65 (1974); Chem. Abstr., 80, 96,387 (1974). 11. J. F. Ditter, A. J. Gotcher, in The Synthesis of New Polymers: Modern Methods, N. Yoda, ed., New York, 1975; U.S. Clearinghouse Fed. Sci. Tech. Inform., AD No. 770625/2GA (1973). 12. M. B. Roller, J. K. Gillham, Polym. Eng. Sci., 14, 567 (1974). 13. J. F. Ditter, Gmelin Handbuch der Anorganischen Chemie,Borverbindungen 6, Vol. 27, SpringerVerlag, Berlin, 69 (1975). 14. W. Stumpf, Themen Chem. Bors., 121 (1976); Chem. Abstr., 88, 23,433 (1976). 15. N. I. Bekasova, Usp. Khim., 53, 107 (1984); 105, 79,482 (1986). 16. E. N. Peters, Ind. Eng. Chem. Prod. Res. Dev., 23, 28 (1984). 17. V. V. Korshak, N. I. Bekasova, L. I. Komarova, M. A. Surikova, Vysokomol.Soedin., B, 30, 116 (1988); Chem. Abstr. 108, 187,376 (1988). 18. 0. A. Mel’nik, A. A. Sakharova, T. M. Frunze, Usp. Khim., 57, 1529 (1988); Chem. Abstr. 109, 231,584 (1988).

15.2.7.1. Polyborane Ollgomers and Polymers; Syntheses.

There have been no definitive studies reporting the preparation and characterization of well-defined polymeric cyclic or cage boranes; however, many synthetic preparations of monomeric polyboranes give rise to poorly defined high molecular weight solid side products with ratios close to 1:1’. In all probability these solids contain oligomeric units of BH clusters. Some plausible structural suggestions have been advanced for polymeric boranes produced from the condensation of polyborane units (each condensation involves formal loss of B,H,, or two BH, units). This type of condensation leads to the arachno-B,H,+, + following generalizations: hypho-B,H,+, + (B,-zH,+2),, (B,-,HJx, nido-B,H,+, + (Bn-2Hn-2)x and closo-B,H,+, + (B,H,-,),. Thus, onlyfor the nido class is a 1:l ratio of B and H predicted; however, for large values of n the distinctions among the classes of polymers become smaller. A (B,H,), polymer is then

Figure 1. The proposed structure for a (B,H,), polymer2.

Figure 2. The proposed structure for a (Bl6H1& polymer2.

99

15.2. Ring-Ring and Ring-Polymer Interconversions 15.2.7. in Cage Polyboranes and Carboranes. 15.2.7.2. Preparation of Carborane-Siloxane Polymers. ~

considered to be formally composed of nido-B, units (Fig. 1) and a (B16H16)xpolymer is based on decaborane( 14) structural units (Fig. 2),. (T. ONAK)

1. A. Stock, Hydrides ojBoron and Silicon, Cornell University Press, Ithaca, NY, 1933. 2. W. N. Lipscomb, Znorg. Chem., 19, 1415 (1980).

15.2.7.2. Preparation of CarboraneSiloxane Polymers.

Carborane siloxanes are a family of polymers with the general structure in which R and R can or typically be alkyl, fluoroalkyl or aryl. Incorporation of the closo-C,B,,H,, C,B,H,3-S group into the siloxane backbone in many instances results in improved thermal and oxidative stability. Polymers of this type in which n is 3 and 5 can be prepared, for example, by hydrolysis-condensation of the carborane,C1. based siloxane Cl(R'MeSiO),- lSi(R)(CH,)CB,,H,,CSi(R)(CH,)(OSiRMe)n~ The cohydrolysis of C1(SiR'Me)0Si(R)-(CH,)CB,,H1,CSi(R)(CH3)0SiRMeC1and Cl(SiR'Me)(OSiR'Me), - ,C1 forms6 polymers with n = 4,5,6. Reaction of H~Si(R)(Me)CB,,H,,CSi(R)(Me)OHand E,Si(R)CH, (E is an N-phenyl-"-tetramethyleneureido group) forms7 a polymer with n = 2. The synthesis of the polymer in which n=1 is accomplished by treating a mixture of CISi(R)(Me)CB,,Hl,CSi(R)(Me)Cl and CH,OSi(R)(Me)CB,,Hl,CSi(R)(Me)OCH, with ferric chloride'. Analogous FeC1,-catalyzed copolymerization of closo-2,4-C,B,H7-based silyl monomers produces more tractable and lower melting polymers than those based on 1,7-C,BlOH,, ',lo; but polymers prepared in this general fashion, whether based on C,B, or C,B,, units, are often intractable and lack consistency in physical properties4. Treatment of a mixture of 1,7-[Cl(CH,),Si],closo-1,7-C,B,,Hl, and 1,7-[(CH,),(CH,0)Si],-closo-1,7-C,Bl,H,, with FeCl, at temperatures up to 190°C produces the homopolymer [-Si(CH,),-CBloH,,CSi(CH,),O-1, (trade names: Dexsil 100; 10-SiB-1; Fig. l)"-13.

[-Si(R)(CH,)CB,H,CSi(R)(CH,)O(SiR'MeO), - ,-I,,,

'9'

r

L

R

n

Figure 1. The structure of the polymer [-Si(CH3),-CB,oH,oC-Si(CH3),0-]n.

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc.

99

15.2. Ring-Ring and Ring-Polymer Interconversions 15.2.7. in Cage Polyboranes and Carboranes. 15.2.7.2. Preparation of Carborane-Siloxane Polymers. ~

considered to be formally composed of nido-B, units (Fig. 1) and a (B16H16)xpolymer is based on decaborane( 14) structural units (Fig. 2),. (T. ONAK)

1. A. Stock, Hydrides ojBoron and Silicon, Cornell University Press, Ithaca, NY, 1933. 2. W. N. Lipscomb, Znorg. Chem., 19, 1415 (1980).

15.2.7.2. Preparation of CarboraneSiloxane Polymers.

Carborane siloxanes are a family of polymers with the general structure in which R and R can or typically be alkyl, fluoroalkyl or aryl. Incorporation of the closo-C,B,,H,, C,B,H,3-S group into the siloxane backbone in many instances results in improved thermal and oxidative stability. Polymers of this type in which n is 3 and 5 can be prepared, for example, by hydrolysis-condensation of the carborane,C1. based siloxane Cl(R'MeSiO),- lSi(R)(CH,)CB,,H,,CSi(R)(CH,)(OSiRMe)n~ The cohydrolysis of C1(SiR'Me)0Si(R)-(CH,)CB,,H1,CSi(R)(CH3)0SiRMeC1and Cl(SiR'Me)(OSiR'Me), - ,C1 forms6 polymers with n = 4,5,6. Reaction of H~Si(R)(Me)CB,,H,,CSi(R)(Me)OHand E,Si(R)CH, (E is an N-phenyl-"-tetramethyleneureido group) forms7 a polymer with n = 2. The synthesis of the polymer in which n=1 is accomplished by treating a mixture of CISi(R)(Me)CB,,Hl,CSi(R)(Me)Cl and CH,OSi(R)(Me)CB,,Hl,CSi(R)(Me)OCH, with ferric chloride'. Analogous FeC1,-catalyzed copolymerization of closo-2,4-C,B,H7-based silyl monomers produces more tractable and lower melting polymers than those based on 1,7-C,BlOH,, ',lo; but polymers prepared in this general fashion, whether based on C,B, or C,B,, units, are often intractable and lack consistency in physical properties4. Treatment of a mixture of 1,7-[Cl(CH,),Si],closo-1,7-C,B,,Hl, and 1,7-[(CH,),(CH,0)Si],-closo-1,7-C,Bl,H,, with FeCl, at temperatures up to 190°C produces the homopolymer [-Si(CH,),-CBloH,,CSi(CH,),O-1, (trade names: Dexsil 100; 10-SiB-1; Fig. l)"-13.

[-Si(R)(CH,)CB,H,CSi(R)(CH,)O(SiR'MeO), - ,-I,,,

'9'

r

L

R

n

Figure 1. The structure of the polymer [-Si(CH3),-CB,oH,oC-Si(CH3),0-]n.

100

15.2. Ring-Ring and Ring-Polymer Interconversions 15.2.7. in Cage Polyboranes and Carboranes. 15.2.7.2. Preparation of Carborane-Siloxane Polymers.

,

Copolymerization of 1,7-[(CH,O)(CH,),Si],-closo- 1,7-C2B ,H, with Cl,Si(CH,), produces the 10-SiB-2 (random) Combinations of 1,7-[(CH,0)(CH,),Si],-closo-l,7-C,B,,Hl, with silanes and siloxanes in a variety of ratios yield a family of (Dexsil, or SiB) polymers (an example is seen in Fig. 2). The experimental procedure involves the heating, with controlled stirring, of a mixture of the comonomers along with approximately 2 mol % of FeCl,; purification of the resulting polymer is accomplished by precipitation from hot xylene. Such polymers are relatively low molecular weight gummy resins with extensive cro~slinking~,'~. The bulk polymerization of 1,7-C,BloHl,-1,7-[Si(CH3)~OCH3]2 and (CH,),SiCI,, catalyzed by FeCl,, to produce poly(m-carboranylene-siloxane) has also been studied at various temperatures16. Linear polymers with relatively low molecular weights (< 20,000) are found from the RT alcoholysis of a bischlorosilyl monomer, e.g., 2,4-[Cl(CH,),Si],-2,4-C,BsH, + (CH,),COH. Copolymers obtained from the alcoholysis of mixtures containing approximately 67 mol% 2,4-[C1(CH3),Si],-2,4-C2B5H5, 25 mol% (CH,),SiCl, and 8 mol% CH,SiCl, yield elastomeric g ~ m s ~ , ~ , ' ~ . A high molecular weight (average molecular weight of 182,000 daltons) polycarboranesiloxane (Fig. 3) based on the C,B, carborane unit is obtained from the hydrolysis of a highly purified sample of 2,4-[C1(CH,),Si],-2,4-C2B5H5 '. Reduced-temperature (ca. in a O'C) hydrolytic polycondensation of 1,7-[C1(CH,)2SiOSi(CH3)2-]-l,7-C2BloHlo H,O-EtO-THF solution gives rise to polymers of the type [-Si(CH,),CB,,H,,CSi(CH,),0Si(CH,)20Si(CH,),0-]n (industrial name: Dexsil 300) with molecular weights in the 10,000-20,000 range6. Variations of the hydrolytic condensation of silyl carboranes include the use of the monomeric starting materials 1,7-R,-1,7-C,B1,Hlo, R = -Si(CH,),OSi(CH,),OSi(CH3),C1, -Si(CH,)(CH,CH,CF,)OSi(CH,)(CH,CH,CF,)Cl, and 2,4-Rz-2,4-C2B5H7,R = -Si(CH,)(CH,CH,CF,)Cl 6317,18. In some instances acid catalysts are used to facilitate this type of polycondensation p r o c e ~ s ~ Hydrolysis , ' ~ ~ ~ ~ .of dichlorosilyl compounds such as C1,Si(i-C,H7)CBloHloCR

?

Figure 2. The structure of the [-Si(CH3),CBloHl,CSi(CH3)20Si(CH3),0Si(CH3),0-]n polymer.

15.2. Ring-Ring and Ring-Polymer Interconversions 15.2.7. in Cage Polyboranes and Carboranes. 15.2.7.2. Preparation of Carborane-Siloxane Polymers.

101

n

Figure 3. The structure of [-Si(CH,),-CB,H,C-Si(CH,),O-I,.

(R = CH,, C,HJ produces oligomersZ0.Several random copolymers are products from the hydrolysis of mixtures containing silylcarboranes and silanes20*21. High molecular weight (approximately 250,000 daltons) linear polycarboranesiloxanes incorporating C,B,, carborane units are produced by reacting a disilanol derivative of the carborane with difunctional silanes having facile leaving g r o ~ p s ~ When ~-~~. certain ureidosilanes are used in this respect, the leaving group is an isoluble neutral urea byproduct that is removed from the polymer (which stays dissolved in a solvent such as chlorobenzene) by filtration; the polymer is precipitated with a more polar solvent such as methanol:

+ [(RCO)(C,H ,)N-],Si(CH,), (a) + [-Si(CH3)2-CBl,HloC-Si(CH3)zOSi(CH3)z-]n + (RCO)(C,H ,)NH

1,7-[HOSi(CH3),]-closo-1,7-C,B,,H,,

where R = an N-pyrrolidine, i.e., (CH2)&N-, group. Reaction of with (-CH,),NCON(C,H ,)SiR'R" (R', R" = 1,7-C,B1,Hl 1,7-[Si(CH,),OH] phenyl, methyl) gives elastomeric polymers containing -OSi(CH,),C,BloH,,Si(CH,),OSiR'R"Owith molecular weights up to 250,0007*28. In a related study a high-temperature-resistant polymer is prepared from 1,7-CzB,,H,o-1,7[Si(CH,),OH], and (CH,),Si[N(CH,)2],2.29. A carborane-siloxane block copolymer with 1,7-C,BloHl, units is prepared by the reaction of a ureidosilane-terminated polysulfone hard block with a carborane-silanol-terminated carborane-siloxane 01igomer~~. Synthesis of crystalline polymers containing

,-

-Si(CH,),-E-Si(CH,),-0-Si(CH,)(R)-0units (E = 1,7-C,B,,Hl0) are found to be more elastomeric when 30-50% of the 1,7-CZB,,Hl, cages are replaced with 1,2-CzB,,Hlo polyhedra or when phenyl groups (R = C,H,) are introduced onto the polymer backbone31. Carborane-siloxane polymers are also prepared from Liz[ 1,7-C2B,,H, 0] and [ClSi(CH,),O],Si(CH,), in ethyl ether3',,,. The thermal stability of a carborane-siloxane polymer containing 1,7-C,B,,Hlo units are improved upon by introducing a ferric oxide stabilizer,,.

102

15.2. Ring-Ring and Ring-Polymer Interconversions 15.2.7. in Cage Polyboranes and Carboranes. 15.2.7.2. Preparation of Carborane-Siloxane Polymers.

Recent studies concern the utilization of a polycarboranesiloxane for a thermally stable liquid phase in packed gas chromatographic columns35, identification of some polycarboranesiloxanes using pyrolysis gas c h r ~ m a t o g r a p h y ~Dexsil ~, 202 polymer pyrolysis37 and the structure and oxidative thermal stabilities of carborane-containing siloxane~~~. (T ONAK)

1. E. N. Peters, ACS Symp. Ser. No. 121,449 (1980). 2. E. N. Peters, Ind. Eng. Chem. Prod. Res. Dev., 23, 28 (1984). 3. R. E. Williams, Carborane Polymers, Pure Appl. Chem., 29,569 (1972); U.S. Natl. Tech. Inform. Serv., AD No. 732,031, 33 pp. (1971); Chem. Abstr., 76, 154,375 (1972). 4. J. F. Ditter, Gmelin Handbuch der Anorganischen Chemie, Borverbindungen 6, Springer-Verlag, Berlin, Vol. 27, 69 (1975). 5. J. F. Ditter, A. J. Gotcher, in The Synthesis of New Polymers: Modern Methods, N. Yoda, ed., New York, 1975; US. Clearinghouse Fed. Sci. Tech. Inform., AD No. 770,625/2GA, (1973). 6. K. 0.Knollmueller, R. N. Scott, H. Kwasnick, J. F. Sieckhaus, J. Polym. Sci., A - I , 9, 1071 (1971). 7. E. N. Peters, E. Hedaya, J. H. Kawakami, G. T. Kwiatkowski, D. W. McNeil, R. W. Tulis, Rubber Chem. Technol., 48, 14 (1975). 8. S . Papetti, B. B. Schaeffer, A. P. Gray, T. L. Heying, J. Polym. Sci., A-I, 4, 1623 (1966). 9. R. E. Kesting, K. F. Jackson, E. B. Klusmann, F. J. Gerhart, J. Appl. Polym. Sci., 14,2525 (1970). 10. R. E. Kesting, K. F. Jackson, J. M. Newman, J. Appl. Polym. Sci., 15, 1527 (1971). 11. S . Papetti, B. B. Schaeffer, A. P. Gray, T. L. Heying, J. Polyrn. Sci., A-1, 4, 1623 (1966). 12. S.Papetti, B. B. Schroeder, US. Pat. 3,463,801 (1967/69). 13. L. P. Dorofeenko, V. F. Gridina, A. L. Klebanskii, L. E. Krupnova, N. I. Shkambarnaya, I. F. Ermachkova, L. M. Vasil'eva, A. F. Zhigach, V. N. Siryatskaya, V. V. Korol'ko, Russian Pat. 319,622 (1969/71); Chem. Abstr., 76, 100,353 (1972). 14. H. A. Schroeder, 0. G. Schaffling, T. B. Larchan, F. F. Trulla, T. L. Heying, Rubber Chem. Technol., 39, 1184 (1966). 15. H. A. Schroeder, U.S. Natl. Tech. Inform. Serv., AD No. 703,677, 37 pp. (1970). 16. H. J. Dietrich, R. P. Alexander, T. L. Heying, H. Kwasnik, C . B. Obenland, H. J. A. Schroeder, Makromol. Chem., 175,425 (1974). 17. R. E. Kestmg, K. F. Jackson, J. M. Newman, J. Appl. Polym. Sci., 15, 2645 (1971). 18. R. N. Scott, K. 0. Knollmueller, H. Hooks, J. F. Sieckhaus, J. Polym. Sci., A-1, 10,2303 (1972). 19. D. R. Chapman, Ger. Pat. 2,021,304 (1969/70); Chem. Abstr., 74, 54,440 (1971). 20. A. S . Shapatin, T. A. Krasovskaya, S . A. Golubtsov, F. N. Vishnevskii, M. P. Smazhok, Kremniiorg. Soedin, Tr. Soveshch, No. 3, 162 (1967); Chem. Abstr., 69, 52,535 (1968). 21. J. Green, N. Mayes, A. Kotloby, M. M. Fein, E. L. O'Brien, M. S . Cohen, J. Polym. Sci., B, 2, 109 (1964). 22. G. B. Dunks, E. Hedaya, J. H. Kawakami, P. W. Kopf, G. T. Kwiatkowski, U.S. Natl. Tech. Inform. Serv., AD No. 766,147, 56 pp. (1973); Chem. Abstr., 80, 60,742 (1974). 23. E. N. Peters, J. J. Bohan, E. Hedaya, J. H. Kawakami, G. T. Kwiatkowski, D. W. McNeil, U.S. Off. Nav. Res. Tech. Rept., No. CRL-T-794 (1974). 24. E. N. Peters, E. Hedaya, J. H. Kawakami, G. T. Kwiatkowski, R. W. Tulis, D. W. McNeil, U.S. Off. Nav. Res. Tech. Rept., No. CRL-T-795 (1974); US.Natl. Tech. Inform. Serv., AD/A Rep., No. 003531/1GA, 20 pp. (1974). 25. K. 0. Knollmueller, U.S. Pat. 3,733,928 (1972/73); Chem. Abstr., 79, 92,831 (1973). 26. E. Hedaya, J. H. Kawakami, P. W. Kopf, G. T. Kwiatkowski, D. W. McNeil, D. A. Owens, E. N. Peters, R. W. Tulis, J. Polym. Sci., Polym. Chem. Ed., 15, 2229 (1977). 27. K. A. Barker, C. D. Beard, J. J. Bohan, G. B. Dunks, E. Hedaya, Gout. Rept. Announce. US.,79, No. 1, 119 (1979); Chem. Abstr., 90, 153,247 (1979). 28. D. D. Stewart, E. N. Peters, C . D. Beard, G. B. Dunks, E. Hedaya, G. T. Kwiatkowski, R. B. Moffit, J. J. Bohan, Macromolecules, 12, 373 (1979). 29. E. Hedaya, J. H. Kawakami, G. T. Kwiatkowski, U S . Pat. 4,145,504 (1973); Chem. Abstr., 91, 22,180 (1979). 30. D. D. Stewart, E. N. Peters, C. D. Beard, R. B. Moffitt, G. T. Kwiatkowski, J. J. Bohan, E. Hedaya, J. Appl. Polym. Sci., 24, 115 (1979).

15.2. Ring-Ring and Ring-Polymer Interconversions 103 15.2.7.in Cage Polyboranes and Carboranes. 15.2.7.3. Preparation of Other Carborane-Incorporated Inorganic Polymers. 31. E. N. Peters, J. H. Kawakami, G. T. Kwiatkowski, E. Hedaya, B. L. Joesten, D. W. McNeil, D. A. Owens, J. Polym. Sci., Polym. Phys. Ed., 15, 723 (1977). 32. E. N. Peters, D. D. Stewart, J. Polym. Sci.,Polym. Letf.Ed., 17, 405 (1979). 33. E. N. Peters, US. Pat. 4,235,987 (1987); Chem. Abstr., 94, 66,439 (1981). 34. E. N. Peters, D. D. Stewart, J. J. Bohan, D. W. McNeil, J. E1ustomer.r Plust., 10, 29 (1978). 35. J. A. Yancey, J . Chromatogr. Sci., 23, 370 (1985). 36. E. E. Sotnikov, Zh. K. Torosyan, Zh. Anal. Khim., 40, 1887 (1985). 37. W. S. Coblenz, G. H. Wiseman, P. B. Davis, R. W. Rice, Mater. Sci. Res., 17, 271 (1984). 38. I. I. Miroshnikova, L. N. Sakharova, V. T. Minakov, A. B. Blyumenfel'd, N. L. Fedorova, L. I. Golubenkova, Plust. Massy, 38 (1987); Chem. Abstr., 108, 38,882 (1988).

15.2.7.3. Preparation of Other Carborane-Incorporated Inorganic Polymers.

An oligomer consisting of a number of phosphorus-linked closo-1,7-C,B,,Hl, polyhedra is obtained by treating a solid product from the reaction of Li,[1,7-C,Bl,Hl,] and PCl, (in a 1:2 ratio) with methanol. The structure of the oligomer is believed to be Cl,P[-CB,oHloCP(C1)-]50CH3 Without the presence of methanol (n z 5) is formed'-3. [-CBl,H,,C-PCl], l-Methyl-l-[(o-carboranyl)methylene]-3,3,5,5-tetrachlorocyclotriphosphazene(in which o-carboranyl = closo-1,2-C,Bl,H, 1) polymerizes, when heated at 250°C, to A. Polymer A reacts with piperidine to give the polymer B, in which the chlorines have been

'.

-'n

n

B

A

replaced with piperidine groups and opening of the carborane cage has also occurred. A THF solution of tris(triphenylphosphine)rhodium(I) chloride reacts with B at 66°C to give polymer C. The latter catalyzes the hydrogenation of 1-hexene. Deprotonation of B PPh,

/

C with sodium hydride, followed by the addition of W(CO), or Mo(CO),, exposure of the resulting solution to ultraviolet irradiation and the addition of trimethylamine hydrochloride, gives rise to polymer D (M = W, M o ) ~ .

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc. 15.2. Ring-Ring and Ring-Polymer Interconversions 103 15.2.7.in Cage Polyboranes and Carboranes. 15.2.7.3. Preparation of Other Carborane-Incorporated Inorganic Polymers. 31. E. N. Peters, J. H. Kawakami, G. T. Kwiatkowski, E. Hedaya, B. L. Joesten, D. W. McNeil, D. A. Owens, J. Polym. Sci., Polym. Phys. Ed., 15, 723 (1977). 32. E. N. Peters, D. D. Stewart, J. Polym. Sci.,Polym. Letf.Ed., 17, 405 (1979). 33. E. N. Peters, US. Pat. 4,235,987 (1987); Chem. Abstr., 94, 66,439 (1981). 34. E. N. Peters, D. D. Stewart, J. J. Bohan, D. W. McNeil, J. E1ustomer.r Plust., 10, 29 (1978). 35. J. A. Yancey, J . Chromatogr. Sci., 23, 370 (1985). 36. E. E. Sotnikov, Zh. K. Torosyan, Zh. Anal. Khim., 40, 1887 (1985). 37. W. S. Coblenz, G. H. Wiseman, P. B. Davis, R. W. Rice, Mater. Sci. Res., 17, 271 (1984). 38. I. I. Miroshnikova, L. N. Sakharova, V. T. Minakov, A. B. Blyumenfel'd, N. L. Fedorova, L. I. Golubenkova, Plust. Massy, 38 (1987); Chem. Abstr., 108, 38,882 (1988).

15.2.7.3. Preparation of Other Carborane-Incorporated Inorganic Polymers.

An oligomer consisting of a number of phosphorus-linked closo-1,7-C,B,,Hl, polyhedra is obtained by treating a solid product from the reaction of Li,[1,7-C,Bl,Hl,] and PCl, (in a 1:2 ratio) with methanol. The structure of the oligomer is believed to be Cl,P[-CB,oHloCP(C1)-]50CH3 Without the presence of methanol (n z 5) is formed'-3. [-CBl,H,,C-PCl], l-Methyl-l-[(o-carboranyl)methylene]-3,3,5,5-tetrachlorocyclotriphosphazene(in which o-carboranyl = closo-1,2-C,Bl,H, 1) polymerizes, when heated at 250°C, to A. Polymer A reacts with piperidine to give the polymer B, in which the chlorines have been

'.

-'n

n

B

A

replaced with piperidine groups and opening of the carborane cage has also occurred. A THF solution of tris(triphenylphosphine)rhodium(I) chloride reacts with B at 66°C to give polymer C. The latter catalyzes the hydrogenation of 1-hexene. Deprotonation of B PPh,

/

C with sodium hydride, followed by the addition of W(CO), or Mo(CO),, exposure of the resulting solution to ultraviolet irradiation and the addition of trimethylamine hydrochloride, gives rise to polymer D (M = W, M o ) ~ .

]

104 15.2. Ring-Ring and Ring-Polymer Interconversions 15.2.7. in Cage Polyboranes and Carboranes. 15.2.7.3. Preparation of Other Carborane-Incorporated Inorganic Polymers. B H M(CO),'-

43

k-YJ+-CH

(Me 3NH'),

n

D

, ,

Condensing Li[C,B ,,H ,] and Li[C,B ,H ,CHJ with dihalophosphazene polymers in inert solvents and replacing residual C1 atoms with hydrocarbyloxy groups produces polymers giving a high char yield when heated at approximately 1000°C5. Other polyphosphazenes that contain carborane moieties have also been r e p ~ r t e d ' ~ - ' ~ . Other polymers, or oligomers, prepared from the direct reaction of Li2[C2B,,H,,] with a di- or polyhalogenated species include:

-

+ C1,Si(CH3), Cl[-Si(CH,)2-CB,oH,,C-Si(CH,),-I,CI Li,[C,Bl,Hl,I + Cl,Sn(CH,), [-CBloHloC-Sn(CH3),-ln

Li,[C,B,,H,,]

+

LizCCzBioHioI + C1,Ge(CH3), LizCGBioHiol Li,[C,B,,H,,]

+ HgC12

+ 1, x-(-SCl),-l,

[-CB,OH,OC-G~(CH~)Z-]" C-CB1oH1oC-Hg-I"

x-C,B,,H,,

+ [-CB,,H,,CS-1,

(aI6 (C)' (d)339 (e)3310911

-

where x = 1, 12. A few polyureacarboranes have been synthesized from cage-bonded nitrogen moieties; e.g., 1,7-(H,N),-c1oso-1,7-C2B1,H1, + OCN-(CH,),-NCO

[-CBloHloCNHCONH(CH,)6NHCONH-]n 12.

(T. ONAK)

1. 2. 3. 4. 5. 6. 7.

8. 9. 10. 11. 12. 13. 14. 15.

R. P. Alexander, H.A. Schroeder, Inorg. Chem., 5, 493 (1966). J. R. Reiner, R. P. Alexander, H. A. Schroeder, Inorg. Chem., 5, 1460 (1966). H. A. Schroeder, Inorg. Macrornol. Reo., I , 45 (1970). H. R. Allcock, A. G. Scopelianos, R. R. Whittle, N. M. Tollefson, J. Am. Chern. Soc., 105, 1316 (1983). L. L. Fewell, H. R. Allcock, J. P. OBrian, A. G. Scopelianos, US. Pat. Appl. 129798 (1980); Chem. Abstr., 93, 187,268 (1980). M. V. Sobolevskii, A. V. Zhigach, Z. M. Frolova, I. G. Sarishvili, Vysokomol.Soedin., B, 14,650 (1972); Chem. Abstr., 78, 44,026 (1973). H. A. Schroeder, S. Papetti, R. P. Alexander, J. F. Sieckhaus, T. L. Heying, Inorg. Chem.,8,2444 (1969). S. Bresadola, F. Rossetto, G. Tagliavini, Chem. Commun., 673 (1966). S. Bresadola, F. Rossetto, G. Tagliavini, Chim. Ind. (Milan),50, 452 (1968). N. S. Semenuk, S. Papetti, H. A. Schroeder, Znorg. Chem., 8, 2441 (1969). H. D. Smith, C. 0. Obenland, S. Papetti, Inorg. Chern., 5, 1013 (1966). N. I. Bekasova, V. V. Korshak, M. P. Prigozhina, A. I. Solomatina, Vysokomol.Soedin., B, 15, 629 (1973); Chem. Abstr., 80, 48,429 (1974). V. V. Korshak, N. I. Bekasova, M. P. Prigozhina, E. G. Bulychova, S. V. Vinogradova, Vysokomol.Soedin., B, 27, 847 (1985); Chem. Abstr., 104, 168,963 (1986). V. V. Korshak, N. I. Bekasova, S. V. Vinogradova, A. I. Solomatina, M. P. Prigozhina, E. G. Bulychova, Acta Polym., 38, 622 (1987) V. V. Korshak, N. I. Bekasova, S. V. Vinogradova, A. I. Solomatina, E. G. Bulychova, Vysokomol.Soedm., A , 29, 844 (1987); Chem. Abstr., 107, 23,845 (1987).

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc. 15.2. Ring-Ring and Ring-Polymer Interconversions 15.2.8. Silicon-Oxygen Systems

105

15.2.8. Silicon-Oxygen Systems Since the skeletal framework of the commercially important silicones is composed of siloxane (Si-0) bonds, their ring-ring and ring-polymer interconversions are not only of academic interest but of practical concern as well. Information has been selected electrically here and is meant to help understand the much larger body of information that cannot be included. Of first consideration are the key parameters that govern the ring-chain equilibria of siloxane systems: concentration of Si-0 skeletal components, functionality of the Si components, ring size, steric effects of Si substituents, temperature and catalytic conditions. These factors are seldom independent. The concentration of the Si-0 skeletal components is the parameter of greatest importance because of kinetic considerations; i.e., ring-opening chain growth polymerizations are second order (bimolecular) processes, retarded to a greater extent by dilution than are first order (monomolecular) cyclic generating depolymerization processes. Therefore, anything that dilutes the Si-0 content shifts its siloxane ring-chain equilibria toward rings. Although such dilution can be accomplished by solvent addition, this is not the only way to lower the concentration of the skeletal Si-0 components. The silicon’s organic substituents themselves exert a diluent effect, so the use of larger substituents leads to increased amounts of cyclics at equilibrium. Another less obvious way to reduce the siloxane concentration is to raise the temperature of the system, whereupon thermal expansion dilutes the skeletal components’. The functionality of the skeletal components, which is inversely related to the degree of organosubstitution, has important ramifications for these equilibria; i.e., trifunctional components RSiO,,, can lead to more complicated cyclic oligomeric and polymeric structures than can the R,SiO,,, components, in which ring size is the only significant degree of freedom. Furthermore, systems with more highly functional components require larger amounts of diluents to avoid the complication of gelation, which results when the mol wt of any part of the system approaches infinity. Such systems can be characterized by their equilibrium gel points; i.e., the maximum concentration at which polymeric systems can be maintained without gelation under skeletal bond-rearranging conditions2. Siloxane chemists employ a set of convenient structural abbreviations based on the functionality of the Si. Thus, monofunctional (R,SiOl,z), difunctional (RzSi02/2), tri-functional (RSiO,/,) and quadrifunctional (Si04/,) moieties are represented by the symbols M, D, T, and Q, respectively, and the R substituents are assumed to be methyl unless otherwise noted. Ring size determines the probability of ring closure from linear precursors; e.g., the probabilities of forming various cyclosiloxanes in the dimethylsiloxane system parallel those for aliphatic cyclics containing one-half as many ring atoms, because the -CH2and -Si(Me)zOl,2repeating units have similar shapes and interunit angle preferences, as well as similar steric requirements relative to interunit distances,. Cyclics with fewer than four siloxane units are not strain free. Although many cyclotrisiloxanes are known, they are usually not in systems in equilibrium unless the steric requirements of the organic substituents are very large. Steric effects of the silicon’s organic substituents are not limited to the simple diluent effect. If the organic substituents of neighboring Si atoms are sufficiently bulky, they can restrict the freedom of segmental rotation about the Si-0 bonds of the chain. To the

106

15.2. Ring-Ring and Ring-Polymer lnterconversions 15.2.8. Silicon-Oxygen Systems

extent that the chains lose this degree of segmental rotational freedom, their thermodynamic stability is decreased relative to that of cyclic structures, and the equilibrium cyclics content is correspondingly increased4. Thus, when R = t-Bu, (R,SiO), is preferred exclusively over higher cyclics or linears”. The most dramatic illustration of such steric enhancement of smaller rings is the synthesis of the long-sought cyclodisilonane (Mes,SiO), . Increasing temperature shifts ring-chain systems toward increased equilibrium cyclics The thermodynamic reasons for this are first the thermal expansion, which in effect decreases the concentration of the skeletal components. Second, increasing temperature increases the level of rotation about the C-Si and C-C bonds of the silicon’s organic substituents. This contributes to thermal expansion, but it also increases the interaction of substituents on neighboring Si sites, restricting rotation about the Si-0 bonds and favoring cyclic structures. Finally, at some point a temperature is reached at which cyclics are favored simply because of their thermodynamic advantage with respect to the degree of freedom relating to translational energy. Equilibration of cyclosiloxanes, whether it be to form other cyclic oligomers or linear polymers, requires the use of alkaline or acidic catalysts’. Triorganosiloxy (R,SiO,/,) components are more susceptible to acid-catalyzed siloxane redistribution, whereas monoorganosilsesquioxy (RSiO,,,) components are more susceptible to alkaline catalysts. Although diorganosiloxy (R,SiO,/,) components are amenable to rearrangement by either acids or bases, the inductive effects of the substituents result in preferential catalyst requirements; e.g., the electron-withdrawing trifluoropropyl (F,CCH,CH,) group enhances base-catalyzed reactivity and retards acid-catalyzed reactivity. Although siloxane redistributions are usually homogeneous, with or without diluents, heterogeneous catalysis is sometimes employed, most notably with clays6 and sulfonated polystyrene resins7. Thermally unstable, transient bases are employed to facilitate deactivation of the catalyst upon completion of the desired redistribution; quaternary ammonium and phosphonium bases* are useful in this regard. Dipolar aprotic solvents such as dimethylsulfoxide markedly enhance the rates of otherwise sluggish processesg. The term “equilibration” must be examined in each context not only to determine the actual intent, but to evaluate whether the intent actually has been achieved (or even is achievable!). “Equilibration,” as used here, refers to the process whereby bonds of a particular type, such as Si-0 bonds, are subjected to redistribution until their thermodynamically most stable state is reached. (c.L. FRYE)

1. For an excellent, highly detailed, review entitled Cyclic Siloxanes and Silazanes, consult the chapter by 0. K. Johannson, C. L. Lee, in High Polymers, Vol. 26, Cyclic Monomers, K. C . Frisch, ed., John Wiley & Sons, 1972, p 459. 2. C . L. Frye, J. M. Klosowski, J. Am. Chem. SOC.,93,4599 (1971). 3. J. F. Brown Jr., G. M. J. Slusarczuk, J. Am. Chem. Soc., 87, 931 (1965). 4. Steric effects of this type are noted in the ring-chain equilibria of orthosilicates and organosiliconates derived from dihydroxyalkanes. Although not dealing specifically with cyclosiloxanes, the study did involve closely related 0-Si-0 structural elements and is hence pertinent to this subject. For details, consult C. L. Frye, J. Org. Chem., 34, 2496 (1969). 5. C. Eaborn, Organosilicon Compounds,Butterworths, London 1960, p 255. 6. W. Noll, The Chemistry and Technology of Silicones, Academic Press, New York, 1968. 7. C. J. Litteral, U. S. Pat. 3,694,405 (1972); Chem. Abstr., 78, 17,047d (1973). 8. A. R. Gilbert, S.W. Kantor, J. Polym. Sci., 40,35 (1959).

15.2. Ring-Ring and Ring-Polymer Interconversions 15.2.8. Si Iicon-Oxygen Systems 15.2.8.1. Formation of Cyclic Oligomers.

107

9. G. D. Cooper, J. R. Elliott, J. Polym. Sci., A-1, 4, 603 (1966); C. L. Lee, C. L. Frye, 0. K. Johannson, Polym. Preprints, 10, 1361 (1969) and literature cited therein. 10. U. Klingebiel, Angew. Chem., Znt. Ed. Engl., 20(8), 678 (1981). 11. M. J. Michalczyk, M. J. Fink, K. J. Haller, R. West, J. Michl, Orgonornetallics, 5, 431 (1986).

15.2.8.1. Formation of Cyclic Oligomers.

Chlorosilanes are the most common precursor to siloxanes, cyclic or otherwise, because of their ready availability from the direct processing of alkyl chlorides with Si. Upon hydrolysis, chlorosilanes are converted to polysiloxanes and their cyclic oligomers. The hydrolysis of chlorosilanes can involve not only silylation of H,O by the Sic1 moiety to yield SiOH, but also the subsequent and competing silylation of the silanol by still more of the Sic1 moiety to yield siloxanes. Siloxane bond formation can result not only from such chlorosilane silanolyses, but also from silanol-silanol condensation. The resulting siloxane bonds are themselves also susceptible to acid-catalyzed redistributions, which, if allowed to occur, alter the initially formed product composition. The relative importance of these competing reactions depends on controlling factors such as manner of reactant combination (H,O-chlorosilane ratio, order of addition, batch vs. flow mode), presence of solvents o r surfactants, contact time, temperature, etc. Thus the cyclic oligomer content of the resulting hydrolyzate depends on the above parameters, especially since they determine whether kinetics or thermodynamics are controlling; e.g., although direct hydrolysis of Me,SiCl, can afford large amounts of cyclic trimer, this initially formed substance polymerizes unless steps are taken to preserve it. Although direct hydrolysis of mixtures of different chlorosilanes yield random mixtures of copolymers and cocyclics, synthetic schemes can be designed to prepare specific isomers e.g., although direct cohydrolysis of equimolar Me,SiCl, and PhMeSiCl, affords a complicated array of cyclotrimers, -tetramers, -pentamers and -hexamers, as well as linears, it is nevertheless possible to prepare high yields of specific desired species by scrupulous adherence to synthetic schemes designed to preclude randomization. Such an approach leads to the efficient synthesis of the isomeric 2,6diphenylhexamethylcyclotetrasiloxanes,free of the related 2,4-diphenyl isomers'. Me

I I

2 HSiCl

+ PhMeSiCl,

Me

Hz0

PhMeSi(OSiMe,H),

buffered transitionmetal catalyst

(4

PhMeSi(OSiMe,H), (H~o) PhMeSi(OSiMe,OH), Ph \

Me','

Si-0-Si

I 0 I

Me,

Me,

'si-0-si

I Me'" I o + o I ,ph I 'Me

cis

I I

0

Si-O-Si,<

Si-O-Si,%

1 Phh44gC12 I

Ph ~

,Me Ph

trans

(b)

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc. 15.2. Ring-Ring and Ring-Polymer Interconversions 15.2.8. Si Iicon-Oxygen Systems 15.2.8.1. Formation of Cyclic Oligomers.

107

9. G. D. Cooper, J. R. Elliott, J. Polym. Sci., A-1, 4, 603 (1966); C. L. Lee, C. L. Frye, 0. K. Johannson, Polym. Preprints, 10, 1361 (1969) and literature cited therein. 10. U. Klingebiel, Angew. Chem., Znt. Ed. Engl., 20(8), 678 (1981). 11. M. J. Michalczyk, M. J. Fink, K. J. Haller, R. West, J. Michl, Orgonornetallics, 5, 431 (1986).

15.2.8.1. Formation of Cyclic Oligomers.

Chlorosilanes are the most common precursor to siloxanes, cyclic or otherwise, because of their ready availability from the direct processing of alkyl chlorides with Si. Upon hydrolysis, chlorosilanes are converted to polysiloxanes and their cyclic oligomers. The hydrolysis of chlorosilanes can involve not only silylation of H,O by the Sic1 moiety to yield SiOH, but also the subsequent and competing silylation of the silanol by still more of the Sic1 moiety to yield siloxanes. Siloxane bond formation can result not only from such chlorosilane silanolyses, but also from silanol-silanol condensation. The resulting siloxane bonds are themselves also susceptible to acid-catalyzed redistributions, which, if allowed to occur, alter the initially formed product composition. The relative importance of these competing reactions depends on controlling factors such as manner of reactant combination (H,O-chlorosilane ratio, order of addition, batch vs. flow mode), presence of solvents o r surfactants, contact time, temperature, etc. Thus the cyclic oligomer content of the resulting hydrolyzate depends on the above parameters, especially since they determine whether kinetics or thermodynamics are controlling; e.g., although direct hydrolysis of Me,SiCl, can afford large amounts of cyclic trimer, this initially formed substance polymerizes unless steps are taken to preserve it. Although direct hydrolysis of mixtures of different chlorosilanes yield random mixtures of copolymers and cocyclics, synthetic schemes can be designed to prepare specific isomers e.g., although direct cohydrolysis of equimolar Me,SiCl, and PhMeSiCl, affords a complicated array of cyclotrimers, -tetramers, -pentamers and -hexamers, as well as linears, it is nevertheless possible to prepare high yields of specific desired species by scrupulous adherence to synthetic schemes designed to preclude randomization. Such an approach leads to the efficient synthesis of the isomeric 2,6diphenylhexamethylcyclotetrasiloxanes,free of the related 2,4-diphenyl isomers'. Me

I I

2 HSiCl

+ PhMeSiCl,

Me

Hz0

PhMeSi(OSiMe,H),

buffered transitionmetal catalyst

(4

PhMeSi(OSiMe,H), (H~o) PhMeSi(OSiMe,OH), Ph \

Me','

Si-0-Si

I 0 I

Me,

Me,

'si-0-si

I Me'" I o + o I ,ph I 'Me

cis

I I

0

Si-O-Si,<

Si-O-Si,%

1 Phh44gC12 I

Ph ~

,Me Ph

trans

(b)

108

15.2. Ring-Ring and Ring-Polymer lnterconversions 15.2.8. Silicon-Oxygen Systems 15.2.8.2. Cyclosiloxane Structural Diversity.

Chlorosilanes are also converted to siloxanes by reactions not involving hydrolysis’. Most are highly exothermic, and appropriate measures for heat dissipation are recommended for safety. Thus chlorosilanes can be converted to siloxanes by reaction with DMSO or with Na,CO, or ZnO in suitable solvents such as ethyl acetate or dioxane. Siloxanes can also be obtained by the reaction of alcohols with chlorosilanes, but this is really a kind of hydrolysis in which the water is generated in situ as a by-product of the formation of alkyl chloride from the alcohol and HCl. Siloxanes can of course be prepared from the reaction of H,O with many other kinds of hydrolyzable silanes (e.g., sulfato, iodo, bromo, fluoro, alkoxy, aryloxy, acyloxy, amino, amido, ketoximo) but such intermediates are themselves derived from chlorosilane precursors. Acetoxysilanes undergo thermolysis to yield siloxane bonds. Thermal-catalytic cracking of siloxane polymers is used to prepare cyclosiloxane 01igomers’-~. Also useful is the catalyzed solution equilibration of linear polymers to yield cyclic oligomers2. In addition to affording simple monocyclic oligomers from diorganosiloxane linear polymers, these techniques can also be applied to the preparation of polycyclic oligomers from more branched polymers such as the organosilsesqui~xanes~~~. (C.L.FRYE) 1. D. E. Spielvogel, C. L. Frye, J. Organomet. Chem., 161, 165 (1978). 2. 0. K. Johannson, C. L. Lee, in High Polymers, Vol. 26, Cyclzc Monomers, K. C. Frisch, ed., John Wiley & Sons, 1972, p 459. 3. C. Eaborn, Organosilicon Compounds, Butterworths, London 1960, p 255. 4. W. Noll, The Chemistry and Technology of Silicones, Academic Press, New York, 1968. 5. A. J. Barry, W. H. Daudt, J. J. Domicone, J. W. Gilkey, J. Am. Chem. Soc., 77,4248 (1955). 6. J. F. Brown Jr., L. H. Vogt Jr., P. I. Prescott, J. Am. Chem. Soc., 86, 1120 (1964).

15.2.8.2. Cyclosiloxane Structural Diversity.

Cyclosiloxanes derived from diorganosilanes such as Me,SiCl, form a homologous series ranging from the strained cyclotrisiloxanes to macrocyclics comprised of 40 or more R,Si02,, units1. One macrocyclic synthesis’ involves the selective conversion of (Me,SiO),, i.e., D,, to cyclopolymers of D, simply upon standing in Me,SiCl, solution: D,

MezSiClz

D,

+ D, + D,,,

etc.

Catenanes (i.e., systems containing two or more interpenetrating rings) can be formed in which one of the rings is a macrocyclosiloxane3. Catenanes may also be present in the macrocyclics fraction of diorganosiloxane equilibrates. There are many cyclocarbosiloxanes in which Si atoms are joined by intervening organic structural moieties4. Whereas these intervening organic structural units can range from 1,4-dihydroarylenes to carboranylenes (e.g., -CB loH,oC-), the most studied series contains aliphatic moieties, e.g., fMe,Si(CH,),SiMe,Oj-,. When n = 1, the smallest strain-free cyclosiloxane is the dimeric eight-membered ring, ; four-membered monomeric disiloxetane would be highly tMe,SiCH,SiMe,O j zthe strained and is not yet known. Because of their availability from the hydrosilylation of vinylsilanes, structures with n = 2 (i.e., a two-carbon intervening moiety) are best known, and because of the greater length of the two-carbon structural elements, more stable

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc. 108

15.2. Ring-Ring and Ring-Polymer lnterconversions 15.2.8. Silicon-Oxygen Systems 15.2.8.2. Cyclosiloxane Structural Diversity.

Chlorosilanes are also converted to siloxanes by reactions not involving hydrolysis’. Most are highly exothermic, and appropriate measures for heat dissipation are recommended for safety. Thus chlorosilanes can be converted to siloxanes by reaction with DMSO or with Na,CO, or ZnO in suitable solvents such as ethyl acetate or dioxane. Siloxanes can also be obtained by the reaction of alcohols with chlorosilanes, but this is really a kind of hydrolysis in which the water is generated in situ as a by-product of the formation of alkyl chloride from the alcohol and HCl. Siloxanes can of course be prepared from the reaction of H,O with many other kinds of hydrolyzable silanes (e.g., sulfato, iodo, bromo, fluoro, alkoxy, aryloxy, acyloxy, amino, amido, ketoximo) but such intermediates are themselves derived from chlorosilane precursors. Acetoxysilanes undergo thermolysis to yield siloxane bonds. Thermal-catalytic cracking of siloxane polymers is used to prepare cyclosiloxane 01igomers’-~. Also useful is the catalyzed solution equilibration of linear polymers to yield cyclic oligomers2. In addition to affording simple monocyclic oligomers from diorganosiloxane linear polymers, these techniques can also be applied to the preparation of polycyclic oligomers from more branched polymers such as the organosilsesqui~xanes~~~. (C.L.FRYE) 1. D. E. Spielvogel, C. L. Frye, J. Organomet. Chem., 161, 165 (1978). 2. 0. K. Johannson, C. L. Lee, in High Polymers, Vol. 26, Cyclzc Monomers, K. C. Frisch, ed., John Wiley & Sons, 1972, p 459. 3. C. Eaborn, Organosilicon Compounds, Butterworths, London 1960, p 255. 4. W. Noll, The Chemistry and Technology of Silicones, Academic Press, New York, 1968. 5. A. J. Barry, W. H. Daudt, J. J. Domicone, J. W. Gilkey, J. Am. Chem. Soc., 77,4248 (1955). 6. J. F. Brown Jr., L. H. Vogt Jr., P. I. Prescott, J. Am. Chem. Soc., 86, 1120 (1964).

15.2.8.2. Cyclosiloxane Structural Diversity.

Cyclosiloxanes derived from diorganosilanes such as Me,SiCl, form a homologous series ranging from the strained cyclotrisiloxanes to macrocyclics comprised of 40 or more R,Si02,, units1. One macrocyclic synthesis’ involves the selective conversion of (Me,SiO),, i.e., D,, to cyclopolymers of D, simply upon standing in Me,SiCl, solution: D,

MezSiClz

D,

+ D, + D,,,

etc.

Catenanes (i.e., systems containing two or more interpenetrating rings) can be formed in which one of the rings is a macrocyclosiloxane3. Catenanes may also be present in the macrocyclics fraction of diorganosiloxane equilibrates. There are many cyclocarbosiloxanes in which Si atoms are joined by intervening organic structural moieties4. Whereas these intervening organic structural units can range from 1,4-dihydroarylenes to carboranylenes (e.g., -CB loH,oC-), the most studied series contains aliphatic moieties, e.g., fMe,Si(CH,),SiMe,Oj-,. When n = 1, the smallest strain-free cyclosiloxane is the dimeric eight-membered ring, ; four-membered monomeric disiloxetane would be highly tMe,SiCH,SiMe,O j zthe strained and is not yet known. Because of their availability from the hydrosilylation of vinylsilanes, structures with n = 2 (i.e., a two-carbon intervening moiety) are best known, and because of the greater length of the two-carbon structural elements, more stable

15.2. Ring-Ring and Ring-Polymer lnterconversions 15.2.8. Silicon-Oxygen Systems 15.2.8.2. Cyclosiloxane Structural Diversity.

109

cyclic monomers result; e.g., the preparation of the strained five-membered ring from the hydrolysis product of 1,2-bis-(chlorodimethylsilyl)ethane uses thermal depolymerization in the presence of an alkaline catalyst5:

HzO

ClMe,SiCH,CH,SiMe,Cl

polymeric hydrolyzate

A

SiO -

Me,Si

/CHZCH,,

' O /

SiMe,

(b)

Also known6 is the five-membered ring system in which the Si sites are functionally substituted with hydrolyzable chlorine or alkoxy groups as well as some spiro derivatives, e.g.: Me

CH,CH,

c1

\

OR

SiMe, RO

Me2 CHZCH,, / 0-c, CH2 / Me,Si \ O / ~ ~ \ ~ - ~ , C H , I Me2 With still longer intervening organic moieties, strain-free cyclosiloxanes result, e.g., and its polymers7. the seven-membered 2,2,7,7-tetramethyl-l-oxa-2,7-disilacycloheptane The ultimate extension of this cyclocarbosiloxane synthesis is illustrated by the next example. Hydrosilylation of both ends of 1,7-octadiene with HMe,SiCl gives an adduct that upon hydrolysis yields fMe,Si(CH,),SiMe,O~~, which in turn affords the 22-membered dimeric and 1I-membered monomeric cyclosiloxanes upon alkaline thermol ysis 8. If monomers of functionality greater than two are used, then polycyclic carbosiloxanes can arise. Thus the hydrolysis and condensation of C1,MeSiCH2CH,SiMeC1, leads to a tricyclo cage structure' having four bridgehead Si sites: Me Si Me

I

Me

I

Cl-SiCH,CH,Si-Cl

l

c1 I

/CH2 - i- - - ~. O -\'-i- -0- - - - - - -.- --SiMe MeSi .- - .- /c-H

/\jfH2 ~

\

o

/

CH,

0

low mol wt siloxane hydrolyzate This crystalline product distills as it forms from the dropwise addition of the low mol wt hydrolyzate to hot (350°C) hydrogenated terphenyl containing an alkaline catalyst

15.2. Ring-Ring and Ring-Polymer lnterconversions 15.2.8. Silicon-Oxygen Systems 15.2.8.2. Cyclosiloxane Structural Diversity.

110 ~

(sodium 2-ethylhexoxide). A distorted adamantane structure may also have been involved but does not appear to have been considered, for the cyclic dimer obtained upon hydrolysis of Cl,MeSi(CH,),SiMeCl, (ref. 4, p. 114) is formulated: CH, -CH,-CH,

I I 0 I Me-Si-0-Si-Me I

Me-Si-0-Si-Me

I I 0 I

CH2-CH2-CH2

The hydrolysis of Cl,MeSiCH,SiMeCl, forms the prismatic tetracyclo cage shown on the rightlo:

R

R I

I

The alternative on the left was not formed (as shown by NMR), owing to its two strained cyclotrisiloxane rings. Bicyclo structures are also prepared in high yields by equilibrating appropriate hydrolyzates, e.g.' :

HSiC1,

I

R

+ ViMe'SiCl

Pt

R

R

I Me, C1,Si-CH,CH,SiCl

hydrolysis

1

polymeric hydrolyzate

In the above cyclosiloxane synthesis sufficient solvent is added during catalytic equilibration to favor cyclic formation. After equilibrium is established, the catalyst is neutralized or removed, and the oligomeric products isolated by distillation or crystallization.

15.2. Ring-Ring and Ring-Polymer Interconversions 15.2.8. Silicon-Oxygen Systems 15.2.8.2. Cyclosiloxane Structural Diversity.

111

~~

~~

Available cyclosiloxanes include the prismatic polycyclic organosilsesquioxanes, e.g. :

R '

T,

TlO

T12

ISO-T,,

As partially illustrated for the cubic octamer, an organo-substituted Si occupies each vertex, and oxygen atoms are located at the midpoint of each edge; T is a trifunctional RSiO,,, moiety. These structures can be prepared via alkaline-solution equilibration of the polymeric hydrolyzates of organotrichlorosilanes, and many (RSi03,2)8-16mixtures are made in this way. Certain of these silsesquioxanes are obtained in almost quantitative yield when they can selectively crystallize from equilibrating solutions; e.g., Me-T, , Ph-T,, Ph-T,,. For some, however, direct hydrolysis and silanol condensation can lead directly to octameric silsesquioxane cage structures free of other oligomeric forms without the use of alkaline equilibration; e.g., Vi-T, free of its other oligomers can be obtained by the controlled hydrolysis of ViSi(OMe), with hydrochloric acid and t-butanol' Furthermore, several such octameric silsesquioxanes can be prepared from RSiCl, with hydrochloric acid in methanol". The formation of solely octameric products is kinetically controlled, since they are not thermodynamically favored. The rare hexameric triangular prism is known' , but the reported pyramidal tetrameric (t-BuSi03,2)4l 4 is actually octameric. In the HSiO,,, family of oligomers alkaline equilibration must be scrupulously avoided if cage species are to be obtained, because alkaline catalysis leads to redistribution, not only of the oxygen but also of the hydrido ligands. Fortunately, however, acidic hydrolysis of suitable precursors can be adapted for the synthesis. Thus traces of (HSiO,,,), are i ~ o l a t e d 'from ~ the cohydrolysis of HSiC1, and Me3SiC1 in strong H2S0, (80%). Much better yields of this octamer may also be prepared via the hydrolysis of dilute cyclohexane-acetic acid solutions of HSi(OMe), with conc. HC116. The higher homologs (HSi03,2)10-16are also obtained by HSiC1, hydrolysis and condensation with H 2 0 generated in situ via the sulfonation of an aromatic solvent16. Metallosiloxanes are known12 in which certain metals form stable mixed oxides with suitably substituted Si atoms. Thus the reaction of AlCl, with (Me,SiO), results in the structure' ':

'.

Me,? /'\SiMe,

Me2Si \O/

SiMe2

112

15.2. Ring-Ring and Ring-Polymer Interconversions 15.2.8. Silicon-Oxygen Systems 15.2.8.3. Ring-Ring Equilibration of Cyclosiloxanes.

-

Finally, oligomeric cyclosiloxanes containing SiO,,, units are also known”; cg., the octameric silsesquioxane (RSiO,,,), where R is Me,SiO--, as well as the corresponding dodeca- and tetradecameric compoundsz0. (C.L. FRYE)

1. 2. 3. 4.

5. 6. 7. 8. 9.

10.

11. 12. 13. 14. 15. 16. 17. 18. 19. 20.

J. A. Semlyen, Makromol. Chem., Mapromol. Symp., 6, 155 (1986); see also ref. 3 from 515.2.8. J. F. Hampton, U.S. Pat. 3,465,016 (1969); Chem. Abstr. 71, 81,5178 (1969). G. Karagounis, Proc. Int. Conf. Colloid Surf Sci.,671; 1975 Chem. Abstr., 85, 6 3 , 3 1 2 ~(1976). N. S. Nametkin, T. Kh. Islamor, L. E. Guselnikov, V. M. Vdovin, Russ. Chem. Rev., 41, 111 (1972). W. A. Piccoli, G. Haberland, R. L. Merker, J. Am. Chem. Soc., 82, 1883 (1960). C. L. Frye, W. T. Collins, J. Org. Chem., 35, 2964 (1970). L. H. Sommer, G. R. Ansul, J. Am. Chem Soc., 77, 2482 (1955). J. W. Ryan, C. A. Roth, unpublished results, Dow Corning Corp. C. L. Frye, unpublished study. This compound (mp 160°C) is prepared in 37% yield by taking advantage of kinetic factors, since its tetrafunctional structural element strongly favors polymeric structures under siloxane bond equilibrating conditions, even at high dilutions. The alternative structural possibility containing strained five-membered rings was excluded by the absence of characteristic infrared absorption at 935 cm-’ (10.7 p), For the preparation and isolation of this compound (R = CH,) see H. A. Clark, Br. Pat. 672,824 (1952); Chem. Abstr., 47, 7862 (1953). This and related materials fail to undergo polymerization under siloxane rearranging conditions: P. J. Wang, Y. J. Zeng, C. T. Huang, Y. Lin, J. Polym. Sci., 30,525 (1958). The structure is established by NMR: D. J. Cooke, N. C. Lloyd, W. J. Owen, J. Organomet. Chem., 22, 55 (1970). C. L. Frye, W. T. Collins, unpublished study at Dow Corning Corp. K. Olsson, Ark. Kemi, 13, 367, (1958); Chem. Abstr., 53, 17,887e (1959). M. M. Sprung, F. 0.Guenther, J. Am. Chem. SOC.,77,3990,3996 (1955). See also J. F. Brown Jr., L. H. Vogt Jr., J. Am. Chem. Soc., 87,4313 (1965). E. Wiberg, W. Simmler, 2. Anorg. Allg. Chem., 282, 330 (1955). R. Muller, R. Kohne, S. Sliwinski, J. Prakt. Chem., 9, 71 (1959). C. L. Frye, W. T. Collins, J. Am. Chem. SOC., 92, 5586 (1970). A. A. Zhdanov, K. A. Andrianov, M. M. Levitskii, Izv.Akad. Nauk. SSSR.,Ser. Khim., 25, 376 (1976). CB Translation; Chem. Abstr., 85, 5748q (1976). C. Ercolani, A. Camilii, G. Sartori, J. Chem. Soc., A , 606 (1966). D. Hoebbel, W. Wieker, Z. Anorg. Allg. Chem., 384, 43 (1971). P. A. Agaskar, V W. Day, W. G. Klemperei, J . Am. Chem. SOC.,109, 5554 (1987).

15.2.8.3. Ring-Ring Equilibration of Cyclosiloxanes. If the position of equilibrium between rings and chains is in favor of rings, then the neat (i.e., diluent-free) system can be studied. If not, then it is necessary to add a diluent to shift the equilibrium toward the cyclics. The most studied’ system is the (Me,SiO), family because of its commercial importance. Although the neat system is ca. 83 % polymeric, dilution of the system to ca. 20 % Me,SiO with solvents (e.g., THF, toluene) precludes h e a r s at equilibrium. The equilibrium cyclics are predominantly tetramer and pentamer, with decreasing amounts of the larger oligomeric cyclics2. The abundances parallel their probabilities, determined by the same factors that are operative in other cyclic distributions. The distribution of tetramers, pentamers, hexamers, etc., may vary with the substituents (i.e., polarity, bulk), but this has not been studied definitively. Ring-ring interconversions can interconvert specific pairs of cyclosiloxanes selectively without causing unwanted wholesale structural proliferation3; e.g., conditions are defined that interconvert cis- and trans-2,6-diphenylhexamethylcyclotetrasiloxane with-

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc. 112

15.2. Ring-Ring and Ring-Polymer Interconversions 15.2.8. Silicon-Oxygen Systems 15.2.8.3. Ring-Ring Equilibration of Cyclosiloxanes.

-

Finally, oligomeric cyclosiloxanes containing SiO,,, units are also known”; cg., the octameric silsesquioxane (RSiO,,,), where R is Me,SiO--, as well as the corresponding dodeca- and tetradecameric compoundsz0. (C.L. FRYE)

1. 2. 3. 4.

5. 6. 7. 8. 9.

10.

11. 12. 13. 14. 15. 16. 17. 18. 19. 20.

J. A. Semlyen, Makromol. Chem., Mapromol. Symp., 6, 155 (1986); see also ref. 3 from 515.2.8. J. F. Hampton, U.S. Pat. 3,465,016 (1969); Chem. Abstr. 71, 81,5178 (1969). G. Karagounis, Proc. Int. Conf. Colloid Surf Sci.,671; 1975 Chem. Abstr., 85, 6 3 , 3 1 2 ~(1976). N. S. Nametkin, T. Kh. Islamor, L. E. Guselnikov, V. M. Vdovin, Russ. Chem. Rev., 41, 111 (1972). W. A. Piccoli, G. Haberland, R. L. Merker, J. Am. Chem. Soc., 82, 1883 (1960). C. L. Frye, W. T. Collins, J. Org. Chem., 35, 2964 (1970). L. H. Sommer, G. R. Ansul, J. Am. Chem Soc., 77, 2482 (1955). J. W. Ryan, C. A. Roth, unpublished results, Dow Corning Corp. C. L. Frye, unpublished study. This compound (mp 160°C) is prepared in 37% yield by taking advantage of kinetic factors, since its tetrafunctional structural element strongly favors polymeric structures under siloxane bond equilibrating conditions, even at high dilutions. The alternative structural possibility containing strained five-membered rings was excluded by the absence of characteristic infrared absorption at 935 cm-’ (10.7 p), For the preparation and isolation of this compound (R = CH,) see H. A. Clark, Br. Pat. 672,824 (1952); Chem. Abstr., 47, 7862 (1953). This and related materials fail to undergo polymerization under siloxane rearranging conditions: P. J. Wang, Y. J. Zeng, C. T. Huang, Y. Lin, J. Polym. Sci., 30,525 (1958). The structure is established by NMR: D. J. Cooke, N. C. Lloyd, W. J. Owen, J. Organomet. Chem., 22, 55 (1970). C. L. Frye, W. T. Collins, unpublished study at Dow Corning Corp. K. Olsson, Ark. Kemi, 13, 367, (1958); Chem. Abstr., 53, 17,887e (1959). M. M. Sprung, F. 0.Guenther, J. Am. Chem. SOC.,77,3990,3996 (1955). See also J. F. Brown Jr., L. H. Vogt Jr., J. Am. Chem. Soc., 87,4313 (1965). E. Wiberg, W. Simmler, 2. Anorg. Allg. Chem., 282, 330 (1955). R. Muller, R. Kohne, S. Sliwinski, J. Prakt. Chem., 9, 71 (1959). C. L. Frye, W. T. Collins, J. Am. Chem. SOC., 92, 5586 (1970). A. A. Zhdanov, K. A. Andrianov, M. M. Levitskii, Izv.Akad. Nauk. SSSR.,Ser. Khim., 25, 376 (1976). CB Translation; Chem. Abstr., 85, 5748q (1976). C. Ercolani, A. Camilii, G. Sartori, J. Chem. Soc., A , 606 (1966). D. Hoebbel, W. Wieker, Z. Anorg. Allg. Chem., 384, 43 (1971). P. A. Agaskar, V W. Day, W. G. Klemperei, J . Am. Chem. SOC.,109, 5554 (1987).

15.2.8.3. Ring-Ring Equilibration of Cyclosiloxanes. If the position of equilibrium between rings and chains is in favor of rings, then the neat (i.e., diluent-free) system can be studied. If not, then it is necessary to add a diluent to shift the equilibrium toward the cyclics. The most studied’ system is the (Me,SiO), family because of its commercial importance. Although the neat system is ca. 83 % polymeric, dilution of the system to ca. 20 % Me,SiO with solvents (e.g., THF, toluene) precludes h e a r s at equilibrium. The equilibrium cyclics are predominantly tetramer and pentamer, with decreasing amounts of the larger oligomeric cyclics2. The abundances parallel their probabilities, determined by the same factors that are operative in other cyclic distributions. The distribution of tetramers, pentamers, hexamers, etc., may vary with the substituents (i.e., polarity, bulk), but this has not been studied definitively. Ring-ring interconversions can interconvert specific pairs of cyclosiloxanes selectively without causing unwanted wholesale structural proliferation3; e.g., conditions are defined that interconvert cis- and trans-2,6-diphenylhexamethylcyclotetrasiloxane with-

15.2. Ring-Ring and Ring-Polymer Interconversions 15.2.8. Silicon-Oxygen Systems 15.2.8.3. Ring-Ring Equilibration of Cyclosiloxanes.

113

out generating the thermodynamically more probable 2,4-diphenyl isomers. Furthermore, cis and trans forms of cyclotrisiloxanes such as (PhMeSiO), can also be interconverted without polymerization, remarkable considering the ring strain of such systems. These highly selective cyclosiloxane interconversions employ electrophilic catalysts such as ZnCl,, FeCl,, and polar nitroalkane solvents3. The mechanism for these selective stereoisomeric interconversions involves metal halide cleavage of siloxane bonds to yield chlorophenylmethylsiloxy and chlorozinc siloxanolate termini; the asymmetric chlorine-bearing site undergoes racemization before regeneration of the cyclosiloxane, i.e.: Ph

I

Me,Si-O-Si\ ZnC1,

+ Me

I

I

P

0

\ ISi-0

Ph

I

Me, Si -O-S!

‘,Me<

+ (cleavage)

M;

o

i

P

Si P h ” ‘c1

-SiMe,

Ph”

i “Me

SiMe,

I

0,

Despite their increased complexity, systems involving trifunctional moieties RSiO,,, have also been studied. The ViSiO,,, system4 steric requirements favor cyclization and allow higher concentration to be employed than with MeSiO,,,, which gels at very(!) low concentration. Furthermore, because of inductive effects, equilibration takes place more quickly than with EtSiO,,,; and, also, the products are more volatile than with the comparable phenyl oligomers, thus facilitating GLC assay of oligomer content. The distribution of cyclics relative to one another is independent of the concentration of the equilibrating solution (although this generalization may break down at very low concentration where an increase of the simplest unstrained species, Vi-T, can be expected at the expense of Vi-T(lo-14J, The T-10 and T-12 oligomers are strongly favored at equilibrium, e.g.: Oligomer

T,

T,

T,,

TI,

TI,

T,,

Wt. ratio

0

1

9.1

13.6

2.7

-0

114

15.2. Ring-Ring and Ring-Polymer lnterconversions 15.2.8. Silicon-Oxygen Systems 15.2.8.4. Ring-Chain Equilibration: Polymerization of Cyclosiloxanes.

~~

On the other hand, the total equilibrium concentration of the combined oligomers relative to polymer is inversely dependent upon total RSiO,,, levels; e.g., with 87% solvent in the system, 66% of the ViSiO,i, is present in discrete oligomeric species. The oligomeric fraction decreased to 5 % with 72% solvent and less than 1 % with 68% solvent. (C.L. FRYE)

1. For dilution effects on dimethylsiloxane ring-chain equilibria see, e.g., J. B. Carmichael, D. J. Gordon, F. J. Isackson, J. Phys. Chem., 71, 2011 (1967) and references therein. 2. J. F. Brown Jr., G. M. J. Slusarczuk, J. Am. Chem. SOC.,87, 931 (1965). 3. D. E. Spielvogel, C. L. Frye, J. Organornet. Chem., 161, 165 (1978). 4. C. L. Frye, W. T. Collins, unpublished study at Dow Corning Corp.

15.2.8.4. Ring-Chain Equilibration: Polymerization of Cyclosiloxanes.

Although silicone polymers are prepared by direct hydrolysis of chlorosilanes, cyclosiloxanes may be preferred, especially for the production of high rnol wt gums suitable for manufacturing silicone rubber. Cyclosiloxanes are more easily purified than the chlorosilane precursors; i.e., it is easier to obtain cyclics free of MeSiO,/, than to remove tiny amounts of MeSiCl, from the closely boiling Me,SiCl,. Therefore, to prepare high mol wt linear polymers free of unwanted branching, cyclics are often the preferred intermediates. There is of course a voluminous literature (mostly patents) dealing with this area of siloxane chemistry and the interested reader should consult an appropriate text for details's2. Polymer yields are optimized by avoiding the use of diluents. If solution polymerization is desired, then in order to avoid the formation of increased amounts of equilibrium cyclics it is necessary to employ more reactive cyclic reactants and selective polymerization conditions, e.g., strained cyclics such as (Me,SiO), or Me,SiCH,CH,SiMe,O can be selectively transformed to linear polymers using conditions that are so mild that little of the resulting polymers are depolymerized to other cyclics'. The use of strained cyclics and selective redistribution conditions is also mandatory for preparing polymers that have high equilibrium cyclics contents even in the absence of diluents. Thus (F,CCH,CH,MeSiO), is the preferred intermediate for a family of commercial fluorosilicones, since the use of the less reactive tetramers or pentamers results in products containing ca. 50% equilibrium cyclics at 25"C3. In this system the equilibrium cyclics content is temperature dependent, increasing to ca. 75 % when the equilibration is performed at 150°C'. In contrast, the Me,SiO system is only slightly shifted (not over 2 %) over the 25-200°C range. This illustrates the substituent segmental rotation/temperature effect discussed in 415.2.8). Studies of the silsesquioxane (i.e., RSiO,,,) systems exemplify the effects of dilution, temperature, and steric factors on siloxane ring-chain equilibria. Whereas solvents are avoided with unstrained diorganosiloxane polymerizations, their use is essential for the preparation of most resinous branched polymers from oligomeric monoorganosilsesquioxanes. Thus, (PhSiO,,z)l may be converted to stable high polymeric solutions by alkaline siloxane redistribution in suitable organic diluents (e.g., xylene, diphenylether) provided that concentrations below the equilibrium gel point (EGP) are employed5. The EGP is a sensitive indicator of the degree of cyclization in the highly branched

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc. 114

15.2. Ring-Ring and Ring-Polymer lnterconversions 15.2.8. Silicon-Oxygen Systems 15.2.8.4. Ring-Chain Equilibration: Polymerization of Cyclosiloxanes.

~~

On the other hand, the total equilibrium concentration of the combined oligomers relative to polymer is inversely dependent upon total RSiO,,, levels; e.g., with 87% solvent in the system, 66% of the ViSiO,i, is present in discrete oligomeric species. The oligomeric fraction decreased to 5 % with 72% solvent and less than 1 % with 68% solvent. (C.L. FRYE)

1. For dilution effects on dimethylsiloxane ring-chain equilibria see, e.g., J. B. Carmichael, D. J. Gordon, F. J. Isackson, J. Phys. Chem., 71, 2011 (1967) and references therein. 2. J. F. Brown Jr., G. M. J. Slusarczuk, J. Am. Chem. SOC.,87, 931 (1965). 3. D. E. Spielvogel, C. L. Frye, J. Organornet. Chem., 161, 165 (1978). 4. C. L. Frye, W. T. Collins, unpublished study at Dow Corning Corp.

15.2.8.4. Ring-Chain Equilibration: Polymerization of Cyclosiloxanes.

Although silicone polymers are prepared by direct hydrolysis of chlorosilanes, cyclosiloxanes may be preferred, especially for the production of high rnol wt gums suitable for manufacturing silicone rubber. Cyclosiloxanes are more easily purified than the chlorosilane precursors; i.e., it is easier to obtain cyclics free of MeSiO,/, than to remove tiny amounts of MeSiCl, from the closely boiling Me,SiCl,. Therefore, to prepare high mol wt linear polymers free of unwanted branching, cyclics are often the preferred intermediates. There is of course a voluminous literature (mostly patents) dealing with this area of siloxane chemistry and the interested reader should consult an appropriate text for details's2. Polymer yields are optimized by avoiding the use of diluents. If solution polymerization is desired, then in order to avoid the formation of increased amounts of equilibrium cyclics it is necessary to employ more reactive cyclic reactants and selective polymerization conditions, e.g., strained cyclics such as (Me,SiO), or Me,SiCH,CH,SiMe,O can be selectively transformed to linear polymers using conditions that are so mild that little of the resulting polymers are depolymerized to other cyclics'. The use of strained cyclics and selective redistribution conditions is also mandatory for preparing polymers that have high equilibrium cyclics contents even in the absence of diluents. Thus (F,CCH,CH,MeSiO), is the preferred intermediate for a family of commercial fluorosilicones, since the use of the less reactive tetramers or pentamers results in products containing ca. 50% equilibrium cyclics at 25"C3. In this system the equilibrium cyclics content is temperature dependent, increasing to ca. 75 % when the equilibration is performed at 150°C'. In contrast, the Me,SiO system is only slightly shifted (not over 2 %) over the 25-200°C range. This illustrates the substituent segmental rotation/temperature effect discussed in 415.2.8). Studies of the silsesquioxane (i.e., RSiO,,,) systems exemplify the effects of dilution, temperature, and steric factors on siloxane ring-chain equilibria. Whereas solvents are avoided with unstrained diorganosiloxane polymerizations, their use is essential for the preparation of most resinous branched polymers from oligomeric monoorganosilsesquioxanes. Thus, (PhSiO,,z)l may be converted to stable high polymeric solutions by alkaline siloxane redistribution in suitable organic diluents (e.g., xylene, diphenylether) provided that concentrations below the equilibrium gel point (EGP) are employed5. The EGP is a sensitive indicator of the degree of cyclization in the highly branched

15.2. Ring-Ring and Ring-Polymer Interconversions 15.2.8. Si Iicon-Oxygen Systems 15.2.8.5. End-Capping Reactions.

115

RSiO,,,-polymer systems; i.e., higher EGP values reflect higher degrees of cyclization in the polymer itself. If R is sufficiently large (e.g., C18H3,-), no solvent is required to keep the equilibrium mol wt distribution within discrete bounds; in fact, such systems at equilibrium may contain little or no material with mol wt higher than the favored polycyclic cage compounds; i.e. (RSiO,,,), = 8--14. Such systems may have no (EGP’s). This is not to say that gelation can never be observed; e.g., hydrolysis of the parent chlorosilanes under non-bond-rearranging conditions can lead to gelation owing to kinetic factors; however, such kinetic gels will undergo transformation to lower mol wt gel-free distributions under bond-equilibrating conditions. With a substituent as small as methyl, neighboring group interactions are minimal, and the probability of equilibrium gelation is so high that it can be avoided only by the use of large amounts of solvents; thus the EGP for MeSiO,,, is substantially less than 2 %. Increasing sharply with size, the EGP is ca. 33 % for ViSiO,,, 6 , and about 50 % for PhSiO,/, ’. (c.L. FRYE)

1. W. Noll, The Chemistry and Technology of Silicones, Academic Press, New York, NY, (1968). 2. J. E. McGrath, J. S. Riffle, A. K. Banthia, I. Yilgor, G. L. Wilkes, Chemistry of Synthetic High Polymers, ACS Symp. Ser. 212, F. E. Bailey, ed., American Chemical Society, Washington, DC, 1983, p. 145. 3. C. L. Lee, C. L. Frye, 0. K. Johannson, Polym. Preprints, 10, 1361 (1969). 4. The literature contains widely divergent values for the equilibrium cyclics content of this fluorosilicone system. Thus, 0. R. Pierce, G. W. Holbrook, 0. K. Johannson, J. C . Saylor, E. D. Brown, Ind. Eng. Chem.,52, 783 (1960) report a value of 85 %, while J. S . Razzano [e.g., U.S. Pat. 3,937,684 (1976); Chem. Abstr., 84, 60,868h (1976)l claims a value of only 20-25%, while unpublished work by C. L. Frye at Dow Corning Corp. indicates a value of ca. 50% at RT and also reveals the extremely pronounced effect of temperature on this equilibrium. 5. C. L. Frye, J. M. Klosowski, J. Am. Chem. Soc., 93,4599 (1971). 6. G. H. Wagner, D. L. Bailey, A. N. Pines, M. L. Dunham, D. B. McIntyre, Ind. Eng. Chem.,45,367 (1953).

15.2.8.5. End-Capping Reactions. End-blocking moieties, being minor components in polymeric systems, are difficult to analyze. The reaction of chlorosilanes produces hydrolyzates consisting of cyclics and linears with hydroxy and/or chlorine ends, depending upon conditions. Silanol functional linears are easily obtained and can be end-capped by silylation. More pertinent to this discussion, however, is the nature of end blocks that result from siloxane redistribution reactions. Conversion of cyclosiloxanes to equilibrium ring-chain distributions affords chains with ends arising from the catalyst; i.e., if KOH is used, the chains have ends bearing silanol and K silanolate functionality. Neutralization with CO, and H,O then converts the silanolates to silanols. Alternatively, the equilibrates, as well as the above hydrolyzates, can be silylated to convert the silanols and silanolates to other kinds of ends. Ring-chain equilibrations are often performed in the presence of other components capable of supplying chain ends. If their concentration is much greater than that of the catalyst, then the ends arising from the catalyst become negligible. Components that give

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc. 15.2. Ring-Ring and Ring-Polymer Interconversions 15.2.8. Si Iicon-Oxygen Systems 15.2.8.5. End-Capping Reactions.

115

RSiO,,,-polymer systems; i.e., higher EGP values reflect higher degrees of cyclization in the polymer itself. If R is sufficiently large (e.g., C18H3,-), no solvent is required to keep the equilibrium mol wt distribution within discrete bounds; in fact, such systems at equilibrium may contain little or no material with mol wt higher than the favored polycyclic cage compounds; i.e. (RSiO,,,), = 8--14. Such systems may have no (EGP’s). This is not to say that gelation can never be observed; e.g., hydrolysis of the parent chlorosilanes under non-bond-rearranging conditions can lead to gelation owing to kinetic factors; however, such kinetic gels will undergo transformation to lower mol wt gel-free distributions under bond-equilibrating conditions. With a substituent as small as methyl, neighboring group interactions are minimal, and the probability of equilibrium gelation is so high that it can be avoided only by the use of large amounts of solvents; thus the EGP for MeSiO,,, is substantially less than 2 %. Increasing sharply with size, the EGP is ca. 33 % for ViSiO,,, 6 , and about 50 % for PhSiO,/, ’. (c.L. FRYE)

1. W. Noll, The Chemistry and Technology of Silicones, Academic Press, New York, NY, (1968). 2. J. E. McGrath, J. S. Riffle, A. K. Banthia, I. Yilgor, G. L. Wilkes, Chemistry of Synthetic High Polymers, ACS Symp. Ser. 212, F. E. Bailey, ed., American Chemical Society, Washington, DC, 1983, p. 145. 3. C. L. Lee, C. L. Frye, 0. K. Johannson, Polym. Preprints, 10, 1361 (1969). 4. The literature contains widely divergent values for the equilibrium cyclics content of this fluorosilicone system. Thus, 0. R. Pierce, G. W. Holbrook, 0. K. Johannson, J. C . Saylor, E. D. Brown, Ind. Eng. Chem.,52, 783 (1960) report a value of 85 %, while J. S . Razzano [e.g., U.S. Pat. 3,937,684 (1976); Chem. Abstr., 84, 60,868h (1976)l claims a value of only 20-25%, while unpublished work by C. L. Frye at Dow Corning Corp. indicates a value of ca. 50% at RT and also reveals the extremely pronounced effect of temperature on this equilibrium. 5. C. L. Frye, J. M. Klosowski, J. Am. Chem. Soc., 93,4599 (1971). 6. G. H. Wagner, D. L. Bailey, A. N. Pines, M. L. Dunham, D. B. McIntyre, Ind. Eng. Chem.,45,367 (1953).

15.2.8.5. End-Capping Reactions. End-blocking moieties, being minor components in polymeric systems, are difficult to analyze. The reaction of chlorosilanes produces hydrolyzates consisting of cyclics and linears with hydroxy and/or chlorine ends, depending upon conditions. Silanol functional linears are easily obtained and can be end-capped by silylation. More pertinent to this discussion, however, is the nature of end blocks that result from siloxane redistribution reactions. Conversion of cyclosiloxanes to equilibrium ring-chain distributions affords chains with ends arising from the catalyst; i.e., if KOH is used, the chains have ends bearing silanol and K silanolate functionality. Neutralization with CO, and H,O then converts the silanolates to silanols. Alternatively, the equilibrates, as well as the above hydrolyzates, can be silylated to convert the silanols and silanolates to other kinds of ends. Ring-chain equilibrations are often performed in the presence of other components capable of supplying chain ends. If their concentration is much greater than that of the catalyst, then the ends arising from the catalyst become negligible. Components that give

116

15.2. Ring-Ring and Ring-Polymer lnterconversions 15.2.9. in Silicon-Nitrogen and Silicon-Sulfur Systems 15.2.9.1. Silicon-Nitrogen Syntheses.

rise to end blocks range from protic materials such as H,O and alcohols to aprotic substances such as (Me,Si),O or alkoxysilanes such as Me,Si(OMe),. The effect of end blocker concentration upon mol wt varies with the functionality of the skeletal components. Thus, in ring-chain equilibria involving diorganosiloxanes, the mol wt of the chains is very dependent upon the end blocker concentration; e.g., if the average degree of polymerization of a dimethylsiloxane were 1000 units, then the additional introduction of only 0.001 Me,SiOSiMe, per Me,SiO,,, moiety would reduce the average degree of polymerization to only 500 units. By contrast, the equilibrium mol wt distribution of organosilsesquioxane systems (i.e., RSiO,,,) is very insensitive to end blocker concentration. The average mol wt of these highly branched materials depends much more on the amount of inert diluent present during bond equilibration’. This is because silsesquioxane polymers have large amounts of builtin cyclic structural elements that can open up and add on the end-blocker moieties, resulting in an actual slight increase of mol wt rather than the drastic reductions observed for linear unbranched systems.

(c.L.FRYE) 1. C. L. Frye, J. M. Klosowski, J. Am. Chem. SOC.,93, 4599 (1971).

15.2.9. in Silicon-Nitrogen and Silicon-Sulfur Systems 15.2.9.1. Silicon-Nitrogen Syntheses.

-

The ammonolysis of dichlorosilanes forms cyclotri- and cycl~tetrasilazanes~-~. \

+ 3n NH,

n SiC1, /

(4

where n = 3, 4. Other classical preparation methods of four- to eight-membered cyclosilazanes are the condensation of 3,w-difunctional building unit^^-^ :

2

\ /

Si(NHR),

-

R I

N \ / \ /

Si

+ 2 RNH,

Si

/ \ / \

N

I

R

I

l

-Si-N-Si-C1

I

\

l

-N-Si-N-H(Li) / \ I

, : : I (

__t

\S/ /

\ /

N-Si

\

N-Si

/

\

7-

/ \

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc. 116

15.2. Ring-Ring and Ring-Polymer lnterconversions 15.2.9. in Silicon-Nitrogen and Silicon-Sulfur Systems 15.2.9.1. Silicon-Nitrogen Syntheses.

rise to end blocks range from protic materials such as H,O and alcohols to aprotic substances such as (Me,Si),O or alkoxysilanes such as Me,Si(OMe),. The effect of end blocker concentration upon mol wt varies with the functionality of the skeletal components. Thus, in ring-chain equilibria involving diorganosiloxanes, the mol wt of the chains is very dependent upon the end blocker concentration; e.g., if the average degree of polymerization of a dimethylsiloxane were 1000 units, then the additional introduction of only 0.001 Me,SiOSiMe, per Me,SiO,,, moiety would reduce the average degree of polymerization to only 500 units. By contrast, the equilibrium mol wt distribution of organosilsesquioxane systems (i.e., RSiO,,,) is very insensitive to end blocker concentration. The average mol wt of these highly branched materials depends much more on the amount of inert diluent present during bond equilibration’. This is because silsesquioxane polymers have large amounts of builtin cyclic structural elements that can open up and add on the end-blocker moieties, resulting in an actual slight increase of mol wt rather than the drastic reductions observed for linear unbranched systems.

(c.L.FRYE) 1. C. L. Frye, J. M. Klosowski, J. Am. Chem. SOC.,93, 4599 (1971).

15.2.9. in Silicon-Nitrogen and Silicon-Sulfur Systems 15.2.9.1. Silicon-Nitrogen Syntheses.

-

The ammonolysis of dichlorosilanes forms cyclotri- and cycl~tetrasilazanes~-~. \

+ 3n NH,

n SiC1, /

(4

where n = 3, 4. Other classical preparation methods of four- to eight-membered cyclosilazanes are the condensation of 3,w-difunctional building unit^^-^ :

2

\ /

Si(NHR),

-

R I

N \ / \ /

Si

+ 2 RNH,

Si

/ \ / \

N

I

R

I

l

-Si-N-Si-C1

I

\

l

-N-Si-N-H(Li) / \ I

, : : I (

__t

\S/ /

\ /

N-Si

\

N-Si

/

\

7-

/ \

15.2. Ring-Ring and Ring-Polymer Interconversions 15.2.9. in Silicon-Nitrogen and Silicon-Sulfur Systems 15.2.9.1. Silicon-Nitrogen Syntheses.

117

and the reaction of aminofunctional compounds with halogenosilyl functional

compound^^-^ :

Cyclodisilazanes are obtained in thcreactions of aminofluorosilanes with organolithiums : via thermal LiF elimination of Li aminofluor~silanes~-~' R

I

R where R = alkyl, anyl. Stepwise synthesis of cyclodi-; cyclotri- and cyclotetrasilazanes via acyclic compounds is achieved from difluorosilanes and LiNH,'2s'3 : H

F

A

R \ Si / -R2Si - LiF R/ \ NLi H

/N \

\

N H

/

SIR,,

R2Si

H /N\

HN\

I

Si R,

HN

SiR,

I

/

NH

-(

3

R2 H \ ,Si-N

I

R,Si,

SIR,

I

N -Si' H R2

NH

where R = CHMe,; CMe,. Reactions of Si halides with Li salts of silylamines form fourand six-membered cyclosilazanes14~1 5: /%Me, \ ->Si-NL) n SiHal, + nLiN /

where n

=

\

"

2, 3.

(u. KLINGEBIEL) 1. 2. 3. 4.

5.

6.

7. 8. 9. 10. 11. 12.

S . D. Brewer, C. P. Haber, J. Am. Chem. Soc., 70, 3888 (1948). W. Fink, Angew. Chem., Int. Ed. Engl., 5, 760 (1966). J. Haiduc, The Chemistry ofhorganic Ring Systems, Wiley-Interscience, New York, 1970. I. Haiduc, D. B. Sowerby, The Chemistry of Inorganic Homo- and Heterocycles Vol. 1, Academic Press, London, 1987. U. Wannagat, Chem. Ztg., 97, 105 (1973). U. Klingebiel, A. Meller, Chem. Ber., 109, 2430 (1976); Z . Naturforsch., Teil B, 32, 537 (1977); Z . Anorg. Allg. Chem. 428, 27 (1977). U. Klingebiel, H. Hluchy, A. Meller, Chem. Ber., I l l , 90 (1978). W. Clegg, U. Klingebiel, G. M. Sheldrick, N. Vater, Z . Anorg. Allg. Chem., 428, 88 (1981). W. Clegg, U. Klingebiel, C. Krampe, G. M. Sheldrick, Z . Nuturforsch., Ted B, 35, 275 (1980). W. Clegg, U. Klingebiel, G. M. Sheldrick, Z . Nuturforsch., Ted B, 37 423 (1982). D. Stalke, N. Kewcloh, U. Klingebiel, M. Noltemeyer, G. M. Sheldricke, Z . Naturforsch., Teil B, 42, 1237 (1987). U. Klingebiel, N. Vater, Chem. Ber., 116, 3277 (1983).

15.2. Ring-Ring and Ring-Polymer lnterconversions 15.2.9. in Silicon-Nitrogen and Silicon-Sulfur Systems 15.2.9.2. Silicon-Nitrogen Ring-Ring Interconversions.

118

13. U. Kliebisch, U. Klingenbiel, N. Vater, Chem. Ber., 118, 4561 (1985). 14. U. Klingebiel, J. Neemann, A. Meller, 2. Anorg. Allg. Chem., 429, 63 (1977). 15. S. Bartholmei, U. Klingebiel, G. M. Sheldrick, D. Stalke, 2.Anorg. Allg. Chem., 556, 129 (1988).

15.2.9.2. Silicon-Nitrogen Ring-Ring Interconversions.

Molecular rearrangements with change of the ring size include all the possible interconversions of the four-, six- and eight-membered rings''2. N-Methylated cyclodisilazanes react with catalytic amounts of NH,Br quantitatively to give cyclotrisiIazanes 'g3s4:

Ring contractions of N-trimethylsilylated cyclotetra- to cyclotrisilazanes occur on heating at higher than 350°C without a catalyst or at 300°C in the presence of ammonium salts5p7. Ring contraction with formation of N-chlorosilylcyclodisilazanes occurs in the reaction of cyclotrisilazanes with diorganodichlorosilanes in a mole ratio of 1:3:

c1

I

H

-Si-

c1

or in the reaction of cyclotetra- or cyclotrisilazanes with hydrogen halidess-' 3. These cyclodisilazanes are also obtained from six-membered ring compounds with dichlorosilazanes in a 1: 1 mol ratio: however, the yields of the cyclodisilazanes decrease, and dimethylsilyl-coupled cyclodisilazanes are also f ~ r m e d ' ~ . ~ ~ :

where n = 2-6. The reaction of cyclotetrasilazane with phenylisocyanate forms the cyclotrisilazane'6 : ! H -Si-N-Si-

I HN I

-Si-N-SiI H

\ /

I

I NH + 2 PhNCOI I

HN'

\I

'h ' Si

Si

'NH I/+

H

PhN

s'i'

/

\

I

NPh

I

H

(c)

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc. 15.2. Ring-Ring and Ring-Polymer lnterconversions 15.2.9. in Silicon-Nitrogen and Silicon-Sulfur Systems 15.2.9.2. Silicon-Nitrogen Ring-Ring Interconversions.

118

13. U. Kliebisch, U. Klingenbiel, N. Vater, Chem. Ber., 118, 4561 (1985). 14. U. Klingebiel, J. Neemann, A. Meller, 2. Anorg. Allg. Chem., 429, 63 (1977). 15. S. Bartholmei, U. Klingebiel, G. M. Sheldrick, D. Stalke, 2.Anorg. Allg. Chem., 556, 129 (1988).

15.2.9.2. Silicon-Nitrogen Ring-Ring Interconversions.

Molecular rearrangements with change of the ring size include all the possible interconversions of the four-, six- and eight-membered rings''2. N-Methylated cyclodisilazanes react with catalytic amounts of NH,Br quantitatively to give cyclotrisiIazanes 'g3s4:

Ring contractions of N-trimethylsilylated cyclotetra- to cyclotrisilazanes occur on heating at higher than 350°C without a catalyst or at 300°C in the presence of ammonium salts5p7. Ring contraction with formation of N-chlorosilylcyclodisilazanes occurs in the reaction of cyclotrisilazanes with diorganodichlorosilanes in a mole ratio of 1:3:

c1

I

H

-Si-

c1

or in the reaction of cyclotetra- or cyclotrisilazanes with hydrogen halidess-' 3. These cyclodisilazanes are also obtained from six-membered ring compounds with dichlorosilazanes in a 1: 1 mol ratio: however, the yields of the cyclodisilazanes decrease, and dimethylsilyl-coupled cyclodisilazanes are also f ~ r m e d ' ~ . ~ ~ :

where n = 2-6. The reaction of cyclotetrasilazane with phenylisocyanate forms the cyclotrisilazane'6 : ! H -Si-N-Si-

I HN I

-Si-N-SiI H

\ /

I

I NH + 2 PhNCOI I

HN'

\I

'h '

Si

'NH I/+

Si

H

PhN

s'i'

/

\

I

NPh

I

H

(c)

15.2. Ring-Ring and Ring-Polymer Interconversions 15.2.9. in Silicon-Nitrogen and Silicon-Sulfur Systems 15.2.9.2. Silicon-Nitrogen Ring-Ring Interconversions.

119

-

whereas (Me,SiNPh), is formed from the tetrameric (Me,SiNH), with aniline7,I7: (Me,SiNH),

+ PhNH,

(4

(Me,SiNPh),

where Me = CH,. N-Trimethylsilylhexamethylcyclotrisilazaneis prepared by the condensation of N-sodiohexamethylcyclotrisilazanewith chlorotrimethylsilane’s. The analogous reaction of N,N,’-dilithiohexamethylcyclotrisilazane and chlorotrimethylsilane at 160°Cyields not N,N’-bis-(trimethylsilyl)hexamethylcyclotrisilazane1g,but N-trimethylsilyl-N’-1-(1,1,3,3-pentamethyldisilazanyl)tetramethylcyclodisilazanezo: H

Me,

/N\

Me/? LiN\ Me

Me

,Me

si\ + 2 ClSiMe, I Me -Me,Si-N - 2 LiCl

NLi

/si\

P\’

Me

\

Me

I. N-SI-N

I hi/ Me Me’ \Me

Me

/

SiMe, (e)

\H

The formation of the four-membered ring is explained by an anionic rearrangement accompanied by ring contraction20,21.The isomerization of cyclotri- and cyclodisilazanes depends on the condition^^^^^^. The substitution of silylgroups is preferred at low T in diglyme as ~ o l v e n t Isomerization ~ ~ ~ ~ ~ . reactions are observed at higher T, e.g., at + 6 0 T in diglyme22,23: Me,Si \

Me,

/si\ N / SiMe, N I I

Me$\ + 2 BuLi

H

L + 60°C

HN Me$\

Me,

/si\ N-Si-N Me2

Me,Si-N

\ /

Si Me,

NH

I

SiMe,

N

I

/ \

%Me, H (f)

SiMe,

Me, Si

N H

Me,SiN

/

\

I Me,Si\

___)

-‘O”‘

NSiMe,

I

,SiMe, N

I

t 3 BuLi

Si Me3

-.1 3 Me3SiC

+ 60°C

Me, Me2 N-Si-N

Me,Si-N \si/ Me2

/

\

SiMe, SiMe,

120

15.2. Ring-Ring and Ring-Polymer Interconversions 15.2.9. in Silicon-Nitrogen and Silicon-Sulfur Systems 15.2.9.2. Silicon-Nitrogen Ring-Ring Interconversions.

The isomerizations are base-catalyzed and take place after metallationz3. In contrast to the reaction of cyclotrisilazanes with chlorosilanes, substitution is preferred with f l ~ o r o s i l a n e s ~at~RT. - ~ ~Contraction to form four-membered rings occurs with lithiated 1-fluorosilyl-5-trimethylsilylcyclotrisilazanes and RH compounds containing an acidic H atom3'. Substitution is observed with the substituents R = CH30, C,H,O, n-C,H,C-C and isomeric cyclodisilazanes are formed with the substituents R = C H 3 0 , C,H,O, n-C,H,C=C, C,H,C-C:

I

L

Me, Me,Si-N-Si-N H

Me., /si\ \ /

Me, N-Si-R

Si Me2

where R = CH,O, C,H,O, n-C,H,C-C, C,H,C=C. The first step of the reaction is ~ ~ ~ ~ ~ . alcoholates and alkynes the lithiation of the H-acid c o r n p ~ u n d Nucleophilic catalyze the ring contraction. I

/

N(3) Figure 1. Bond lengths

(A) and angles (").

15.2. Ring-Ring and Ring-Polymer Interconversions 15.2.9. in Silicon-Nitrogen and Silicon-Sulfur Systems 15.2.9.2. Silicon-Nitrogen Ring-Ring Interconversions.

121

Isomeric four-membered rings are not observed. The cleavage of the Si-N-ring bond occurs between the atoms N-1 and Si-6 2 5 , 3 0 * 3 1 . The mechanism of In the isomerization of cyclotri- and cyclodisilazanes is studied by the reaction of the lithiated trimethylsilylsubstituted cyclotrisilazane with Me,CSiF, 2 5 9 3 0 , 3 1 , four compounds are obtained in a 1: 1.5: 1:0.2 ratio: Me, Si\ Me, /SiF2CMe3 Me3SiN N-Si-N \si/ \H

,

1.5

Me,Si

Me3 Si

HN

I

Me,Si

I

/

HN\

N

\

SiMe,

I / NLi Si Me2

/

N

\

SiMe,

I

N

LSi/ \

Me2

+ Me3CSiF3

Me2

- LiF

SiF,CMe,

Me, N-Si-N

1

+ Me,SiN

,SiF2CMe3

\si/ Me2

Me,Si

I

1 0 . 2 ,

Me3CF,Si

/

N

\SiF,CMe,

/

N

\

\si/ Me2

SiMe,

I

N

\

SiF,CMe,

Bulky groups increase the tendency of ring contraction. Lithiated cyclotrisilazanes crystallize with T H F as dimers in boat conformation24s25. THF H Me, I Me2 / R ,N-Si Li Si-N Me,Si N ‘’ ‘N6# j:]$SiMe, \ N-Si / ‘Li/ \Si-iq R’ Me, 1 Me, H THF where R = H Z 4 , and the Si(l)-N(2)H-Si(3) angle is 125.5” and the Si(1)N(2)SiMe3-Si(3) angle is 116.7”. Because of the bulkiness of the SiMe, group of the lithiated SiMe,-substituted ringz5 (Fig. 1) the Si(l)-N(2)-Si(3) angle gets smaller than in the H-substituted compoundz4 and a contraction of the Si(2)-N(3) distance occurs above the ring”. The disubstituted isomeric rings C and D are formed by the intermediate lithiation of A and B and reaction with CMe,SiF, 28-32:

15.2. Ring-Ring and Ring-Polymer Interconversions 15.2.9. in Silicon-Nitrogen and Silicon-Sulfur Systems 15.2.9.2. Silicon-Nitrogen Ring-Ring Interconversions.

I\ \I I

FSi

D

/s1\

N

- LiF

I

-Si-

I

A

B

-Si-

I

-Si-

I

I

+ -SiF2

-

I / N\ . /

I

/N\ \

-Si-

I

Si

/

I

C

SiF

122

123

15.2. Ring-Ring and Ring-Polymer lnterconversions 15.2.9. in Silicon-Nitrogen and Silicon-Sulfur Systems 15.2.9.2. Silicon-Nitrogen Ring-Ring Interconversions.

Ring contraction is used to prepare silyl-bridged cyclotri- and c y c l o d i ~ i l a z a n e s The ~~~~~. isomerizations occur in boiling n-hexane or THF. The cleavage of the Si-N bond of the six-membered ring depends on the substituted fluorosilyl groups: Me,Si-R

I

I

)

D

I

- LiF

-LiF

Q

Me,Si-R

I

N

Me,%

/ \ \ /

SiMe,

N

I SiMe, I

Me-Si-F

I

N

Me&

/ \

R”

N R“ I SiMe,

I

SiMe,

I

SiMe,

I

f Me,%

r;r / \ \ /

SiMe,

N

N

Me,Si(

\ /

N

R’ N R’ 1 \ / /Si\ \ / 2 /Si\

I I

\iMe,

SiMe,

N

N

I

Me,&-R

Me,%

/ \ \N/

Me-Si-F

D

SiMe,

15.2. Ring-Ring and Ring-Polymer lnterconversions 15.2.9. in Silicon-Nitrogen and Silicon-Sulfur Systems 15.2.9.2. Silicon-Nitrogen Ring-Ring Interconversions.

124

Figure 2 Ring contraction reactions are used for the preparation of silyl-bridged cyclotri- and cyclodisilazanes 3,34. The dimerization of the isomeric cyclodisilazanes occurs with elimination of LiF. Monoclinic and triclinic crystalline modifications of one compound are described.33 Figure 2 shows the structure of compound A. In attempts to synthesize a silane with three cyclotrisilazane substituents, ring contraction is observed and two cyclodisilazanes connected by a (SiNSiNSi)- bridge and a cyclodisilazane connected by a (SiNSi)bridge to a cyclotrisilazane are f ~ r r n e d ~ ~ , ~ ~ . 393

Me3 Si I

3

Me,Si-N

Me,-/ N-Si-NLi

/ \ \ /

SiF

+

Si

Me3Si-N

Si

/

\ /

\

Si

Me2

Me3Si-N

Me2 N-Si-N

/ \

/si\

Me,SiF Me2 N-Si-

‘si’ Me2

I

I I

SiF,

Me2

F

N-Si-N-Si-N

SiF

Me2

/si\

1

F Me,SiF

Me2

N-SiMe,

(k)

15.2. Ring-Ring and Ring-Polymer Interconversions 15.2.9. in Silicon-Nitrogen and Silicon-Sulfur Systems 15.2.9.2. Silicon-Nitrogen Ring-Ring Interconversions.

Me,SiF

Me,Si

\

125

Me,SiF Me, I

/

N-Si-F

N-Si I . Me2 Me,SiF

I

F -

+ LiN

\

/SiMe2

Si -N Me2 Me, SiF

1

I

- LIF

Me2SiF I Me, Me, N-Si-N-Si--N/

Me,%

\N-Si /

1. Me2 Me,SiF

I F 1SiF Me2

Me, Si ‘N-SiFMe, ‘Si’ Me2

with the The reaction of 1,3-bis-(fluorodimethylsilyl-5-trimethylsilyl-cyclotrisilazane dilithiated cyclotrisilazane produces ring contraction, to form a silyl-bridged tris(cyclodisilazane)3s (see eq. (m) on page 126). With bis(difluoromethylsily1)-substituted cyclotrisilazane with the dilithiated ring, however, a fused-ring system, a bicyc10[4.2.0]octane, is formed35 (see Fig. 3) (see eq. (n) on page 127).

Figure 3

126

15.2. Ring-Ring and Ring-Polymer lnterconversions 15.2.9. in Silicon-Nitrogen and Silicon-Sulfur Systems 15.2.9.2. Silicon-Nitrogen Ring-Ring Interconversions.

15.2. Ring-Ring and Ring-Polymer Interconversions 15.2.9. in Silicon-Nitrogen and Silicon-Sulfur Systems 15.2.9.2. Silicon-Nitrogen Ring-Ring Interconversions.

m

33-

h

127

128

15.2. Ring-Ring and Ring-Polymer lnterconversions 15.2.9. in Silicon-Nitrogen and Silicon-Sulfur Systems 15.2.9.2. Silicon-Nitrogen Ring-Ring Interconversions.

Steric influence on ring contractions control reactions of Li salts of cyclotrisilazanes R‘ /

with the bulky aminofluorosilanes, R-SiF,-N

\

SiMe,)31,36:

(R = F, Me; R ’ = CMe,, SiMe,

SiMe,

1

Me, ,Si\ HN

+

Me

SiMe,

/Si\

N’ / \R,

N

I

I

Me2 Si

F

LiN’

\ NLi

I

I , SiMe,

Me,Si,

\

2) +F2Si

NR’SiMe3

(0)

- LiF

,N-Si-N

R‘

\si/

N-SiMe,-NH

F

Me2

Figure 4. A view of the molecule B showing the atom numbering and torsions angles (") for the ring bonds. The four-membered ring is almost planar. The dimensions are typical for Si ,N, rings with N-Si-N angles less than and LSi-N-Si greater than 90". The six-membered ring is far from planar. The ring does not correspond to any of the ideal conformations3'.

15.2. Ring-Ring and Ring-Polymer Interconversions 15.2.9.in Silicon-Nitrogen and Silicon-Sulfur Systems 15.2.9.2. Silicon-Nitrogen Ring-Ring Interconversions.

129

Substitution products are obtained with the aminofluorosilanes with the monolithiated ring36,and four-membered rings are formed in a further reaction31 as well as with the dilithium salt and the aminofluorosilanes in a 1:2 mole ratio. The Li salt of the disubstituted four-membered ring equilibrates to the Li salt of the six-membered ring3’ : Me3

:i

‘N-Si--P

R’: +FzSi

c

- LiF

/

R

\

/

\

SiMes

‘\R,

Me

SiMe,

R’:

‘SiMe,

- LiF

I

*, B

R’

Me, Si

i

IMeQ -%Me3

,.,

Me,Si

Si R’

F

Si-N Me,

\s[

/ \

R

Me

The silyl-bridged four-membered rings are not A, B observed in these reactions with aminofluorosilyl-substituted rings. These Li salts react by one route to triply substituted products; they also form bicyclo[4.2.0]octanes by an intramolecular reaction with LiF elimination and silyl group and methanide ion r n i g r a t i ~ n ~Figure ~ . ~ ~ .4 shows the structure of compound B37.

15.2. Ring-Ring and Ring-Polymer Interconversions 15.2.9. in Silicon-Nitrogen and Silicon-Sulfur Systems 15.2.9.2. Silicon-Nitrogen Ring-Ring Interconversions.

130

Anionic rearrangements of cyclotetrasilazanes to form cyclotrisilazanes are observed in the preparation and methylation of eight-membered Me

I

Me, CI-Si

Me,%-N-Li

I I

+

MeN

I

Me,%-N-Li

I

NH CI-Si

Me

- LiCl

"'

Me

I

- Lil

Me,Si-N-SiMe,

I

I NH I

LiN

I

Me Me,Si-NSiMe, Mek

I

I

NH

I

-

;;2

MeN/

Me,Si\

Me,Si-N-%Me, Me

\ NMe

I

I

/SiMe,

N

I

SiMe,

I

N

Me,Si-N-SiMe, Me I

H'

\Me (9)

Ring expansion of four-membered silylaminsubstituted rings to six-membered rings occurs with catalytic amounts of baseZ2,23,31,40:

where R = H, SiMe,. Three products are obtained from trimethylcyclodisilazane4' in the presence of acids or bases: Me, Si

I

NH

Me3 Si

I Me&-N-%Me,

Me2

/si\

Me,SiN\

/NH-

Si Me2

bases

I I

3;:

I I

NH +Me&

HN

Me,Si-N-%Me,

I

Si Me3

/

I I

SiMe,

I

N

\

I

SiMe, +Me,Si

I NH

HN, Si Me2

/N\

SiMe,

"/

I I NH I

SiMe,

15.2. Ring-Ring and Ring-Polymer Interconversions 15.2.9. in Silicon-Nitrogen and Silicon-Sulfur Systems 15.2.9.2. Silicon-Nitrogen Ring-Ring Interconversions.

131

Bis(dimethy1amino)cyclodisilazane catalyzed by a silanolate anion forms two

products4’:

(H,NSiMe,NSiMe,),

-

(H,NSiMe,)(NSiMe,)(NHSiMe,)z+ (Me,SiNH),

(t)

No reaction of the eight-membered ring compound (Me,SiNH4 with retention of the ring moietyZ3 is reported (before 1989). All attempts to carry out substitution reactions lead to the formation of isomeric six- and four-membered ringsz3. The Li salt of the eight-membered ring equilibrates to the Li salt of the six- and fourmembered rings43:

The products formed in the reactions with fluorosilanes depend very strongly on reaction conditions and attacking ligands. Substitution is preferred at low temperatures. Route 1 is preferred in reactions with trifluorosilanes and route 2 with bulky difluor~silanes~~. The known monosilyl-substituted eight-membered rings crystallize in boat c o n f ~ r m a t i o n Figure ~ ~ . 5a shows the structure of the SiF,Me-substituted ring. Figure 5b shows the same ring with H-F bridges in the crystal The diLi44, di-Na4’ and di-K45 salts crystallize from THF as dimers. The eight-membered rings are bonded via perpendicularly oriented planar (LiN),, (NaN), or (KN), four-membered rings. Figure 6 shows the structure of the dilithium compound. Lithium atoms Li2 and Li2a are each coordinated to two T H F molecules. The dilithium compound reacts with fluorosilanes in a molar ratio 1:2 to give disubstituted eight-membered rings43-45.Reactions in mole ratio 1: 1 lead to the novel bicyclic systrem 1,3,5,7-tetraaza-2,4,6,8,9-pentasilabicyclo[3.3. llnonane.

132

15.2. Ring-Ring and Ring-Polymer Interconversions 15.2.9. in Silicon-Nitrogen and Silicon-Sulfur Systems 15.2.9.2. Silicon-Nitrogen Ring-Ring Interconversions.

a

b

Figure 5

15.2. Ring-Ring and Ring-Polymer Interconversions 15.2.9. in Silicon-Nitrogen and Silicon-Sulfur Systems 15.2.9.2. Silicon-Nitrogen Ring-Ring Interconversions.

Figure 6

Figure 7

133

134

+

15.2. Ring-Ring and Ring-Polymer lnterconversions 15.2.9. in Silicon-Nitrogen and SiliconSulfur Systems 15.2.9.2. Silicon-Nitrogen Ring-Ring Interconversions.

(Me,SiNH),

+ 2 BuLi

- 2 BuH

+F2Si

+ 2 FZSi,/H N- 2 L iF

I

Me S i /

f

FSi-N \ I Me,Si

\SiMe,

\

-2LiF

H N

I

N-SiF SiMe, / I

"/

H

Me,%

/

N-Si-N MkSi

/ \

I

SiMe,

\

/

\

N H

/

(v)

SiMe,

The known disilyl-substituted eight-membered rings crystallize in chair c ~ n f o r m a t i o n ~Figure ~ , ~ ~7. shows the crystal structure of the CMe,SiF,-disubstituted ring45and Figure 8 the crystal structure of the C,H, SiF-bridged eight-membered ring44.

Figure 8.

135

15.2. Ring-Ring and Ring-Polymer lnterconversions 15.2.9. in Silicon-Nitrogen and Silicon-Sulfur Systems 15.2.9.2. Silicon-Nitrogen Ring-Ring Interconversions.

Six-membered rings are obtained in reactions of lithiated disilyl-substituted eightmembered rings with f l u o r ~ s i l a n e s ~ ~ ~ ~ ~ : H

R \

/N-si\

Me,Si

Me,

Me,

,R

F-sl-NSiF, -

\ N-Si H Me,

\

/

::$Me,Si I

+F$iMe -LiF

Me,

/N-si

\

N-R

N

' T T -G iiiz 2LiF

SiMe,

\Si-N Me, H

R'

2 BuLi -2BuH

/

Me

/

FSi / \N-y2 Me / \ Me, /R Me,Si N-Si-N\Si6Me

\ / N-Si R Me,

(w)

1

Me/

Figure 9 shows the structure of a trisilyl-substituted ring (R = CMe,SiF,). The isomerization of a cyclodisilazane to a cyclic hydrazine derivative is demonstrated in the reaction of an Li salt with triflu~rophenylsilane~~:

u c5

Figure 9.

136

15.2. Ring-Ring and Ring-Polymer Interconversions 15.2.9. in Silicon-Nitrogen and Silicon-Sulfur Systems 15.2.9.2. Silicon-Nitrogen Ring-Ring Interconversions.

A four-membered ring can be enlarged to the corresponding eight-membered ring' : Me,SiNHMe NI

/ \

175°C

H Me,Si-N-SiMe,

I I

/SiMez (NH4)2S04MeN

MezSi,

N

I

Me,SiNHMe

Me,Si-N-SiMe, H

i

NMe

i

(Y)

(U. KLINGEBIEL)

1. W. Fink, Angew. Chem., Int. Ed. Engl., 5, 760 (1963). 2. J. Haiduc, The Chemistry of Inorganic Ring Systems, Wiley-Interscience, New York, 1970. 3. K. Lienhard, E. G. Rochow, Angew. Chem., Inc. Ed. Engl., 2, 325 (1963). 4. K. Lienhard, E. G. Rochow, 2. Anorg. Allg. Chem., 331, 316 (1964). 5. K. A. Andrianov, B. A. Ismailov, J . Gen. Chem. U S S R (Engl. Transl.), 35, 334 (1965). 6. K. A. Andrianov, B. A. Ismailov, A. M. Kononov, G. V. Kotrelev, J. Organomet. Chem., 3, 129 (1965). 7. K. A. Andrianov, G. Ya. Rumba, Proc. Acad. Sci. USSR(Eng1. Transl.) 145, 709 (1962). 8. U. Wannagat, Angew. Chem., Int. Ed. Engl., 4, 605 (1965). 9. U. Wannagat, Adv. Inorg. Chem. Radiochem., 6, 225 (1964). 10. U. Wannagat, Pure Appl. Chem., 13, 263 (1966). 11. L. W. Breed, W. L. Budde, R. L. Elliott, J. Organomet. Chem., 6, 676 (1966). 12. J. Silbiger, J. Fuchs, N. Gesundheit, Inorg. Chem., 6, 399 (1967). 13, U. Wannagat, E. Bogusch, P. Geymayer, Monatsh. Chem., 102, 1825 (1971). 14. W. Fink, Helv. Chim. Acta, 51, 1011 (1968). 15. W. Fink, Helv. Chim. Acta, 51, 978 (1968). 16. W. Fink, Chem. Ber., 97, 1424 (1964). 17. K. A. Andrianov, G. Ya. Rumba, J. Gen. Chem. USSR (Engl. Transl.), 32, 1993 (1962). 18. L. W. Breed, R. L. Elliott, Inorg. Chem., 2, 1069 (1963). 19. W. Fink, Helv. Chim. Acta, 45, 1081 (1962). 20. L. W. Breed, Inorg. Chem., 7, 1940 (1968). 21. U. Wannagat, Chem. Eng. News, 46, 38 (1968). 22. W. Fink, Angew. Chem., Int. Ed. Engl., 8, 521 (1969). 23. W. Fink, Helo. Chim. Acta, 52, 2261 (1969). 24. M. Haase, G. M. Sheldrick, Acta Crystallogr. Sect A , C42, 1009. 25. E. Egert, U. Kliebisch, U. Klingebiel, D. Schmidt, Z. Anorg. Allg. Chem., 548, 86 (1987). 26. U. Klingebiel, D. Enterling, L. Skoda, A. Meller; J . Organomet. Chem., 135, 167 (1977). 27. U. Klingebiel, D. Enterling, A. Meller, Chem. Ber., 110, 1277 (1977). 28. D. Enterling, U. Klingebiel, A. Meller, 2.Naturforsch., Teil B, 33, 527 (1978). 29. U. Klingebiel, L. Skoda, A. Meller, 2. Anorg. Allg. Chem., 441, 113 (1978). 30. L. Skoda, U. Klingebiel, A. Meller, Chem. Ber., 113, 1444 (1980). 31. L. Skoda, Ph.D. Diss. Gottingen Univ., 1981. 32. L. Skoda, U. Klingebiel, A. Meller; 2 Anorg. Allg. Chem., 467, 131 (1980). 33. W. Clegg, M. Hesse, U. Klingebiel, G. M. Sheldrick, L. Skoda, Z. Naturforsch., TeilB,35, 1359 (1980). 34. M. Hesse, U. Klingebiel; J. Organomet. Chem., 221, C1 (1981).

15.2. Ring-Ring and Ring-Polymer Interconversions 15.2.9. in Silicon-Nitrogen and Silicon-Sulfur Systems 15.2.9.3. Cyclic Silicon-Nitrogen Oligorners Polymerization.

137

35. W. Clegg, U. Klingebiel, G. M. Sheldrick, L. Skoda, N. Vater, 2. Naturforsch., Ted B, 35, 1503 (1980). 36. L. Skoda, U. Klingebiel, A. Meller, Chem. Ber., 113, 2342 (1980). 37. W. Clegg, Acta Crystallogr., Sect. B, 36 2830 (1980). 38. U. Wannagat, V. Paul, Monatsh. Chem., 105, 1240 (1974). 39. U. Wannagat, F. Rabet, H.-J. Wismar, Monatsh. Chem., 102, 1429 (1971). 40. H. Burger, R. Mellies, K. Wiegel, J. Organomet. Chem., 142, 55 (1977). 41. R. P. Bush, C. A. Pearce, J. Chem. Soc., A , 808 (1969). 42. K. A. Andrianov, G. V. Kotrelev, V. V. Kazakova, N. A. Tebeneva, Dokl. Akad. Nauk SSSR, 233, 353 (1977). 43. K. Dippel, U. Klingebiel, T. Kottke, G. M. Sheldrick, D. Stalke, unpublished. 44. K. Dippel, U. Klingebiel, M. Noltemeyer, F. Pauer, G. M. Sheldrick, Angew. Chem., Int. Ed. Engl., 27, 1074 (1988). 45. K. Dippel, U. Klingebiel, F. Pauer, G. M. Sheldrick, D. Stalke, unpublished. 46. W. Clegg, M. Haase, H. Hluchy, U. Klingebiel, G. M.Sheldrick, Chem. Ber.,ll6, 290 (1983).

15.2.9.3. Cyclic Silicon-Nitrogen Oligomers Polymerization.

Organocyclosilazanes undergo ring opening and cleavage to form linear fragments. u p - Difunctional di- and trisilazanes are obtained in reactions with acids, alcohols,

phenols, thiols, hydrogen dichlorodimethylsilane'-6 : (Me,SiNH),,,

-

chloride,

+ Me,SiCl,

silicon

tetrafluoride

diammoniate

Cl(Me,SiNH),,,SiMe,Cl

or

(4

Decomposition of organocyclosilazanes occurs with many reagents'. The polymerization of (SiN)- rings does not occur as easily as the polymerization of (Si0)- rings. In attempts to polymerize cyclosilazanes with basic catalysts, waxy and resinous crosslinked polymers are f ~ r m e d ~but - ~ ,tricyclic compounds as ell^,'^. Me

RMe

\

Me

Si

R

/

/ \

R

Linear polysilazane oligomers H,N(SiMe,NH),-SiMe,NH, with a degree of polymerization of ca. 15 are obtained in the reaction of cyclosilazanes in liquid NH, 11312. Polymers A and B are formed in reactions of cyclosilazanes with hot ammonium halides (160-300°C, 1-45% ammonia halid)'2,'3: (R,SiNH),

A

NH4Br

[(R,Si),N,I,

+ NH,

B

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc. 15.2. Ring-Ring and Ring-Polymer Interconversions 15.2.9. in Silicon-Nitrogen and Silicon-Sulfur Systems 15.2.9.3. Cyclic Silicon-Nitrogen Oligorners Polymerization.

137

35. W. Clegg, U. Klingebiel, G. M. Sheldrick, L. Skoda, N. Vater, 2. Naturforsch., Ted B, 35, 1503 (1980). 36. L. Skoda, U. Klingebiel, A. Meller, Chem. Ber., 113, 2342 (1980). 37. W. Clegg, Acta Crystallogr., Sect. B, 36 2830 (1980). 38. U. Wannagat, V. Paul, Monatsh. Chem., 105, 1240 (1974). 39. U. Wannagat, F. Rabet, H.-J. Wismar, Monatsh. Chem., 102, 1429 (1971). 40. H. Burger, R. Mellies, K. Wiegel, J. Organomet. Chem., 142, 55 (1977). 41. R. P. Bush, C. A. Pearce, J. Chem. Soc., A , 808 (1969). 42. K. A. Andrianov, G. V. Kotrelev, V. V. Kazakova, N. A. Tebeneva, Dokl. Akad. Nauk SSSR, 233, 353 (1977). 43. K. Dippel, U. Klingebiel, T. Kottke, G. M. Sheldrick, D. Stalke, unpublished. 44. K. Dippel, U. Klingebiel, M. Noltemeyer, F. Pauer, G. M. Sheldrick, Angew. Chem., Int. Ed. Engl., 27, 1074 (1988). 45. K. Dippel, U. Klingebiel, F. Pauer, G. M. Sheldrick, D. Stalke, unpublished. 46. W. Clegg, M. Haase, H. Hluchy, U. Klingebiel, G. M.Sheldrick, Chem. Ber.,ll6, 290 (1983).

15.2.9.3. Cyclic Silicon-Nitrogen Oligomers Polymerization.

Organocyclosilazanes undergo ring opening and cleavage to form linear fragments. u p - Difunctional di- and trisilazanes are obtained in reactions with acids, alcohols,

phenols, thiols, hydrogen dichlorodimethylsilane'-6 : (Me,SiNH),,,

-

chloride,

+ Me,SiCl,

silicon

tetrafluoride

diammoniate

Cl(Me,SiNH),,,SiMe,Cl

or

(4

Decomposition of organocyclosilazanes occurs with many reagents'. The polymerization of (SiN)- rings does not occur as easily as the polymerization of (Si0)- rings. In attempts to polymerize cyclosilazanes with basic catalysts, waxy and resinous crosslinked polymers are f ~ r m e d ~but - ~ ,tricyclic compounds as ell^,'^. Me

RMe

\

Me

Si

R

/

/ \

R

Linear polysilazane oligomers H,N(SiMe,NH),-SiMe,NH, with a degree of polymerization of ca. 15 are obtained in the reaction of cyclosilazanes in liquid NH, 11312. Polymers A and B are formed in reactions of cyclosilazanes with hot ammonium halides (160-300°C, 1-45% ammonia halid)'2,'3: (R,SiNH),

A

NH4Br

[(R,Si),N,I,

+ NH,

B

138

15.2. Ring-Ring and Ring-Polymer lnterconversions 15.2.9. in Silicon-Nitrogen and Silicon-Sulfur Systems 15.2.9.3. Cyclic Silicon-Nitrogen Oligomers Polymerization. ~~~

When the hydrolysis of bis(chlorodimethylsily1)-substituted cyclodisilazane is conducted both with excess and with stoichiometric amounts of H,O, up to 30% of six-membered ring is formed, but also a mixture of polydimethylcyclosiloxanes (Me2SiO), and linear polydimethylsiloxane - a,o-diols1 4 :

+ HO

4

Me2 -Si-0

y e , Si-OH

(d)

Organocyclosilazanes are used as precursors to SiN polymers and these polymers in the pyrolytic preparation of Si3N415; e.g. the ammonolysis of organochlorosilicon hydrides gives cyclic products (RSiHNH),, n 2 3. The cyclosilazanes are converted to polymers of higher molecular weight with basic catalysts such as KH. Below is an example of linkage of eight-membered rings via Si,N, bridges in a ladder polymer:

Pyrolyses of these polysilazanes give ceramic Organosiliconcompounds such as hexaphenylcyclotrisilazaneandmethylphenylpolysilane are used to infiltrate porous reaction-sintered silicon nitride”. The cyclosilazane is thermallty decomposed slowly and carefully in an inert atmosphere at 600-1400°C; very large white whisker bundles of a-Si,N4 are synthesized”. Organocyclosilazanes are used as precursors to SiN polymers and these polymers in the pyrolytic preparation of Si,N, 5 ; e.g. the ammonolysis of organochlorosilicon hydrides gives cyclic products (RSiHNH),, n 2 3. The cyclosilazanes are converted to polymers of higher molecular weight with basic catalysts such as KH. Below is an example of linkage of eight-membered rings via Si,N, bridges in a ladder polymer:

139

15.2. Ring-Ring and Ring-Polymer lnterconversions 15.2.9. in Silicon-Nitrogen and Silicon-Sulfur Systems 15.2.9.4. Silicon-Sulfur Syntheses.

Pyrolyses of these polysilazanes give ceramic materials' 5-16. Organosilicon compounds such as hexaphenylcyclotrisilazane and methylphenylpolysilane are used to infiltrate porous reaction-sintered silicon nitride17. The cyclosilazane is thermally decomposed slowly and carefully in an inert atmosphere at 600- 1400°C; very large white whisker bundles of a-Si,N, are synthesized". 1. 2. 3. 4. 5. 6.

(u KLINGEBIEL)

I. Haiduc, The Chemistry of Inorganic Ring Systems, Wiley-Interscience, New York, 1970. U. Wannagat, Angew. Chem., Int. Ed. Engl., 4, 605 (1965). U. Wannagat, Adv. Inorg. Chem. Radiochem., 6, 225 (1964). U. Wannagat, E. Bogusch, P. Geymayer, Monatsh. Chem., 95, 801 (1964). P. Geymayer, E. G. Rochow, Angew. Chem., Int. Ed. Engl., 4, 592 (1965). K. A. Andrianov, Zh. S. Syrtsova, V. M. Kopylov, J. Gen. Chem. USSR (Engl. Transl.),40,1655

(1970). 7. K. A. Andrianov, B. A. Ismailov, A. M. Kononov, G . V. Kotrelev, J. Organomet. Chem., 3, 129 (1965). 8. K. A. Andrianov, G. V. Kotrelev, Vysokomol. Soedin., 6, 691 (1964); Chem. Abstr., 61, 12,100 (1964). 9. K. A. Andrianov, G. Ya. Rumba, Vysokomol. Soedin., 4, 1060 (1962); Chem. Abstr., 58, 12,683h (1962). 10. K. A. Andrianov, G. V. Kotrelev, J. Organomet. Chem., 7, 217 (1967). 11. G. Redl, E. G. Rochow, Angew. Chem., Int. Ed. EngL, 3,650 (1964). 12. E. G. Rochow, Pure Appl. Chem., 13,247 (1966). 13. C. R. Kruger, E. G. Rochow, Angew. Chem.,Int. Ed. Engl., I , 588 (1962); J. Polym. Sci., 2 A , 3179 (1964). 14. Yu. M. Varezhkin, D. Ya. Zhinkin, M. M. Morgunova, V. N. Bochkarev, J. Gen. Chem. USSR (Engl. Transl.), 45, 2410 (1975). 15. D. Seyferth, G. H. Wiseman, Commun. Am. Ceram. Soc., C-132 (1984). 16. R. R. Wills, R. A. Markle, S . P. Mukherjee, Ceram. Bull. 62, 904 (1983). 17. K. S. Mazdiyasni, R. West, L. D. David, J. Am. Ceram. SOC.,61, 504 (1978).

15.2.9.4. SiliconSulfur Syntheses. Silicon-sulfur bonds may be formed by the same methods utilized to generate Si-N bonds'-,. Four-, six- and eight-membered cyclosilthianes rings based on a regular alternation of Si and S are known. The reaction of SiCl, with H,S forms the dimeric Si thiochloride (Cl,SiS), &':

SiCl,

+ H,S-

C1,Si-S

I

180°C

I

C1,Si-S-SiCl,

1 1

s

1 1

+ I 5

S-SiCl,

C1,Si-S-SiCl,

(a)

The nature of the products obtained depends on temperature and conditions5. A polymer with mol wt close to that of the tetramer is observed when (Cl,SiS), is heated above 180°C Eq. (a)6. Organic derivatives of four- and six-membered cyclosilthianes are formed by passing H,S through a solution of an organodichlorosilane in the presence of a base7-" :

I

RSiCl,

+ H,S

-

1 n

1

- (RSiS),

+ 2 HC1

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc.

139

15.2. Ring-Ring and Ring-Polymer lnterconversions 15.2.9. in Silicon-Nitrogen and Silicon-Sulfur Systems 15.2.9.4. Silicon-Sulfur Syntheses.

Pyrolyses of these polysilazanes give ceramic materials' 5-16. Organosilicon compounds such as hexaphenylcyclotrisilazane and methylphenylpolysilane are used to infiltrate porous reaction-sintered silicon nitride17. The cyclosilazane is thermally decomposed slowly and carefully in an inert atmosphere at 600- 1400°C; very large white whisker bundles of a-Si,N, are synthesized". 1. 2. 3. 4. 5. 6.

(u KLINGEBIEL)

I. Haiduc, The Chemistry of Inorganic Ring Systems, Wiley-Interscience, New York, 1970. U. Wannagat, Angew. Chem., Int. Ed. Engl., 4, 605 (1965). U. Wannagat, Adv. Inorg. Chem. Radiochem., 6, 225 (1964). U. Wannagat, E. Bogusch, P. Geymayer, Monatsh. Chem., 95, 801 (1964). P. Geymayer, E. G. Rochow, Angew. Chem., Int. Ed. Engl., 4, 592 (1965). K. A. Andrianov, Zh. S. Syrtsova, V. M. Kopylov, J. Gen. Chem. USSR (Engl. Transl.),40,1655

(1970). 7. K. A. Andrianov, B. A. Ismailov, A. M. Kononov, G . V. Kotrelev, J. Organomet. Chem., 3, 129 (1965). 8. K. A. Andrianov, G. V. Kotrelev, Vysokomol. Soedin., 6, 691 (1964); Chem. Abstr., 61, 12,100 (1964). 9. K. A. Andrianov, G. Ya. Rumba, Vysokomol. Soedin., 4, 1060 (1962); Chem. Abstr., 58, 12,683h (1962). 10. K. A. Andrianov, G. V. Kotrelev, J. Organomet. Chem., 7, 217 (1967). 11. G. Redl, E. G. Rochow, Angew. Chem., Int. Ed. EngL, 3,650 (1964). 12. E. G. Rochow, Pure Appl. Chem., 13,247 (1966). 13. C. R. Kruger, E. G. Rochow, Angew. Chem.,Int. Ed. Engl., I , 588 (1962); J. Polym. Sci., 2 A , 3179 (1964). 14. Yu. M. Varezhkin, D. Ya. Zhinkin, M. M. Morgunova, V. N. Bochkarev, J. Gen. Chem. USSR (Engl. Transl.), 45, 2410 (1975). 15. D. Seyferth, G. H. Wiseman, Commun. Am. Ceram. Soc., C-132 (1984). 16. R. R. Wills, R. A. Markle, S . P. Mukherjee, Ceram. Bull. 62, 904 (1983). 17. K. S. Mazdiyasni, R. West, L. D. David, J. Am. Ceram. SOC.,61, 504 (1978).

15.2.9.4. SiliconSulfur Syntheses. Silicon-sulfur bonds may be formed by the same methods utilized to generate Si-N bonds'-,. Four-, six- and eight-membered cyclosilthianes rings based on a regular alternation of Si and S are known. The reaction of SiCl, with H,S forms the dimeric Si thiochloride (Cl,SiS), &':

SiCl,

+ H,S-

C1,Si-S

I

180°C

I

C1,Si-S-SiCl,

1 1

s

1 1

+ I 5

S-SiCl,

C1,Si-S-SiCl,

(a)

The nature of the products obtained depends on temperature and conditions5. A polymer with mol wt close to that of the tetramer is observed when (Cl,SiS), is heated above 180°C Eq. (a)6. Organic derivatives of four- and six-membered cyclosilthianes are formed by passing H,S through a solution of an organodichlorosilane in the presence of a base7-" :

I

RSiCl,

+ H,S

-

1 n

1

- (RSiS),

+ 2 HC1

140

15.2. Ring-Ring and Ring-Polymer lnterconversions 15.2.9. in Silicon-Nitrogen and Silicon-Sulfur Systems 15.2.9.4. Silicon-Sulfur Syntheses.

-

where R = Me, Et, Vi, n-Pr, Ph; n = 2, 3. By the reaction of Si disulfide with phenols or t-butanol aryl- and alkyloxicyclodisilthianes are ~ b t a i n e d ' ~ ~ ' ~ : 2 SiS,

+ 4 ROH

+ 2 H,S

[(RO),SiS],

(c)

where R = CMe,, 2,6-Me,C,H4. Cyclodisilthianes are also formed from R,SiI, and Ag,S 2 R,SiJ,

+ Ag,S

or from F,SiI, and HgS 15: F,SiI,

+ 4 AgI

(R,SiS),

+ HgS

(F,SiS),

(4

+ 2 HgI

(e)

The condensation of mercaptosilanes forms mercaptocyclodisilthianes' ,: Si(SR),

$ [(RS),SiS],

+ R,S

(f)

Other methods of preparing cyclosilthianes SiF,

+ SiS,

(F,SiS),

(€9

+ SF, (Br,SiS), Si,Me, + SF, (Me,SiS), Me,SiSS SiMe, + Cl,SiPh, - ~ e s ~ (Ph,SiS), i~i 120°C

Si,Br,

Me,_Me, Si Si

'\ /

Me2

\

hv

/ s-s-CH2CH2

(Me,SiNH), MePhSi(NEt,),

The oligomeric (H,SiS), cy~lotrisilthiane~~:

/si\

\si/ Me,

+ Me,SiCl,

s' s +Me,SiI

-

+ MePhSiCl,

H2S

(9

(8

Me, Si

S

\ \

H2S

(h)

S

/

I

SiMe,

(Me,SiS), (MePhSiS),

w0 (1)

(m)

can be depolymerized in vacuum at 210°C to give

15.2. Ring-Ring and Ring-Polymer Interconversions 15.2.9. in Silicon-Nitrogen and Silicon-Sulfur Systems 15.2.9.4. Si I icon-Su If ur Syntheses.

-

The polycyclic SiS compounds (RSiS, organotrichlorosilanes5~8~1g~z5~z6:

4 RSiC1,

+ 6 H,S

5)n

141

are formed by the thiohydrolysis of

(RSiS,,,),

-

+ 12 HC1

(P)

o r dichlorosilanes28. Sulfur reacts with phenylsilane to give or alkoxidi~hlorosilanes~~ (PhSi),S, ”:

4 PhSiH, The compound (MeSi),S,

30

+ 12 S

(PhSi),S,

+ 6 H,O

(9)

has an adamantanoid structure:

Me

(u KLINGEBIEL)

1. I. Haiduc, The Chemistry of Inorganic Ring Systems, Wiley-Interscience, New York, 1970; I. Haiduc, D. B. Sowerby, The chemistry of Inorganic Homo- and Heterocycles, Vol. 1, Academic Press, London, 1987. 2. Gmelin’s Handbuch der Anorganischen Chemie, System No. 15 C, Verlag Chemie, Weinheim, 1958. 3. A. Haas, Anaew. Chem., Int. Ed. E n d , 4, 1014 (1965). 4. M. Blix, W.Wirbelauer, Chem. Be;, 36, 4220 (1903). 5. Y. Etienne, Bull. Soc. Chim. Fr., 791 (1953). 6. W. C. Schumb, W. J. Bernhard, J. Am. Chem. Soc., 77, 862 (1955). 7. G. Champetier, Y. Etienne, R. Kullmann, C. R. Hebd. Seances Acad. Sci. (Paris), 234, 1985 (1952). 8. Y. Etienne, C. R. Hebd. Seances Acad. Sci. (Paris), 235, 966 (1952). 9. H. Kriegsmann, H. Clauss, Z. Anorg. Allg. Chem., 300, 210 (1959). 10. D. L. Mayfield, R. A. Flath, L. R. Best, J. Org. Chem., 29, 2444 (1964). 11. D. J. Panckhurst, C. J. Wilkins, P. W. Craighead, J. Chem. Soc., 3395 (1955). 12. R. Piekos, W. Wojnowski, Z. Anorg. Allg. Chem., 318, 212 (1962). 13. W. Wojnowski, M. Wojnowski, Z. Anorg. Allg. Chem.. 389, 302 (1972); 397, 69 (1973). 14. C. Eaborn, J . Chem. Soc., 3077 (1950). 15. B. J. Aylett, J. A. Ellis, J. R. Richmond; J . Chem. Sac., Dalton Trans., 981 (1973). 16. M. Schmeisser, H. Muller, W. Burgmeister; Angew. Chem., 69, 781 (1957). 17. M. Weidenbruch, G. Rottig, Znorg. Nucl. Chem. Lett., 13, 85 (1977). 18. V. 0. Reikhsfel‘d, E. P. Lebedev, J. Gen. Chem. U S S R (Engl. Transl.), 40, 586 (1970). 19. E. P. Lebedev, M. M. Frenkel’, V. 0. Reikhsfel’d, D. V. Fridland, J. Gen. Chem. USSR (Engl. Transl.), 47, 1308 (1977). 20. H. S. Dilanjan Soysa, I1 Nam Jung, W. P. Weber; J. Organomet. Chem., 171, 177 (1979). 21. E. P. Lebedev, D. V. Fridland, V. 0. Reikhsfel’d, E. N. Kovol, J. Gen. Chem. U S S R (Engl. Transl.), 46, 313 (1976). 22. E. P. Lebedev. D. C. Fridland. V. 0.Reikhsfel’d. J. Gen. Chem. U S S R (Engl. Transl.), 44,2737 (1974). 23. E. P. Lebedev, V. 0. Reikhsfel’d, D. V. Fridland, J. Gen. Chem. U S S R (Engl. Transl.), 43, 680 (1973).

15.2. Ring-Ring and Ring-Polymer Interconversions 15.2.9.in Silicon-Nitrogen and Silicon-Sulfur Systems 15.2.9.5.SiliconSulfur Ring-Ring Interconversions.

142

24. 25. 26. 27.

A. Haas, M. Vongehr; Z . Anorg. Allg. Chem., 447, 119 (1978); Chem.-Ztg., 99, 432 (1975). Y. Etienne; Angew. Chem., 67, 753 (1955). J. A. Forstner, E. L. Muetterties, Inorg. Chem., 5, 552 (1966). E. P. Lebedev, D. V. Fridland, V. 0. Reikhsfel'd, J. Gen. Chem. USSR (Engl. T r a i d ) , 45,2603 (1975). 28. V. 0. Reikhsfel'd, E. P. Lebedev, J. Gen. Chem. USSR (Engl. Transl.), 37, 1342 (1967). 29. F. Feher, R. Liipschen, 2. Naturjorsch., Teil E, 26, 1191 (1971). 30. J. C . J. Bart, J. J. Daly, J. Chem. Soc., Chem. Commun., 1207 (1968); J . Chem. SOC., Dalton Pans., 2063 (1975).

15.2.9.5. SiliconSulfur Ring-Ring Interconversions.

Silicon-sulfur rings undergo ring-ring interconversions. Ring contractions and also ring expansions are The cyclotrisilthiane is more stable; however, when the six-membered ring is heated, cyclodisilthianes form.'-* : RT

200 C

3

The equilibria in solution between the tetraatomic ring of (Me,SiS), and the hexaatomic ring of (Me,SiS), can be followed by 'H NMR spectroscopyg.With unlike groups on the Si, cis and trans isomers are possible for the four-membered r i r ~ g ~When ? ~ . trans-1,3,5trimethyl-l,3,5-triphenylcyclotrisilthianeis maintained at 200°C for several weeks, only one cyclodisilthiane isomer is observed2*,: 2 (MePhSiS),

200°C

3 (MePhSiS),

or 1,3,5-trivinylcyclotrisilthiane Continued refluxing of 1,3,5-trimethyl-1,3,5-triethylresults in a 50: 50 mixture of cis- and trans-cyclodisilthianes'.

(u. KLINGEBIEL) 1. J. Haiduc, The Chemistry of Inorganic Ring Systems, Wiley-Interscience, New York, 1970; I. Haiduc, D. B. Sowerby, The Chemistry of Inorganic Homo- and Heterocycles, Vol. 1, Academic Press, London, 1987. 2. M. M. Millard, L. J. Pazdernik, J. Organomet. Chem., 51, 135 (1973). 3. M. M. Millard, K. Steele, L. J. Pazdernik; J. Organornet. Chem., 13, 7 (1968). 4 K. A. Andrianov, J. Haiduc, L. M. Khananashvili Russ. Chem. Rev. (Engl. Transl.), 32,243 (1963). 5. K. Moedritzer, J. R. Van Wazer, C . H. Dungau, J. Chem. Phys., 42, 2478 (19651. 6. K. Moedritzer, J. R. Van Wazer, J. Chem. Phys., 70, 2030 (1966). 7. K. Moedritzer, J. Organornet. Chem., 21, 315 (1970). 8. M. M. Millard, L. J. Pazdernik, W. F. Haddon, R. E. Lundin, J. Organomet. Chem., 52, 283, (1973). 9. K. Moedritzer, J. Organomet. Chem., 21, 315 (1970).

15.2.9.6. Cyclic Silicon-Sulfur Oligomer Polymerization.

A bicyclic spirosilthiane is formed in the reaction of SiS, with SiCI, at 900°C l S 2 : 2 SiS,

+ 2 SiC14- 900°C

s

C1,Si

\ / S

SiCl,

s

+ C1,Si /\ s / \ Si\/s / \SiCl,

(a)

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc. 15.2. Ring-Ring and Ring-Polymer Interconversions 15.2.9.in Silicon-Nitrogen and Silicon-Sulfur Systems 15.2.9.5.SiliconSulfur Ring-Ring Interconversions.

142

24. 25. 26. 27.

A. Haas, M. Vongehr; Z . Anorg. Allg. Chem., 447, 119 (1978); Chem.-Ztg., 99, 432 (1975). Y. Etienne; Angew. Chem., 67, 753 (1955). J. A. Forstner, E. L. Muetterties, Inorg. Chem., 5, 552 (1966). E. P. Lebedev, D. V. Fridland, V. 0. Reikhsfel'd, J. Gen. Chem. USSR (Engl. T r a i d ) , 45,2603 (1975). 28. V. 0. Reikhsfel'd, E. P. Lebedev, J. Gen. Chem. USSR (Engl. Transl.), 37, 1342 (1967). 29. F. Feher, R. Liipschen, 2. Naturjorsch., Teil E, 26, 1191 (1971). 30. J. C . J. Bart, J. J. Daly, J. Chem. Soc., Chem. Commun., 1207 (1968); J . Chem. SOC., Dalton Pans., 2063 (1975).

15.2.9.5. SiliconSulfur Ring-Ring Interconversions.

Silicon-sulfur rings undergo ring-ring interconversions. Ring contractions and also ring expansions are The cyclotrisilthiane is more stable; however, when the six-membered ring is heated, cyclodisilthianes form.'-* : RT

200 C

3

The equilibria in solution between the tetraatomic ring of (Me,SiS), and the hexaatomic ring of (Me,SiS), can be followed by 'H NMR spectroscopyg.With unlike groups on the Si, cis and trans isomers are possible for the four-membered r i r ~ g ~When ? ~ . trans-1,3,5trimethyl-l,3,5-triphenylcyclotrisilthianeis maintained at 200°C for several weeks, only one cyclodisilthiane isomer is observed2*,: 2 (MePhSiS),

200°C

3 (MePhSiS),

or 1,3,5-trivinylcyclotrisilthiane Continued refluxing of 1,3,5-trimethyl-1,3,5-triethylresults in a 50: 50 mixture of cis- and trans-cyclodisilthianes'.

(u. KLINGEBIEL) 1. J. Haiduc, The Chemistry of Inorganic Ring Systems, Wiley-Interscience, New York, 1970; I. Haiduc, D. B. Sowerby, The Chemistry of Inorganic Homo- and Heterocycles, Vol. 1, Academic Press, London, 1987. 2. M. M. Millard, L. J. Pazdernik, J. Organomet. Chem., 51, 135 (1973). 3. M. M. Millard, K. Steele, L. J. Pazdernik; J. Organornet. Chem., 13, 7 (1968). 4 K. A. Andrianov, J. Haiduc, L. M. Khananashvili Russ. Chem. Rev. (Engl. Transl.), 32,243 (1963). 5. K. Moedritzer, J. R. Van Wazer, C . H. Dungau, J. Chem. Phys., 42, 2478 (19651. 6. K. Moedritzer, J. R. Van Wazer, J. Chem. Phys., 70, 2030 (1966). 7. K. Moedritzer, J. Organornet. Chem., 21, 315 (1970). 8. M. M. Millard, L. J. Pazdernik, W. F. Haddon, R. E. Lundin, J. Organomet. Chem., 52, 283, (1973). 9. K. Moedritzer, J. Organomet. Chem., 21, 315 (1970).

15.2.9.6. Cyclic Silicon-Sulfur Oligomer Polymerization.

A bicyclic spirosilthiane is formed in the reaction of SiS, with SiCI, at 900°C l S 2 : 2 SiS,

+ 2 SiC14- 900°C

s

C1,Si

\ / S

SiCl,

s

+ C1,Si /\ s / \ Si\/s / \SiCl,

(a)

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc. 15.2. Ring-Ring and Ring-Polymer Interconversions 15.2.9.in Silicon-Nitrogen and Silicon-Sulfur Systems 15.2.9.5.SiliconSulfur Ring-Ring Interconversions.

142

24. 25. 26. 27.

A. Haas, M. Vongehr; Z . Anorg. Allg. Chem., 447, 119 (1978); Chem.-Ztg., 99, 432 (1975). Y. Etienne; Angew. Chem., 67, 753 (1955). J. A. Forstner, E. L. Muetterties, Inorg. Chem., 5, 552 (1966). E. P. Lebedev, D. V. Fridland, V. 0. Reikhsfel'd, J. Gen. Chem. USSR (Engl. T r a i d ) , 45,2603 (1975). 28. V. 0. Reikhsfel'd, E. P. Lebedev, J. Gen. Chem. USSR (Engl. Transl.), 37, 1342 (1967). 29. F. Feher, R. Liipschen, 2. Naturjorsch., Teil E, 26, 1191 (1971). 30. J. C . J. Bart, J. J. Daly, J. Chem. Soc., Chem. Commun., 1207 (1968); J . Chem. SOC., Dalton Pans., 2063 (1975).

15.2.9.5. SiliconSulfur Ring-Ring Interconversions.

Silicon-sulfur rings undergo ring-ring interconversions. Ring contractions and also ring expansions are The cyclotrisilthiane is more stable; however, when the six-membered ring is heated, cyclodisilthianes form.'-* : RT

200 C

3

The equilibria in solution between the tetraatomic ring of (Me,SiS), and the hexaatomic ring of (Me,SiS), can be followed by 'H NMR spectroscopyg.With unlike groups on the Si, cis and trans isomers are possible for the four-membered r i r ~ g ~When ? ~ . trans-1,3,5trimethyl-l,3,5-triphenylcyclotrisilthianeis maintained at 200°C for several weeks, only one cyclodisilthiane isomer is observed2*,: 2 (MePhSiS),

200°C

3 (MePhSiS),

or 1,3,5-trivinylcyclotrisilthiane Continued refluxing of 1,3,5-trimethyl-1,3,5-triethylresults in a 50: 50 mixture of cis- and trans-cyclodisilthianes'.

(u. KLINGEBIEL) 1. J. Haiduc, The Chemistry of Inorganic Ring Systems, Wiley-Interscience, New York, 1970; I. Haiduc, D. B. Sowerby, The Chemistry of Inorganic Homo- and Heterocycles, Vol. 1, Academic Press, London, 1987. 2. M. M. Millard, L. J. Pazdernik, J. Organomet. Chem., 51, 135 (1973). 3. M. M. Millard, K. Steele, L. J. Pazdernik; J. Organornet. Chem., 13, 7 (1968). 4 K. A. Andrianov, J. Haiduc, L. M. Khananashvili Russ. Chem. Rev. (Engl. Transl.), 32,243 (1963). 5. K. Moedritzer, J. R. Van Wazer, C . H. Dungau, J. Chem. Phys., 42, 2478 (19651. 6. K. Moedritzer, J. R. Van Wazer, J. Chem. Phys., 70, 2030 (1966). 7. K. Moedritzer, J. Organornet. Chem., 21, 315 (1970). 8. M. M. Millard, L. J. Pazdernik, W. F. Haddon, R. E. Lundin, J. Organomet. Chem., 52, 283, (1973). 9. K. Moedritzer, J. Organomet. Chem., 21, 315 (1970).

15.2.9.6. Cyclic Silicon-Sulfur Oligomer Polymerization.

A bicyclic spirosilthiane is formed in the reaction of SiS, with SiCI, at 900°C l S 2 : 2 SiS,

+ 2 SiC14- 900°C

s

C1,Si

\ / S

SiCl,

s

+ C1,Si /\ s / \ Si\/s / \SiCl,

(a)

15.2. Ring-Ring and Ring-Polymer Interconversions 15.2.10. in Phosphorus-Oxygen Systems

143

The higher polymeric compounds (SiSCl,), form at 180°C3. The bi- and tricyclic bromine-substituted Si-S oligomers are obtained from (Br,SiS), with H,S Linear oligomers form by equilibration between cyclodi- and cyclotrisilthianes. 435.

(u. KLINGEBIEL) A. Haas, Angew. Chem., Int. Ed. Engl., 4, 1014 (1965). A. Buschfeld, PhD. Diss., Technische Hochschule, Aachen,1962. D. J. Panckhurst, C. J. Wilkins, P. W. Craighead; J. Chem. SOC.,3395 (1955). J. Haiduc, The Chemistry of Inorganic Ring Systems, Wiley-Interscience, New York, 1970; I. Haiduc, D. B. Sowerby, The Chemistry of Inorganic Homo- and Heterocycles, Vol. 1, Academic Press, London, 1987. 5. M. Blix, Chem. Ber., 36,4218 (1903).

1. 2. 3. 4.

15.2.10. in Phosphorus-Oxygen Systems The condensed phosphoric acids and their alkali-metal and substituted ammonium salts are the reference compounds for the ring, chain, cage and network molecular structures that are based on P(0)-0-P(0) linkages. Compounds exhibiting this structural feature may be treated as derivatives of these ionic substances. In the doubleoxide notation, the family of phosphoric acids was considered in terms of the ortho acid, 3 H,0.P,05 = H 3 P 0 4 ; the pyro acid 2 H,O.P,O, E H4P,0,; the meta acid, H,O.P,O, = HPO,; and the anhydride, P,O,. However, in accord with a speculation’ of the 19th Century that has been shown to be correct,, the composition range between the pyrophosphoric and the metaphosphoric acid is occupied by the series of polyphosphoric acids, H n + 2 P n 0 3 n +=l (n + 2) H,O.n P,O,, and the range between the metaand the acid anhydride by a complicated series of branched and cyclized P-O-P-bridged structures3, a few of which have been made and characterized in solution4. The cyclic phosphoric acids exhibiting simple rings are all metaphosphates; and the chain phosphates without cyclized segments are all polyphosphates. Long-chain phosphates exhibit the metaphosphate composition, since limn+mHn+zP,03,+ = [HPO,],. This discussion deals with the poly and meta acids and their derivatives made by heating appropriate combinations of reactants to form a melt or a labile solid composition from which the desired product crystallizes. Such labile solid compositions often include an amorphous phase, plasticized by water, in which ring-ring and ring-chain interconversions take place5. (J.R VAN WAZER)

1. T. Fleitmann, W. Henneberg, Justus Liebig’s Ann. Chem., 65, 30, 387 (1845). 2. J. R. Van Wazer, Phosphorus and Its Compounds, Vol.I : Chemistry, Interscience Publishers, New York, 1958; also see J. E. Such, in Mellor’s Inorganic and Theoretical Chemistry, WileyInterscience, New York, 1971, Vol. VIII, Suppl. 111, p. 717; and D. C. E. Corbridge, Phosphorus, An Outline of Its Chemistry, Biochemistry and Technology,2nd ed., Elsevier Scientific Publishing Company, Amsterdam, 1980, Ch. 3. 3. J. R. Van Wazer, E. J. Griffith, J. Am. Chem. Soc., 77,6140 (1955); also see D. E. C. Corbridge, The Structural Chemistry of Phosphorus, Elsevier Scientific Publishing Company, Amsterdam, 1974, p. 161. 4. T. Glonek, J. R. Van Wazer, R. A. Kleps, T. C. Myers, Inorg. Chem., 13, 2337 (1974). 5. See pp. 642-647 of the first listing in ref. 2 for examples.

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc.

15.2. Ring-Ring and Ring-Polymer Interconversions 15.2.10. in Phosphorus-Oxygen Systems

143

The higher polymeric compounds (SiSCl,), form at 180°C3. The bi- and tricyclic bromine-substituted Si-S oligomers are obtained from (Br,SiS), with H,S Linear oligomers form by equilibration between cyclodi- and cyclotrisilthianes. 435.

(u. KLINGEBIEL) A. Haas, Angew. Chem., Int. Ed. Engl., 4, 1014 (1965). A. Buschfeld, PhD. Diss., Technische Hochschule, Aachen,1962. D. J. Panckhurst, C. J. Wilkins, P. W. Craighead; J. Chem. SOC.,3395 (1955). J. Haiduc, The Chemistry of Inorganic Ring Systems, Wiley-Interscience, New York, 1970; I. Haiduc, D. B. Sowerby, The Chemistry of Inorganic Homo- and Heterocycles, Vol. 1, Academic Press, London, 1987. 5. M. Blix, Chem. Ber., 36,4218 (1903).

1. 2. 3. 4.

15.2.10. in Phosphorus-Oxygen Systems The condensed phosphoric acids and their alkali-metal and substituted ammonium salts are the reference compounds for the ring, chain, cage and network molecular structures that are based on P(0)-0-P(0) linkages. Compounds exhibiting this structural feature may be treated as derivatives of these ionic substances. In the doubleoxide notation, the family of phosphoric acids was considered in terms of the ortho acid, 3 H,0.P,05 = H 3 P 0 4 ; the pyro acid 2 H,O.P,O, E H4P,0,; the meta acid, H,O.P,O, = HPO,; and the anhydride, P,O,. However, in accord with a speculation’ of the 19th Century that has been shown to be correct,, the composition range between the pyrophosphoric and the metaphosphoric acid is occupied by the series of polyphosphoric acids, H n + 2 P n 0 3 n +=l (n + 2) H,O.n P,O,, and the range between the metaand the acid anhydride by a complicated series of branched and cyclized P-O-P-bridged structures3, a few of which have been made and characterized in solution4. The cyclic phosphoric acids exhibiting simple rings are all metaphosphates; and the chain phosphates without cyclized segments are all polyphosphates. Long-chain phosphates exhibit the metaphosphate composition, since limn+mHn+zP,03,+ = [HPO,],. This discussion deals with the poly and meta acids and their derivatives made by heating appropriate combinations of reactants to form a melt or a labile solid composition from which the desired product crystallizes. Such labile solid compositions often include an amorphous phase, plasticized by water, in which ring-ring and ring-chain interconversions take place5. (J.R VAN WAZER)

1. T. Fleitmann, W. Henneberg, Justus Liebig’s Ann. Chem., 65, 30, 387 (1845). 2. J. R. Van Wazer, Phosphorus and Its Compounds, Vol.I : Chemistry, Interscience Publishers, New York, 1958; also see J. E. Such, in Mellor’s Inorganic and Theoretical Chemistry, WileyInterscience, New York, 1971, Vol. VIII, Suppl. 111, p. 717; and D. C. E. Corbridge, Phosphorus, An Outline of Its Chemistry, Biochemistry and Technology,2nd ed., Elsevier Scientific Publishing Company, Amsterdam, 1980, Ch. 3. 3. J. R. Van Wazer, E. J. Griffith, J. Am. Chem. Soc., 77,6140 (1955); also see D. E. C. Corbridge, The Structural Chemistry of Phosphorus, Elsevier Scientific Publishing Company, Amsterdam, 1974, p. 161. 4. T. Glonek, J. R. Van Wazer, R. A. Kleps, T. C. Myers, Inorg. Chem., 13, 2337 (1974). 5. See pp. 642-647 of the first listing in ref. 2 for examples.

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc. 144

15.2. Ring-Ring and Ring-Polymer Interconversions 15.2.10. in Phosphorus-Oxygen Systems 15.2.10.1. Reorganized Mixtures of Chains and Rings.

15.2.10.1. Reorganized Mixtures of Chains and Rings.

The equilibrium distributions of molecules or molecule ions based on backbones of repeated P-0 bonds' are a source of both chain' and ring3 structures isolated by preparative chromatography. The best known are the polyphosphate glasses (including the amorphous phosphoric acids). These compositions have been made with different (H', Li', N a t , K', Rb', Mg", Ca2+,Zn", A13', etc.), and their molecular compositions have been measured by paper chromatography (all but A13+) of the counterions listed where the glass dissolves in the chromatographic solvent. Dissolving does not disrupt the d i s t r i b ~ t i o n ' ~so~ a, freshly prepared solution containing chain and ring structures represents the composition of the solid. In these reorganizing systems the equilibria between the end, middle and branch PO, building units (see $15.1.3.2) correspond to a relatively high exothermic heat of formation of the ends and middles from the neso (isolated PO,) units and branch units, a situation that leads to unbranched chains in the polyphosphate region of composition and no end groups in the ultraphosphate region. As a result, the chemistry of these polyphosphate glasses is sufficiently simple to be interpretable. The distribution corresponds to a range of p values (see $15.1.3.1)depending on the particular cation or cation mixture chosen, so that if p > 1, the shorter-chain glasses based on Lit or Na' counterions exhibit particularly sharp distributions, much sharper than the random one for sorting end with middle groups into The vitreous phosphoric acid? in the polyphosphate composition range are usually prepared by either dehydrating a Icss condensed acid or combining the desired proportions of a less condensed with a more condensed acid, with the extreme example of the latter consisting of mixing phosphorus pentoxide with orthophosphoric acid or even water'. Condensation of orthophosphoric acid is effected by heating in a gold (or less preferably, platinum) dish. The dish must not be in contact with a reducing atmosphere or furnace linings since the elemental phosphorus thus produced attacks the preciousmetal container, causing it to crumble owing to intergranular corrosion. Mixtures of orthophosphoric acid with phosphorus pentoxide are also best prepared in similar dishes so that the mixture (which should be well stirred at RT) can be heated to equilibrate. The exact composition of such equilibrated mixtures can be established by proportioning the ingredients; but water loss or pickup is hard to avoid, so it is prudent to analyze all products by end-group titration'. The most generally prepared vitreous phosphates, the sodium polyphosphate glasses (1.5 > Na,O-P,O, 2 1.00), are made' by dehydration and melting of acid salts or their mixtures in platinum (again avoiding reducing substances), e.g.: n NaH,PO,

-

(NaPO,), (the metaphosphate compd.) + H,O

(a)

By employing phosphorus pentoxide as a reagent, ultraphosphates (Na,O-P,O, < 1) as well as meta- and polyphosphates may be made this way. Sodium carbonate reacts with phosphoric acid or phosphorus pentoxide similarly to make condensed sodium phosphates, and even mixtures of phosphoric acid with common salt produce these products if sufficient steam is blown through the melt* (hours of steaming). Un-ionized derivatives of the condensed phosphoric acids are also prepared thermally, using various proportions of phosphorus pentoxide with the smallest member of the desired family of compoundsg-", e.g.:

15.2. Ring-Ring and Ring-Polymer Interconversions 15.2.10. in Phosphorus-Oxygen Systems 15.2.10.2. Preparation of Cyclic Polymers.

(n

-

+ 2) (C,H,),PO, + (n - 1) P,O, (n

3 (C2Hs)n+2Pn03n+1 (average compd.) (b)

+ 2) POCl, + (n - 1) P,O,

(n + 2) OP[N(CH3)213

+ p2°5

145

3 PnO,,- 1Clnt2(average compd.)

(c)

pno2n-1[N(cH3)21n+2 (average compd.)(d)

Equilibrium mixtures of the polyphosphoryl chlorides have also been prepared by partial hydrolysis of POCl, followed by thermal equilibration, whereas ethyl esters of the condensed phosphates are made from the attack of phosphorus pentoxide on diethyl ether” : n POCl,

(n

+ (n - 1) H,O

+ 2) C,H,OC,H, + n P,O,

-

Pn0,,- 1Clnt2

+ (2n - 2) HCl

(el

2 (C2H5)n+2Pn03n+1 (average compd.) (f)

Mixtures of oxygen-bridged phosphoryl structures are produced by condensation polymerization under reflux, so that the volatile condensation byproduct (e.g., methyl chloride) can escape, or by heating in a sealed tube (caution: do not exceed the critical temperature) when the condensation byproduct does not have to be expelled for the reaction to proceed, e.g.:’, n RP(O)Cl,

+ (n + 2) RP(O)(OCH,),

-

2CH,O

if I P-0

CH,

+ 2n CH,C1

(g)

where R = CH, or C,H,. Except for the ionized compounds, all of the compositions discussed here can be handled in glass. (J R. VAN WAZER)

1. J. R. Van Wazer, Phosphorus and Its Compounds, Vol. I Chemistry, Interscience Publishers, New York, 1958, pp. 717-734, 744-800; also 419-477. Also see E. Thilo, Adv. Inorg. Chem. Radiochem., 4, 1 (1962). 2. T. Glonek, A. J. R. Costello, T. C. Myers, J. R. Van Wazer, J. Phys. Chem., 79, 1214 (1975). 3. T. Glonek, J. R. Van Wazer, M. Mudgett, T. C. Myers, Inorg. Chem., 11, 567 (1972). 4. R. F Jameson, J. Chem. SOC.,752 (1959). 5. T. R. Meadowcroft, F. D. Richardson, Trans. Faraduy SOC.,61, 54 (1965); see also A. E. R. Westman, P. A. Gartaganis, J. Am. Ceram. Soc., 40, 293 (1957). 6. D. W. Matula, L. C. D. Groenweghe, J. R. Van Wazer, J . Chem. Phys., 41, 3105 (1964). 7. J. R. Van Wazer, E. J. Griffith, J. F. McCullough, Anal. Chem., 26, 1755 (1954); see also E. J. Griffith, Anal. Chem., 28, 525 (1956). 8. C. F. Callis, unpublished investigations. 9. E. Schwarzmann, J. R. Van Wazer, J. Am. Chem. Soc., 83, 365 (1961); also see ref. 12. 10. L. C. D. Groenweghe, J. H. Payne, J. R. Van Wazer, J. Am. Chem. SOC.,82,5305 (1960); also see W. E. Morgan, T. Glonek, J. R. Van Wazer, Inorg. Chem., 13, 1832 (1974). 11. E. Schwarzmann, J R. Van Wazer, J. Am. Chem Soc., 82, 6009 (1960). 12. J. R. Van Wazer, S. Norval, J. Am. Chem. SOC., 88,4415 (1966). 13. D. Grant, J. R. Van Wazer, C. H. Dungan, J. Polym. Sci., 5, 57 (1967).

15.2.10.2. Preparation of Cyclic Polymers.

Depending on the cation, crystalline salts of the cyclic meta- and ultraphosphates are obtained from thermal processes’. Early literature (especially 1920- 1940) describes

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc. 15.2. Ring-Ring and Ring-Polymer Interconversions 15.2.10. in Phosphorus-Oxygen Systems 15.2.10.2. Preparation of Cyclic Polymers.

(n

-

+ 2) (C,H,),PO, + (n - 1) P,O, (n

3 (C2Hs)n+2Pn03n+1 (average compd.) (b)

+ 2) POCl, + (n - 1) P,O,

(n + 2) OP[N(CH3)213

+ p2°5

145

3 PnO,,- 1Clnt2(average compd.)

(c)

pno2n-1[N(cH3)21n+2 (average compd.)(d)

Equilibrium mixtures of the polyphosphoryl chlorides have also been prepared by partial hydrolysis of POCl, followed by thermal equilibration, whereas ethyl esters of the condensed phosphates are made from the attack of phosphorus pentoxide on diethyl ether” : n POCl,

(n

+ (n - 1) H,O

+ 2) C,H,OC,H, + n P,O,

-

Pn0,,- 1Clnt2

+ (2n - 2) HCl

(el

2 (C2H5)n+2Pn03n+1 (average compd.) (f)

Mixtures of oxygen-bridged phosphoryl structures are produced by condensation polymerization under reflux, so that the volatile condensation byproduct (e.g., methyl chloride) can escape, or by heating in a sealed tube (caution: do not exceed the critical temperature) when the condensation byproduct does not have to be expelled for the reaction to proceed, e.g.:’, n RP(O)Cl,

+ (n + 2) RP(O)(OCH,),

-

2CH,O

if I P-0

CH,

+ 2n CH,C1

(g)

where R = CH, or C,H,. Except for the ionized compounds, all of the compositions discussed here can be handled in glass. (J R. VAN WAZER)

1. J. R. Van Wazer, Phosphorus and Its Compounds, Vol. I Chemistry, Interscience Publishers, New York, 1958, pp. 717-734, 744-800; also 419-477. Also see E. Thilo, Adv. Inorg. Chem. Radiochem., 4, 1 (1962). 2. T. Glonek, A. J. R. Costello, T. C. Myers, J. R. Van Wazer, J. Phys. Chem., 79, 1214 (1975). 3. T. Glonek, J. R. Van Wazer, M. Mudgett, T. C. Myers, Inorg. Chem., 11, 567 (1972). 4. R. F Jameson, J. Chem. SOC.,752 (1959). 5. T. R. Meadowcroft, F. D. Richardson, Trans. Faraduy SOC.,61, 54 (1965); see also A. E. R. Westman, P. A. Gartaganis, J. Am. Ceram. Soc., 40, 293 (1957). 6. D. W. Matula, L. C. D. Groenweghe, J. R. Van Wazer, J . Chem. Phys., 41, 3105 (1964). 7. J. R. Van Wazer, E. J. Griffith, J. F. McCullough, Anal. Chem., 26, 1755 (1954); see also E. J. Griffith, Anal. Chem., 28, 525 (1956). 8. C. F. Callis, unpublished investigations. 9. E. Schwarzmann, J. R. Van Wazer, J. Am. Chem. Soc., 83, 365 (1961); also see ref. 12. 10. L. C. D. Groenweghe, J. H. Payne, J. R. Van Wazer, J. Am. Chem. SOC.,82,5305 (1960); also see W. E. Morgan, T. Glonek, J. R. Van Wazer, Inorg. Chem., 13, 1832 (1974). 11. E. Schwarzmann, J R. Van Wazer, J. Am. Chem Soc., 82, 6009 (1960). 12. J. R. Van Wazer, S. Norval, J. Am. Chem. SOC., 88,4415 (1966). 13. D. Grant, J. R. Van Wazer, C. H. Dungan, J. Polym. Sci., 5, 57 (1967).

15.2.10.2. Preparation of Cyclic Polymers.

Depending on the cation, crystalline salts of the cyclic meta- and ultraphosphates are obtained from thermal processes’. Early literature (especially 1920- 1940) describes

146

15.2. Ring-Ring and Ring-Polymer Interconversions 15.2.10. in Phosphorus-Oxygen Systems 15.2.10.2. Preparation of Cyclic Polymers.

metaphosphate mixtures erroneously considered to be pure compounds (mainly monothrough hexametaphosphates). The smallest metaphosphate ring produced by equilibrium processes is the trimeta. The larger rings are difficult to crystallize. Sodium trimetaphosphate, Na,[P,O,], is made' by heating monosodium orthophospate, Na[H,PO,], to 500-600°C and allowing it to be held at that temperature a few minutes for dehydration. This salt recrystallizes from water as the hexahydrate, monohydrate or anhydrous form. Double salts of the trimetaphosphate anion include Na,K,[P,O,],, LiK,[P,O,] and NaBa[P,O,] and its tetrahydrate. Aqueous triphosphoric acid is prepared by passing a solution of the sodium salts through an ion-exchange column; certain acid salts may be made thermally, e.g., Na,H[P,O,]. Tetrametaphosphates of doubly charged metal cations are prepared thermally or by precipitation and may be converted to soluble salts by ion-exchange or action of a sulfide, e.g., Cuz[P,O,,] reacted with Na,S free of polysulfides. However, the main way of making soluble tetrametaphosphates' is by slowly adding P,Ol0 to ice water and then neutralizing promptly. A hydrated salt, Na,[P,O,,].x H,O with x = 4 or 10, is readily crystallized from this solution after addition of sodium chloride in 75 % yield. Crystalline salts with six and eight phosphorus atoms per metaphosphate ring are utilized in preparing soluble salts. Soluble cyclic hexametaphosphates are made by thermal synthesis3 of the lithium salt, Li6[P,01,], which dissolves readily. The cyclic lead(I1) o~tametaphosphate~ is obtained in 70 % yield by heating a mixture of lead(I1) nitrate with sodium tetrametaphosphate and is then converted to the sodium salt, H,O, by treating with a polysulfide-free sodium sulfide solution. A Na,[P,0,,].6 decametaphosphate ring has been found' for the 2 BaO . 3 ZnO 5 P,O, crystalline composition. By the above furnace syntheses soluble ring phosphates are obtained in gram to kilogram amounts. Metaphosphate phase-diagram6 and ring-size studies are needed in the search for crystalline salts containing presently unknown sizes of metaphosphate rings. Until recently there were no solvents other than water for dissolving a range of different phosphate structures. Since the rate of hydrolysis is about as rapid as the phosphate scrambling processes', condensation reactions carried out in water involve too much hydrolytic degradation to be used for preparative purposes. However, tetramethylurea with a tri-n-alkylamine (to act as a counterion) is a good solvent for mixtures of ortho- and condensed phosphoric acids so that condensation and reorganization reactions can be carried out' in solution. Since conversion of the longer chains to rings increases the mole number, dilution favors ring formation. This has been usedg to make and isolate the complete series of simple-ring phosphates, from the tri- through the nonameta. Following preparative chromatographic separation, 1-2 g samples of pure, dry sodium salts are obtained for each ring molecule from a rotary evaporator. By carrying out the solution condensation to the metaphosphate at several selected concentrations, this method could be developed into a general procedure for making cyclic metaphosphates. Several crystalline ultraphosphates have been discovered in phase-diagram studies', and the structures of some determined by x-ray diffraction. Thus, the rare-earth ultraphosphate HoP,O,, consists" of an infinite ladder-like structure in which alternating PO, groups in each of the paired polyphosphate chains forming the sides of the ladder are bridged by a rung consisting of a PO, group. The best known crystalline metal ultraphosphate is Ca[P,O, J,which has a ladder structure". The various forms of phosphorus pentoxide' are also considered ultraphosphates. Of the three crystalline forms, two consist of sheets of fused rings and one is a cage structure. N

147

15.2. Ring-Ring and Ring-Polymer lnterconversions 15.2.10. in Phosphorus-Oxygen Systems 15.2.10.2. Preparation of Cyclic POlyInerS.

The only small-molecule ultraphosphate (necessarily a cage structure) is the P,O form of phosphorus pentoxide, which has been made only by direct oxidation of elemental phosphorus and for which no nondestructive solvent is known. Phosphorus-31 NMR studies' indicate that a soluble organic adduct", P4010 C(NR),, where R = cyclohexyl, of phosphorus pentoxide results from a double-step condensation of H,PO, by dicyclohexylcarbodiimide in tetramethylurea. Use of these organic condensing agents (formerly employed in the preparation of biological phosphates) has led to the preparation and 'P NMR identification12 of small-molecule ultraphosphates that occur in condensation products of phosphoric acid derivatives:

-

0 D

0 I

0 I1

0 2-

I11

0,-

0 2-

/p -9P-0-P

0

\

P-0

0 2

/o

P

P

/\

0 2-

o\

"\

/Po2-

0

0

\ /~

-

P

IV

0 2-

0 2-

V

0 2-

Structure I is postulated to occur as the first step in the hydrolysis of P4010. The characteristic NMR patterns of some of these structures are observedl3 in the POC1,-P20, system of reorganizing molecules, where C1 substitutes for the acid OH group. (J.R. VAN WAZER)

1. J. R. Van Wazer, Phosphorus and Its Compounds, Vol. I : Chemistry, Interscience Publishers, New York, 1958, pp. 601,679, also 419; see also D. E. C. Corbridge, The Structural Chemistry of Phosphorus,.Elsevier Scientific Publishing Company, Amsterdam, 1974, pp. 153-166; E. Thilo, Adv. Inorg. Radiochem., 4, 1 (1962); S. Y. Kalliney, Topics Phos. Chem., 7, 255 (1975); C. Y. Shen, D. R. Dryoff, Prep. Inorg. React., 5,157 (1968); I. Haduc, The Chemistry of Inorganic Ring Systems, Wiley, London, 1970; in two volumes. 2. E. Thilo, W. Wieker, Z. Anorg. ANg. Chem., 277, 27 (1954). 3. E. J. Griffith, R. L. Buxton, Inorg. Chem. 4, 549 (1965). 4. U. Schiilke, Angew. Chem., Int. Ed. Engl., 7, 71 (1968). 5 . M: Bagieu-Beucher, A.. Durif, J. C. Guitel, J . Solid State Chem., 40, 248 (1980); 45, 159 (1982). 6. Phase diagrams of binary mixtures of metaphosphate salts of various metals are reported by A. Durif, Rev. Phos. Chem., 6,109 (1969) and by J. Majling, F. Hanic, Topics Phos. Chem., 10,341 (1980). 7. J. F. McCullough, J. R. Van Wazer, E. J. Griffith, J. Am. Chem. SOC.,78,4528 (1956). 8. A. J. Costello, T. Glonek, T. C. Myers, J. R. Van Wazer, Inorg. Chem., 13, 1225 (1974).

148

15.2. Ring-Ring and Ring-Polymer Interconversions 15.2.10. in Phosphorus-Oxygen Systems 15.2.10.3. Chain Oligomers and Macromolecules.

9. T. Glonek, J. R. Van Wazer, M. Mudgett, T. C. Myers, Znorg. Chem., 11, 567 (1972). 10. A. Durif, Bull, Soc. Fr. Mineral. Crist., 94, 314 (1971), for HoP,O,,; M. Rzaigui, N. Kbis Ariguib, M. T. Averbuch-Pouchot,A. Durif, J . Solid State Chem., 52,61(1984) for CeP,O,,; I. Torjman, M. Bagieu-Beucher,R. Zilber, Z . Krist., 140, 145 (1974) for Ca,LP,O,,]. 11. T. Glonek, T. C. Myers, J. R. Van Wazer, J. Am. Chem. Soc., 97, 206 (1975). 12. T. Glonek, J. R. Van Wazer, R. A. Kleps, T. C. Myers, Inorg. Chem., 13,2337 (1974). 13. W. E. Morgan, T. Glonek, J. R. Van Wazer, Znorg. Chem., 13, 1832 (1974).

15.2.10.3. Chain Oligomers and Macromolecules. Pyrophosphate salts''2 are known for all but the rarest of cations, and other derivatives abound, including metal complexes and esters. However, it was not until 1940 that the existence of linear triphosphates was accepted'. Those linear phosphate^'-^ for which n > 2 in the [Pn03n+1]("+2)anion have been called polyphosphates, as in sodium tripolyphosphate, with the poly implying a linear chain structure. This notation is not employed here. Soluble pyro- and triphosphates are manufactured in tonnage as the anhydrous sodium salts' Na,P,O, and Na,P,O,, . Both appear in the Na,O-P,O, phase diagram' and are made thermally by dehydration of the acid orthophosphates. Since the thermal conversion whereby crystalline sodium triphosphate is formed is a complicated process'34 involving transitions between ca. seven crystalline phases and an amorphous one consisting of both ring and chain structures, the anhydrous salt contains a few percent impurities. Therefore, the triphosphate is recrystallized to get Na,[P,Ol0]~6 H,O, a salt that decomposes in a complicated manner5 to various proportions of other phosphates when dehydrated. Tetraphosphates appear in phase diagrams; and the lead tetraphosphate6 is used to make soluble salts. Since soluble tetraphosphate salts (including the Na and the K) cannot be crystallized the lead compound is reacted with ammonium sulfide to get crystalline [NH,],[P,O,,] .6 H,O, a salt that recrystallizes from water7. An alkaline solution of sodium tetrametaphosphate made from the reaction of P,O,, with cold water ($15.2.10.3)hydrolyses' to the linear tetraphosphate, which can then be precipitated as the guanidinium salt' [HN=C(NH,)2H],[P,01,] .H,O. Tromelite, which occupies a triangular area in the anhydrous calcium phosphate phase diagram, was erroneously claimed9 to be a pentapolyphosphate on the basis of paper chromatography, but it was later shown1° to be a hexapolyphosphate, so that tromelite must be Ca,[P,O,,]. The crystalline tri- through pentapolyphosphates appear in their phase diagrams with incongruent melting points, as expected for a melt undergoing dynamic stuctural reorganization. Longer-chain crystalline phosphates will be found as more phase diagrams of systems involving two or three cations are examined. As for the cyclic metaphosphates, the most readily available source of linear phosphate oligomers is an equilibrated amorphous mixture. Since chains are favored in concentrated systems, a condensed acid or an alkali-metal glass exhibiting a moderate chain length ( - 6) is employed as starting material. Isolation in pure form of gram quantities of individual chain phosphates from the oligomeric glasses is largely unrealized. One approach' is solvent extraction using t-caprylamine in xylene as the extractant for aqueous phosphoric acid solutions. Purification is achieved with an anionexchange column, the use of which introduces KC1 that is then removed by dialysis. Since the precipitated sodium or potassium salts of the separated chain-phosphate anions decompose within weeks, stable guanidinium or acridinium salts are crystallized. Tetra-, penta-, hexa-, hepta- and octaphosphates are obtained.

-

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc. 148

15.2. Ring-Ring and Ring-Polymer Interconversions 15.2.10. in Phosphorus-Oxygen Systems 15.2.10.3. Chain Oligomers and Macromolecules.

9. T. Glonek, J. R. Van Wazer, M. Mudgett, T. C. Myers, Znorg. Chem., 11, 567 (1972). 10. A. Durif, Bull, Soc. Fr. Mineral. Crist., 94, 314 (1971), for HoP,O,,; M. Rzaigui, N. Kbis Ariguib, M. T. Averbuch-Pouchot,A. Durif, J . Solid State Chem., 52,61(1984) for CeP,O,,; I. Torjman, M. Bagieu-Beucher,R. Zilber, Z . Krist., 140, 145 (1974) for Ca,LP,O,,]. 11. T. Glonek, T. C. Myers, J. R. Van Wazer, J. Am. Chem. Soc., 97, 206 (1975). 12. T. Glonek, J. R. Van Wazer, R. A. Kleps, T. C. Myers, Inorg. Chem., 13,2337 (1974). 13. W. E. Morgan, T. Glonek, J. R. Van Wazer, Znorg. Chem., 13, 1832 (1974).

15.2.10.3. Chain Oligomers and Macromolecules. Pyrophosphate salts''2 are known for all but the rarest of cations, and other derivatives abound, including metal complexes and esters. However, it was not until 1940 that the existence of linear triphosphates was accepted'. Those linear phosphate^'-^ for which n > 2 in the [Pn03n+1]("+2)anion have been called polyphosphates, as in sodium tripolyphosphate, with the poly implying a linear chain structure. This notation is not employed here. Soluble pyro- and triphosphates are manufactured in tonnage as the anhydrous sodium salts' Na,P,O, and Na,P,O,, . Both appear in the Na,O-P,O, phase diagram' and are made thermally by dehydration of the acid orthophosphates. Since the thermal conversion whereby crystalline sodium triphosphate is formed is a complicated process'34 involving transitions between ca. seven crystalline phases and an amorphous one consisting of both ring and chain structures, the anhydrous salt contains a few percent impurities. Therefore, the triphosphate is recrystallized to get Na,[P,Ol0]~6 H,O, a salt that decomposes in a complicated manner5 to various proportions of other phosphates when dehydrated. Tetraphosphates appear in phase diagrams; and the lead tetraphosphate6 is used to make soluble salts. Since soluble tetraphosphate salts (including the Na and the K) cannot be crystallized the lead compound is reacted with ammonium sulfide to get crystalline [NH,],[P,O,,] .6 H,O, a salt that recrystallizes from water7. An alkaline solution of sodium tetrametaphosphate made from the reaction of P,O,, with cold water ($15.2.10.3)hydrolyses' to the linear tetraphosphate, which can then be precipitated as the guanidinium salt' [HN=C(NH,)2H],[P,01,] .H,O. Tromelite, which occupies a triangular area in the anhydrous calcium phosphate phase diagram, was erroneously claimed9 to be a pentapolyphosphate on the basis of paper chromatography, but it was later shown1° to be a hexapolyphosphate, so that tromelite must be Ca,[P,O,,]. The crystalline tri- through pentapolyphosphates appear in their phase diagrams with incongruent melting points, as expected for a melt undergoing dynamic stuctural reorganization. Longer-chain crystalline phosphates will be found as more phase diagrams of systems involving two or three cations are examined. As for the cyclic metaphosphates, the most readily available source of linear phosphate oligomers is an equilibrated amorphous mixture. Since chains are favored in concentrated systems, a condensed acid or an alkali-metal glass exhibiting a moderate chain length ( - 6) is employed as starting material. Isolation in pure form of gram quantities of individual chain phosphates from the oligomeric glasses is largely unrealized. One approach' is solvent extraction using t-caprylamine in xylene as the extractant for aqueous phosphoric acid solutions. Purification is achieved with an anionexchange column, the use of which introduces KC1 that is then removed by dialysis. Since the precipitated sodium or potassium salts of the separated chain-phosphate anions decompose within weeks, stable guanidinium or acridinium salts are crystallized. Tetra-, penta-, hexa-, hepta- and octaphosphates are obtained.

-

15.2. Ring-Ring and Ring-Polymer Interconversions 15.2.10. in Phosphorus-Oxygen Systems 15.2.10.3. Chain Oligomers and Macromolecules.

149

Alternatively, preparative column chromatography involving a diethylaminocellulose column (5 x 8 cm) is used for a rough separation, followed by further purification of the crude cuts by passage through a second similarly packed column (2.5 x 240 crn)',. High purity is required since trace-metal impurities broaden the elution peaks corresponding to the various oligomers and inhibit crystallization of the isolated alkali-metal salts13. The starting phosphate glass is passed through the hydrogen form of a cation-exchange resin to remove impurities (particularly dissolved gold from the crucible used to make the glass), all equipment is washed with EDTA, and the polyethylene collection vessels are covered to avoid contamination by dust. Phosphates from tetra through nona are isolated in high purity. High-pressure liquid chromatography (HPLC) separate^'^ oligomeric phosphates and, with avoidance of trace impurities, is the best choice for future work. In addition to the oligomers macromolecular chain phosphates are well established'-,. The long-chain vitreous sodium phosphate can be prepared with an average chain length of several hundred phosphorus atoms per Chain length is limited owing to problems in removing chain-breaking trace amounts of water from the melt so as to bring the (Na,O H2)/P,05 mole ratio exactly to unity. Smaller ratios reduce chain length, and a larger one introduces crosslinking. However, a trace of crosslinking can be detected at average chain lengths as low as -50, (Na,O H,) = 1.04; e.g., at (Na,O + H,O)/P,O, ratios of -0.96, aqueous glass solutions exhibit rheological behavior characteristic of crosslinking. In solution the hydrolytic degradation at branching PO4 groups is rapid, and the branches disappear within 12 h. The alkali metals form crystalline metaphosphates that exhibit high-polymer proper tie^'-^. There are two main macromolecular sodium metaphosphates, I, fibrous, with three reported crystalline forms, and 11, a high- and a low-temperature formI5. Potassium sodium metaphosphate salts are readily prepared. The long chains in these and other macromolecular metaphosphates exhibit characteristic configurations, of which eight arrangements have been identified2. In potassium sodium metaphosphate, salt the size of the phosphate chains can be controlled within wide limits without changing the diffraction pattern of the crystal'-3. If water is not thoroughly eliminated during the thermal conversion of KH,PO, to the (KPO,),, the chain length is shortened'. Other impurities also have the same effect, as does excess potassium. A deficiency of potassium leads to crosslinking, as do traces of silica. As little as one crosslink per lo3 PO, tetrahedra raises the solution viscosity. A chain length of lo4 P atoms per molecule-ion is achieved when avoiding chain terminators and crosslinking. The thermally prepared alkali-metal salts of the long-chain metaphosphates are insoluble in water at RT, but they swell and then form a viscous solution when exposed to an aqueous solution of the salt of a different alkali-metal cation.

+

+

-

-

(J.R. VAN WAZER)

1. J. R. Van Wazer, Phosphorus and Its Compounds, Vol. I : Chemistry, Interscience Publishers, New York, 1958, p. 601, 419; see also C. Y. Shen, D. R. Dyroff, Prep. Inorg. Reactions, 5, 157 (1968). 2. D. E. C. Corbridge, The Structural Chemistry of Phosphorus, Elsevier Scientific Publishing Company, Amsterdam, 1914, pp. 127, 243; Phosphorus, An Outline of Its Chemistry, Biochemistry, and Technology, 2nd ed., Elsevier Scientific Publishing Company, Amsterdam, 1974, Chapt. 3. 3. E. Thilo, Ado. Inorg. Chem. Radiochem., 4, 1 (1962). 4. J. W. Edwards, A. H. Herzog, J. Am. Chem. SOC.,79, 3647 (1957). For another paper on the dehydration mechanism, see J. D. McGilvery, A. E. Scott, Can. J. Chem., 32, 1100 (1955).

150

15.2. Ring-Ring and Ring-Polymer Interconversions 15.2.10. in Phosphorus-Oxygen Systems 15.2.10.4. Related Mixed Systems. ~~

~

~~~

5. E. J. Griffith, Pure Appl. Chem., 44, 173 (1975). 6. R. K. Osterheld, R. P. Langguth, J. Phys. Chem., 59, 76 (1955); for similar work on tetraphosphates of other metals, see I. Schulze, 2. Anorg. Allg. Chem., 287, 106 (1956). 7. E. J. Griffith, J. Inorg. Nucl. Chem., 26, 1381 (1964). 8. 0. T. Quimby, J. Phys. Chem., 58, 603 (1954). 9. S. Ohashi, J. R. Van Wazer, J. Am. Chem. SOC.,XI, 831 (1959); J. R. Van Wazer, S. Ohashi, J. Am. Chem. SOC.,80, 1010 (1958). 10. W. Wieker, A. R. Grimmer, E. Thilo, Z. Anorg. Allg. Chem., 330, 78 (1964). 11. E. J. Griffith, R. L. Buxton, J. Am. Chem. Soc., 89, 2884 (1967). 12. T. Glonek, J. R. Costello, T. C. Myers, J. R. Van Wazer, J. Phys. Chem., 79, 1214 (1975). 13. E. J. Griffith disagrees with T. Glonek about the efficacy of trace-impurity removal in permitting

the sodium or potassium salts of the higher oligomers to crystallize and be stable in storage.

14. Private communication from T. Miyajima to T. Glonek describing a study using the technology of Bull. Chem. SOC.Jpn., 51, 2543 (1978). 15. J. Malling, F. Hanic, Top. Phosphorous Chem., 10, 341 (1980).

15.2.10.4. Related Mixed Systems. There are chain and ring anionic species in which a portion of either the P or the bridging 0 atom of the condensed phosphates is substituted by other atoms, e.g., arsenato-, sulfato- and silicophosphates1S2, as well as other systems with mixed anions. Vitreous sodium arsenatoph~sphates~ are prepared like the phosphate glasses by dehydration of orthophosphate and orthoarsenate mixtures. The glasses in the poly region of composition dissolve readily, with instantaneous scission of all As-0-As and As-0-P linkages, to give distributions of chain segments in which all linkages are P-0-P. These polyphosphate distributions remaining after hydrolytic loss of arsenic correspond to random sorting of the arsenic with the phosphorus atoms in the original arsenatophosphate glasses. Cyclic arsenatophosphates are known3, as are crystals exhibiting macromolecular chains4 Because of rapid crystallization of sodium sulfate from sodium sulfatophosphate melts, a rapid quenching rate (between a pair of heavy copper chill plates) is needed to obtain glasses in the poly composition region'. Glasses in the ultra region are easier to prepare'q5s6. Although S-0-S and S-0-P linkages hydrolyze rapidly upon dissolution in water, mixed sulfatophosphate molecule-ions are isolated' with quickly performed anion-exchange chromatography, and ,'P NMR also gives evidence of mixed molecules5. Sulfatophosphoric acids produced' by combining H,SO,, H , P 0 4 , SO, and P,O, in various proportions contain by 31P N M R mixed molecules that do not dominate the acid until it becomes highly viscous. Crystalline oligomers are also reported7. Calcium silicophosphate glasses' exhibit molecular structures in which SiO, and PO, tetrahedra are interlinked through bridging 0 atoms. As expected, dissolution in water is slow and partial but may be accelerated by EDTA or an acidic cation exchanger. Barium fractionation shows that the average polyphosphate chain length in the dissolved portion decreases with increasing silica content. Oligomeric structures based on dimethylsilyl and methylphosphoryl moieties and prepared* by R T condensation seem to exhibit a regular alternation of P and Si groups in the molecular backbone, attributable to kinetic rather than thermodynamic control of the condensation process. Methylenediphosphonic acid, H,[O,P-CH,-PO,], is condensedg in a nonaqueous medium by a carbodiimide to give the cyclic dimer, the anion of which is a tetrametaphosphate analog having alternate bridging 0 atoms substituted by methylene

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc. 150

15.2. Ring-Ring and Ring-Polymer Interconversions 15.2.10. in Phosphorus-Oxygen Systems 15.2.10.4. Related Mixed Systems. ~~

~

~~~

5. E. J. Griffith, Pure Appl. Chem., 44, 173 (1975). 6. R. K. Osterheld, R. P. Langguth, J. Phys. Chem., 59, 76 (1955); for similar work on tetraphosphates of other metals, see I. Schulze, 2. Anorg. Allg. Chem., 287, 106 (1956). 7. E. J. Griffith, J. Inorg. Nucl. Chem., 26, 1381 (1964). 8. 0. T. Quimby, J. Phys. Chem., 58, 603 (1954). 9. S. Ohashi, J. R. Van Wazer, J. Am. Chem. SOC.,XI, 831 (1959); J. R. Van Wazer, S. Ohashi, J. Am. Chem. SOC.,80, 1010 (1958). 10. W. Wieker, A. R. Grimmer, E. Thilo, Z. Anorg. Allg. Chem., 330, 78 (1964). 11. E. J. Griffith, R. L. Buxton, J. Am. Chem. Soc., 89, 2884 (1967). 12. T. Glonek, J. R. Costello, T. C. Myers, J. R. Van Wazer, J. Phys. Chem., 79, 1214 (1975). 13. E. J. Griffith disagrees with T. Glonek about the efficacy of trace-impurity removal in permitting

the sodium or potassium salts of the higher oligomers to crystallize and be stable in storage.

14. Private communication from T. Miyajima to T. Glonek describing a study using the technology of Bull. Chem. SOC.Jpn., 51, 2543 (1978). 15. J. Malling, F. Hanic, Top. Phosphorous Chem., 10, 341 (1980).

15.2.10.4. Related Mixed Systems. There are chain and ring anionic species in which a portion of either the P or the bridging 0 atom of the condensed phosphates is substituted by other atoms, e.g., arsenato-, sulfato- and silicophosphates1S2, as well as other systems with mixed anions. Vitreous sodium arsenatoph~sphates~ are prepared like the phosphate glasses by dehydration of orthophosphate and orthoarsenate mixtures. The glasses in the poly region of composition dissolve readily, with instantaneous scission of all As-0-As and As-0-P linkages, to give distributions of chain segments in which all linkages are P-0-P. These polyphosphate distributions remaining after hydrolytic loss of arsenic correspond to random sorting of the arsenic with the phosphorus atoms in the original arsenatophosphate glasses. Cyclic arsenatophosphates are known3, as are crystals exhibiting macromolecular chains4 Because of rapid crystallization of sodium sulfate from sodium sulfatophosphate melts, a rapid quenching rate (between a pair of heavy copper chill plates) is needed to obtain glasses in the poly composition region'. Glasses in the ultra region are easier to prepare'q5s6. Although S-0-S and S-0-P linkages hydrolyze rapidly upon dissolution in water, mixed sulfatophosphate molecule-ions are isolated' with quickly performed anion-exchange chromatography, and ,'P NMR also gives evidence of mixed molecules5. Sulfatophosphoric acids produced' by combining H,SO,, H , P 0 4 , SO, and P,O, in various proportions contain by 31P N M R mixed molecules that do not dominate the acid until it becomes highly viscous. Crystalline oligomers are also reported7. Calcium silicophosphate glasses' exhibit molecular structures in which SiO, and PO, tetrahedra are interlinked through bridging 0 atoms. As expected, dissolution in water is slow and partial but may be accelerated by EDTA or an acidic cation exchanger. Barium fractionation shows that the average polyphosphate chain length in the dissolved portion decreases with increasing silica content. Oligomeric structures based on dimethylsilyl and methylphosphoryl moieties and prepared* by R T condensation seem to exhibit a regular alternation of P and Si groups in the molecular backbone, attributable to kinetic rather than thermodynamic control of the condensation process. Methylenediphosphonic acid, H,[O,P-CH,-PO,], is condensedg in a nonaqueous medium by a carbodiimide to give the cyclic dimer, the anion of which is a tetrametaphosphate analog having alternate bridging 0 atoms substituted by methylene

15.2. Ring-Ring and Ring-Polymer Interconversions 15.2.11. in Phosphorus-Nitrogen Oligomers and Polymers 15.2.11.l. Halocyclophosphazene Polymerization

151

groups. Further condensation produces an oxygen bridge between opposing pairs of phosphorus atoms in this ring. The final, fully condensed, product is the analog of phosphorus pentoxide in which a pair of oppositely placed 0 atoms are substituted by methylene groups to give tetraphosphorus dimethylene octaoxide. Mixtures of linear phosphatic oligomers (in which oxygen bridges alternate with methylene bridges) are obtained" by condensation of methylenediphosphonate anions, with chains of 4-20 P atoms per molecule being separated by preparative column chromatography. Although the cage molecules P,O,, P,O,, P,O, and P,O, (which result from burning elemental P in an 0 deficiency) exhibit oxygen bridging between phosphorus atoms each with a lone pair of electrons, there is no evidence for chain and simple-ring strrdures in which such triply connected P atoms are bridged by oxygen". When P-0-P condensation is attempted P-P bonding rather than P-0-P bridging results, owing to the transfer of 0 atoms from the bridge to the phosphine lone pairs. (J.R. VAN WAZER)

1. S. Ohashi, Topics Phos. Chem., 1, 189 (1965). 2. D. E. C. Corbridge, The Structural Chemistry of Phosphorus, Elsevier Scientific Publishing Company, New York, 1974, p. 166. 3. E. Thilo, Adv. Inorg. Chem. Radiochem., 4, 1 (1962). 4. K. Jost, H. Worzala, E. Thilo, Z. Anorg. Allg. Chem., 325, 98 (1963). 5. F. G . Remy, J. R. Van Wazer, J. Inorg. Nucl. Chem.,36, 1905 (1974) for sodium sulfatophosphate glasses; Phosphorus, 5, 37 (1974) for mixed polyacids. 6 . S. Ohashi, K. Ikeda, Bull. Chem. Soc. Jpn., 36, 1530 (1963). 7. F. von Lempe, Z. Anorg. Allg. Chem., 367, 170 (1969); Z. Anorg. Allg. Chem.,368, 93 (1969). 8. S. H. Cook, J. R. Van Wazer, Inorg. Chem., 12, 909 (1973). 9. T. Glonek, R. A. Kleps, J. R. Van Wazer, T. C. Myers, Bioinorg. Chem., 6, 295 (1976). 10. S. 0. Nweke, J. R. Van Wazer, Inorg. Chem., 16, 860 (1977). 11. See pp. 65-72, 166 in ref. 2. 12. K. M. Abraham, J. R. Van Wazer, Inorg. Chem., I3,2346 (1974); Znorg. Chem., 14, 1099 (1975).

15.2.11. in Phosphorus-Nitrogen Oligomers and Polymers 15.2.11.l.Halocyclophosphazene Polymerization (i) Background. Phosphazene monomers (i.e., :N-PR,) or cyclic dimers [+N=PR,+,] have not yet been observed as stable species. The lowest members of the cyclic oligomers series are trimers and tetramers. The polymerization reactions of the trimers (NPF,), and (NPCI,), are the critical processes for the synthesis of high polymers. The analogous bromocyclotriphosphazene, (NPBr,), , also polymerizes, but its behavior has not yet been studied in detail. As long ago as 1897 it was reported' that molten hexachlorocyclotriphosphazene (I) changes slowly to a transparent, rubbery material. During the following 68 years, this material, known as inorganic rubber, received a limited degree of attention as a laboratory curiosity. Although the unusual nature of its physical properties was recognized by a number of investigators2, its hydrolytic instability and insolubility in all solvents discouraged both fundamental studies and attempts to apply this material in technology. In the mid-1960's techniques were developed for the polymerization of

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc. 15.2. Ring-Ring and Ring-Polymer Interconversions 15.2.11. in Phosphorus-Nitrogen Oligomers and Polymers 15.2.11.l. Halocyclophosphazene Polymerization

151

groups. Further condensation produces an oxygen bridge between opposing pairs of phosphorus atoms in this ring. The final, fully condensed, product is the analog of phosphorus pentoxide in which a pair of oppositely placed 0 atoms are substituted by methylene groups to give tetraphosphorus dimethylene octaoxide. Mixtures of linear phosphatic oligomers (in which oxygen bridges alternate with methylene bridges) are obtained" by condensation of methylenediphosphonate anions, with chains of 4-20 P atoms per molecule being separated by preparative column chromatography. Although the cage molecules P,O,, P,O,, P,O, and P,O, (which result from burning elemental P in an 0 deficiency) exhibit oxygen bridging between phosphorus atoms each with a lone pair of electrons, there is no evidence for chain and simple-ring strrdures in which such triply connected P atoms are bridged by oxygen". When P-0-P condensation is attempted P-P bonding rather than P-0-P bridging results, owing to the transfer of 0 atoms from the bridge to the phosphine lone pairs. (J.R. VAN WAZER)

1. S. Ohashi, Topics Phos. Chem., 1, 189 (1965). 2. D. E. C. Corbridge, The Structural Chemistry of Phosphorus, Elsevier Scientific Publishing Company, New York, 1974, p. 166. 3. E. Thilo, Adv. Inorg. Chem. Radiochem., 4, 1 (1962). 4. K. Jost, H. Worzala, E. Thilo, Z. Anorg. Allg. Chem., 325, 98 (1963). 5. F. G . Remy, J. R. Van Wazer, J. Inorg. Nucl. Chem.,36, 1905 (1974) for sodium sulfatophosphate glasses; Phosphorus, 5, 37 (1974) for mixed polyacids. 6 . S. Ohashi, K. Ikeda, Bull. Chem. Soc. Jpn., 36, 1530 (1963). 7. F. von Lempe, Z. Anorg. Allg. Chem., 367, 170 (1969); Z. Anorg. Allg. Chem.,368, 93 (1969). 8. S. H. Cook, J. R. Van Wazer, Inorg. Chem., 12, 909 (1973). 9. T. Glonek, R. A. Kleps, J. R. Van Wazer, T. C. Myers, Bioinorg. Chem., 6, 295 (1976). 10. S. 0. Nweke, J. R. Van Wazer, Inorg. Chem., 16, 860 (1977). 11. See pp. 65-72, 166 in ref. 2. 12. K. M. Abraham, J. R. Van Wazer, Inorg. Chem., I3,2346 (1974); Znorg. Chem., 14, 1099 (1975).

15.2.11. in Phosphorus-Nitrogen Oligomers and Polymers 15.2.11.l.Halocyclophosphazene Polymerization (i) Background. Phosphazene monomers (i.e., :N-PR,) or cyclic dimers [+N=PR,+,] have not yet been observed as stable species. The lowest members of the cyclic oligomers series are trimers and tetramers. The polymerization reactions of the trimers (NPF,), and (NPCI,), are the critical processes for the synthesis of high polymers. The analogous bromocyclotriphosphazene, (NPBr,), , also polymerizes, but its behavior has not yet been studied in detail. As long ago as 1897 it was reported' that molten hexachlorocyclotriphosphazene (I) changes slowly to a transparent, rubbery material. During the following 68 years, this material, known as inorganic rubber, received a limited degree of attention as a laboratory curiosity. Although the unusual nature of its physical properties was recognized by a number of investigators2, its hydrolytic instability and insolubility in all solvents discouraged both fundamental studies and attempts to apply this material in technology. In the mid-1960's techniques were developed for the polymerization of

152

15.2. Ring-Ring and Ring-Polymer Interconversions 15.2.11. in Phosphorus-Nitrogen Oligomers and Polymers 15.2.11.l.Halocyclophosphazene Polymerization

(NPCl,), to a soluble (uncrosslinked) form of high molecular weight, polydichlorophosphazene This material is a key intermediate for the preparation of polyorganophosphazenes by the substitutive method. (ii) Chlorophosphazene Polymerization. The polymerization and depolymerization reactions of chlorophosphazenes form a complex series of interrelated processes. The overall scheme is summarized in Eq. (a):

I

I

II I

Cl-P-N=

P-Cl

I

c1

C1 111

Polymerization of (NPCl,), or (NPCI,), takes place at 230-270 “C in the melt. Because these oligomers are quite volatile, the reactions must be contained within sealed, thickwalled glass tubes or within an autoclave. The rate of polymerization (and, indeed, whether polymerization takes place at all) is critically dependent on the purity of the starting ~ligomer,-~. Impure starting materials either do not polymerize or polymerize rapidly to insoluble, crosslinked products. Multiple recrystallizations from aliphatic hydrocarbon solvents, followed by several vacuum sublimations, are needed before reproducible polymerizations can be achieved. Reports exist5z6 that exhaustive purification yields oligomers that do not polymerize. This is consistent with the finding5 that traces of water exert a catalytic or cocatalytic effect, but that larger quantities cause crosslinking. Phosphorus pentachloride (a possible impurity left over from the synthesis of (NPCl,)30r4from PCl, and NH,C1) is a powerful inhibitor of the polymerization’. A number of deliberately added impurities, such as sulfur or metals, accelerate the polymerization, but at the expense of increasing the rapidity of crosslinking. Avoidance of the crosslinking reaction is perhaps the most important practical requirement in this polymerization. The key to achieving this end is to terminate the polymerization reaction (by cooling to RT) before 100% of the oligomer has been converted to high The higher the degree of conversion, the greater is the chance that crosslinking will occur. Experience has shown that crosslinking takes place rapidly after about 70% of the trimer has been converted to high polymer3n4. Significantly perhaps, polymer isolated from low-yield polymerizations (< 20 % conversion to polymer) appears to be essentially linear’, whereas the polymer formed at higher conversions may be highly branched. Although the polymer formed in this reaction has a

153

15.2. Ring-Ring and Ring-Polymer Interconversions 15.2.11. in Phosphorus-Nitrogen Oligomers and Polymers 15.2.11.1. Halocyclophosphazene Polymerization

very broad molecular weight distribution (often with a bimodal distribution), no cyclic oligomers other than the starting material can be detected. Thus, if the cyclic trimer, (NPCl,),, is polymerized at 250°C, no (NPCI,),,,,,,, 7e,c, can be detected after the polymerization is complete. This suggests either that the polymerization does not involve random ring-ring and ring-chain equilibration of the type found in organocyclosiloxanes, for example, or that the higher oligomers, once formed, are polymerized more rapidly than the cyclic trimer. The answer to this question is not yet known. From a practical point of view, however the final reaction mixture from polymerization of (NPCI,), contains only unchanged (NPCI,), ,uncrosslinked (NPCI,), high polymer that is soluble in benzene or T H F and (under specific circumstances) insoluble, crosslinked (NPCl,), high polymer. The conversion of molten (NPCl,), to (NPCI,), is accompanied by a marked increase in viscosity. Unreacted trimer can be separated from the high polymer either by vacuum sublimation or by extraction-reprecipitation techniques using heptane as a solvent for the trimer but a nonsolvent for the polymer3s4. Depolymerization of (NPCI,), takes place if the polymer is heated above 350°C8~10. The distribution of products depends on the reaction conditions. If equilibration is carried out in a completely sealed system, decreases in pressure favor depolymerization to cyclic oligomers. Under nonequilibrium conditions (continuous removal of oligomers as they are formed), depolymerization takes place at 260-400°C. The oligomer formed in the largest amounts under these conditions is the cyclic hexamer’, (NPCI,), . (iii) Mechanism of Chiorophosphazene Equilibration. The mechanism ofpolymerization and depolymerization is still imperfectly understood. Because replacement of all the chlorine atoms in (NPCI,), by organic groups usually destroys the propensity for polymerization, it appears that halogen atoms participate in the mechanism. A plausible mechanism is one that requires a thermal ionization of chloride ions from phosphorus at elevated temperatures. However, as indicated below, the ring-ring equilibration reactions of organocyclophosphazenes may be more compatible with mechanisms that involve heterocyclic skeletal cleavage to zwitterions or even deoligomerization to a transient monomer. The halide ionization mechanism is outlined in Eq. (b)5,11:

-,

etc.

(b)

c1-P0 -c1 c1 Once the ionization process has occurred, the phosphazenium cation can initiate a cationic propagation process. This mechanism also explains the ease of chain branching,

154

15.2. Ring-Ring and Ring-Polymer Interconversions 15.2.11. in Phosphorus-Nitrogen Oligomers and Polymers 15.2.11.2. Organocyclophosphazenes

since PC1, middle units are, in principle, as prone to ionization as the propagating end units. Depolymerization of the high polymer to cyclic oligomers can occur by back-biting reactions in which active cationic end units attack middle units of the same chain. (iv) Fluoro- and Bromophosphazene Systems. Other halocyclophosphazenes also polymerize at elevated temperatures. Hexafluorocyclotriphosphazene, (NPF,),, polymerizes to (NPF,), at 350°C'2-'5. As in the case of (NPCI,),, the polymerization is a two-step process. The first product is an uncrosslinked, open-chain polymer that is soluble in fluorocarbon media. This polymer crosslinks when held at 350°C for an appreciable length of time. Thus, once again, the preparation of (NPF,), as a reactive polymeric intermediate requires that the polymerization reaction be terminated before the crosslinking process occurs. This requires a subtle control of the reaction time and trimer purity. Like polydichlorophosphazene, polydifluorophosphazene is hydrolytically unstable in contact with a moist atmosphere. Hexabromocyclotriphosphazene, (NPBr,), ,polymerizes when heated at 220"C, but only the crosslinked form of this polymer has been isolatedi6. (v) Solid-state Polymerization. Although the melt polymerization of halophosphazene oligomers is the principle method of polymer formation, polymerization also occurs when the solid crystalline oligomers are irradiated with high-energy radiation (e.g., gamma rays) at moderate temperatures". The yields of these polymerizations are not high ( N 10 %). The rates of polymerization increase with temperature up to the melting point of the oligomer, above which the rate falls to zero. The mechanism of the radiationinduced synthesis is not known. However, the fact that it is clearly a solid-state phenomenon suggests either that the active catalytic moieties are stabilized by site isolation within the crystalline lattice, or that the crystal structure provides a particularly favorable molecular packing arrangement to facilitate a combined ring opening and intermolecular coupling process. (H.R. ALLCOCK)

1. H. N. Stokes, Am. Chem. J., 19, 782 (1897). 2. K. H. Meyer, W. Lotmar, G. W. Pankow, Helv. Chim. Acta, 19, 93Q (1936). 3. H. R. Allcock, R. L. Kugel, J. Am. Chem. SOC.,87, 4216 (1965). 4. H. R. Allcock, R. L. Kugel, K. J. Valan, Inorg. Chem., 5, 1709 (1966). 5. H. R. Allcock, J. E. Gardner, K. M. Smeltz, Macromolecules, 8, 36 (1975). 6. R. 0. Colclough, G. Gee, J. Polym. Sci., Part C, 16, 3639 (1968). 7. H. R. Allcock, R. A. Arcus, E. G. Stroh, Macromolecules, 13, 919 (1980). 8. 0. Schmitz-DuMont, Angew. Chem., 52,498 (1939). 9. H. R. Allcock, W. J. Cook, Macromolecules, 7, 284 (1974). 10. J. R. Soulen, M. S. Silverman, J. Polym. SCL,A I , 823 (1963). 11. H. R. Allcock, R. J. Best, Can. J . Chem., 42, 447 (1964). 12. F. Seel, J. Langer, Angew. Chem., 68, 461 (1956). 13. F. Seel, J. Langer, 2. Anorg. Allg. Chem., 295, 316 (1958). 14. H. R. Allcock, R. L. Kugel, E. G. Stroh, Inorg. Chem., 11, 1120 (1972). 15. H. R. Allcock, D. B. Patterson, T. L. Evans, Macromolecules, 12, 172 (1979). 16. R. L. Kugel, H. R. Allcock, unpublished work (1969). 17. V. Caglioti, D. Cordischi, A. Mele, Nature (London), 195, 491 (1962).

15.2.1 1.2. Organocyclophosphazenes: Ring-Ring and Ring-Polymer Interconversions (i) Reasons for Interest. Halocyclophosphazenes polymerize to open-chain macromolecules when heated in the melt. It has long been recognized that the hydrolytic

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc.

154

15.2. Ring-Ring and Ring-Polymer Interconversions 15.2.11. in Phosphorus-Nitrogen Oligomers and Polymers 15.2.11.2. Organocyclophosphazenes

since PC1, middle units are, in principle, as prone to ionization as the propagating end units. Depolymerization of the high polymer to cyclic oligomers can occur by back-biting reactions in which active cationic end units attack middle units of the same chain. (iv) Fluoro- and Bromophosphazene Systems. Other halocyclophosphazenes also polymerize at elevated temperatures. Hexafluorocyclotriphosphazene, (NPF,),, polymerizes to (NPF,), at 350°C'2-'5. As in the case of (NPCI,),, the polymerization is a two-step process. The first product is an uncrosslinked, open-chain polymer that is soluble in fluorocarbon media. This polymer crosslinks when held at 350°C for an appreciable length of time. Thus, once again, the preparation of (NPF,), as a reactive polymeric intermediate requires that the polymerization reaction be terminated before the crosslinking process occurs. This requires a subtle control of the reaction time and trimer purity. Like polydichlorophosphazene, polydifluorophosphazene is hydrolytically unstable in contact with a moist atmosphere. Hexabromocyclotriphosphazene, (NPBr,), ,polymerizes when heated at 220"C, but only the crosslinked form of this polymer has been isolatedi6. (v) Solid-state Polymerization. Although the melt polymerization of halophosphazene oligomers is the principle method of polymer formation, polymerization also occurs when the solid crystalline oligomers are irradiated with high-energy radiation (e.g., gamma rays) at moderate temperatures". The yields of these polymerizations are not high ( N 10 %). The rates of polymerization increase with temperature up to the melting point of the oligomer, above which the rate falls to zero. The mechanism of the radiationinduced synthesis is not known. However, the fact that it is clearly a solid-state phenomenon suggests either that the active catalytic moieties are stabilized by site isolation within the crystalline lattice, or that the crystal structure provides a particularly favorable molecular packing arrangement to facilitate a combined ring opening and intermolecular coupling process. (H.R. ALLCOCK)

1. H. N. Stokes, Am. Chem. J., 19, 782 (1897). 2. K. H. Meyer, W. Lotmar, G. W. Pankow, Helv. Chim. Acta, 19, 93Q (1936). 3. H. R. Allcock, R. L. Kugel, J. Am. Chem. SOC.,87, 4216 (1965). 4. H. R. Allcock, R. L. Kugel, K. J. Valan, Inorg. Chem., 5, 1709 (1966). 5. H. R. Allcock, J. E. Gardner, K. M. Smeltz, Macromolecules, 8, 36 (1975). 6. R. 0. Colclough, G. Gee, J. Polym. Sci., Part C, 16, 3639 (1968). 7. H. R. Allcock, R. A. Arcus, E. G. Stroh, Macromolecules, 13, 919 (1980). 8. 0. Schmitz-DuMont, Angew. Chem., 52,498 (1939). 9. H. R. Allcock, W. J. Cook, Macromolecules, 7, 284 (1974). 10. J. R. Soulen, M. S. Silverman, J. Polym. SCL,A I , 823 (1963). 11. H. R. Allcock, R. J. Best, Can. J . Chem., 42, 447 (1964). 12. F. Seel, J. Langer, Angew. Chem., 68, 461 (1956). 13. F. Seel, J. Langer, 2. Anorg. Allg. Chem., 295, 316 (1958). 14. H. R. Allcock, R. L. Kugel, E. G. Stroh, Inorg. Chem., 11, 1120 (1972). 15. H. R. Allcock, D. B. Patterson, T. L. Evans, Macromolecules, 12, 172 (1979). 16. R. L. Kugel, H. R. Allcock, unpublished work (1969). 17. V. Caglioti, D. Cordischi, A. Mele, Nature (London), 195, 491 (1962).

15.2.1 1.2. Organocyclophosphazenes: Ring-Ring and Ring-Polymer Interconversions (i) Reasons for Interest. Halocyclophosphazenes polymerize to open-chain macromolecules when heated in the melt. It has long been recognized that the hydrolytic

15.2. Ring-Ring and Ring-Polymer Interconversions 15.2.11. in Phosphorus-Nitrogen Oligomers and Polymers 15.2.11.2. Organocyclophosphazenes

155

instability of halogenophosphazene high polymers could be overcome if organic residues could be introduced as side groups into the polymer in place of halogen atoms. In principle this could be accomplished in one of two ways. First, the halogenophosphazene high polymers could be used as substrates for substitution reactions. This approach is discussed in 815.2.11.3. Likewise, attempts could be made to polymerize organocyclophosphazenes to yield organophosphazene polymers by ring-opening processes analogous to those found for halogenophosphazene systems. A wide variety of organophosphazene cyclic oligomers are known’-3 ($4.2.2.1) and polymerization routes based on these compounds are appealing. Unfortunately, it has been found that replacement of all the halogen atoms in cyclic trimeric and tetrametric phosphazenes by organic groups generally destroys their propensity for polymerization, although not for ring-ring equilibration. This is of considerable interest with respect to the mechanism and thermodynamics of ring-polymer and ring-ring equilibrations. (ii) Methylcyclophosphazenes. The methylcyclophosphazene trimer (I) and tetramer (11) retain their structural integrity at temperatures below 200°C:

CH,

CH,

N

N

I I CH,-P=N-P-CH, I I

CH,

‘P’

CH,

CH, I

CH,

11

However, when pure [NP(CH,),], or [NP(CH,),], is heated above 200°C, equilibration occurs to yield mixtures of the two4. Less than 1% of higher cyclic oligomers are generated in this process and no high polymer is formed. As the temperature is raised from 200°C to 35OoC, the proportion of cyclic trimer in the equilibrate increases markedly. For the reaction:

3 “P(CH,),I4

e

4 “P(CHd213

(c)

AH = + 10.2 kcal and AS = + 14.3 eu. Thus, the dimer is destabilized by about 1 kcal per monomer unit, perhaps by ring strain. This equilibration is retarded by added base but is accelerated by protonic acids. Acids have an additional influence. They bring about cleavage of the rings to short, linear chains, such as HN = P(CH,),fN = P(CH,),+,X4. These species cyclize to form mixtures of [NP(CH,),], and [NP(CH,),], on pyrolysis. A similar cyclization process seems to be operative when (CH,),P(NH,)Cl is converted It should first to linear oligomers at 200°C and then to the cyclic methylpho~phazenes~~~. be noted that the stability of the cyclic trimer and tetramer to random fragmentation is quite high. At 350°C the irreversible decomposition to P(CH,),, HP(CH,),, (CH,),P-PHCH, and (CH,),P-P(CH,), is less than 2 % per week4. However, the presence of only one, two, or three methyl groups per trimer ring in 111, IV, and V does not inhibit polymerization6s7. These species polymerize to open-chain, high molecular weight macromolecules when heated at 250°C. (iii) Phenylcyclophosphazenes. Phenyl groups attached to a cyclotriphosphazene ring also inhibit ring-opening polymerization to open-chain macromolecules’. The effect is more marked than with methyl groups. The presence of one phenyl group per ring, as

15.2. Ring-Ring and Ring-Polymer lnterconversions 15.2.11. in Phosphorus-Nitrogen Oligomers and Polymers 15.2.11.2. Organocyclophosphazenes

156

CH3\p/cI

N4

c1

CH,\ 7 P / \ N N

\N

I11

3

IV

CH,

C1 \P/ N 4 \N c1\ I I1 /CH, P P / \N/ CH, V

in VI, has very little inhibiting effect, but the presence of two to six phenyl groups totally inhibits homopolymerization. Species such as VII or IX retain the ability to copolymerize with (NPCI,), . F\p/Ph N// \

Ph \P/ N//

N

Ph \

N

Ph, N//

P

,Ph \

N

N VI

VII

VIII

IX

X

XI

Ph, ,Ph P N4\ N

Ph\ I P Ph \N/

/I /Ph

P

\Ph

XI1 (iv) Fluoroalkoxycyclophosphazenes. The attachment of trifluoroethoxy substi-

tuents to a cyclophosphazene ring gives rise to a more complex ring-ring and ring-polymer equilibration patterng. If one to three trifluoroethoxy groups are attached to a cyclotriphosphazene ring in a manner that leaves at least one chlorine atom attached to each phosphorus atom, as in XIII-XV: c1 OCH,CF, C1, ,OCH,CF, \P/ P N// \ N N @ \N

c1\ I P c1/ \”

c1 I

/c1 \cl

XI11

C1, N//

CF,CH 0

\P

c1/

I

P

,OCH,CF,

‘N

j /c1

N/ XV

\OCH,CF,

\P CI/ \ N / XTV

j /c1

\OCH,CF,

15.2. Ring-Ring and Ring-Polymer Interconversions 15.2.11. in Phosphorus-Nitrogen Oligomers and Polymers 15.2.11.2. Organocyclophosphazenes

157

polymerization occurs at 2OO"C-225"C to yield open-chain high polymers. However, the presence of 4-,5- or 6-trifluoroethoxy groups, as in XVIII-XX, inhibits the polymerization process: CF,CH,O\p/OCH,CF, N//

\

N CF,CH 0 I II /c1 \P P C1/ \ N / \OCH,CF, XVI

CF,CH,O\p/OCH,CF, N// \ N CF,CH 0 I I1 /OCH,CF, \P P C f N' 'OCH,CF, XVII

CF,CH,O\p/OCH,CF, N 4 \N CF,CH 0 I pcH,cF, \P \OCH,CF, CF,CH,O / \ N / XVIII However, species XVI-XVIII undergo ring expansion reactions at 200-250°C. Such reactions are accompanied by ring-ring or ring-chain coupling reactions brought about by the elimination of CF,CH,Cl whenever both P-CI and P-0-CH,CF, side groups are present in the same reaction systemg. (v) Carborane-Linked Cyclophosphazenes. In view of the role played by bulky side groups in lowering the tendency for polymerization of cyclotriphosphazenes, it is of some interest that a carborane unit attached to the phosphazene ring does not inhibit polymerization". Species XIX and XX undergo a phosphazene ring-opening polymerization at 250°C:

to yield open-chain polymers of type XXI: carborane

c1

1

XXI These, like all the macromolecules prepared from halogenoorganocyclophosphazenes, can be further modified by substitutive techniques. (vi) Copolymerizations. Although many organocyclophosphazenes do not polymerize when heated, they may copolymerize with halogenocyclophosphazenes such as

158

15.2. Ring-Ring and Ring-Polymer lnterconversions 15.2.11. in Phosphorus-Nitrogen Oligomers and Polymers 15.2.11.2. Organocyclophosphazenes

(NPF,), ,(NPCl,),, or (NPBr,), . The underlying idea is that the hexahalogenophosphazene may provide an initiation mechanism and a chain propagation process that can incorporate the organophosphazene into the polymer. Copolymerization reactions take place at 250-300°C between (NPCI,), and N,P,F,Ph (VIII), N,P,F,Ph, (IX), N3P,Cl,Ph, (X), or N,P,Cl,Ph, (XI) but not between (NPCI,), and N,P,Cl,Ph, (XII), N,P,ClPh, (XII) or (NPPh,), (XIV)'. In fact, XII, XI11 and XIV inhibit the polymerization of (NPCI,), . However, if trifluoroethoxy groups are present instead of phenyl groups, more extensive copolymerization processes are possibleg. For example, [NP(OCH,CF,),], copolymerizes with (NPCl,), at 200"C, as do mixed trifluoroethoxychloro-substituted trimers such as XV-XIX. These results are consistent with the idea that P-C1 units must be present before polymerization occurs, and that nonionizable, bulky organic residues such as phenyl groups interfere with both the initiation and the chain propagation processes. (vii) Depolymerization of Polyorganophosphazenes. The polymerization or copolymerization of a cyclic phosphazene is only one aspect of the ring-chain equilibration process. The other aspect is the depolymerization of preformed high polymers to cyclic oligomers. A large number of high molecular weight polyorganophosphazenes have been prepared by the substitutive route and these materials provide substrates for depolymerization experiments. As discussed earlier, polydichlorophosphazene depolymerizes to a spectrum of cyclic oligomers when heated at 250-350°C 11-13. In a helium flow-tube apparatus at 400"C, species (NPCl,),-, are identified, with the cyclic hexamer predominating". It should be noted that these results are for nonequilibrium reaction conditions under which all or most of the polymer is converted to oligomers. In similar nonequilibrium experiments, it has been shown that polybis(trifluoroeth0xy)phosphazene (XXII) depolymerizes at temperatures above 150°C to yield [NP(OCH,CF,),],,,,, , higher cyclic oligomers and medium molecular weight openchain polymers": OCH,CF,

OCH,CF,

XXII Acids accelerate that rate of depolymerization. Some evidence exists that the depolymerization is a two-step process, with random chain cleavage to yield medium chain length polymers preceding unzipping-t ype depolymerization to cyclic oligomers. Polydiphenoxyphosphazene, [NP(OC,H,),],, behaves similarly at temperatures above 100°C It should be noted that these temperatures provide an unduly pessimistic description of the thermal stability of polyphosphazenes. As in most technological macromolecular systems, techniques are available to retard depolymerization processes. These depolymerization reactions are complicated by side reactions. For example, if both P-Cl and P-OCH,CF, or P-OC,H, units are present in the same reaction system, thermolysis causes elimination of CF,CH,Cl or C,H,Cl with simultaneous crosslinking of the chainslO.Clearly, depolymerization and elimination reactions are important in the context of the practical thermal stability of polyphosphazenes. Thus,

15.2. Ring-Ring and Ring-Polymer lnterconversions 15.2.11. in Phosphorus-Nitrogen Oligomers and Polymers 15.2.11.3. Mechanisms and Thermodynamics

159

considerable interest is focused on the reaction mechanisms involved and on the thermodynamic features. (H R. ALLCOCK)

1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14.

H. R. Allcock, Phosphorus-Nitrogen Compounds,Academic Press, New York, 1972. H. R. Allcock, Acc. Chem. Res., 12, 351 (1979). H. R. Allcock, Chem. Rev., 72, 315 (1972). H. R. Allcock, D. B. Patterson, Inorg. Chem., 16, 197 (1977). H. H. Sisler, S. E. Frazier, R. G. Rice, M. G. Sanchez, Inorg. Chew., 5, 326 (1966). H. R. Allcock, R. J. Ritchie, P. J. Harris, Macromolecules, 13, 1332 (1980). V. N. Prons, M. P. Grinblat,A. L. Klebanskii, Zh. Obshch. Khim., 41,482 (1971); J. Gen. Chem. USSR (Engl. Transl.), 41, 475 (1971).

H. R. Allcock, G . Y. Moore, Macromolecules, 8, 377 (1975). H. R. Allcock, J. L. Schmutz, K. M. Kosydar, Macromolecules, lI, 179 (1978). A. G. Scopelianos, J. P. OBrien, H. R. Allcock, J. Chem. Soc., Chzm. Commun., 198 (1980). 0. Schmitz-DuMont, Angew. Chem., 52,498 (1939). H. R. Allcock, W. J. Cook, Macromolecules, 7, 284 (1974). J. R. Soulen, M. S. Silverman, J. Polym. Sci., Al, 823 (1963). H. R. Allcock, G. Y. Moore, W. J. Cook, Macromolecules, 7, 571 (1974).

15.2.11.3. Mechanisms and Thermodynamics of Phosphazene Equilibrations

The equilibration reactions of phosphazene oligomers and high polymers are of interest from two points of view; first, because of their importance in macromolecular syntheses, and second because of the practical need to improve the higher temperature stability of polyphosphazenes. (i) Mechanisms. Many questions remain about the mechanisms of polymerization, ring-ring equlilibration and depolymerization of phosphazenes. Still uncertain is whether all three processes follow the same mechanism. Four mechanisms appear possible', and these are illustrated in Scheme I. They involve (A) an ionization of a side group (especially halogen) from phosphorus, followed by a cationic chain propagation process; (B) hydrolysis of a side group by traces of catalytic water, followed by skeletal cleavage; (C) cleavage of the skeleton by traces of catalytic acid; and (D) a thermal cleavage of the ring or chain to yield open-chain zwitterions or monomer. The side-group ionization mechanism mentioned earlier for the polymerization of (NPCI,), applies equally well to the polymerization, ring-ring equilibration or depolymerization of any cyclic or linear phosphazene that contains P-Cl bonds. This mechanism may also explain the higher temperatures needed for polymerization along the series (NPBr,), , (NPCl,), , (NPF,), , since the phosphorus-halogen bond strengths increase in the same order. It is also significant that the conductivity of molten (NPCI,), is low until the temperature is increased into the range where polymerization takes place2. At this point, a dramatic rise in conductivity occurs. If this mechanism is correct, it implies that initiation of polymerization is a slower step than chain propagation, since high polymer and unchanged trimer are the only products isolated from the polymerization of (NPCI,), . The reluctance of some organohalogenocyclophosphazenes to polymerize may be attributed to an inefficient chain propagation process owing to steric hindrance. However, the ring-ring equilibrations of [NP(CH,)J,, [NP(C,H,),], and [NP(OCH,CF,),], cannot be explained by this mechanism.

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc. 15.2. Ring-Ring and Ring-Polymer lnterconversions 15.2.11. in Phosphorus-Nitrogen Oligomers and Polymers 15.2.11.3. Mechanisms and Thermodynamics

159

considerable interest is focused on the reaction mechanisms involved and on the thermodynamic features. (H R. ALLCOCK)

1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14.

H. R. Allcock, Phosphorus-Nitrogen Compounds,Academic Press, New York, 1972. H. R. Allcock, Acc. Chem. Res., 12, 351 (1979). H. R. Allcock, Chem. Rev., 72, 315 (1972). H. R. Allcock, D. B. Patterson, Inorg. Chem., 16, 197 (1977). H. H. Sisler, S. E. Frazier, R. G. Rice, M. G. Sanchez, Inorg. Chew., 5, 326 (1966). H. R. Allcock, R. J. Ritchie, P. J. Harris, Macromolecules, 13, 1332 (1980). V. N. Prons, M. P. Grinblat,A. L. Klebanskii, Zh. Obshch. Khim., 41,482 (1971); J. Gen. Chem. USSR (Engl. Transl.), 41, 475 (1971).

H. R. Allcock, G . Y. Moore, Macromolecules, 8, 377 (1975). H. R. Allcock, J. L. Schmutz, K. M. Kosydar, Macromolecules, lI, 179 (1978). A. G. Scopelianos, J. P. OBrien, H. R. Allcock, J. Chem. Soc., Chzm. Commun., 198 (1980). 0. Schmitz-DuMont, Angew. Chem., 52,498 (1939). H. R. Allcock, W. J. Cook, Macromolecules, 7, 284 (1974). J. R. Soulen, M. S. Silverman, J. Polym. Sci., Al, 823 (1963). H. R. Allcock, G. Y. Moore, W. J. Cook, Macromolecules, 7, 571 (1974).

15.2.11.3. Mechanisms and Thermodynamics of Phosphazene Equilibrations

The equilibration reactions of phosphazene oligomers and high polymers are of interest from two points of view; first, because of their importance in macromolecular syntheses, and second because of the practical need to improve the higher temperature stability of polyphosphazenes. (i) Mechanisms. Many questions remain about the mechanisms of polymerization, ring-ring equlilibration and depolymerization of phosphazenes. Still uncertain is whether all three processes follow the same mechanism. Four mechanisms appear possible', and these are illustrated in Scheme I. They involve (A) an ionization of a side group (especially halogen) from phosphorus, followed by a cationic chain propagation process; (B) hydrolysis of a side group by traces of catalytic water, followed by skeletal cleavage; (C) cleavage of the skeleton by traces of catalytic acid; and (D) a thermal cleavage of the ring or chain to yield open-chain zwitterions or monomer. The side-group ionization mechanism mentioned earlier for the polymerization of (NPCI,), applies equally well to the polymerization, ring-ring equilibration or depolymerization of any cyclic or linear phosphazene that contains P-Cl bonds. This mechanism may also explain the higher temperatures needed for polymerization along the series (NPBr,), , (NPCl,), , (NPF,), , since the phosphorus-halogen bond strengths increase in the same order. It is also significant that the conductivity of molten (NPCI,), is low until the temperature is increased into the range where polymerization takes place2. At this point, a dramatic rise in conductivity occurs. If this mechanism is correct, it implies that initiation of polymerization is a slower step than chain propagation, since high polymer and unchanged trimer are the only products isolated from the polymerization of (NPCI,), . The reluctance of some organohalogenocyclophosphazenes to polymerize may be attributed to an inefficient chain propagation process owing to steric hindrance. However, the ring-ring equilibrations of [NP(CH,)J,, [NP(C,H,),], and [NP(OCH,CF,),], cannot be explained by this mechanism.

160

15.2. Ring-Ring and Ring-Polymer Interconversions 15.2.11. in Phosphorus-Nitrogen Oligomers and Polymers 15.2.11.3. Mechanisms and Thermodynamics

r

I

X

C

I

I

/

t

x-L-x

It

z x-a-x

II I

z

x--a-x II

z

t

I

? X

Ill

z

5

”\ /x

Z=a

15.2. Ring-Ring and Ring-Polymer Interconversions 15.2.11. in Phosphorus-Nitrogen Oligomers and Polymers 15.2.11.4. Polyorganophosphazenes via Reactions.

161

Evidence for mechanism (B) is provided by the known catalytic activity of traces of water, and the known reactivity of (NPCI,), with water. However, the ring-ring equilibration of hexaorganocyclotriphosphazenes is again incompatible with this mechanism. Mechanism (C) may well operate during the depolymerization of phosphazene high polymers, but evidence has not yet been obtained that skeletal cleavage reagents accelerate polymerization. However, as described earlier, acids do accelerate the rate of ring-ring equilibration in [NP(CH,),], Mechanism (D) is perhaps the most intriguing possibility, but no firm evidence for it has yet been obtained. (ii) Thermodynamic Effects. An alternative approach to these problems is to view them in terms of thermodynamic effects. A promising theory3 is based on the idea that the relative thermodynamic stabilities of small rings and long chains are strongly influenced by the steric size of the side groups. Examination of molecular models and the results of intramolecular nonbonding potential energy calculation^^.^ suggest that open-chain polyphosphazenes are much more sensitive to intramolecular side-group steric hindrance that are the analogous cyclic trimers or tetramers. Hence, cyclophosphazenes with small side groups (F, C1, Br) should form linear high polymers without an increase in enthalpy, whereas those oligomers with large side groups (OCH,CF,, C6H,, OC,H,, etc.) should not. Conversely, high polymers with bulky substituent groups should be more prone to depolymerize to cyclic oligomers than those with small side groups. In general, these predictions are borne out in practice. However, the behavior of methylphosphazene oligomers is anomalous. They undergo ring-ring equilibration but do not polymerize6, in spite of the fact that a methyl group has approximately the same dimensions as a bromine atom. However, polydimethylphosphazene can be prepared by a condensation polymerization route7s8, and this suggests that the cyclic trimer and tetramer are energy traps. Once chain growth has proceeded beyond the stage of a linear trimer or tetramer, chain propagation may perhaps be facile. (H.R ALLCOCK)

1. 2. 3. 4. 5. 6. 7. 8.

H. R. Allcock, Polymer, 21, 673 (1980). H. R. Allcock, R. J. Best, Can. J. Chem., 42, 447 (1964). H. R. Allcock, J. Macromol. Sci., Chem., C4, 149 (1970). H. R. Allcock, R. W. Allen, J. J. Meister, Macromolecules, 9, 950 1976). R. W. Allen, H. R. Allcock, Macromolecules, 9, 956 (1976). H. R. Allcock, D. B. Patterson, Inorg. Chem., 16, 197 (1977). P. Wisian-Neilson, R. H. Neilson, J. Am. Chem. Soc., 102, 2848 ( 980). R. H. Neilson, P. Wision-Neilson, Chem. Rev., 88, 541 (1988).

15.2.11.4. Polyorganophosphazenes via the Reactions of Polydihalophosphazenes. (i) General Principles Although high molecular weight polyorganophosphazenes can be prepared by the polymerization of organohalocyclophosphazenes ($15.2.1 1.2) or by condensation polymerization methods', by far the largest number of polymers have been synthesized by substitutive techniques. In these syntheses a reactive, preformed polyphosphazene is used as a substrate for substitution by organic or organometallic reagents. Three general principles underly the success of this synthesis approach. In the first, techniques exist for the facile preparation of reactive high polymers, such as (NPCI,), or

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc. 15.2. Ring-Ring and Ring-Polymer Interconversions 15.2.11. in Phosphorus-Nitrogen Oligomers and Polymers 15.2.11.4. Polyorganophosphazenes via Reactions.

161

Evidence for mechanism (B) is provided by the known catalytic activity of traces of water, and the known reactivity of (NPCI,), with water. However, the ring-ring equilibration of hexaorganocyclotriphosphazenes is again incompatible with this mechanism. Mechanism (C) may well operate during the depolymerization of phosphazene high polymers, but evidence has not yet been obtained that skeletal cleavage reagents accelerate polymerization. However, as described earlier, acids do accelerate the rate of ring-ring equilibration in [NP(CH,),], Mechanism (D) is perhaps the most intriguing possibility, but no firm evidence for it has yet been obtained. (ii) Thermodynamic Effects. An alternative approach to these problems is to view them in terms of thermodynamic effects. A promising theory3 is based on the idea that the relative thermodynamic stabilities of small rings and long chains are strongly influenced by the steric size of the side groups. Examination of molecular models and the results of intramolecular nonbonding potential energy calculation^^.^ suggest that open-chain polyphosphazenes are much more sensitive to intramolecular side-group steric hindrance that are the analogous cyclic trimers or tetramers. Hence, cyclophosphazenes with small side groups (F, C1, Br) should form linear high polymers without an increase in enthalpy, whereas those oligomers with large side groups (OCH,CF,, C6H,, OC,H,, etc.) should not. Conversely, high polymers with bulky substituent groups should be more prone to depolymerize to cyclic oligomers than those with small side groups. In general, these predictions are borne out in practice. However, the behavior of methylphosphazene oligomers is anomalous. They undergo ring-ring equilibration but do not polymerize6, in spite of the fact that a methyl group has approximately the same dimensions as a bromine atom. However, polydimethylphosphazene can be prepared by a condensation polymerization route7s8, and this suggests that the cyclic trimer and tetramer are energy traps. Once chain growth has proceeded beyond the stage of a linear trimer or tetramer, chain propagation may perhaps be facile. (H.R ALLCOCK)

1. 2. 3. 4. 5. 6. 7. 8.

H. R. Allcock, Polymer, 21, 673 (1980). H. R. Allcock, R. J. Best, Can. J. Chem., 42, 447 (1964). H. R. Allcock, J. Macromol. Sci., Chem., C4, 149 (1970). H. R. Allcock, R. W. Allen, J. J. Meister, Macromolecules, 9, 950 1976). R. W. Allen, H. R. Allcock, Macromolecules, 9, 956 (1976). H. R. Allcock, D. B. Patterson, Inorg. Chem., 16, 197 (1977). P. Wisian-Neilson, R. H. Neilson, J. Am. Chem. Soc., 102, 2848 ( 980). R. H. Neilson, P. Wision-Neilson, Chem. Rev., 88, 541 (1988).

15.2.11.4. Polyorganophosphazenes via the Reactions of Polydihalophosphazenes. (i) General Principles Although high molecular weight polyorganophosphazenes can be prepared by the polymerization of organohalocyclophosphazenes ($15.2.1 1.2) or by condensation polymerization methods', by far the largest number of polymers have been synthesized by substitutive techniques. In these syntheses a reactive, preformed polyphosphazene is used as a substrate for substitution by organic or organometallic reagents. Three general principles underly the success of this synthesis approach. In the first, techniques exist for the facile preparation of reactive high polymers, such as (NPCI,), or

162

15.2. Ring-Ring and Ring-Polymer lnterconversions 15.2.11. in Phosphorus-Nitrogen Oligomers and Polymers 15.2.11.4. Polyorganophosphazenes via Reactions.

(NPF,),. These species are so reactive to nucleophilic replacement of halogen that the usual restrictions to macromolecular substitution, found in many organic polymers, do not apply. Second, the nucleophilic substitution reactions of polydihalophosphazenes are generally clean in the sense that they are free from side reactions that lead to crosslinking or chain cleavage. And third, advances in this field have been facilitated by the ready availability of cyclic oligomeric analogs, such as (NPC12)30r4.These can be used as models for the exploration of new substitution reactions that are difficult to pioneer initially with the high polymeric systems'. Once new reactions have been studied in detail at the cyclic oligomeric level, extrapolation to the high polymeric systems is feasible. However, experience has shown that attempts to develop new reactions with the polymers without prior experience with the oligomers invariably leads to complications. ( i i ) General Synthesis Route. The general synthesis routes used for the preparation of most polyorganophosphazenes are summarized in Scheme 13,4,5-7. The isolation of soluble (i.e., uncrosslinked) polydichlorophosphazene is a key step in this reaction394. The insoluble, crosslinked end product, even when swelled in suitable solvents, does not allow penetration of the reagent into the inner recesses of the polymer matrix. Hence, complete substitution is difficult or impossible to achieve with the crosslinked form. In general, with sodium alkoxides or aryloxides, or primary or secondary amines used as nucleophiles, all the halogen atoms can be r e p l a ~ e d ~Unlike - ~ . the polydichlorophosphazene precursor, most organophosphazene polymers are stable to water. The physical and chemical properties of each individual polymer depend on the structure of the side groups. Hence, different polymers may be elastomers, fiber-forming materials, films or glasses. They may be soluble in water or soluble in organic media. The solid polymers may be hydrophilic or hydrophobic, bioresistant or biodegradable, etc. Given the enormous variety of available nucleophilic reagents, it is obvious that an extremely wide range of different structures can be generated by this route. Add to this the fact that two or more different types of side groups can be attached to the same chain, geminally, nongeminally, cis, trans or randomly, and it is clear that the structural diversity in this polymer system is comparable to or exceeds that found in organic polymer chemistry.

Scheme I

15.2. Ring-Ring and Ring-Polymer lnterconversions 15.2.11. in Phosphorus-Nitrogen Oligomers and Polymers 15.2.11.4. Polyorganophosphazenes via Reactions.

163

(iii) Synthetic Diversity. A few examples illustrate this diversity. Polymers I and I1 are rubbery elastomers that remain flexible down to temperatures near - 80°C. Species I11 is a water-repellant, flexible, film- and fiber-forming polymer with surface properties similar to those of polytetrafluoroethylene. It does not burn. Polymers IV and V are also flexible, film-forming materials, both soluble in high-boiling solvents. Species VI and VII are substrates for substitution reactions carried out on the side groups structures. Polybismethylaminophosphazene (VII) is soluble in water to yield basic solutions. It is a prototype carrier molecule for transition metals and biologically active species. Polymer VIII is a hydrophilic film-forming material with possible biomedical uses. Polymer IX is a nonflexible, glasslike material. Macromolecules of this type can be considered firstgeneration polyphosphosphazenes. Some of them are now used in technology.

OCH,CF,

I

I11

I1

"=ia OI

-

0

IV

VII

V

VIII

VI

IX

(iv) Alternative Substitutive Approaches. First, it is possible to replace organic side groups attached to a polyphosphazene by treatment with a second organic nucleophile. Exchange of fluoroalkoxy groups can be accomplished in this way','. Second, polymers that bear two or more different side groups can be prepared by either the simultaneous or the sequential reaction of (NPCI,), with two different nucleophiles'0-'2. Sequential reactions are possible when the first nucleophile is sufficiently sterically hindered, e.g., (C,H,),NH or C,H,ONa, that replacement of the second chlorine atom at a particular phosphorus atom is slower than replacement of the f i r ~ t ' ~ .This ' ~ . allows the first stage of the reaction to be terminated at a stage that approximates to nongeminal partial substitution. Treatment with a second nucleophile then brings about replacement of the remaining chlorine. Third, polyorganophosphazenes can be prepared by the reactions of organometallic reagents with polydihalophosphazenes. The reactions are complex and they are considered next.

164

15.2. Ring-Ring and Ring-Polymer Interconversions 15.2.11. in Phosphorus-Nitrogen Oligomers and Polymers 15.2.11.4. Polyorganophosphazenes via Reactions.

Figure 1. Fibers of polybistrifluoroethoxyphosphazene,[NP(OCHZCF3)2]n. (v) Organometallic Reactions (General). A number of advantages can be foreseen for the synthesis of polyorganophosphazenes that bear alkyl or aryl side groups attached to the skeleton through phosphorus-carbon bonds. Such polymers would be structural analogs of polyorganosiloxanes. Lacking P-0-C or P-N-C linkages to the side group, they may be expected to be stable at elevated temperatures. A logical route to the synthesis of such polymers is via the reactions of organometallic reagents with polydihalophosphazenes. Thus, alkylmagnesium halide reagents, organolithium reagents, organocopper compounds and lithiocarboranes have been examined as possible nucleophiles. The reactions of alkylmagnesium halide reagents with cyclophosphazenes, such as (NPCI,), o r 4 , are quite complex. Halogen replacement, skeletal cleavage and ring-ring coupling reactions are all possible1 These complexities are amplified with the high polymers and are probably exacerbated by crosslinkage brought about by coordination of the skeletal nitrogen atoms to magnesium salts. Thus, at present, this is not a viable route for polymer synthesis. Organolithium reagents react with cyclic and polymeric halophosphazenes by slightly less complex pathways. However, skeletal cleavage is concurrent with halogen replacement. This is illustrated by the broken curve in Figure 2, which shows the decline in average chain length that occurs when (NPCI,), high polymer reacts with phenyllithi~m~. Phenylchlorophosphazenes are formed, but increasing substitution is accompanied by progressive chain cleavages. By the time that 90-100% of the chlorine atoms have been replaced by phenyl groups, the average chain length corresponds to linear oligomers only. It has been speculated' that chain cleavage in this system is a consequence of the ability of the skeletal nitrogen atoms in the polymer to coordinate to the organolithium species. In theory, this binding should be weakened if more electro-

'.

165

15.2. Ring-Ring and Ring-Polymer Interconversions 15.2.11. in Phosphorus-Nitrogen Oligomers and Polymers 15.2.11.4. Polyorganophosphazenes via Reactions.

25%

50%

75%

100%

Percent Aryl Groups in Polymer

Figure 2. Variation of molecular weight of the products formed by the reactions of (NPCl,),

and (NPF,), with phenyllithium, showing the relationship between halogen substitution and chain cleavage.

negative fluorine atoms are being replaced rather than chlorine. In fact, polydifluorosphazene reacts with phenyllithium in fluorocarbon-THF media, mainly by a substitutive mechanism until roughly 75 % of the fluorine atoms have been replaced by pheny17 (Figure 2). After this point, chain cleavage becomes serious, presumably because the phenyl groups now present have insufficient electron-withdrawing power to prevent coordination and phosphorus-nitrogen bond rupture. It is also possible that steric crowding, resulting from the introduction of bulky aryl groups, induces a thermodynamic destabilization of the high polymer. Partially arylated polymers can, of course, be treated with an alkoxide, aryloxide or amine to generate mixed-substituent polymer:

OCH,CF,

OCH,CF,

Methyllithium reacts with (NPF,), by a complex process that includes substitution, chain cleavage and crosslinking via metallation of methyl groups attached to the skeletonL6.

166

15.2. Ring-Ring and Ring-Polymer Interconversions 15.2.11. in Phosphorus-Nitrogen Oligomers and Polymers 15.2.11.4. Polyorganophosphazenes via Reactions. ~

~

Many of the problems encountered with the reactions of alkyl- or arylorganometallic reagents with halogenophosphazene high polymers can be circumvented by the direct synthesis of alkyl- or arylphosphazenes via condensation'. Methyl groups introduced in this way are susceptible to lithiation, and these sites can then be used for reactions with electrophiles'. Some of the most promising reagents for the synthesis of alkylphosphazenes by the substitutive route are organocopper species derived from an alkylmagnesium halide reagent and [Bu,PCuI], ",". These reagents have not yet been used in high polymer synthesis reactions, but in small-molecule compound systems they allow the replacement of chlorine in (NPCI,), by a wide variety of alkyl groups or by hydrogen or iodine. The principles of this approach are summarized in Eq. (b): CLP/

c1

The introduction of unsaturated alkyl groups into the high polymeric system is of some interest since it would provide sites for crosslinking or for transition metal coordination. Finally, carborane units have been attached to phosphazene high polymers by reactions of lithiocarboranes with (NPCI,),. These reactions are summarized in Eq. (c)'9: R

-

r

{

N=P-

c1

]

[

,-

R--:2;--Li

I

>B'oH'oI -N=P

N=P-

I

c1

n

[

-N-TH2"'] OCH,CF,

c1

1

R

-

I

I I

N=P-

OCH,CF,

y'

15.2. Ring-Ring and Ring-Polymer lnterconversions 15.2.11. in Phosphorus-Nitrogen Oligomers and Polymers 15.2.11.4. Polyorganophosphazenes via Reactions.

167

(vi) Utilization of Polyphosphazenes. Apart from their fundamental scientific importance as new and unusual macromolecules, polyphosphazenes are the subject of considerable interest as useful materials in research and technology. These uses are a consequence of the novel structure of these polymers, which gives rise to physical and chemical characteristics that cannot be found in conventional organic macromolecules. Three areas of application are of special interest

1. the use of polyphosphazenes as technological polymers, often as replacements for organic polymers in high performance, new technology areas 2. utilization of polyphosphazenes as biopolymers in medicine and biochemical research 3. use of phosphazene high polymers as “carriers” for transition-metal compounds, especially those used in catalysis

These three aspects are considered in turn. ( v i i ) Technological Polymers. Most synthetic polymers are used in technology because of their solid-state properties, such as elasticity, strength, fiber-forming properties, which depend on molecular entanglements. Polyorganophosphazenes have a number of attributes that are of interest in technology. Depending on the side group structure, they may be elastomeric, rigid, flexible or fiber-forming, stable to oxidation or to ultraviolet light, resistant to oils and other hydrocarbons, hydrophobic or hydrophilic. Most polyphosphazenes are resistant to burning. These properties can be tailored into the system by the attachment of appropriate side groups to the skeleton. For example, the introduction of one type of side group, such as trifluoroethoxy, yields a polymer in which the solid-state properties are dominated by the presence of microcrystalline domains4. However, the random introduction of two different fluoroalkoxy substituents, such as CF3CHzO- and HCF,(CF,),CH,O--, lowers the symmetry of each macromolecule to the point where ordered packing of chains cannot occur, and the material becomes an elastomer’0-’2. Polymers of this types are now manufactured as nonburning, oil-resistant elastomers that remain elastomeric at low temperature. Such polymers are used in high-technology aerospace and automotive applications. A related use of polymers with mixed aryloxy groups is as nonburning, elastomeric materials. A number of polyalkoxy- and aryloxyphosphazenes are possible bioreconstruction polymers (heart valves, pumps, etc.) because of their inertness in a biological environment. However, the synthetic versatility of the phosphazene system allows a wide range of other biologically interesting polymers to be prepared, polymers that are important for their molecular and chemical behavior rather than their chain entanglement characteristics. These are discussed in the following section. (viii) Biopolymers. A growing interest exists in the synthesis of macromolecules as carrier molecules for biologically active units, especially for chemotherapeutic drugs or naturally occurring bioactive species such as enzymes or metalloporphyrins. The macromolecular carrier serves to immobilize the bioactive agent either within a solid phase or in solution. In the latter situation, dissolved macromolecules cannot migrate through semipermeable membranes. Polyorganophosphazenes are of special value in this respect because of the subtlety with which the side-group structure can be modified, and the various alternative ways that exist for the attachment of bioactive species. Bioactive units can be attached to a polyphosphazene chain by coordination or by covalent linkage. These two methods are considered separately. An attempt has been made to utilize a prototype, water-soluble polyphosphazene as a carrier for Pt antitumor agents”. Such species are synthesized by the interaction of [NP(NHCH,),], with K,[PtC14] in CHC1, in the presence of 18-crown-6 ether.

168

15.2. Ring-Ring and Ring-Polymer lnterconversions 15.2.1 1. in Phosphorus-Nitrogen Oligomers and Polymers 15.2.11.4. Polyorganophosphazenes via Reactions.

Spectroscopic and oligomeric model compound studies suggest that the coordination is through skeletal nitrogen atoms, as shown in Xzl. NHMe -N=P-

I

NHMe

\

/Pt\

c1

I

/

c1

NHMe

X

XI The biological activity of the platinum is apparently retained in the polymer. A modified polybis(methy1amino)phosphazene has also been used as a water-solubilizing carrier for heme and hemin''. Eeaction of (NPCl,), with H,N(CH,),N,C,H,, followed by methylamine, yields XI. The pendent imidazolyl groups constituted 4-5% of the total side groups. Polymer XI coordinates to heme or hemin via the imidazolyl function and the complex has been used as a model for oxidation-oxygenation studies in hemoproteins The linkage of bioactive substituents to polyphosphazenes via covalent bonds has been investigated. Steroid molecules in particular can be attached via the phenolic sites at the 3 positionz3.The overall reaction pathway is shown in Scheme 11. Steroid salts of the type shown in Scheme I1 react with (NPCI,), in a manner similar to simple aryloxides. However, the steric bulk of the steroid unit prevents total replacement of the halogen atoms. From a practical point of view this is no disadvantage because the non-bioactive substituent group should remain in excess because it determines the solubility and other physical properties. In different polymers, 0.5-40 % of the substituent groups are steroidoxy. Linkage via ONa groups at the 17 position of steroids is more difficult, presumably for steric reasons. Steroids with alicyclic A rings cannot be linked through hydroxy groups at the 3 position because (NPCl,), dehydrates the steroid. In any pharmacologically active polymer, it is an advantage if the polymer itself can be designed to biodegrade eventually to harmless small molecules. Very few synthetic organic macromolecules possess this attribute. Hydrolytic biodegradability can be

15.2. Ring-Ring and Ring-Polymer lnterconversions 15.2.11. in Phosphorus-Nitrogen Oligomers and Polymers 15.2.11.4. Polyorganophosphazenes via Reactions.

"=A,-

c1 NaO

169

e, J

n

n

Scheme I1

ensured in polyorganophosphazenes by a choice of suitable substituents or cosubstituents. Two substituent groups are especially effective-amino acid ester unitsz4 and imidazolyl groupsz5linked directly to the skeleton. Two homopolymers based on these structures are illustrated in XI1 and XIII:

XI11

170

15.2. Ring-Ring and Ring-Polymer Interconversions 15.2.11. in Phosphorus-Nitrogen Oligomers and Polymers 15.2.11.4. Polyorganophosphazenes via Reactions.

Polymer XI1 (and model oligomers based on the same repeating structure) hydrolyze to ethanol, amino acid, phosphate and ammonia. Species of type XI1 hydrolyze to imidazole, phosphate and ammonia. The utility of this mechanism for polymers such as those shown in'Scheme I1 is obvious. (ix) Carriers for Transition Metals. Macromolecules are of interest as carriers for transition metals for two reasons: 1. because a polymer can solubilize a catalyst unit, ensure site isolation and facilitate recovery of the catalyst, 2. because polymers may function as one-dimensional templates for the construction of metal-metal-linked, electrically conducting systems.

The synthesis of polyorganophosphazenes with transition metals linked to the side groups has been accomplished with the use of aryloxy-substituted polyphosphazenes. The over-all synthetic pathways are summarized in Scheme IIII4. The key principle involved in this synthetic route is a metal-halogen exchange reaction carried out with a p-halogeno-phenoxy-substituted polyphosphazene. For catalyst-binding purposes, it would be disadvantageous if every side group held a coordination site. Hence, the preferred polymer system is one in which phenoxy andphalogenophenoxy side groups are present in the initial reaction substrate. The ratio of these can be modified by the conditions of substitution used for (NPCI,),. Aryloxy spacer groups are used in order to shield bound transition metals from the coordinating power of the skeletal nitrogen atoms. The metal-halogen exchange reaction must be carried out at low temperatures (- SOT), and the lithiated polymer can then be allowed to react with ClPPh,, ClSnPh,, ClAuPPh, or CO,. The triarylphosphine groups then function as coordination sites for a variety of transition-metal systems. It should be noted that complex reaction sequences of this type require prior model compound studies with the analogous cyclic oligomersZ6. Although phosphazene macromolecules have not yet been synthesized with outrigger chains of metal-metal-bonded polymers, a prototype reaction has been discovered at the oligomeric model compound level. Hexafluorocyclotriphosphazene, (NPF,), , reacts with sodium cyclopentadienyldicarbonylferrateto yield compound XIV. This compound undergoes decarbonylation on photolysis to yield XV, which is a stable metal-metalbonded specie^^^,^^. Compounds XIV and XV were the first phosphazenes synthesized with P-metal side-group bonds. Related Ru derivatives have also been prepared. 0

XIV

15.2. Ring-Ring and Ring-Polymer lnterconversions 15.2.11. in Phosphorus-Nitrogen Oligomers and Polymers 15.2.11.4. Polyorganophosphazenesvia Reactions.

I

n-BuLi, - n-BuBr

-f

171

- LiCl

J

n-RuLi -n-BuBr

--

ClPPhz

- LiCl

Scheme I11

In addition, a series of cyclophosphazenes with metallocenyl side groups has been proposed, and some of these have been polymerized to high polymer^*^^^^. The organometallic side groups confer a variety of electroactive properties on the polymers, including their use as electrode mediator materials3‘. (H.R. ALLCOCK)

1. P. Wisian-Neilson, R. H. Neilson, J. Am. Chem. SOC.,102,2848 (1980); R. H. Nielson, P. WisianNielson, Chem. Rev., 8, 541 (1988). 2. H. R. Allcock, Acc. Chem. Rex, 12, 351 (1979). 3. H. R. Allcock, R. L. Kugel, J. Am. Chem. Soc., 87,4216 (1965). 4. H. R. Allcock, R. L. Kugel, K. J. Valan, Znorg. Chem., 5, 1709 (1966). 5. H. R. Allcock, R. L. Kugel, Znorg. Chem., 5, 1716 (1966).

172

15.2. Ring-Ring and Ring-Polymer Interconversions 15.2.12. in Sulfur-Nitrogen Oligomers and Polymers 15.2.12.1. Formation of the Monomer SN.

H. R. Allcock, C. T.-W. Chu, Macromolecules, 12, 551 (1979). H. R. Allcock, T. L. Evans, D. B. Patterson, Macromolecules, 13, 201 (1980). H. R. Allcock, G. Y. Moore, Macromolecules, 5, 231 (1972). H. R. Allcock, L. A. Smeltz, J. Am. Chem. Soc., 98, 4143 (1976). S. H. Rose, J. Polym. Sci.,B6, 837 (1968). D. P. Tate, J. Polym. Sci., Polym. Symp., 48, 33 (1974). R. E. Singler, N. S. Schneider, G. L. Hagnauer, Polym. Eng. Sci.,15, 321 (1975). H. R. Allcock, W. J. Cook, D. P. Mack, Inorg. Chem., 11, 2584 (1972). H. R. Allcock, T. J. Fuller, T. L. Evans, Macromolecules, 13, 1325 (1980). T. L. Evans, P. R. Suszko, P. J. Harris, R. J. Ritchie, M. S. Connolly, J. L. DeSorcie, H. R. Allcock, unpublished work. 16. T. L. Evans, D. B. Patterson, P. R. Suszko, H. R. Allcock, Macromolecules, in press. 17. P. J. Harris, H. R. Allcock, J. Am. Chem. Soc., ZOO, 6512 (1978). 18. H. R. Allcock, P. J. Harris,. J. Am. Chem. Soc., 101, 6221 (1979). 19. A. G. Scopelianos, J. P. OBrien, H. R. Allcock, J. Chem. Soc., Chem. Commun., 198 (1980). 20. H. R. Allcock, R. W. Allen, J. P. O’Brien, J. Am. Chem. Soc., 99, 3984 (1977). 21. R. W. Allen, J. P. O’Brien, H. R. Allcock, J. Am. Chem. Soc., 99, 3987 (1977). 22. H. R. Allcock, P. P. Greigger, J. E. Gardner, J. L. Schmutz, J. Am. Chem. Soc., 101,606 (1979). 23. H. R. Allcock, T. J. Fuller, Macromolecules, 13, 1338 (1980). 24. H. R. Allcock, T. J. Fuller, D. P. Mack, K. Matsumura, K. M. Smeltz, Macromolecules, 10, 824 6. 7. 8. 9. 10. 11. 12. 13. 14. 15.

25. 26. 27. 28. 29. 30. 31.

(1977). H. R. Allcock, T. J. Fuller, J. Am. Chem. Soc., 103, 2250 (1981). H. R. Allcock, T. L. Evans, T. J. Fuller, Inorg. Chem., 19 1026 (1980). P. P. Greigger, H. R. Allcock, J. Am. Chem. Soc., 101,2492 (1979). H. R. Allcock, P. P. Greigger, L. J. Wagner, M. Y. Bernheim, Inorg. Chem., in press (1981). H. R. Allcock, K. D. Lavin, G. H. Riding, Macromolecules, 18, 1340 (1985). H. R. Allcock, J. L. Desorcie, G. H. Riding, Polyhedron, 6, 119 (1987). R. A. Saraceno, G . H. Riding, H. R. Allcock, A. G. Ewing, J. Am. Chem. Soc., 110,7254 (1988).

15.2.12. in Sulfur-Nitrogen Oligomers and Polymers The major interest in sulfur-nitrogen oligomers and polymers centers on the polymeric metal (SN), and its cyclic precursors (SN), and (SN),. The sulfur atoms in these molecules are in the 3 + formal oxidation state and 2-coordinate [ignoring the transannular S-S interactions in (SN),]. There is also an extensive chemistry of cyclic oligomers of the type (NSX), (X = C1, n = 3; X = F, n = 3,4), in which the 3-coordinate sulfur atoms are in the 4 oxidation state, and of the cyclic sulfanuric halides [NS(O)X], (X = C1, F, alkyl, aryl or R,N) containing 4-coordinate sulfur in the 6+ oxidation state. The nitrogen atoms are 2-coordinate in all these cyclic oligomers and in the linear polymer (SN), and its derivatives. The cyclic oligomer (SNH), contains 3-coordinate nitrogen and 2-coordinate sulfur atoms.

+

(T CHIVERS)

15.2.12.1. Formation of the Monomer

SN.

Thiazyl monomer, SN, polymerizes readily in the solid or liquid phases and has only a transient existence in the gaseous phase. It is formed by the passage of an electric discharge through sulfur vapor and nitrogen or SF, and nitrogen, from the reaction of active nitrogen with H,S or SCl, or from flash photolysis of a mixture of COS and NF, 1. H. G. Heal, Adv. Inorg. Chem. Radiochem., IS, 375 (1972).

(T CHIVERS)

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc. 172

15.2. Ring-Ring and Ring-Polymer Interconversions 15.2.12. in Sulfur-Nitrogen Oligomers and Polymers 15.2.12.1. Formation of the Monomer SN.

H. R. Allcock, C. T.-W. Chu, Macromolecules, 12, 551 (1979). H. R. Allcock, T. L. Evans, D. B. Patterson, Macromolecules, 13, 201 (1980). H. R. Allcock, G. Y. Moore, Macromolecules, 5, 231 (1972). H. R. Allcock, L. A. Smeltz, J. Am. Chem. Soc., 98, 4143 (1976). S. H. Rose, J. Polym. Sci.,B6, 837 (1968). D. P. Tate, J. Polym. Sci., Polym. Symp., 48, 33 (1974). R. E. Singler, N. S. Schneider, G. L. Hagnauer, Polym. Eng. Sci.,15, 321 (1975). H. R. Allcock, W. J. Cook, D. P. Mack, Inorg. Chem., 11, 2584 (1972). H. R. Allcock, T. J. Fuller, T. L. Evans, Macromolecules, 13, 1325 (1980). T. L. Evans, P. R. Suszko, P. J. Harris, R. J. Ritchie, M. S. Connolly, J. L. DeSorcie, H. R. Allcock, unpublished work. 16. T. L. Evans, D. B. Patterson, P. R. Suszko, H. R. Allcock, Macromolecules, in press. 17. P. J. Harris, H. R. Allcock, J. Am. Chem. Soc., ZOO, 6512 (1978). 18. H. R. Allcock, P. J. Harris,. J. Am. Chem. Soc., 101, 6221 (1979). 19. A. G. Scopelianos, J. P. OBrien, H. R. Allcock, J. Chem. Soc., Chem. Commun., 198 (1980). 20. H. R. Allcock, R. W. Allen, J. P. O’Brien, J. Am. Chem. Soc., 99, 3984 (1977). 21. R. W. Allen, J. P. O’Brien, H. R. Allcock, J. Am. Chem. Soc., 99, 3987 (1977). 22. H. R. Allcock, P. P. Greigger, J. E. Gardner, J. L. Schmutz, J. Am. Chem. Soc., 101,606 (1979). 23. H. R. Allcock, T. J. Fuller, Macromolecules, 13, 1338 (1980). 24. H. R. Allcock, T. J. Fuller, D. P. Mack, K. Matsumura, K. M. Smeltz, Macromolecules, 10, 824 6. 7. 8. 9. 10. 11. 12. 13. 14. 15.

25. 26. 27. 28. 29. 30. 31.

(1977). H. R. Allcock, T. J. Fuller, J. Am. Chem. Soc., 103, 2250 (1981). H. R. Allcock, T. L. Evans, T. J. Fuller, Inorg. Chem., 19 1026 (1980). P. P. Greigger, H. R. Allcock, J. Am. Chem. Soc., 101,2492 (1979). H. R. Allcock, P. P. Greigger, L. J. Wagner, M. Y. Bernheim, Inorg. Chem., in press (1981). H. R. Allcock, K. D. Lavin, G. H. Riding, Macromolecules, 18, 1340 (1985). H. R. Allcock, J. L. Desorcie, G. H. Riding, Polyhedron, 6, 119 (1987). R. A. Saraceno, G . H. Riding, H. R. Allcock, A. G. Ewing, J. Am. Chem. Soc., 110,7254 (1988).

15.2.12. in Sulfur-Nitrogen Oligomers and Polymers The major interest in sulfur-nitrogen oligomers and polymers centers on the polymeric metal (SN), and its cyclic precursors (SN), and (SN),. The sulfur atoms in these molecules are in the 3 + formal oxidation state and 2-coordinate [ignoring the transannular S-S interactions in (SN),]. There is also an extensive chemistry of cyclic oligomers of the type (NSX), (X = C1, n = 3; X = F, n = 3,4), in which the 3-coordinate sulfur atoms are in the 4 oxidation state, and of the cyclic sulfanuric halides [NS(O)X], (X = C1, F, alkyl, aryl or R,N) containing 4-coordinate sulfur in the 6+ oxidation state. The nitrogen atoms are 2-coordinate in all these cyclic oligomers and in the linear polymer (SN), and its derivatives. The cyclic oligomer (SNH), contains 3-coordinate nitrogen and 2-coordinate sulfur atoms.

+

(T CHIVERS)

15.2.12.1. Formation of the Monomer

SN.

Thiazyl monomer, SN, polymerizes readily in the solid or liquid phases and has only a transient existence in the gaseous phase. It is formed by the passage of an electric discharge through sulfur vapor and nitrogen or SF, and nitrogen, from the reaction of active nitrogen with H,S or SCl, or from flash photolysis of a mixture of COS and NF, 1. H. G. Heal, Adv. Inorg. Chem. Radiochem., IS, 375 (1972).

(T CHIVERS)

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc. 15.2. Ring-Ring and Ring-Polymer Interconversions 15.2.12. in Sulfur-Nitrogen Oligomers and Polymers 15.2.12.2. Formation of Cyclic Oligorners (SN), and (SN),.

15.2.12.2. Formation of Cyclic Oligomers (SN), and (SN),

173

.

The standard synthesis of (SN), involves the reaction of sulfur halides with ammonia in carbon tetrachloride’,’: S,Cl, is first treated with chlorine and then ammonia is passed into the reaction mixture, which is maintained at 20-50°C. The reaction proceeds via a number of stages, including the formation of the volatile monomer, NSCl, which reacts with S,C1, to give (SN),. In xs S,Cl, this product is converted to [S,N,]Cl, which reacts with ammonia to give (SN),. An excess of ammonia should be avoided, since this reagent converts (SN), to S,N;. The crude product is washed with water (to remove NH,C1) dried and extracted with dioxan in a continuously recycling extractor to give (SN), which can be purified by recrystallization from chloroform or toluene or by vacuum sublimation at 80-90°C lo-’ torr. CAUTION: (SN), may explode when struck, ground or suddenly heated. The purest samples are the most sensitive to explosion, and great care should be exercised in handling or storing this compound‘,*. An alternative synthesis of (SN),, which avoids the use of NH,, is the reaction of S,Cl, with sulfur and NH,CI at 150-160°C. The NH,Cl should be dry and finely ground and the reaction flask filled to the neck in order to optimize yields. The [S,N,Cl]Cl is collected as a sublimate in an air condenser and is treated with xs C1, to give (NSCl), . Reduction of (NSCl), with Ph,Sb in CH,CN gives (SN), quickly and in high yield3. Iron or Hg metal have also been used for this The synthesis of (SN), was first achieved by passing the vapors of impure (SN), over finely divided silver to remove elemental sulfur6. This reaction is the most widely used method for the preparation of high-purity (SN), . Various modifications to the experimental procedure involving variations in the pyrolysis temperature (200-300°C) and the purification of (SN), have been suggested’-’’. Alternative routes for the preparation of (SN), include the thermal decomposition of S,N,Cl 1 2 , Ph,AsN-S,N, l 3 or S,N,.2 AlCl, which is obtained from (SN), and AlCl, (see Scheme I)’5. None of these methods is clearly superior to the traditional approach to (SN),. CAUTION. (SN), is a colorless crystalline solid that is

MesSiNs

(NSCl), --N

/S-N

\

I

S-N

/S-N\

S-

eC-

O T , solid state

(SN),*2 AlCl,

Ph,AsNS,N,

Scheme I

(SN);

174

15.2. Ring-Ring and Ring-Polymer Interconversions 15.2.12.in Sulfur-Nitrogen Oligomers and Polymers 15.2.12.3. Polymerization of (SN), to (SN),.

susceptible to detonation from thermal or mechanical shock. Appropriate precautions should be taken in handling this material. It is stable indefinitely in organic solvents but rapidly dimerizes in the presence of trace amounts of nucleophiles or reducing agents. The overlap properties of the HOMO and LUMO of two cofacially aligned (SN), molecules have been invoked to explain this dimerization16. This interpretation has been questioned, however, and an alternative involving a centrosymmetric approach of two (SN), units, one of which is reduced to the radical anion (SN),T, has been proposed17. (T. CHIVERS)

1. M. Villeno-Blanca, W. L. Jolly, Inorg. Synth., 9, 98 (1967). 2. A. J. Banister, Inorg. Synth., 17, 197 (1977). Safety precautions that must be observed when handling S,N,. 3. T. Chivers, R. T. Oakley, unpublished results. 4. A. J. Banister, A. J. Fielder, R. G. Hey, N. R. M. Smith, J. Chem. Soc., Dalton Trans., 1457 (1980). 5. M. P. Berthet, H. Vincent, Y. Monteil, 2. Nuturforsch., 356, 329 (1980). 6. F. P. Burt, J. Chem. Soc., 1171 (1910). 7. M. Becke-Goehring, Inorg. Synth., 6, 123 (1960). 8. C. Hsu, M. M. Labes, J. Chem. Phys., 61,4640 (1974). 9. V. V. Walatka Jr., M. M. Labes, J. H. Perlstein, Phys. Rev. Lett., 31, 1139 (1973). 10. G. B. Street, H. Arnal, W. D. Gill, P. M. Grant, R. L. Greene, Mater. Res. Bull., 10, 877 (1975). 11. C . M. Mikulski, P. J. Russo, M. S. Saran, A. G. MacDiarmid, A. F. Garito, A. J. Heeger, J . Am. Chem. Soc., 97, 6358 (1975). 12. A. J. Banister, Z. V. Hauptman, J. Chem. Soc., Dalton Trans., 731 (1980). 13. T. Chivers, A. W. Cordes, R. T. Oakley, P. N. Swepston, Inorg. Chem., 20, 2376 (1981). 14. H. W. Roesky, J. Anhaus, Chem. Ber., 115, 3682 (1982). 15. U. Thewalt, M. Burger, Angew. Chem., Int. Ed. Engl., 21, 634 (1982). 16. K. Tanaka, T. Yamabe, K. Noda, K. Fukui, H. Kato, J. Phys. Chem., 82, 1453 (1978). 17. R. T. Oakley, Progress Znorg. Chem., 36, 299 (1988).

15.2.12.3. Polymerization of (SN), to (SN),.

The spontaneous topochemical polymerization of (SN), to (SN), at 0°Cl,', the most prominent example of a polymerization reaction in S-N chemistry, has been reviewed3. Crystals of (SN), belong to the space group P2,/c and it has been proposed that the a axis of the dimer converts into the b axis of the polymer with chain extension occurring ~ * the ~ . basis of EHMO calculations, this process is both parallel to this a x i ~ ' ~On thermally and photochemically allowed5, and the photoinduced polymerization occurs at -65°C in THF6. ESR studies indicate a radical process, but ab initio MO and configuration interaction studies suggest that the polymerization involves a cascade effect initiated by only a few radical centers7. Some routes to (SN), that do not involve (SN), as an intermediate are available (see 515.2.12.3). The polymer is obtained in high yield from the reaction of (NSCl), with trimethylsilyl azide in a~etonitrile'~~, by the electrochemical reduction of [(SN),] saltsi0-' ', and by radiofrequency discharge through (SN), vapors in a helium plasma13. Other reactions that produce (SN), include the reduction of S,N+ with azide ion',, the solid state polymerization of impure S,N, (recrystallized S,N, does not polymerize)16, and the oxidation of (SN); with certain electrophiles". The polymer (SN), is a shiny metallic solid consisting of highly oriented parallel fibers. Well-formed crystals of (SN), are bright gold and lustrous. The ends of the crystals are blue-black, and most chemical syntheses, other than the polymerization of (SN),, produce (SN), as a blue-black powder3. The RT conductivity of (SN), is 1-4 x +

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc. 174

15.2. Ring-Ring and Ring-Polymer Interconversions 15.2.12.in Sulfur-Nitrogen Oligomers and Polymers 15.2.12.3. Polymerization of (SN), to (SN),.

susceptible to detonation from thermal or mechanical shock. Appropriate precautions should be taken in handling this material. It is stable indefinitely in organic solvents but rapidly dimerizes in the presence of trace amounts of nucleophiles or reducing agents. The overlap properties of the HOMO and LUMO of two cofacially aligned (SN), molecules have been invoked to explain this dimerization16. This interpretation has been questioned, however, and an alternative involving a centrosymmetric approach of two (SN), units, one of which is reduced to the radical anion (SN),T, has been proposed17. (T. CHIVERS)

1. M. Villeno-Blanca, W. L. Jolly, Inorg. Synth., 9, 98 (1967). 2. A. J. Banister, Inorg. Synth., 17, 197 (1977). Safety precautions that must be observed when handling S,N,. 3. T. Chivers, R. T. Oakley, unpublished results. 4. A. J. Banister, A. J. Fielder, R. G. Hey, N. R. M. Smith, J. Chem. Soc., Dalton Trans., 1457 (1980). 5. M. P. Berthet, H. Vincent, Y. Monteil, 2. Nuturforsch., 356, 329 (1980). 6. F. P. Burt, J. Chem. Soc., 1171 (1910). 7. M. Becke-Goehring, Inorg. Synth., 6, 123 (1960). 8. C. Hsu, M. M. Labes, J. Chem. Phys., 61,4640 (1974). 9. V. V. Walatka Jr., M. M. Labes, J. H. Perlstein, Phys. Rev. Lett., 31, 1139 (1973). 10. G. B. Street, H. Arnal, W. D. Gill, P. M. Grant, R. L. Greene, Mater. Res. Bull., 10, 877 (1975). 11. C . M. Mikulski, P. J. Russo, M. S. Saran, A. G. MacDiarmid, A. F. Garito, A. J. Heeger, J . Am. Chem. Soc., 97, 6358 (1975). 12. A. J. Banister, Z. V. Hauptman, J. Chem. Soc., Dalton Trans., 731 (1980). 13. T. Chivers, A. W. Cordes, R. T. Oakley, P. N. Swepston, Inorg. Chem., 20, 2376 (1981). 14. H. W. Roesky, J. Anhaus, Chem. Ber., 115, 3682 (1982). 15. U. Thewalt, M. Burger, Angew. Chem., Int. Ed. Engl., 21, 634 (1982). 16. K. Tanaka, T. Yamabe, K. Noda, K. Fukui, H. Kato, J. Phys. Chem., 82, 1453 (1978). 17. R. T. Oakley, Progress Znorg. Chem., 36, 299 (1988).

15.2.12.3. Polymerization of (SN), to (SN),.

The spontaneous topochemical polymerization of (SN), to (SN), at 0°Cl,', the most prominent example of a polymerization reaction in S-N chemistry, has been reviewed3. Crystals of (SN), belong to the space group P2,/c and it has been proposed that the a axis of the dimer converts into the b axis of the polymer with chain extension occurring ~ * the ~ . basis of EHMO calculations, this process is both parallel to this a x i ~ ' ~On thermally and photochemically allowed5, and the photoinduced polymerization occurs at -65°C in THF6. ESR studies indicate a radical process, but ab initio MO and configuration interaction studies suggest that the polymerization involves a cascade effect initiated by only a few radical centers7. Some routes to (SN), that do not involve (SN), as an intermediate are available (see 515.2.12.3). The polymer is obtained in high yield from the reaction of (NSCl), with trimethylsilyl azide in a~etonitrile'~~, by the electrochemical reduction of [(SN),] saltsi0-' ', and by radiofrequency discharge through (SN), vapors in a helium plasma13. Other reactions that produce (SN), include the reduction of S,N+ with azide ion',, the solid state polymerization of impure S,N, (recrystallized S,N, does not polymerize)16, and the oxidation of (SN); with certain electrophiles". The polymer (SN), is a shiny metallic solid consisting of highly oriented parallel fibers. Well-formed crystals of (SN), are bright gold and lustrous. The ends of the crystals are blue-black, and most chemical syntheses, other than the polymerization of (SN),, produce (SN), as a blue-black powder3. The RT conductivity of (SN), is 1-4 x +

15.2. Ring-Ring and Ring-Polymer Interconversions 15.2.12. in Sulfur-Nitrogen Oligomers and Polymers 15.2.12.4. Halogenated Derivatives of (SN), . ~~~~

~

~

~

~

~~

~

175

~

lo3R - ' cm-I. The exact value depends on the quality of the crystals. The conductivity increases by about two orders of magnitude at 4 K. At even lower T the polymer becomes superconducting with a transition T of 0.3 K3. The sublimation of (SN), at 135°C and reduced pressure produces a gas-phase species that reforms the polymer as epitaxial fibers upon condensation. An open-chain (SN), structure has been suggested for the vapor-phase intermediate, but more recent spectroscopic investigations together with ab initio MO and CI calculations support the formulation of this species as the cyclic (SN),*radical'*. (T. CHIVERS)

1. G. B. Street, H. Arnal, W. D. Gill, P. M. Grant, R. L. Greene, Mater. Res. Bull., 10, 877 (1975). 2. M. J. Cohen, A. F. Garito, A. J. Heeger, A. G. MacDiarmid, C. M. Mikulski, M. S. Saran, J. Kleppinger, J. Am. Chem. SOC., 98, 3844 (1976). 3. M. M. Labes, P. Love, L. F. Nichols, Chem. Ren, 79,l (1979). This review gives a comprehensive account of the preparation and physical properties of (SN),. 4. M. Boudeulle, Cryst. Struc. Commun.,4, 9 (1975). 5. J. K. Burdett, J. Am. Chem. Soc., 102, 5458 (1980). 6. P. Love, H. I. Kao, G. H. Myer, M. M. Labes, J. Chem. Soc., Chem. Commun.,301 (1978). 7. M. H. Palmer, R. H. Findlay, J. Mol. Struc., Theochem.,92, 373 (1983). 8. F. A. Kennett, G. K. MacLean, J. Passmore, M. N. S. Rao, J. Chem. SOC., Dalton Trans., 851 (1 982) 9. A. J. Banister, Z. V. Hauptman, J. Passmore, C.-M. Wong, P. S. White, J. Chem. SOC.,Dalton Trans., 2371 (1986). 10. A. J. Banister, Z. V. Hauptman, A. G. Kendrick, J . Chem. Soc., Chem. Commun., 1017 (1983). 11. A. J. Banister, 2.V. Hauptman, A. G. Kendrick, R. W. H. Small, J. Chem. SOC.,Dalton Trans., 915 (1987). 12. H. P. Fritz, R. Bruchhaus, 2. Naturforsch., 38b, 1375 (1983). 13. M. W. R. Witt, W. I. Bailey, R. J. Lagow, J. Am. Chem. Soc., 105, 1668 (1983). 14. F. A. Kennett, G. K. MacLean, J. Passmore, M. N. S. Rao, J. Chem. Soc., Dalton Trans., 851 (1982). Dalton Trans., 2188 (1981). 15. R. W. Small, A. J. Banister, Z . V. Hauptman, J. Chem. SOC., 16. T. Chivers, P. W. Codding, W. G. Laidlaw, S. W. Liblong, R. T. Oakley, M. Trsic, J. Am. Chem. Soc., 105, 1186 (1983). 17. T. Chivers, M. N. S. Rao, Can. J. Chem., 61, 1957 (1983). 18. W. M. Lau, N. P. C. Westwood, M. H. Palmer, J. Am. Chew SOC.,108, 3229 (1986).

15.2.12.4. Halogenated Derivatives of (SN),

.

The reactions of powdered (SN), with bromine or the interhalogens, ICl and IBr, yield highly conducting polymers of approximate composition (SNBr,,,), and [(SN)(IX),,J,, respectively' -3. Vibrational scattering and x-ray diffraction studies, of brominated (SN), indicate that bromine is present as Br; and Br,. The polymer (SNBr,,,), has also been prepared by the reaction of (NSCl), with Me,SiBr in CH,Cl, at - 60°C '. An iodinated conducting polymer of approximate composition (S3.0N2,110,5) is obtained when vaporized S,N,Cl is passed over sodium iodide supported on glass wool at 250"C6. This polymer decomposes above 40°C in vacuo to give sulfur, (SN), and iodine. (T. CHIVERS)

1. G. B. Street, R. L. Bingham, J. I. Crowley, J. Kuyper, J. Chem. Soc., Chem. Commun.,464 (1977). 2. M. Akhtar, C. K. Chiang, A. J. Heeger, A. G. MacDiarmid, J. Chem. Soc., Chem. Commun., 846 (1977).

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc. 15.2. Ring-Ring and Ring-Polymer Interconversions 15.2.12. in Sulfur-Nitrogen Oligomers and Polymers 15.2.12.4. Halogenated Derivatives of (SN), . ~~~~

~

~

~

~

~~

~

175

~

lo3R - ' cm-I. The exact value depends on the quality of the crystals. The conductivity increases by about two orders of magnitude at 4 K. At even lower T the polymer becomes superconducting with a transition T of 0.3 K3. The sublimation of (SN), at 135°C and reduced pressure produces a gas-phase species that reforms the polymer as epitaxial fibers upon condensation. An open-chain (SN), structure has been suggested for the vapor-phase intermediate, but more recent spectroscopic investigations together with ab initio MO and CI calculations support the formulation of this species as the cyclic (SN),*radical'*. (T. CHIVERS)

1. G. B. Street, H. Arnal, W. D. Gill, P. M. Grant, R. L. Greene, Mater. Res. Bull., 10, 877 (1975). 2. M. J. Cohen, A. F. Garito, A. J. Heeger, A. G. MacDiarmid, C. M. Mikulski, M. S. Saran, J. Kleppinger, J. Am. Chem. SOC., 98, 3844 (1976). 3. M. M. Labes, P. Love, L. F. Nichols, Chem. Ren, 79,l (1979). This review gives a comprehensive account of the preparation and physical properties of (SN),. 4. M. Boudeulle, Cryst. Struc. Commun.,4, 9 (1975). 5. J. K. Burdett, J. Am. Chem. Soc., 102, 5458 (1980). 6. P. Love, H. I. Kao, G. H. Myer, M. M. Labes, J. Chem. Soc., Chem. Commun.,301 (1978). 7. M. H. Palmer, R. H. Findlay, J. Mol. Struc., Theochem.,92, 373 (1983). 8. F. A. Kennett, G. K. MacLean, J. Passmore, M. N. S. Rao, J. Chem. SOC., Dalton Trans., 851 (1 982) 9. A. J. Banister, Z. V. Hauptman, J. Passmore, C.-M. Wong, P. S. White, J. Chem. SOC.,Dalton Trans., 2371 (1986). 10. A. J. Banister, Z. V. Hauptman, A. G. Kendrick, J . Chem. Soc., Chem. Commun., 1017 (1983). 11. A. J. Banister, 2.V. Hauptman, A. G. Kendrick, R. W. H. Small, J. Chem. SOC.,Dalton Trans., 915 (1987). 12. H. P. Fritz, R. Bruchhaus, 2. Naturforsch., 38b, 1375 (1983). 13. M. W. R. Witt, W. I. Bailey, R. J. Lagow, J. Am. Chem. Soc., 105, 1668 (1983). 14. F. A. Kennett, G. K. MacLean, J. Passmore, M. N. S. Rao, J. Chem. Soc., Dalton Trans., 851 (1982). Dalton Trans., 2188 (1981). 15. R. W. Small, A. J. Banister, Z . V. Hauptman, J. Chem. SOC., 16. T. Chivers, P. W. Codding, W. G. Laidlaw, S. W. Liblong, R. T. Oakley, M. Trsic, J. Am. Chem. Soc., 105, 1186 (1983). 17. T. Chivers, M. N. S. Rao, Can. J. Chem., 61, 1957 (1983). 18. W. M. Lau, N. P. C. Westwood, M. H. Palmer, J. Am. Chew SOC.,108, 3229 (1986).

15.2.12.4. Halogenated Derivatives of (SN),

.

The reactions of powdered (SN), with bromine or the interhalogens, ICl and IBr, yield highly conducting polymers of approximate composition (SNBr,,,), and [(SN)(IX),,J,, respectively' -3. Vibrational scattering and x-ray diffraction studies, of brominated (SN), indicate that bromine is present as Br; and Br,. The polymer (SNBr,,,), has also been prepared by the reaction of (NSCl), with Me,SiBr in CH,Cl, at - 60°C '. An iodinated conducting polymer of approximate composition (S3.0N2,110,5) is obtained when vaporized S,N,Cl is passed over sodium iodide supported on glass wool at 250"C6. This polymer decomposes above 40°C in vacuo to give sulfur, (SN), and iodine. (T. CHIVERS)

1. G. B. Street, R. L. Bingham, J. I. Crowley, J. Kuyper, J. Chem. Soc., Chem. Commun.,464 (1977). 2. M. Akhtar, C. K. Chiang, A. J. Heeger, A. G. MacDiarmid, J. Chem. Soc., Chem. Commun., 846 (1977).

176

15.2. Ring-Ring and Ring-Polymer Interconversions 15.2.12. in Sulfur-Nitrogen Oligomers and Polymers 15.2.12.5. Formation of Cyclic Oligorners (NSX), (X = CI, F; n = 3,4).

3. M. Akhtar, C. K. Chiang, A. J. Heeger, J. Milliken, A. G. MacDiarmid, Inorg. Chem., 17, 1539 (1978). 4. Z. Iqbal, R. H. Baughman, J. Kleppinger, A. G. MacDiarmid, Solid State Commun., 25, 409 (1978). 5. U. Demant, K. Dehnicke, 2. Naturforsch., 41b, 929 (1986). 6. A. J. Banister, N. R. M. Smith, J. Chem. SOC., Dalton Trans., 937 (1980).

15.2.12.5. Formation of Cyclic Oligomers (NSX), (X = CI,

-

F; n = 3,4).

A safe and convenient procedure for the preparation of (NSCl), is the chlorination of [S,N,Cl]Cl with either C1, or SO,C1, ': 3 S,N,Cl,

+ 3 SO,Cl,

2 (NSCl),

+ 3 SCl, + 3 SO,

(a)

The cyclic tetramer (NSCl), may be formed, after S,N,Cl,, in the direct chlorination of (SN), in a solvent, but it is apparently unstable with respect to elimination of NSCl to give the cyclic trimer,. Above 100°C (NSCl), gives a vapor containing much NSCl monomer: (NSCl), (g)

3 NSCl (g)

(b)

the AH' for the forward reaction has been estimated at 92 & 12 kJ mol-' and ASo at 320 40 Jmo1-IK-I ', The dissociation is accompanied by considerable decomposition to (SN),, SC1, and S,Cl, Dissociation of the trimer into monomer also occurs in such solvents as CC1, or CH,Cl,. Equilibrium (b) can be conveniently monitored by I4N NMR spectroscopy since the chemical shifts of NSCl and (NSCl), are well separated (ca. +350 and -264 ppm, respectively, relative to external CH,NO, at 22°C)6. Monomerization is favored for impure samples of (NSCl), and by increasing the temperature of the solution. No evidence for the formation of a cyclic dimer, (NSCI),, was found from I4N NMR studies6. The cyclic dimers (NSX), (X = C1, F) are predicted to be thermodynamically unstable with respect to 2 NSX but should be kinetically stable at low temperatures on the basis of ab initio SCF calculations7. Esters of the hypothetical parent acid (NSOH), are prepared by the reaction of (NSCl), with oxiranes' or NaOR (R = alkyl, b e n ~ y l )in~ the corresponding alcohol. Linear polymers based on an NSX repeating unit are unknown. Monomer NSF is most conveniently generated by decomposition of FC(O)NSF, or Hg(NSF,), ', but the combination of moisture sensitivity and thermal instability make it a difficult substance to handle. The cyclic trimer (NSF), is prepared in 90% yield by fluorination of (NSCI), with AgF, in CCl, l o or by polymerization of NSF for 3 d in a sealed glass tube". The cyclic tetramer (NSF), is prepared in 14 % yield by the reaction of (SN), with AgF, in dry CCl, under reflux". The tetramer is thermodynamically unstable with respect to the trimer, which is also obtained in this reaction. The tetramer depolymerizes to NSF in vacuo at 250°C. The reaction of (SN), with (CF,),NO produces {NS[ON(CF,),]}, as a white crystalline solid',. The related cyclic tetramer {NS[OC(CF,),]}, is obtained in 98% yield by treating (SN), with (CF,),COCl at 0°C 13. The thermally unstable tetramer (NSCF,), is prepared from CF,SCl and Me,SiN, at -30°C14. It decomposes at RT to give the polymer (CF,SN),, which is also obtained by the decomposition of CF,S(F)NC(O)F in the presence of HgF, at 25°C.

'.

(T CHIVERS)

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc. 176

15.2. Ring-Ring and Ring-Polymer Interconversions 15.2.12. in Sulfur-Nitrogen Oligomers and Polymers 15.2.12.5. Formation of Cyclic Oligorners (NSX), (X = CI, F; n = 3,4).

3. M. Akhtar, C. K. Chiang, A. J. Heeger, J. Milliken, A. G. MacDiarmid, Inorg. Chem., 17, 1539 (1978). 4. Z. Iqbal, R. H. Baughman, J. Kleppinger, A. G. MacDiarmid, Solid State Commun., 25, 409 (1978). 5. U. Demant, K. Dehnicke, 2. Naturforsch., 41b, 929 (1986). 6. A. J. Banister, N. R. M. Smith, J. Chem. SOC., Dalton Trans., 937 (1980).

15.2.12.5. Formation of Cyclic Oligomers (NSX), (X = CI,

-

F; n = 3,4).

A safe and convenient procedure for the preparation of (NSCl), is the chlorination of [S,N,Cl]Cl with either C1, or SO,C1, ': 3 S,N,Cl,

+ 3 SO,Cl,

2 (NSCl),

+ 3 SCl, + 3 SO,

(a)

The cyclic tetramer (NSCl), may be formed, after S,N,Cl,, in the direct chlorination of (SN), in a solvent, but it is apparently unstable with respect to elimination of NSCl to give the cyclic trimer,. Above 100°C (NSCl), gives a vapor containing much NSCl monomer: (NSCl), (g)

3 NSCl (g)

(b)

the AH' for the forward reaction has been estimated at 92 & 12 kJ mol-' and ASo at 320 40 Jmo1-IK-I ', The dissociation is accompanied by considerable decomposition to (SN),, SC1, and S,Cl, Dissociation of the trimer into monomer also occurs in such solvents as CC1, or CH,Cl,. Equilibrium (b) can be conveniently monitored by I4N NMR spectroscopy since the chemical shifts of NSCl and (NSCl), are well separated (ca. +350 and -264 ppm, respectively, relative to external CH,NO, at 22°C)6. Monomerization is favored for impure samples of (NSCl), and by increasing the temperature of the solution. No evidence for the formation of a cyclic dimer, (NSCI),, was found from I4N NMR studies6. The cyclic dimers (NSX), (X = C1, F) are predicted to be thermodynamically unstable with respect to 2 NSX but should be kinetically stable at low temperatures on the basis of ab initio SCF calculations7. Esters of the hypothetical parent acid (NSOH), are prepared by the reaction of (NSCl), with oxiranes' or NaOR (R = alkyl, b e n ~ y l )in~ the corresponding alcohol. Linear polymers based on an NSX repeating unit are unknown. Monomer NSF is most conveniently generated by decomposition of FC(O)NSF, or Hg(NSF,), ', but the combination of moisture sensitivity and thermal instability make it a difficult substance to handle. The cyclic trimer (NSF), is prepared in 90% yield by fluorination of (NSCI), with AgF, in CCl, l o or by polymerization of NSF for 3 d in a sealed glass tube". The cyclic tetramer (NSF), is prepared in 14 % yield by the reaction of (SN), with AgF, in dry CCl, under reflux". The tetramer is thermodynamically unstable with respect to the trimer, which is also obtained in this reaction. The tetramer depolymerizes to NSF in vacuo at 250°C. The reaction of (SN), with (CF,),NO produces {NS[ON(CF,),]}, as a white crystalline solid',. The related cyclic tetramer {NS[OC(CF,),]}, is obtained in 98% yield by treating (SN), with (CF,),COCl at 0°C 13. The thermally unstable tetramer (NSCF,), is prepared from CF,SCl and Me,SiN, at -30°C14. It decomposes at RT to give the polymer (CF,SN),, which is also obtained by the decomposition of CF,S(F)NC(O)F in the presence of HgF, at 25°C.

'.

(T CHIVERS)

15.2. Ring-Ring and Ring-Polymer Interconversions 15.2.12. in Sulfur-Nitrogen Oligomers and Polymers 15.2.12.6. Formation of the Cyclic Oligomers [NS(O)X],.

177

1. W. L. Jolly, K. D. Maguire, Inorg. Synth., 9, 102 (1967). 2. G. G. Alange, A. J. Banister, B. Bell, J. Chem. Sot., Dalton Trans., 2399 (1972). 3. H. G. Heal, The Inorganic Heterocyclic Chemistry ofsulfur, Nitrogen and Phosphorus, Academic Press, London, 1980. This book has chapters on the cyclic oligomers formed by thiazyl halides, NSX; sulfanuric halides, [NS(O)X]; and thionyl imide, HNSO. 4. R. C. Patton, W. L. Jolly, Inorg. Chem., 9, 1079 (1970). 5. H. Vincent, Y. Monteil, Synth. React. Inorg. Met. Org. Chem., 8, 51 (1978). 6. J. Passmore, M. J. Schriver, Znorg. Chem., 27, 2749 (1988). 7. R. Ahlrichs, C. Ehrhardt, Chem. Phys., 107, l(1986). 8. G. G. Alange, A. J. Banister, B. Bell, J. Znorg. Nucl. Chem., 41, 1421 (1979). 9. 0. Glemser, R. Mews, Angew. Chem., Int. Ed. Engl., 19, 883 (1980). This review gives a detailed account of the chemistry of the nonomer NSF and the cyclic oligomers (NSF), and (NSF),. 10. H. Schroder, 0. Glemser, Z. Anorg. Allg. Chem., 298, 78 (1959). 11. 0. Glemser, H. Meyer, A. Haas, Chem. Ber., 97, 1704 (1964). 12. H. J. Emeleus, R. A. Forder, R. J. Poulet, G. M. Sheldrick, Chem. Commun., 1483 (1970). 13. G. L. Card, J. M. Shreeve, J. Am. Chem. Soc., 104, 5566 (1982). 14. D. Bielefeldt, A. Haas, Chem. Ber., 116, 1257 (1983).

15.2.12.6. Formation of the Cyclic Oligomers [NS(O)X],.

Unlike the isoelectronic cyclophosphazenes only six-membered rings have been well characterized for oligomers of the -NS(O)E(E = F, C1, alkyl, aryl or R,N) monomer unit. The trimeric chloride is best prepared by treating of SOCl, with sodium azide in acetonitrile at - 35°C ': 3 SOCl,

+ 3 NaN,

-

[NS(O)Cl],

+ 3 NaCl + 3 N,

(4

The azide route has also been used to make [NS(O)Ph], from PhS(0)Cl'. Sulfanuric chloride may also be obtained as a mixture of LY and /3 isomers in a twostage reaction from sulfamic acid or, preferably, sulfamoyl chloride,. The /3 isomer is stable in the solid state or in nonpolar solvents but changes to the LY isomer in ca. 1 h in acetonitrile: H,NSO,H

PClS

CI,P=NSO,Cl

- OPC13

[NS(O)Cl],

(b)

A third, but less convenient, way of making [NS(O)Cl], is by the oxidation of (NSCI), with SO, at 150°C and 20 atm4. The fluoride [NS(O)F], is obtained as a mixture of so-called cis and trans isomers by the fluorination of [NS(O)Cl],, preferably with SbF, This cis isomer shows a singlet in the "F N M R spectrum consistent with a structure in which all three fluorine atoms are on the same side of the S,N, ring {cf. [NS(O)Cl],), while the trans isomer exhibits an AB, pattern, indicating that one fluorine is on the side of the ring opposite to the other two6. Trimeric sulfanuric fluoride has also been obtained by the thermal decomposition of Cs[NS(O)F,], Hg[NS(O)F,], ', B[NS(O)F,], * or Me,Sn[NS(O)F,] ', which occurs at RT for the Sn compound. The aq hydrolysis of [NS(O)Cl], in the presence of Ag+ produces Ag,[(NSO,),], containing a six-membered ring in a chair conformation". The reaction of [NS(O)F], with benzene at reflux in the presence of AlCl, gives two isomers of [NS(O)Ph], Complete substitution of fluorines in [NS(O)F], by dimethylamino groups occurs upon reaction of the trimer with Me,NH at 80°C in the absence of a solvent12. Poorly characterized materials believed to be linear polymers, [N=S(O)F],, have been obtained as viscous yellow residues when the reaction product of ammonia and

'.

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc. 15.2. Ring-Ring and Ring-Polymer Interconversions 15.2.12. in Sulfur-Nitrogen Oligomers and Polymers 15.2.12.6. Formation of the Cyclic Oligomers [NS(O)X],.

177

1. W. L. Jolly, K. D. Maguire, Inorg. Synth., 9, 102 (1967). 2. G. G. Alange, A. J. Banister, B. Bell, J. Chem. Sot., Dalton Trans., 2399 (1972). 3. H. G. Heal, The Inorganic Heterocyclic Chemistry ofsulfur, Nitrogen and Phosphorus, Academic Press, London, 1980. This book has chapters on the cyclic oligomers formed by thiazyl halides, NSX; sulfanuric halides, [NS(O)X]; and thionyl imide, HNSO. 4. R. C. Patton, W. L. Jolly, Inorg. Chem., 9, 1079 (1970). 5. H. Vincent, Y. Monteil, Synth. React. Inorg. Met. Org. Chem., 8, 51 (1978). 6. J. Passmore, M. J. Schriver, Znorg. Chem., 27, 2749 (1988). 7. R. Ahlrichs, C. Ehrhardt, Chem. Phys., 107, l(1986). 8. G. G. Alange, A. J. Banister, B. Bell, J. Znorg. Nucl. Chem., 41, 1421 (1979). 9. 0. Glemser, R. Mews, Angew. Chem., Int. Ed. Engl., 19, 883 (1980). This review gives a detailed account of the chemistry of the nonomer NSF and the cyclic oligomers (NSF), and (NSF),. 10. H. Schroder, 0. Glemser, Z. Anorg. Allg. Chem., 298, 78 (1959). 11. 0. Glemser, H. Meyer, A. Haas, Chem. Ber., 97, 1704 (1964). 12. H. J. Emeleus, R. A. Forder, R. J. Poulet, G. M. Sheldrick, Chem. Commun., 1483 (1970). 13. G. L. Card, J. M. Shreeve, J. Am. Chem. Soc., 104, 5566 (1982). 14. D. Bielefeldt, A. Haas, Chem. Ber., 116, 1257 (1983).

15.2.12.6. Formation of the Cyclic Oligomers [NS(O)X],.

Unlike the isoelectronic cyclophosphazenes only six-membered rings have been well characterized for oligomers of the -NS(O)E(E = F, C1, alkyl, aryl or R,N) monomer unit. The trimeric chloride is best prepared by treating of SOCl, with sodium azide in acetonitrile at - 35°C ': 3 SOCl,

+ 3 NaN,

-

[NS(O)Cl],

+ 3 NaCl + 3 N,

(4

The azide route has also been used to make [NS(O)Ph], from PhS(0)Cl'. Sulfanuric chloride may also be obtained as a mixture of LY and /3 isomers in a twostage reaction from sulfamic acid or, preferably, sulfamoyl chloride,. The /3 isomer is stable in the solid state or in nonpolar solvents but changes to the LY isomer in ca. 1 h in acetonitrile: H,NSO,H

PClS

CI,P=NSO,Cl

- OPC13

[NS(O)Cl],

(b)

A third, but less convenient, way of making [NS(O)Cl], is by the oxidation of (NSCI), with SO, at 150°C and 20 atm4. The fluoride [NS(O)F], is obtained as a mixture of so-called cis and trans isomers by the fluorination of [NS(O)Cl],, preferably with SbF, This cis isomer shows a singlet in the "F N M R spectrum consistent with a structure in which all three fluorine atoms are on the same side of the S,N, ring {cf. [NS(O)Cl],), while the trans isomer exhibits an AB, pattern, indicating that one fluorine is on the side of the ring opposite to the other two6. Trimeric sulfanuric fluoride has also been obtained by the thermal decomposition of Cs[NS(O)F,], Hg[NS(O)F,], ', B[NS(O)F,], * or Me,Sn[NS(O)F,] ', which occurs at RT for the Sn compound. The aq hydrolysis of [NS(O)Cl], in the presence of Ag+ produces Ag,[(NSO,),], containing a six-membered ring in a chair conformation". The reaction of [NS(O)F], with benzene at reflux in the presence of AlCl, gives two isomers of [NS(O)Ph], Complete substitution of fluorines in [NS(O)F], by dimethylamino groups occurs upon reaction of the trimer with Me,NH at 80°C in the absence of a solvent12. Poorly characterized materials believed to be linear polymers, [N=S(O)F],, have been obtained as viscous yellow residues when the reaction product of ammonia and

'.

178

15.2. Ring-Ring and Ring-Polymer lnterconversions 15.2.12. in Sulfur-Nitrogen Oligomers and Polymers 15.2.12.7. Formation of the Cyclic Oligomer (SNH),.

SOF, is heated" and in the preparation of [NS(O)F], from the chloride and K F in acetonitrile 13. (T CHIVERS)

1. 2. 3. 4. 5. 6.

H. Kluver, 0. Glemser, 2. Naturforsch.,Teil B, 32, 1209 (1977). T. J. Maricich, J. Am. Chem. Soc., 90, 7179 (1968). T. J. Maricich, M. H. Khalil, Znorg. Chem.; 18, 912 (1974). T. Moeller, T.-H. Chang, A. Ouchi, A. Vandi, A. Failli, Inorg. Synth., 13, 9 (1972). T.-P. Lin, U. Klingebiel, 0. Glemser, Angew. Chem., Int. Ed. Engl., 11, 1095 (1972).

H. G. Heal, The Inorganic Heterocyclic Chemistry of Sulfur, Nitrogen and Phosphorus, Academic Press, London, 1980. This book has chapters on the cyclic oligomers formed by thiazyl halides, NSX, sulfanuric halides, [NS(O)X], and thionyl imide, HNSO. 7. 0. Glemser, Z. Naturforsch., Teil B, 31, 610 (1976). 8. A. Roland, W. Sundermeyer, 2. Naturforsch., Teil B, 27, 1102 (1972). 9. R. Hofer, 0. Glemser, Z. Naturforsch., Teil B, 27, 1106 (1972). 10. G. A. P. Dalgaard, A. C. Hazell, R. G. Hazell, Acta Crystallogr., 30B, 2721 (1974). 11. T. Moeller, A. Ouchi, J. Inorg. Nucl. Chem., 28, 2147 (1966). 12. H. Wagner, R. Mews, T.-P. Lin, 0. Glemser, Chem. Ber., 107, 584 (1974). 13. F. Seel, G. Simon, Angew. Chem., 72, 709 (1960).

15.2.12.7. Formation of the Cyclic Ollgomer (SNH),.

The parent molecule SNH, of the unstable organic thionitroso compounds is not known in the free state. The cyclic tetramer (SNH), is obtained in ca. 60 % yield by the reduction of S,N, in benzene or 1,2-dichloroethane at 80°C with methanolic SnCl,-2 H,O'. It is also produced in > 80 % yield by the electrolytic reduction of (SN), in acetonitrile in the presence of acetic acid'. (SN),

+ 4 HA + 4 e-

-

(SNH),

+ 4 A-

(a)

A variety of alkyl derivatives (SNR), (R = Me, Et, Bz, P-PhEt) are prepared by the

condensation reaction of the appropriate primary amine with SC1, '. Derivatives of the type (SNCH,OR), (R = H, Me, p-NO,C,H,) and (SNCOR), (R = Me, PhNH) are also available by standard procedures4. The pyrolysis of (NSR), (R = H, Me) at temperatures above 750°C produces the corresponding sulfur diimides RNSNR according to analysis of the vapors by field ionization mass and photoelectron spectros~opies~. (T. CHIVERS)

1. G. Brauer, Handbook of Preparative Inorganic Chemistry, Vol. 1, 2nd ed., Academic Press, New York, 1963, p. 411. 2. M. Hojo, Bull. Chem. SOC.Jpn., 53, 2856 (1980). 3. A. L. MacDonald, J. Trotter, Can. J. Chem., 51, 2504 (1973). 4. H. G. Heal, The Inorganic Heterocyclic Chemistry of Sulfur, Nitrogen and Phosphorus, Academic Press, London, 1980. This book has chapters on the cyclic oligomers formed by thiazyl halides, NSX, sulfanuric halides, [NS(O)X], and thionyl imide, HNSO. 5. L. Carlson, H. Egsgaard, S. Elbel, Sulfur Lett., 3, 87 (1985).

15.2.12.8. Formation of Thionyl lmide Polymers (HNSO),.

The monomer HNSO is obtained by reacting SOCl, with ammonia in the gas phase'. Liquid HNSO polymerizes above -70°C to give initially the cyclic tetramer

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc. 178

15.2. Ring-Ring and Ring-Polymer lnterconversions 15.2.12. in Sulfur-Nitrogen Oligomers and Polymers 15.2.12.7. Formation of the Cyclic Oligomer (SNH),.

SOF, is heated" and in the preparation of [NS(O)F], from the chloride and K F in acetonitrile 13. (T CHIVERS)

1. 2. 3. 4. 5. 6.

H. Kluver, 0. Glemser, 2. Naturforsch.,Teil B, 32, 1209 (1977). T. J. Maricich, J. Am. Chem. Soc., 90, 7179 (1968). T. J. Maricich, M. H. Khalil, Znorg. Chem.; 18, 912 (1974). T. Moeller, T.-H. Chang, A. Ouchi, A. Vandi, A. Failli, Inorg. Synth., 13, 9 (1972). T.-P. Lin, U. Klingebiel, 0. Glemser, Angew. Chem., Int. Ed. Engl., 11, 1095 (1972).

H. G. Heal, The Inorganic Heterocyclic Chemistry of Sulfur, Nitrogen and Phosphorus, Academic Press, London, 1980. This book has chapters on the cyclic oligomers formed by thiazyl halides, NSX, sulfanuric halides, [NS(O)X], and thionyl imide, HNSO. 7. 0. Glemser, Z. Naturforsch., Teil B, 31, 610 (1976). 8. A. Roland, W. Sundermeyer, 2. Naturforsch., Teil B, 27, 1102 (1972). 9. R. Hofer, 0. Glemser, Z. Naturforsch., Teil B, 27, 1106 (1972). 10. G. A. P. Dalgaard, A. C. Hazell, R. G. Hazell, Acta Crystallogr., 30B, 2721 (1974). 11. T. Moeller, A. Ouchi, J. Inorg. Nucl. Chem., 28, 2147 (1966). 12. H. Wagner, R. Mews, T.-P. Lin, 0. Glemser, Chem. Ber., 107, 584 (1974). 13. F. Seel, G. Simon, Angew. Chem., 72, 709 (1960).

15.2.12.7. Formation of the Cyclic Ollgomer (SNH),.

The parent molecule SNH, of the unstable organic thionitroso compounds is not known in the free state. The cyclic tetramer (SNH), is obtained in ca. 60 % yield by the reduction of S,N, in benzene or 1,2-dichloroethane at 80°C with methanolic SnCl,-2 H,O'. It is also produced in > 80 % yield by the electrolytic reduction of (SN), in acetonitrile in the presence of acetic acid'. (SN),

+ 4 HA + 4 e-

-

(SNH),

+ 4 A-

(a)

A variety of alkyl derivatives (SNR), (R = Me, Et, Bz, P-PhEt) are prepared by the

condensation reaction of the appropriate primary amine with SC1, '. Derivatives of the type (SNCH,OR), (R = H, Me, p-NO,C,H,) and (SNCOR), (R = Me, PhNH) are also available by standard procedures4. The pyrolysis of (NSR), (R = H, Me) at temperatures above 750°C produces the corresponding sulfur diimides RNSNR according to analysis of the vapors by field ionization mass and photoelectron spectros~opies~. (T. CHIVERS)

1. G. Brauer, Handbook of Preparative Inorganic Chemistry, Vol. 1, 2nd ed., Academic Press, New York, 1963, p. 411. 2. M. Hojo, Bull. Chem. SOC.Jpn., 53, 2856 (1980). 3. A. L. MacDonald, J. Trotter, Can. J. Chem., 51, 2504 (1973). 4. H. G. Heal, The Inorganic Heterocyclic Chemistry of Sulfur, Nitrogen and Phosphorus, Academic Press, London, 1980. This book has chapters on the cyclic oligomers formed by thiazyl halides, NSX, sulfanuric halides, [NS(O)X], and thionyl imide, HNSO. 5. L. Carlson, H. Egsgaard, S. Elbel, Sulfur Lett., 3, 87 (1985).

15.2.12.8. Formation of Thionyl lmide Polymers (HNSO),.

The monomer HNSO is obtained by reacting SOCl, with ammonia in the gas phase'. Liquid HNSO polymerizes above -70°C to give initially the cyclic tetramer

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc. 178

15.2. Ring-Ring and Ring-Polymer lnterconversions 15.2.12. in Sulfur-Nitrogen Oligomers and Polymers 15.2.12.7. Formation of the Cyclic Oligomer (SNH),.

SOF, is heated" and in the preparation of [NS(O)F], from the chloride and K F in acetonitrile 13. (T CHIVERS)

1. 2. 3. 4. 5. 6.

H. Kluver, 0. Glemser, 2. Naturforsch.,Teil B, 32, 1209 (1977). T. J. Maricich, J. Am. Chem. Soc., 90, 7179 (1968). T. J. Maricich, M. H. Khalil, Znorg. Chem.; 18, 912 (1974). T. Moeller, T.-H. Chang, A. Ouchi, A. Vandi, A. Failli, Inorg. Synth., 13, 9 (1972). T.-P. Lin, U. Klingebiel, 0. Glemser, Angew. Chem., Int. Ed. Engl., 11, 1095 (1972).

H. G. Heal, The Inorganic Heterocyclic Chemistry of Sulfur, Nitrogen and Phosphorus, Academic Press, London, 1980. This book has chapters on the cyclic oligomers formed by thiazyl halides, NSX, sulfanuric halides, [NS(O)X], and thionyl imide, HNSO. 7. 0. Glemser, Z. Naturforsch., Teil B, 31, 610 (1976). 8. A. Roland, W. Sundermeyer, 2. Naturforsch., Teil B, 27, 1102 (1972). 9. R. Hofer, 0. Glemser, Z. Naturforsch., Teil B, 27, 1106 (1972). 10. G. A. P. Dalgaard, A. C. Hazell, R. G. Hazell, Acta Crystallogr., 30B, 2721 (1974). 11. T. Moeller, A. Ouchi, J. Inorg. Nucl. Chem., 28, 2147 (1966). 12. H. Wagner, R. Mews, T.-P. Lin, 0. Glemser, Chem. Ber., 107, 584 (1974). 13. F. Seel, G. Simon, Angew. Chem., 72, 709 (1960).

15.2.12.7. Formation of the Cyclic Ollgomer (SNH),.

The parent molecule SNH, of the unstable organic thionitroso compounds is not known in the free state. The cyclic tetramer (SNH), is obtained in ca. 60 % yield by the reduction of S,N, in benzene or 1,2-dichloroethane at 80°C with methanolic SnCl,-2 H,O'. It is also produced in > 80 % yield by the electrolytic reduction of (SN), in acetonitrile in the presence of acetic acid'. (SN),

+ 4 HA + 4 e-

-

(SNH),

+ 4 A-

(a)

A variety of alkyl derivatives (SNR), (R = Me, Et, Bz, P-PhEt) are prepared by the

condensation reaction of the appropriate primary amine with SC1, '. Derivatives of the type (SNCH,OR), (R = H, Me, p-NO,C,H,) and (SNCOR), (R = Me, PhNH) are also available by standard procedures4. The pyrolysis of (NSR), (R = H, Me) at temperatures above 750°C produces the corresponding sulfur diimides RNSNR according to analysis of the vapors by field ionization mass and photoelectron spectros~opies~. (T. CHIVERS)

1. G. Brauer, Handbook of Preparative Inorganic Chemistry, Vol. 1, 2nd ed., Academic Press, New York, 1963, p. 411. 2. M. Hojo, Bull. Chem. SOC.Jpn., 53, 2856 (1980). 3. A. L. MacDonald, J. Trotter, Can. J. Chem., 51, 2504 (1973). 4. H. G. Heal, The Inorganic Heterocyclic Chemistry of Sulfur, Nitrogen and Phosphorus, Academic Press, London, 1980. This book has chapters on the cyclic oligomers formed by thiazyl halides, NSX, sulfanuric halides, [NS(O)X], and thionyl imide, HNSO. 5. L. Carlson, H. Egsgaard, S. Elbel, Sulfur Lett., 3, 87 (1985).

15.2.12.8. Formation of Thionyl lmide Polymers (HNSO),.

The monomer HNSO is obtained by reacting SOCl, with ammonia in the gas phase'. Liquid HNSO polymerizes above -70°C to give initially the cyclic tetramer

15.2. Ring-Ring and Ring-Polymer lnterconversions 15.2.12. in Sulfur-Nitrogen Oligomers and Polymers 15.2.12.9. S-N Oligomers with Terminal Organic Groups.

179

(HNSO), 2 , 3 , which is also formed by air oxidation of molten (HNS), ’. At RT the yellow, metastable polymer changes to a stable, yellow-brown high polymer, which is a semicond~ctor~~~. (T. CHIVERS)

1. P. W. Schenk, Chem. Ber., 75, 94 (1942). 2. P. W. Schenk, Monatsh. Chem., 95, 710 (1964). 3. J. F. May, G. Vallet, Rev. Gen. Elec., 81, 255 (1972). 4. E. Fluck, M. Becke-Goehring, Z. Anorg. Allg. Chem., 292, 229 (1957).

15.2.12.9. S-N

Oligomers with Terminal Organic Groups.

The reaction of (SN), with diphenyldiazomethane (or fluorenyl diazomethane) produces bis(diphenylmethy1ene)trisulfur tetranitride, Ph,CNSNSNSNCPh, I, in which the central S,N, unit of the S-N chain adopts a planar, cis,cisconformation2. These materials are deeply colored, photoconductive pigments,. An alternative synthesis is: 2 Ph,CNSiMe,

+ 2 SCl,

-

I

2 Ph,CNSCl

+ 2 Me,SiCl

(a 1

MezSiNSNSiMe,

(Ph2C)2S3N4

-

Structurally similar derivatives of the S,N4 chain can be prepared:

NSF + LiN(SiMe,)R’ where R’ = t-Bu, SiMe,; (SN),

+ PhSO,NCl,

-

(PhSO,),S,N,

R;S,N,

(b)

+ S4N4Cl, + S,N,Cl

(c)~

The compound (Me,Si),S,N, may also be synthesized from Me,SiNSNSiMe, and SCl, in a 2:l molar ratio’. The reaction of Me,SiNSNSiMe, with aryl sulfenyl chlorides produces neutral and cationic S-N chains end capped by aryl groupss3g. Me,SiNSNSiMe,

+ ArSCl

Me,SiNSNSiMe,

-

-

ArSNSNSiMe,

+ 3 ArSCl

sc12

ArSNSNSNSNSAr

[ArSNSNSNSArlCl

(4 (e) (T. CHIVERS)

1. E. Fluck, Z. Anorg. Chem., 312, 195 (1961). 2. E. M. Holt, S. L. Holt, K. J. Watson, J . Chern. SOC.,Dalton Trans., 1357 (1974). 3. B. Grushkin, US. Pat. 3,615,409 and 3,616,393, October 25,1971; Chem.Abstr., 74,40,262 (1972); 76, 40,283 (1972). 4. T. Chivers, R. T. Oakley, R. Pieters, J. F. Richardson, Can. J. Chern., 43, 1063 (1985). 5. W. Isenberg, R. Mews, G. M. Sheldrick, 2. Anorg. Allg. Chem., 525, 54 (1985). 6. H. W. Roesky, J. Sundermeyer, M. Noltemeyer, G. M. Sheldrick, K. Meyer-Base, P. G. Jones, Z . Naturforsch., Teil B, 41, 53 (1986). 7. W. Lidy, W. Sundermeyer, W. Verbeek, 2. Anorg. Allg. Chem., 404, 228 (1974). 8. J. Kuyper, G. B. Street, J. Am. Chem. SOC.,99, 7848 (1979). 9. J. J. Mayerle, J. Kuyper, G. B. Street, J. Am. Chem. Soc., 17, 2610 (1978).

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc. 15.2. Ring-Ring and Ring-Polymer lnterconversions 15.2.12. in Sulfur-Nitrogen Oligomers and Polymers 15.2.12.9. S-N Oligomers with Terminal Organic Groups.

179

(HNSO), 2 , 3 , which is also formed by air oxidation of molten (HNS), ’. At RT the yellow, metastable polymer changes to a stable, yellow-brown high polymer, which is a semicond~ctor~~~. (T. CHIVERS)

1. P. W. Schenk, Chem. Ber., 75, 94 (1942). 2. P. W. Schenk, Monatsh. Chem., 95, 710 (1964). 3. J. F. May, G. Vallet, Rev. Gen. Elec., 81, 255 (1972). 4. E. Fluck, M. Becke-Goehring, Z. Anorg. Allg. Chem., 292, 229 (1957).

15.2.12.9. S-N

Oligomers with Terminal Organic Groups.

The reaction of (SN), with diphenyldiazomethane (or fluorenyl diazomethane) produces bis(diphenylmethy1ene)trisulfur tetranitride, Ph,CNSNSNSNCPh, I, in which the central S,N, unit of the S-N chain adopts a planar, cis,cisconformation2. These materials are deeply colored, photoconductive pigments,. An alternative synthesis is: 2 Ph,CNSiMe,

+ 2 SCl,

-

I

2 Ph,CNSCl

+ 2 Me,SiCl

(a 1

MezSiNSNSiMe,

(Ph2C)2S3N4

-

Structurally similar derivatives of the S,N4 chain can be prepared:

NSF + LiN(SiMe,)R’ where R’ = t-Bu, SiMe,; (SN),

+ PhSO,NCl,

-

(PhSO,),S,N,

R;S,N,

(b)

+ S4N4Cl, + S,N,Cl

(c)~

The compound (Me,Si),S,N, may also be synthesized from Me,SiNSNSiMe, and SCl, in a 2:l molar ratio’. The reaction of Me,SiNSNSiMe, with aryl sulfenyl chlorides produces neutral and cationic S-N chains end capped by aryl groupss3g. Me,SiNSNSiMe,

+ ArSCl

Me,SiNSNSiMe,

-

-

ArSNSNSiMe,

+ 3 ArSCl

sc12

ArSNSNSNSNSAr

[ArSNSNSNSArlCl

(4 (e) (T. CHIVERS)

1. E. Fluck, Z. Anorg. Chem., 312, 195 (1961). 2. E. M. Holt, S. L. Holt, K. J. Watson, J . Chern. SOC.,Dalton Trans., 1357 (1974). 3. B. Grushkin, US. Pat. 3,615,409 and 3,616,393, October 25,1971; Chem.Abstr., 74,40,262 (1972); 76, 40,283 (1972). 4. T. Chivers, R. T. Oakley, R. Pieters, J. F. Richardson, Can. J. Chern., 43, 1063 (1985). 5. W. Isenberg, R. Mews, G. M. Sheldrick, 2. Anorg. Allg. Chem., 525, 54 (1985). 6. H. W. Roesky, J. Sundermeyer, M. Noltemeyer, G. M. Sheldrick, K. Meyer-Base, P. G. Jones, Z . Naturforsch., Teil B, 41, 53 (1986). 7. W. Lidy, W. Sundermeyer, W. Verbeek, 2. Anorg. Allg. Chem., 404, 228 (1974). 8. J. Kuyper, G. B. Street, J. Am. Chem. SOC.,99, 7848 (1979). 9. J. J. Mayerle, J. Kuyper, G. B. Street, J. Am. Chem. Soc., 17, 2610 (1978).

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc. 180

15.2. Ring-Ring and Ring-Polymer Interconversions 15.2.13. Polymers Containing Metal-Oxygen and Metal-Carbon Bonds 15.2.13.1. Introduction.

15.2.12.10. S-Polymers

with Organic Groups in the Backbone.

The band structure of the hypothetical polymer [cis-(R)CNSN], indicates that it should be a conducting material like (SN), but such polymers have not been prepared. Conducting polymers with p-phenylene groups in the backbone are obtained by extending the metathetical reactions described in the previous section':

',

C1SCsH4SCl

Me,SiNSN-C,H,-NSNSiMe,

I

SCl2

1

-

[-C,H,NSNSC,H,SNSN-1,

(a)

,1 [-C,H,NSNSNSN-], -

X

Doping of these polymers with acceptors such as I,, Br, or AsF, increases the conductivity to ca. lo-, Q-' cm-' '. Polymers with -PhSNSNSPhor -PhSNSNSPhSNSNSPnsegments separated by flexible spaces have been synthesize3. These insulators are converted to semiconductors when doped with Br, '. (T. CHIVERS)

1. M. H. Whangbo, R. Hoffman, R. B. Woodward, Proc. Roy. SOC.London, A, 366, 23 (1979). 2. 0. J. Scherer, G. Wolmershauser, R. Jotter, Z . Narurforsch., Teil B, 37, 432 (1982). 3. J. C. W. Chien, S . Ramakrishnan, Macromolecules, 21, 2007 (1988).

15.2.13. Polymers Containing Metal-Oxygen and Metal-Carbon Bonds 15.2.13.1. Introduction. Discussion of polymers containing the M-C, M-0 and M-0-C moieties within the polymer backbone suffers from poorly, and at times, inaccurately characterized products; products with structures that vary according to the specific reaction conditions-a dependency typically not recognized-and whose structures may vary even within a given chain; and few definitive or sufficiently broad studies to permit generalizations. The synthetic routes to these polymers are analogous to those employed in inorganic and organometallic small-molecule reactions, but reaction conditions are more stringent since the presence of monofunctional materials (resulting from side reactions such as hydrolysis) can alter the resulting chain length or structure of the product. Further, certain basic synthetic tools are excluded. Thus, the use of temperature to drive a reaction to produce polymer may be excluded where either the monomers or desired product is thermally unstable, or where the high temperature encourages side reactions such as solvolysis or oxidation, the subject has been reviewed'-' (C E CARRAHER, JR )

1. C . Carraher, J. Schats, C. Pittman, Advances in Organometallic and Inorganic Polymers, Marcel Dekker, New York, 1982.

2. M. Zeldin, K. Wynne, H. Allcock, Inorganic and Organometallic Polymers, Am. Chem. SOC., Washington, DC, 1988.

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc. 180

15.2. Ring-Ring and Ring-Polymer Interconversions 15.2.13. Polymers Containing Metal-Oxygen and Metal-Carbon Bonds 15.2.13.1. Introduction.

15.2.12.10. S-Polymers

with Organic Groups in the Backbone.

The band structure of the hypothetical polymer [cis-(R)CNSN], indicates that it should be a conducting material like (SN), but such polymers have not been prepared. Conducting polymers with p-phenylene groups in the backbone are obtained by extending the metathetical reactions described in the previous section':

',

C1SCsH4SCl

Me,SiNSN-C,H,-NSNSiMe,

I

SCl2

1

-

[-C,H,NSNSC,H,SNSN-1,

(a)

,1 [-C,H,NSNSNSN-], -

X

Doping of these polymers with acceptors such as I,, Br, or AsF, increases the conductivity to ca. lo-, Q-' cm-' '. Polymers with -PhSNSNSPhor -PhSNSNSPhSNSNSPnsegments separated by flexible spaces have been synthesize3. These insulators are converted to semiconductors when doped with Br, '. (T. CHIVERS)

1. M. H. Whangbo, R. Hoffman, R. B. Woodward, Proc. Roy. SOC.London, A, 366, 23 (1979). 2. 0. J. Scherer, G. Wolmershauser, R. Jotter, Z . Narurforsch., Teil B, 37, 432 (1982). 3. J. C. W. Chien, S . Ramakrishnan, Macromolecules, 21, 2007 (1988).

15.2.13. Polymers Containing Metal-Oxygen and Metal-Carbon Bonds 15.2.13.1. Introduction. Discussion of polymers containing the M-C, M-0 and M-0-C moieties within the polymer backbone suffers from poorly, and at times, inaccurately characterized products; products with structures that vary according to the specific reaction conditions-a dependency typically not recognized-and whose structures may vary even within a given chain; and few definitive or sufficiently broad studies to permit generalizations. The synthetic routes to these polymers are analogous to those employed in inorganic and organometallic small-molecule reactions, but reaction conditions are more stringent since the presence of monofunctional materials (resulting from side reactions such as hydrolysis) can alter the resulting chain length or structure of the product. Further, certain basic synthetic tools are excluded. Thus, the use of temperature to drive a reaction to produce polymer may be excluded where either the monomers or desired product is thermally unstable, or where the high temperature encourages side reactions such as solvolysis or oxidation, the subject has been reviewed'-' (C E CARRAHER, JR )

1. C . Carraher, J. Schats, C. Pittman, Advances in Organometallic and Inorganic Polymers, Marcel Dekker, New York, 1982.

2. M. Zeldin, K. Wynne, H. Allcock, Inorganic and Organometallic Polymers, Am. Chem. SOC., Washington, DC, 1988.

15.2.13. Polymers Containing Metal-Oxygen and Metal-Carbon Bonds 15.2.13.2. M-C-Containing Polymers 15.2.13.2.1. Metallocenes.

181

3. C. Pittman, C. Carraher, J. Reynolds, in Encyclopedia ofpolymer Science and Engineering, Vol. 10, 2nd ed., J. I. Kraschwitz, ed., New York, 1987, pp. 541-594. 4. J. Sheats, C. Carraher, C. Pittman, Metal-Containing Polymeric Systems, Plenum Press, New York, 1985. 5. P. Brooke, H. Schurmans. J. Verhoest, Inorganic Fibers and Composite Materials, Pergamon Press, Oxford, 1984. 6. R. D. Archer, in Encyclopedia of Materials Science and Engineering, Vol. 3, M. Bover, ed., Pergamon Press, Oxford, 1986, p. 235. 7 . A. Cowley, Rings, Clusters and Polymers of the Main Group Elements, Am. Chem. SOL, Washington, DC, 1983. 8. B. Culbertson, C. Pittman, New Monomers and Polymers, Plenum Press, New York, 1984. 9. F. Hartley, Support Metal Complexes, Reidel, Boston, 1985. 10. G. Wilkinson, F. Stone, Comprehensive Organometallic Chemistry, Pergamon Press, Oxford, 1982. 11. H. Carraher, C. Pittman, in Application ofPolymers, R. Seymour, ed., Plenum Press, New York, 1988.

15.2.13.2. M-C-Containing

Polymers

15.2.13.2.1. Metallocenes.

Metallocene polymers, which contain a metal sandwiched usually between the cyclopentadienyl moieties, are the most investigated M-C-containing polymers'-4. Ferrocene containing metallocenes are by far the most studied, followed by cobalticenium, with some work where the metal is Ni, Rh, Ro or Ru. Polymer formation can occur through a variety of reactions. Metallocene polyamides, pol yethers, polyesters, polyhydrazides and polyurethanes have been synthesized through condensation reactions of 1,l'-difunctional metallocenes (typically ferrocene derivatives) with diacid halides (if the metallocene derivative contains functional electron-pair donor bases) and electron-pair donor bases (as diamines, hydrazines, diisocyanates, diols) if the metallocene derivative contains functional electron-pair acceptor acids 3 , 5-9. CP H O H , C O Fe o C H , O H

I

+ Cl-Ti-CI I

CP

-

CP

f H 2 C 0 Fe o C H , O - T ' i I- O ~

I

CP

0 0

0

C 1 g O Fe o ! C l +

HOROH

-

0

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc. 15.2.13. Polymers Containing Metal-Oxygen and Metal-Carbon Bonds 15.2.13.2. M-C-Containing Polymers 15.2.13.2.1. Metallocenes.

181

3. C. Pittman, C. Carraher, J. Reynolds, in Encyclopedia ofpolymer Science and Engineering, Vol. 10, 2nd ed., J. I. Kraschwitz, ed., New York, 1987, pp. 541-594. 4. J. Sheats, C. Carraher, C. Pittman, Metal-Containing Polymeric Systems, Plenum Press, New York, 1985. 5. P. Brooke, H. Schurmans. J. Verhoest, Inorganic Fibers and Composite Materials, Pergamon Press, Oxford, 1984. 6. R. D. Archer, in Encyclopedia of Materials Science and Engineering, Vol. 3, M. Bover, ed., Pergamon Press, Oxford, 1986, p. 235. 7 . A. Cowley, Rings, Clusters and Polymers of the Main Group Elements, Am. Chem. SOL, Washington, DC, 1983. 8. B. Culbertson, C. Pittman, New Monomers and Polymers, Plenum Press, New York, 1984. 9. F. Hartley, Support Metal Complexes, Reidel, Boston, 1985. 10. G. Wilkinson, F. Stone, Comprehensive Organometallic Chemistry, Pergamon Press, Oxford, 1982. 11. H. Carraher, C. Pittman, in Application ofPolymers, R. Seymour, ed., Plenum Press, New York, 1988.

15.2.13.2. M-C-Containing

Polymers

15.2.13.2.1. Metallocenes.

Metallocene polymers, which contain a metal sandwiched usually between the cyclopentadienyl moieties, are the most investigated M-C-containing polymers'-4. Ferrocene containing metallocenes are by far the most studied, followed by cobalticenium, with some work where the metal is Ni, Rh, Ro or Ru. Polymer formation can occur through a variety of reactions. Metallocene polyamides, pol yethers, polyesters, polyhydrazides and polyurethanes have been synthesized through condensation reactions of 1,l'-difunctional metallocenes (typically ferrocene derivatives) with diacid halides (if the metallocene derivative contains functional electron-pair donor bases) and electron-pair donor bases (as diamines, hydrazines, diisocyanates, diols) if the metallocene derivative contains functional electron-pair acceptor acids 3 , 5-9. CP H O H , C O Fe o C H , O H

I

+ Cl-Ti-CI I

CP

-

CP

f H 2 C 0 Fe o C H , O - T ' i I- O ~

I

CP

0 0

0

C 1 g O Fe o ! C l +

HOROH

-

0

182

15.2. Ring-Ring and Ring-Polymer Interconversions 15.2.13. Polymers Containing Metal-Oxygen and Metal-Carbon Bonds 15.2.13.2. M-C-Containing Polymers

Metallocene moieties can be inserted into already formed polymers as crosslinking agents" but no industrial applications have resulted. Polymetal metallocenes have been synthesized utilizing a variety of reactants to aive: v

where M is Zn, Sn, Zr, Ti, Hg, Sb or Si. The exact structures can be complex, varying even within a given chain. Polymers have been formed through reaction of the metallocene itself with the cyclopentadienyl moiety acting in an aromatic manner3,4,9:

0 0+ Fe

0

11

RCH

electron-pair acceptor acid

(el H

Cobalticenium salt containing polymers offer the advantages of good resistance to strong acids, bases and oxidation, but the disadvantages of salt exchange. The metallocenes utilized exhibit some catalytic properties that may be transmitted into polymers containing these m e t a l l o ~ e n e s ~ ~ ~ ~ ~ ~ ~ '8'.

Where Y = PF,, C1, Br, I, NO,, 0-C,H,. Polysilicone-linked ferrocenes have been synthesized from silyldiolbis(aminosilany1)ferrocene condensed with a d i ~ l ' ~ , ' ~ : R,NZ R

00 Fe

ZNR, R

+ HOR'OH

-

A C6H5CH3

by reaction of cyclic ferrocene siloxane with phenyllithium followed by reaction with a difunctional chlorosilane' 5,16:

Fe

'0

+ C,H5Li + Fe

15.2.13. Polymers Containing Metal-Oxygen and Metal-Carbon Bonds 15.2.13.2. M-C-Containing Polymers 15.2.13.2.1. Metallocenes.

by self-condensation of suitable diols' ','*

4

Fe @R'SiOSiR% R R

R

183

(i)

R R

by reaction of 1,l'-dilithioferrocene and dihalo~rganosilanes~~~~'~: R

R

and through hydrosilylation" :

0 0

R,SiCH=CH, Fe

R,SiCH=CH,

R,SiH

+

0 0 Fe

HzPtC16

-f(CH,%Si

0 0: 0 "-0 R

Fe

Sij,

(k)

R,SiH

Selected products' 3,14 give flexible films, are fiber forming and exhibit reasonable thermal hydrolytic stability. Additional reviews have (C.E CARRAHER, JR.)

1. C. Carraher, J. Sheats, C. Pittman, Organometallic Polymers, Academic Press, New York, 1978. 2. C. Pittman, Organic Reactions, Vol.6, E. Becker and M. Tsutsui, eds., Plenum Press, New York, 1977. 3. E. Neuse, H. Rosenberg, Metallocene Polymers, Marcel Dekker, New York, 1970. 4. A. H. Gerber, E. F. McInerney, Survey of Inorganic Polymers, NASA Cr-159563 HRI-396, Contract No. NA53-21369, 1979; Chem. Abstr. 92,94,921 (1980). 5. C. Carraher, Interfacial Synthesis, Vol. II, Applications and Technology, F. Millich and C. Carraher, eds., Marcel Dekker, New York, 1977, Ch. 21. 6. C. Carraher, G. Burrish, J. Macromol. Sci. Chem., AIOC87, 1457 (1976). 7. C. Carraher, J. Sheats, Makromol. Chem., 166, 23 (1973). 8. C. Carraher, M. J. Christensen, Angew. Makromol. Chem., 69, 61 (1978). 9. J. Sheats, Chem. Rev., in press. 10. A. Volozhin, L. Verhovodka, A. Rozmyslova, Y. Paushkin, Vysokomol.Soedin., Ser. B, 19, 593 (1977); Chem. Abstr., 87, 1,532,354 (1977). 11. C. Carraher, Inorg. Macromol. Revs., I , 271 (1972). 12. C. Carraher, G. Peterson, J. Sheats, Organic Coatings and Plastics Chemistry, 33, 427 (1973). 13. W. Patterson, S. McManns, C. Pittman, J. Polymer Sci., 12, 837 (1974). 14. C. Pittman, W. Patterson, S. McManus, J. Polymer Sci., Polymer Chem. Ed., 14, 1715 (1976). 15. P. Kan, C. Lenk, R. Schaaf, J. Org. Chem., 26,4038 (1961). 16. R. Schaaf, US. Pat; 3,036,105 (1962); Chem. Abstr., 57, 16,656 (1962). 17. E. V. Wilkus, A. Berger, Fr. Pat. 1,396,271 (1965): Chem. Abstr., 63, 3077 (1965). 18. E. V. Wilkus, W. Rauscher, J. Org. Chem., 30,2889 (1965). 19. H. Rosenberg, U.S. Patent 3,426,053 (1969); Chem Abstr., 70, 78,551 (1972). 20. G. Gerber, M. L. Hallensleben, Makromol. Chem., 104, 77 (1967). 21. C. Carraher, L. Tisinger, M. Williams, I. Lopez, Polym. Muter., 58, 239 (1988). 22. C. Carraher, R. Linville, D. Stevison, V. Foster, M. Williams, M. J. Aloi, Polym. Muter., 58, 85 (1988). 23. Y. Naoshima, C. Carraher, S. Iwamore, H. Shido, Appl. Organomet. Chem., 1, 245 (1987).

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc. 184

15.2. Ring-Ring and Ring-Polymer Interconversions 15.2.13. Polymers Containing Metal-Oxygen and Metal-Carbon Bonds 15.2.13.2. M-C-Containing Polymers

15.2.13.2.2. Polycarbosilanes, Polysilarylenes, Polycarbosiloxanes and Poly(carborane-siloxanes).

Polycarbosilanes can exhibit thermal stabilities greater than the analogous polysiloxane1,2,contrary to their relative bond energies (Si-C, 78 kcalmol- ; Si-0, 106 kcal mol- '). Thus polydimethylsilane is polymerized by heating the hexamer to 3O0-5OOcC, forming a product with thermal stability greater than 500°C and that can be spun into fiber, which in turn can be thermally oxidized, forming /%silicon carbide3, fSi(CH,),-fn. Polysilarylenes are synthesized using organomagnesium or organolithium intermediates or by use of Na coupling A more direct, and less costly, route utilizes electron-pair acceptor acid ~atalysis"~:

'

With the exception of thermal data, the properties of polycarbosilanes and polysilarylenes are not well documented, although preliminary work indicates such products may offer thermal, oxidative and chemical stabilities above those of the polysilonanes. Polycarbosiloxanes are derived by the condensation of bisphenols: R,SiCl,

+ HOR'OH

-€:1 -tP, za SiOR'O

(b)

by reaction of silyldiamines with bisphenols": R,Si(NH,),

+ HOR'OH

SiOR'O

(c)

and by condensation of diols and bis(ani1ino)diphenylsilanes'' : H (RNf,SiR,

+ HOR'OH

The products show good thermal stabilities and are film forming. The strong interest in poly(carborane-siloxanes) is primarily due to their thermal The synthesis of elastomers with a 50°C gain in the use temperature over corresponding polysiloxanes has been a technical success, yet, owing to their high cost, industrial applicability is doubtful. They are synthesized by condensation of the carborane-siloxane halide with a diether15: R R ClSi-CB,,H,,C-Sic1 R R

R + ROSi-CB,,H,,C-SiOR' R

FeCls

R

Further reviews have appeared'6-'8. (C E CARRAHER, JR.)

15.2.13. Polymers Containing Metal-Oxygen and Metal-Carbon Bonds 15.2.13.2. M-C-Contai ni ng Polymers 15.2.13.23. Other Metallorganic Compounds.

185

~

1. J. P. Wesson, T. C. Williams, Organosilane Polymers. I. Poly(dimethylsilylene), Tech. Rept. Contract No. N00014-75-C-1024, ADA050 337/5ST, 1977. 2 N. S. Nametkin, V. M. Vdovin, V. I. Zav’yalov, Vysokomol.Soedin., 7, 757 (1965); 3. S. Yajima, Y. Hasegama, K. Okamura, T. Matsuzawa, Nature (London),273, 525 (1978). 4. E. A. Hoess, E. L. O’Brien, Preparation of Polysilanes and Polysilarylenea, Tech. Rept. No. 3289, Picatinny Arsenal, Dover, NJ, 1966. 5. A. Meston, US. Pat. 3,136,730 (1964). 6. C. A. Pearce, K. A. Hadd, P. G. Chantrell, Br. Pat. 896,301 (1962); Chem. Abstr., 57,4881 (1963). 7. H. A. Clark, US. Pat. 2,557,782 (1951). 8. R Moroni, E. Dumont, Ger. Pat. 1,145,798 (1963); Chem. Abstr., 59, 4063 (1963). 9. M. Jaboric, 2. Anorg. Allg. Chem., 288, 324 (1956). 10. W. R. Dunnavant, R. A. Markle, R. G. Sinclair, P. B. Stickney, J. E. Curry, J. D. Bird, Macromolecules, I, 249 (1968). 11. Y. K. Kim, D. B. Bourrie, 0. R. Pierce, J . Polymer Sci., Polymer Chem. Ed., 16, 483 (1978). 12. H. A. Schroeder, horg. Macromol. Revs., I , 45 (1970). 13. H. A Schroeder, Carborane Polymers, A Summary, Contract No. N0014-71-C-0003, Ad 742444, 1972. 14. W. von Strumpf, Chem. Z. 99, 416 (1975). 15. S. Papetti, B. B. Schaeffer, A. P. Gray, T. L. Heyling, J. Polym. Scz., 4, 1623 (1966). 16. J. Tretzel, H. Achtsnit, J. Bohler, M. Grimpel, H. Herwig, A. Wegerhoff, Production, Processing and Application of Enka Silicon Fibers, 57th Annual TRI Conference, Charlotte, NC, April 21-22, 1987. 17. B. Arkles, W. Peterson, R. Anderson, Organosilicon Compounds, Petrarch Systems, Bristol, PA, 1982. 18. C. Carraher, Polyrn. News, 13 ( l l ) , 341 (1988). 15.2.13.2.3. Other Metallorganic Compounds.

Metallorganic carborane polymers a r e synthesized by reaction of group-IV organometallic dihalides with dialkali-metal derivatives of meta- and para-carboranes1.2.

Hydride addition reactions a r e utilized t o synthesize a number of g r o u p IVAcontaining p ~ l y r n e r s ~: - ~

R RtCH=CHzj2

1

CHz-CHz-M-CH,-CH,-R

1

R or no catalyst

R

1

CH=CH-M-CH=CH-R

1

R

T h e products a r e oligomeric t o high polymers (molecular weights a b o u t lo5).

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc. 15.2.13. Polymers Containing Metal-Oxygen and Metal-Carbon Bonds 15.2.13.2. M-C-Contai ni ng Polymers 15.2.13.23. Other Metallorganic Compounds.

185

~

1. J. P. Wesson, T. C. Williams, Organosilane Polymers. I. Poly(dimethylsilylene), Tech. Rept. Contract No. N00014-75-C-1024, ADA050 337/5ST, 1977. 2 N. S. Nametkin, V. M. Vdovin, V. I. Zav’yalov, Vysokomol.Soedin., 7, 757 (1965); 3. S. Yajima, Y. Hasegama, K. Okamura, T. Matsuzawa, Nature (London),273, 525 (1978). 4. E. A. Hoess, E. L. O’Brien, Preparation of Polysilanes and Polysilarylenea, Tech. Rept. No. 3289, Picatinny Arsenal, Dover, NJ, 1966. 5. A. Meston, US. Pat. 3,136,730 (1964). 6. C. A. Pearce, K. A. Hadd, P. G. Chantrell, Br. Pat. 896,301 (1962); Chem. Abstr., 57,4881 (1963). 7. H. A. Clark, US. Pat. 2,557,782 (1951). 8. R Moroni, E. Dumont, Ger. Pat. 1,145,798 (1963); Chem. Abstr., 59, 4063 (1963). 9. M. Jaboric, 2. Anorg. Allg. Chem., 288, 324 (1956). 10. W. R. Dunnavant, R. A. Markle, R. G. Sinclair, P. B. Stickney, J. E. Curry, J. D. Bird, Macromolecules, I, 249 (1968). 11. Y. K. Kim, D. B. Bourrie, 0. R. Pierce, J . Polymer Sci., Polymer Chem. Ed., 16, 483 (1978). 12. H. A. Schroeder, horg. Macromol. Revs., I , 45 (1970). 13. H. A Schroeder, Carborane Polymers, A Summary, Contract No. N0014-71-C-0003, Ad 742444, 1972. 14. W. von Strumpf, Chem. Z. 99, 416 (1975). 15. S. Papetti, B. B. Schaeffer, A. P. Gray, T. L. Heyling, J. Polym. Scz., 4, 1623 (1966). 16. J. Tretzel, H. Achtsnit, J. Bohler, M. Grimpel, H. Herwig, A. Wegerhoff, Production, Processing and Application of Enka Silicon Fibers, 57th Annual TRI Conference, Charlotte, NC, April 21-22, 1987. 17. B. Arkles, W. Peterson, R. Anderson, Organosilicon Compounds, Petrarch Systems, Bristol, PA, 1982. 18. C. Carraher, Polyrn. News, 13 ( l l ) , 341 (1988). 15.2.13.2.3. Other Metallorganic Compounds.

Metallorganic carborane polymers a r e synthesized by reaction of group-IV organometallic dihalides with dialkali-metal derivatives of meta- and para-carboranes1.2.

Hydride addition reactions a r e utilized t o synthesize a number of g r o u p IVAcontaining p ~ l y r n e r s ~: - ~

R RtCH=CHzj2

1

CHz-CHz-M-CH,-CH,-R

1

R or no catalyst

R

1

CH=CH-M-CH=CH-R

1

R

T h e products a r e oligomeric t o high polymers (molecular weights a b o u t lo5).

186

15.2. Ring-Ring and Ring-Polymer Interconversions 15.2.13. Polymers Containing Metal-Oxygen and Metal-Carbon Bonds 15.2.13.2. M-C-Containi ng Polymers

~~~

Oligomeric products are formed from transition-metal carbonyls with diacid chlorides3.

(possible structure) Metal dimethylphosphonium bis(methy1ide) polymers are prepared6 from trimethylphosphorus methylide, (CH,),P=CH,, and an organometallic CR,M:

CH, ‘CH, These products are hydrolytically unstable and decompose < 200°C. Additional reviews have appeared7-I9. (C.E. CARRAHER, JR.)

1. H. A. Schroeder, U.S. Patent 3,533,968 (1970).

S . Bresadola, F. Rossetto, G. Tagliavini, Eur. Poly. J., 4, 75 (1968). E. G. Rochow, P.. L. Stern, U.S. Patent 3,291,783 (1966). L. Lunera, A. Sladkov, V. V. Korshak, Vysokomol. Soedin., 7,427 (1965). CA 63, 1879e (1965). J. Noltes, G. J. M. van der Kerk, Recl. Trao. Chim. Pays Bas, 80,623 (1961); 81,41 (1962). H. Schmidbam, J. Eberlein, Z . Anorg. Allg. Chern., 434, 145 (1977). C. Carraher, J. Schats, C. Pittman, Adljances in Organometallic and Inorganic Polymers, Marcel Dekker, New York, 1982. 8. M. Zeldin, K. Wynne, H. Allcock, Inorganic and Organometallic Polymers, Am. Chem. SOC., Washington, DC, 1988. 9. C. Pittman, C. Carraher, J. Reynolds, in Encyclopedia of Polymer Science and Engineering, Vol. 10, 2nd ed., J. I. Kraschwitz, ed., Wiley, New York, 1987, pp. 541-594. 10. J. Sheats, C. Carraher, C. Pittman, Metal-Containing Polymeric Systems, Plenum Press, New York, 1985. 11. P. Brooke, H. Schurmans. J. Verhoest, Inorganic Fibers and Composite Materials, Pergamon Press, Oxford, 1984. 12. R. D. Archer, in Encyclopedia of Materials Science and Engineering, Vol. 3, M. Bever, ed., Pergamon Press, Oxford, 1986, p. 235. 13. A. Cowley, Rings, Clusters and Polymers of the Main Group Elements, Am. Chem. SOC., Washington, DC, 1983. 14. B. Culbertson, C. Pittman, New Monomers and Polymers, Plenum Press, New York, 1984. 15. F. Hartley, Support Metal Complexes, Reidel, Boston, 1985. 2. 3. 4. 5. 6. 7.

15.2. Ring-Ring and Ring-Polymer Interconversions 15.2.13. Polymers Containing Metal-Oxygen and Metal-Carbon Bonds 15.2.13.3. M-0-Containing Polymers

187

16. G. Wilkinson, F. Stone, Comprehensive Organometallic Chemistry, Pergamon Press, Oxford, 1982. 17. H. Carraher, C. Pittman, in Application ofPolymers, R. Seymour, ed., Plenum Press, New York, 1988.

18. A. Meller, Gremlin's Handbuch der Anorganische Chemie, Boron Compounds, Suppl. 2, SpringerVerlag, Berlin, 1983. 19. J. Carlsson, Encyclopedia of Materials Science and Engineering, M. Bever, ed., Pergamon Press, Oxford, 1986.

15.2.13.3. M-0-Containing

Polymers

15.2.13.3.1. Inorganic Polysilicas.

Sand and rock are network inorganic silica polymers containing Si-0 moieties whose structured formula approximates S O , . Glass is also such a polymer derived from sand. Glasses are three-dimensional polymers whereas fibrous or ladder silica is linear and unidimensional. Fibrous silica readily reacts with water or alcohol, forming metasilicic acid or a polymeric ester of metasilic acid:

OH

I

OH

l

OH

l

Reactions between particular silica and electron-pair donar bases are the basis of attempts to form polysiloxanes from natural sources1'2. (C.E. CARRAHER)

1. R. Atwal, B. R. Curell, C . B. Cook, H. G. Midgley, J. R. Parsonage, in Organometallic Polymers C. Carraher, J. Sheats, C. Pittman, eds., Academic Press, New York, 1978, Ch. 23. 2. B. R. Currell, J. R. Parsonage, Advances in Organometallic Polymers, C. Carraher, J. Sheats, and C. Pittman, eds., Marcel Dekker, New York, 1981. 3. C . Carraher, Polym. News, I3(11), 341 (1988). 15.2.13.3.2. Polymetallosiloxanes.

Siloxanes are considered in $152.8, but other polymers contain siloxane-like structures. Metallosiloxane polymers have been synthesized in attempts to enhance the thermal properties of the product beyond the siloxanes themselves. Elements from groups IIIB (Al), IVB [Ge, Sn(2+,4+), Pb], VB [As(3+, 4+), Sb], and transition metals [Co, Cu(2+), Cr(6+), Fe(3+), Hg(2+), Mn, Ni, Ti(4+), Zn, Zr(4')l have been employed

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc. 15.2. Ring-Ring and Ring-Polymer Interconversions 15.2.13. Polymers Containing Metal-Oxygen and Metal-Carbon Bonds 15.2.13.3. M-0-Containing Polymers

187

16. G. Wilkinson, F. Stone, Comprehensive Organometallic Chemistry, Pergamon Press, Oxford, 1982. 17. H. Carraher, C. Pittman, in Application ofPolymers, R. Seymour, ed., Plenum Press, New York, 1988.

18. A. Meller, Gremlin's Handbuch der Anorganische Chemie, Boron Compounds, Suppl. 2, SpringerVerlag, Berlin, 1983. 19. J. Carlsson, Encyclopedia of Materials Science and Engineering, M. Bever, ed., Pergamon Press, Oxford, 1986.

15.2.13.3. M-0-Containing

Polymers

15.2.13.3.1. Inorganic Polysilicas.

Sand and rock are network inorganic silica polymers containing Si-0 moieties whose structured formula approximates S O , . Glass is also such a polymer derived from sand. Glasses are three-dimensional polymers whereas fibrous or ladder silica is linear and unidimensional. Fibrous silica readily reacts with water or alcohol, forming metasilicic acid or a polymeric ester of metasilic acid:

OH

I

OH

l

OH

l

Reactions between particular silica and electron-pair donar bases are the basis of attempts to form polysiloxanes from natural sources1'2. (C.E. CARRAHER)

1. R. Atwal, B. R. Curell, C . B. Cook, H. G. Midgley, J. R. Parsonage, in Organometallic Polymers C. Carraher, J. Sheats, C. Pittman, eds., Academic Press, New York, 1978, Ch. 23. 2. B. R. Currell, J. R. Parsonage, Advances in Organometallic Polymers, C. Carraher, J. Sheats, and C. Pittman, eds., Marcel Dekker, New York, 1981. 3. C . Carraher, Polym. News, I3(11), 341 (1988). 15.2.13.3.2. Polymetallosiloxanes.

Siloxanes are considered in $152.8, but other polymers contain siloxane-like structures. Metallosiloxane polymers have been synthesized in attempts to enhance the thermal properties of the product beyond the siloxanes themselves. Elements from groups IIIB (Al), IVB [Ge, Sn(2+,4+), Pb], VB [As(3+, 4+), Sb], and transition metals [Co, Cu(2+), Cr(6+), Fe(3+), Hg(2+), Mn, Ni, Ti(4+), Zn, Zr(4')l have been employed

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc. 15.2. Ring-Ring and Ring-Polymer Interconversions 15.2.13. Polymers Containing Metal-Oxygen and Metal-Carbon Bonds 15.2.13.3. M-0-Containing Polymers

187

16. G. Wilkinson, F. Stone, Comprehensive Organometallic Chemistry, Pergamon Press, Oxford, 1982. 17. H. Carraher, C. Pittman, in Application ofPolymers, R. Seymour, ed., Plenum Press, New York, 1988.

18. A. Meller, Gremlin's Handbuch der Anorganische Chemie, Boron Compounds, Suppl. 2, SpringerVerlag, Berlin, 1983. 19. J. Carlsson, Encyclopedia of Materials Science and Engineering, M. Bever, ed., Pergamon Press, Oxford, 1986.

15.2.13.3. M-0-Containing

Polymers

15.2.13.3.1. Inorganic Polysilicas.

Sand and rock are network inorganic silica polymers containing Si-0 moieties whose structured formula approximates S O , . Glass is also such a polymer derived from sand. Glasses are three-dimensional polymers whereas fibrous or ladder silica is linear and unidimensional. Fibrous silica readily reacts with water or alcohol, forming metasilicic acid or a polymeric ester of metasilic acid:

OH

I

OH

l

OH

l

Reactions between particular silica and electron-pair donar bases are the basis of attempts to form polysiloxanes from natural sources1'2. (C.E. CARRAHER)

1. R. Atwal, B. R. Curell, C . B. Cook, H. G. Midgley, J. R. Parsonage, in Organometallic Polymers C. Carraher, J. Sheats, C. Pittman, eds., Academic Press, New York, 1978, Ch. 23. 2. B. R. Currell, J. R. Parsonage, Advances in Organometallic Polymers, C. Carraher, J. Sheats, and C. Pittman, eds., Marcel Dekker, New York, 1981. 3. C . Carraher, Polym. News, I3(11), 341 (1988). 15.2.13.3.2. Polymetallosiloxanes.

Siloxanes are considered in $152.8, but other polymers contain siloxane-like structures. Metallosiloxane polymers have been synthesized in attempts to enhance the thermal properties of the product beyond the siloxanes themselves. Elements from groups IIIB (Al), IVB [Ge, Sn(2+,4+), Pb], VB [As(3+, 4+), Sb], and transition metals [Co, Cu(2+), Cr(6+), Fe(3+), Hg(2+), Mn, Ni, Ti(4+), Zn, Zr(4')l have been employed

188

15.2. Ring-Ring and Ring-Polymer interconversions 15.2.13. Polymers Containing Metal-Oxygen and Metal-Carbon Bonds 15.2.13.3. M-0-Containing Polymers

but the resulting poly(e1ement siloxane) typically exhibits poorer thermal and hydrolytic stabilities than the parent polysiloxanei-3. R

R

fS i O M M O

fSiOM 0% R

R

Introduction of other metal-oxygen linkages effects increased polarization above that found for the S i - 0 moiety, resulting in an increased tendency toward nucleophilic attack and subsequent polymer degradation. Tin-containing siloxane polymers are prepared throueh reaction of a disilanol with a substituted Sn oxide or organotin dihalide4:

R ClSiCl R

R

+ HOSiOH R

-

R R' fSnO-SiO+" R R'

Similar products are formed from the metathesis of Na salts of organosilane triols with tin(1V) chloride5. Much research effort is focusing on the synthesis and application of silicate sols and gels in glasses, ceramics, ceramers, protective coatings for glass, adhesives for glass surfaces, semiconductive films and fibers6-I6. (C.E. CARRAHER, JR.)

1. V. V. Krivonishchenkor, G. A. Semerneva, A. L. Suvorov, Zh. Prikl. Spektrosk, 23,475 (1978); Chem. Abstr., 88, 191,740 (1978). 2. K. A. Andrianov, N. A. Kurashera, L. I. Kuteinikova, B. D. Lavrukhin, E. D. Gurrits, Vysokomol.Soedin. Ser. A , 17, 1049 (1975). 3. D. C. Bradley, J. W. Lorimer, C. Prevedorou-Demas, Can. J. Chem., 49, 2310 (1971). 4. W. E. Foster, P. E. Koenig, U.S. Pat 2,998,440 (1956). 5. E. 2. Asnovich, K. A. Andrianov, Vysokomol.Soedin., 4,216,26 (1962) Chem. Abstr., 56, 15,664b (1962). 6. D. Hofer, R. Miller, C. Willson, J. SOC.Photo.-Opt. Instrum. SOC.,469, 16 (1984). 7. D. Hofer, R. Miller, C. Willson, A. Neureuther, SPIE J., 469, 108 (1984). 8. C. Brinker, D. Clark, D. Ulrich, Bettcr Ceramics Through Chemistry, 11, Mater. Res. Soc., Pittsburgh, PA, 1986. 9. L. Hench, D. Ulrich, Science of Ceramic Chemical Processing, Wiley, New York, 1986. 10. M. Zeldin, K. Wynn, H. Allcock, Inorganic and Organometallic Polymers, Am. Chem. Soc., Washington, DC, 1988. 11. M. Schmidt, H. Scholze, G. Tunker, J . Non-Cryst. Solids, 80, 557 (1986). 12. G. Phillip, H. Schmidt, J . Non-Cryst. Solids, 63, 261 (1984). 13 S. Satka, Polyrn. Preprints, 28(1), 430 (1987). 14. G. Tunker, H. Patzelt, H. Schmidt, H. Scholze, Glastechn. Ber., 59, 272 (1986). 15. H. Dislich, Angew. Chew., Int. Ed. Engl., 106, 363 (1970). 16. B. Yoldas, J. Met. Sci., 12, 203 (1977); 14, 1843 (1979). 15.2.13.3.3. Poly(silylary1ene siloxanes).

Poly(silylary1ene-siloxanes) can be synthesized as alternating, block or segmental polymers'. R R +SiArSiOfn R R

R R R R R € S i A r ~ i O ~ ~ ~ ( S i O ~ S ~ A r S i O ) ~ ~ R R R R R

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc. 188

15.2. Ring-Ring and Ring-Polymer interconversions 15.2.13. Polymers Containing Metal-Oxygen and Metal-Carbon Bonds 15.2.13.3. M-0-Containing Polymers

but the resulting poly(e1ement siloxane) typically exhibits poorer thermal and hydrolytic stabilities than the parent polysiloxanei-3. R

R

fS i O M M O

fSiOM 0% R

R

Introduction of other metal-oxygen linkages effects increased polarization above that found for the S i - 0 moiety, resulting in an increased tendency toward nucleophilic attack and subsequent polymer degradation. Tin-containing siloxane polymers are prepared throueh reaction of a disilanol with a substituted Sn oxide or organotin dihalide4:

R ClSiCl R

R

+ HOSiOH R

-

R R' fSnO-SiO+" R R'

Similar products are formed from the metathesis of Na salts of organosilane triols with tin(1V) chloride5. Much research effort is focusing on the synthesis and application of silicate sols and gels in glasses, ceramics, ceramers, protective coatings for glass, adhesives for glass surfaces, semiconductive films and fibers6-I6. (C.E. CARRAHER, JR.)

1. V. V. Krivonishchenkor, G. A. Semerneva, A. L. Suvorov, Zh. Prikl. Spektrosk, 23,475 (1978); Chem. Abstr., 88, 191,740 (1978). 2. K. A. Andrianov, N. A. Kurashera, L. I. Kuteinikova, B. D. Lavrukhin, E. D. Gurrits, Vysokomol.Soedin. Ser. A , 17, 1049 (1975). 3. D. C. Bradley, J. W. Lorimer, C. Prevedorou-Demas, Can. J. Chem., 49, 2310 (1971). 4. W. E. Foster, P. E. Koenig, U.S. Pat 2,998,440 (1956). 5. E. 2. Asnovich, K. A. Andrianov, Vysokomol.Soedin., 4,216,26 (1962) Chem. Abstr., 56, 15,664b (1962). 6. D. Hofer, R. Miller, C. Willson, J. SOC.Photo.-Opt. Instrum. SOC.,469, 16 (1984). 7. D. Hofer, R. Miller, C. Willson, A. Neureuther, SPIE J., 469, 108 (1984). 8. C. Brinker, D. Clark, D. Ulrich, Bettcr Ceramics Through Chemistry, 11, Mater. Res. Soc., Pittsburgh, PA, 1986. 9. L. Hench, D. Ulrich, Science of Ceramic Chemical Processing, Wiley, New York, 1986. 10. M. Zeldin, K. Wynn, H. Allcock, Inorganic and Organometallic Polymers, Am. Chem. Soc., Washington, DC, 1988. 11. M. Schmidt, H. Scholze, G. Tunker, J . Non-Cryst. Solids, 80, 557 (1986). 12. G. Phillip, H. Schmidt, J . Non-Cryst. Solids, 63, 261 (1984). 13 S. Satka, Polyrn. Preprints, 28(1), 430 (1987). 14. G. Tunker, H. Patzelt, H. Schmidt, H. Scholze, Glastechn. Ber., 59, 272 (1986). 15. H. Dislich, Angew. Chew., Int. Ed. Engl., 106, 363 (1970). 16. B. Yoldas, J. Met. Sci., 12, 203 (1977); 14, 1843 (1979). 15.2.13.3.3. Poly(silylary1ene siloxanes).

Poly(silylary1ene-siloxanes) can be synthesized as alternating, block or segmental polymers'. R R +SiArSiOfn R R

R R R R R € S i A r ~ i O ~ ~ ~ ( S i O ~ S ~ A r S i O ) ~ ~ R R R R R

15.2.13. Polymers Containing Metal-Oxygen and Metal-Carbon Bonds 15.2.13.3. M-0-Containing Polymers 15.2.13.3.3. Poly(silylarylene si loxanes).

They can be synthesized diary1)hydroxy~ilylarylenes~~~ :

by

R R HOSiArSiOH R R

self-condensation

-H~O

of

p-bis(dialky1-

189

or

R R +SiArSiO+” R R

by condensation of hydrides with hydroxyl-containing ~ilylarylenes~: R R HOSiArSiOH R R

R

R

R R fSiArSiO3 R R

+ HSiArSiH R

R

by condensation of dihalosilanes with silyldiols5: RLSiCI,

R

R

R

R

R R R’ fOSiArSiOSi3 R R R’

+ HOSiArSiOH

by reaction with silazanes and silamines6: R R HOSiArSiOH R R

R” R”

+ RiNSiRSiNR; R” R”

-

R R R”R” +OSiArSiOSiRSi+n R R R”R”

( 4

by reaction of bis(ureidosi1anes) with ~ i l y l d i o l:s ~ ~ ~ R R HOSiArSiOH R R

0

+ R;Si+NCN R“ll

D2

R

R R

--f 0gArFO$+

and by condensation of silanolacetoxysilanes with silyldiolsg: R R HOSiArSiOH R R

0

0

IIR R‘ 11 R R R + RCOSiO+Si jmOCR-+SiArSiOSi*O-Si R‘ R‘ R R R’

R‘

jmIn

R’

(f)

(C E CARRAHER, JR.)

1. W. R. Dunnarant, Inorg. Macromol. Rev., 1, 165 (1971). 2. R. L. Merker, M. J. Scort, J. Polym. Sci., 2, 15 (1964). 3. R. M. Pike, J. Polym. Sci., 50, 151 (1961). 4. A. L. Klebanski, L. P. Fomina, S. B. Dolgoplaosk, Zh. Uses, Khim. Obscheh D.J.Mendeleeva, 7, 594 (1962); Chem. Abstr., 58, 4708g (1963). 5. E. G. Rochow, An Introduction to the Chemistry of Silicones, 2nd ed., John Wiley, New York, 1951. 6. L. W. Breed, R. L. Elliott, M. E. Whitehead, J. Polym. Sci., A l , 5, 2745 (1967). 7. R. W. Lenz, P. R. Dvornic, Polymer Preprints, 19, 857 (1978). 8. J. B. Davison, K. I. Wynne, Silicon-Phthaiocyanine Siioxane Polymers: Synthesis and Proton Nuclear Magnetic Resonance Study, USNTIS, AD A036 274, 1977. 9. H. Rosenberg, B. D. Nahlovsky, Polym Preprints, 19, 625 (1978).

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc. 190

15.2. Ring-Ring and Ring-Polymer Interconversions 15.2.13. Polymers Containing Metal-Oxygen and Metal-Carbon Bonds 15.2.13.3. M-0-Containing Polymers

15.2.13.3.4. Polyarylsilsesquioxanes.

The exact structure of polyarylsilsesquioxanes varies within a given chain and with synthesis but approximates a ladder s t r ~ c t u r e l - ~ .

They have reasonable cost, outstanding oxidative stability and good resistance to acidic and neutral hydrolysis. They are synthesized by the hydrolysis of aryltriethoxysilane in the presence of a quaternary base4 or the hydrolysis of trichlorosilanes5'6. Additives such as CuSiO, and variation in solvent and extent and temperature of post-heating change product properties. The polymers are soluble in typical organic solvents some have tensile strength of several thousand psi, elongation greater than 15% and a glass temperature in excess of 250°C. (C.E. CARRAHER, JR.)

1. R. J. Cotter, M. Matsner, Ring-Forming Polymerizations, Part A-Carbocyclic and Metallorganic Rings, Academic Press, New York, 1969. 2. K. A. Andrianov, Metalorganic Polymers (Eng. Transl.), Interscience, New York, 1965. 3. K. A. Andrianov, A. A. Zhdanov, V. Y. Levin, Ann. Rev. Mater. Sci., 8, 313 (1978). 4. J. F. Brown, L. H. Vogt, US.Pat. 3,017,386 (1962); Chem. Abstr., 57 13,993 (1963). 5. C . L. Frye, J. M. Klosowski, J. Am. Chem. SOC.,93,4599 (1971). 6. K. W. Krantz, Fr. Pat. 1,423,143 (1966); Chem. Abstr., 65, 12,361 (1966). 15.2.13.3.5. Polymetal Phosphinates and Polyphosphonatolanes.

Properties of poly(meta1 phosphinates) are dependent on the coordination metal and any accompanying ligand (such as H,O), the nature of the substituents on phosphorus and the number of bridging sites per center'. Poly(meta1 phosphinates) have been synthesized by reacting solutions of the metal ions with dialkyl- and diarylphosphinates.

where M = Be, Co, Zn, Al, Cr, Ni, Ti or Zn. Some poly(meta1 phosphinates) are high polymers with molecular weights < 1.5 x los, exhibit tensile strengths of several thousand psi, are soluble in organic solvents and can be cast as films and exhibit superior thermal stabilities to 500°C 2, yet no one member of this class has combined any three major properties. Related to the poly(meta1 phosphinates) are the Al-0-P products, polyphos-

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc. 190

15.2. Ring-Ring and Ring-Polymer Interconversions 15.2.13. Polymers Containing Metal-Oxygen and Metal-Carbon Bonds 15.2.13.3. M-0-Containing Polymers

15.2.13.3.4. Polyarylsilsesquioxanes.

The exact structure of polyarylsilsesquioxanes varies within a given chain and with synthesis but approximates a ladder s t r ~ c t u r e l - ~ .

They have reasonable cost, outstanding oxidative stability and good resistance to acidic and neutral hydrolysis. They are synthesized by the hydrolysis of aryltriethoxysilane in the presence of a quaternary base4 or the hydrolysis of trichlorosilanes5'6. Additives such as CuSiO, and variation in solvent and extent and temperature of post-heating change product properties. The polymers are soluble in typical organic solvents some have tensile strength of several thousand psi, elongation greater than 15% and a glass temperature in excess of 250°C. (C.E. CARRAHER, JR.)

1. R. J. Cotter, M. Matsner, Ring-Forming Polymerizations, Part A-Carbocyclic and Metallorganic Rings, Academic Press, New York, 1969. 2. K. A. Andrianov, Metalorganic Polymers (Eng. Transl.), Interscience, New York, 1965. 3. K. A. Andrianov, A. A. Zhdanov, V. Y. Levin, Ann. Rev. Mater. Sci., 8, 313 (1978). 4. J. F. Brown, L. H. Vogt, US.Pat. 3,017,386 (1962); Chem. Abstr., 57 13,993 (1963). 5. C . L. Frye, J. M. Klosowski, J. Am. Chem. SOC.,93,4599 (1971). 6. K. W. Krantz, Fr. Pat. 1,423,143 (1966); Chem. Abstr., 65, 12,361 (1966). 15.2.13.3.5. Polymetal Phosphinates and Polyphosphonatolanes.

Properties of poly(meta1 phosphinates) are dependent on the coordination metal and any accompanying ligand (such as H,O), the nature of the substituents on phosphorus and the number of bridging sites per center'. Poly(meta1 phosphinates) have been synthesized by reacting solutions of the metal ions with dialkyl- and diarylphosphinates.

where M = Be, Co, Zn, Al, Cr, Ni, Ti or Zn. Some poly(meta1 phosphinates) are high polymers with molecular weights < 1.5 x los, exhibit tensile strengths of several thousand psi, are soluble in organic solvents and can be cast as films and exhibit superior thermal stabilities to 500°C 2, yet no one member of this class has combined any three major properties. Related to the poly(meta1 phosphinates) are the Al-0-P products, polyphos-

15.2.13. Polymers Containing Metal-Oxygen and Metal-Carbon Bonds 15.2.13.3. M-0-Containing Polymers 15.2.13.3.6. Miscellaneous.

191

~~~

phonatolanes. These are synthesized through reaction of etherated aluminum hydride or dialkyl or dialkoxy haloalane with phosphinic acids’ or phosphinic esters3.

0

-

0

R,AlX

II

+ HO-P-OH I

R

II

R’-P-R’

X

moiety are Metal-containing polymers containing the M-0 and M-0-R synthesized with oxygen-containing chelating agents. These are considered in $15.2.15 and are reviewed in refs 4-6. Other studies also have (C.E. CARRAHER, JR.)

B. P. Block, Inorg. Macromol. Revs., 1, 115 (1970). D. L. Schmidt, E. E. Flagg, J. Polym. Sci., A-1, 6, 3235 (1968). R. F. Monroe, D. L. Schmidt, U.S. Pat. 3,497,464 (1970); Chem Abstr., 72, 122,177 (1970). C. Carraher, J. Chem. Ed., 58 ( l l ) , 921 (1981). C. Carraher, J. Sheats, C. Pittman, Organometallic Polymers, Academic Press, New York, 1978. A. H. Gerber, E. F. McInerney, Survey of Inorganic Polymers, Natl. Aeronaut. Space Admin., Cr0159563 HRJ-396, Contract No. NA53-21369, 1979. 7. C. Gebelein, C. Carraher, Advances in Biomedical Polymers, Plenum Press, New York, 1987. 8. L. Mathais, C. Carraher, Crown Ethers and Phase Transfer Catalysts for Polymer Applications, Plenum Press, New York, 1984, Ch. 8. 9. C. Carraher, R. Linville, H. Blaxall, Polym. Preprmts, 25, 31 (1984).

1. 2. 3. 4. 5. 6.

15.2.13.3.6. Miscellaneous.

Short-chain group-IVB products analogous to polysiloxanes can be formed from the hydrolysis of the dihaloorganometallic monomer’,’. The reaction of dialkyltin oxide with R,Si(OH), results in mixed polystannosil~xanes~~~: R,SnO

+ R,Si(OH),

-

R R f S i O ~ S n O ~ n R R

Similar reactions have been employed to form polystannoxanes and mixed lead-silicon polymers5. The lead-containing polymer, glycerin-litharge cement, exists as a mixture of chains and crosslinks6. (C.E. CARRAHER, JR.)

1. R. D. Grain, P. F. Koenig, WADC Tech. Rept. 69-427 (Jan. 1960). 2. F. Henglein, R. Lang, L. Schmack, Makromol. Chem., 22, 103 (1957). 3. D. L. Venezky, Encyclopedia ofPolymer Science and Tehnology, Vol. 7, John Wiley, New York, 1967. 4. F. G. A. Stone, W. A. G. Graham, Inorganic Polymers, Academic Press, New York, 1962.

0

II

15.2.13.4. M-0-C-Contalning

Polymers.

Compounds that are loosely referred to as metal polycarbonates have been synthesized but not well characterized’s2.

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc.

15.2.13. Polymers Containing Metal-Oxygen and Metal-Carbon Bonds 15.2.13.3. M-0-Containing Polymers 15.2.13.3.6. Miscellaneous.

191

~~~

phonatolanes. These are synthesized through reaction of etherated aluminum hydride or dialkyl or dialkoxy haloalane with phosphinic acids’ or phosphinic esters3.

0

-

0

R,AlX

II

+ HO-P-OH I

R

II

R’-P-R’

X

moiety are Metal-containing polymers containing the M-0 and M-0-R synthesized with oxygen-containing chelating agents. These are considered in $15.2.15 and are reviewed in refs 4-6. Other studies also have (C.E. CARRAHER, JR.)

B. P. Block, Inorg. Macromol. Revs., 1, 115 (1970). D. L. Schmidt, E. E. Flagg, J. Polym. Sci., A-1, 6, 3235 (1968). R. F. Monroe, D. L. Schmidt, U.S. Pat. 3,497,464 (1970); Chem Abstr., 72, 122,177 (1970). C. Carraher, J. Chem. Ed., 58 ( l l ) , 921 (1981). C. Carraher, J. Sheats, C. Pittman, Organometallic Polymers, Academic Press, New York, 1978. A. H. Gerber, E. F. McInerney, Survey of Inorganic Polymers, Natl. Aeronaut. Space Admin., Cr0159563 HRJ-396, Contract No. NA53-21369, 1979. 7. C. Gebelein, C. Carraher, Advances in Biomedical Polymers, Plenum Press, New York, 1987. 8. L. Mathais, C. Carraher, Crown Ethers and Phase Transfer Catalysts for Polymer Applications, Plenum Press, New York, 1984, Ch. 8. 9. C. Carraher, R. Linville, H. Blaxall, Polym. Preprmts, 25, 31 (1984).

1. 2. 3. 4. 5. 6.

15.2.13.3.6. Miscellaneous.

Short-chain group-IVB products analogous to polysiloxanes can be formed from the hydrolysis of the dihaloorganometallic monomer’,’. The reaction of dialkyltin oxide with R,Si(OH), results in mixed polystannosil~xanes~~~: R,SnO

+ R,Si(OH),

-

R R f S i O ~ S n O ~ n R R

Similar reactions have been employed to form polystannoxanes and mixed lead-silicon polymers5. The lead-containing polymer, glycerin-litharge cement, exists as a mixture of chains and crosslinks6. (C.E. CARRAHER, JR.)

1. R. D. Grain, P. F. Koenig, WADC Tech. Rept. 69-427 (Jan. 1960). 2. F. Henglein, R. Lang, L. Schmack, Makromol. Chem., 22, 103 (1957). 3. D. L. Venezky, Encyclopedia ofPolymer Science and Tehnology, Vol. 7, John Wiley, New York, 1967. 4. F. G. A. Stone, W. A. G. Graham, Inorganic Polymers, Academic Press, New York, 1962.

0

II

15.2.13.4. M-0-C-Contalning

Polymers.

Compounds that are loosely referred to as metal polycarbonates have been synthesized but not well characterized’s2.

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc.

15.2.13. Polymers Containing Metal-Oxygen and Metal-Carbon Bonds 15.2.13.3. M-0-Containing Polymers 15.2.13.3.6. Miscellaneous.

191

~~~

phonatolanes. These are synthesized through reaction of etherated aluminum hydride or dialkyl or dialkoxy haloalane with phosphinic acids’ or phosphinic esters3.

0

-

0

R,AlX

II

+ HO-P-OH I

R

II

R’-P-R’

X

moiety are Metal-containing polymers containing the M-0 and M-0-R synthesized with oxygen-containing chelating agents. These are considered in $15.2.15 and are reviewed in refs 4-6. Other studies also have (C.E. CARRAHER, JR.)

B. P. Block, Inorg. Macromol. Revs., 1, 115 (1970). D. L. Schmidt, E. E. Flagg, J. Polym. Sci., A-1, 6, 3235 (1968). R. F. Monroe, D. L. Schmidt, U.S. Pat. 3,497,464 (1970); Chem Abstr., 72, 122,177 (1970). C. Carraher, J. Chem. Ed., 58 ( l l ) , 921 (1981). C. Carraher, J. Sheats, C. Pittman, Organometallic Polymers, Academic Press, New York, 1978. A. H. Gerber, E. F. McInerney, Survey of Inorganic Polymers, Natl. Aeronaut. Space Admin., Cr0159563 HRJ-396, Contract No. NA53-21369, 1979. 7. C. Gebelein, C. Carraher, Advances in Biomedical Polymers, Plenum Press, New York, 1987. 8. L. Mathais, C. Carraher, Crown Ethers and Phase Transfer Catalysts for Polymer Applications, Plenum Press, New York, 1984, Ch. 8. 9. C. Carraher, R. Linville, H. Blaxall, Polym. Preprmts, 25, 31 (1984).

1. 2. 3. 4. 5. 6.

15.2.13.3.6. Miscellaneous.

Short-chain group-IVB products analogous to polysiloxanes can be formed from the hydrolysis of the dihaloorganometallic monomer’,’. The reaction of dialkyltin oxide with R,Si(OH), results in mixed polystannosil~xanes~~~: R,SnO

+ R,Si(OH),

-

R R f S i O ~ S n O ~ n R R

Similar reactions have been employed to form polystannoxanes and mixed lead-silicon polymers5. The lead-containing polymer, glycerin-litharge cement, exists as a mixture of chains and crosslinks6. (C.E. CARRAHER, JR.)

1. R. D. Grain, P. F. Koenig, WADC Tech. Rept. 69-427 (Jan. 1960). 2. F. Henglein, R. Lang, L. Schmack, Makromol. Chem., 22, 103 (1957). 3. D. L. Venezky, Encyclopedia ofPolymer Science and Tehnology, Vol. 7, John Wiley, New York, 1967. 4. F. G. A. Stone, W. A. G. Graham, Inorganic Polymers, Academic Press, New York, 1962.

0

II

15.2.13.4. M-0-C-Contalning

Polymers.

Compounds that are loosely referred to as metal polycarbonates have been synthesized but not well characterized’s2.

192

15.2. Ring-Ring and Ring-Polymer Interconversions 15.2.13. Polymers Containing Metal-Oxygen and Metal-Carbon Bonds 15.2.13.5. M-0-R (Polyether)-Containing Polymers.

Di- and polyvalent metal ions and divalent metal-containing moieties are reacted with organic di- (and higher functionality) carboxylate (acid, acid chloride or salt form) compounds forming these polycarboxylates. Divalent ions favor formation of linear products, while the use of trivalent ions leads to crosslinked products. Divalent cations utilized include Mg, Ca, Ba, Zn, Cd, Pb, Sn, Be and Mn. The metal polycarboxylates undergo reorganization or interchange, giving products whose structure varies with the particular reaction conditions and with time. The condensation of salts of dicarboxylic acids in aqueous based phases (includes dipolar aprotic-H,O and salt-H,O mixes) with a variety of metal-containing electronpair acceptor acids in an organic phase (interfacial) or water miscible phase, including water itself as the solvent (solution condensation) was effectively Metalcontaining reactants include R,TiX,, R,HDC,, R,SnX,, R,MnX,, MoO,Cl,, WO,Cl,, UO;', R,BiX, and R,SbX,. The structures of these polymers depend on the metal. Thus the tendency for internal carbonyl bridging increases for group-IVA polyesters in the order Pb, Sn, Ge, Si. This has been extended to include reactions with carboxylatecontaining polymers such as poly(rodium acrylate):

non bridging

bridging

Additional reviews have appeared'0'". (C.E. CARRAHER, JR.)

1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11.

J. Economy, J. H. Mason, L. C. 'Wohrer, Polym. Preprints, 7, 586 (1966).

L. Holliday, Inorg. Macromol. Rev., I , 3 (1970). S. Migdal, D. Gertner, A. Kilkha, J . Organomet. Chem., 11, 441 (1968). M. Iskenderor, K. Plekhanova, N. Adigezalora, Uch. Zap. Azerb. Gas. Unfr., Ser. Khim. Nauk, 4, 71 (1965). C. Carraher, R. L. Dammeier, Makromol. Chem., 141,251 (1971). C. Carraher, J. T. Reimer, J. Polym. Sci., Polymer Chem. Ed., 10, 3367 (1972). C. Carraher, Inorg. Macromol. Rev., I , 271 (1972). C. Carraher, Interfacial Synthesis, Vol. ZI, F. Millich, C. Carraher, eds., Marcel Dekker, New York, 1972. C. Carraher, J. Sheats, C. Pittman, Organometallic Polymers, Academic Press. New York, 1978. C. Carraher, L. Tisinger, W. Tisinger, Polyrn. Mater. 52, 177 (1985). J. Sheats, C. Carraher, C. Pittman, Metal-Containing Polymeric Systems, Plenum Press, New York, 1985.

15.2.13.5. M-0-R

(Polyether)-Containing Polymers.

Organometallic polyethers are formed utilizing low T condensation and addition systems'-6. Group-IVA and -B polymetalethers are synthesized utilizing nonaqueous interfacial systems and group-IVB polymetalethers are synthesized employing classical and inverse interfacial systems and aqueous solution systems: R,MX,

+ HOROH

-

R +MORO+"

R

(a)

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc. 192

15.2. Ring-Ring and Ring-Polymer Interconversions 15.2.13. Polymers Containing Metal-Oxygen and Metal-Carbon Bonds 15.2.13.5. M-0-R (Polyether)-Containing Polymers.

Di- and polyvalent metal ions and divalent metal-containing moieties are reacted with organic di- (and higher functionality) carboxylate (acid, acid chloride or salt form) compounds forming these polycarboxylates. Divalent ions favor formation of linear products, while the use of trivalent ions leads to crosslinked products. Divalent cations utilized include Mg, Ca, Ba, Zn, Cd, Pb, Sn, Be and Mn. The metal polycarboxylates undergo reorganization or interchange, giving products whose structure varies with the particular reaction conditions and with time. The condensation of salts of dicarboxylic acids in aqueous based phases (includes dipolar aprotic-H,O and salt-H,O mixes) with a variety of metal-containing electronpair acceptor acids in an organic phase (interfacial) or water miscible phase, including water itself as the solvent (solution condensation) was effectively Metalcontaining reactants include R,TiX,, R,HDC,, R,SnX,, R,MnX,, MoO,Cl,, WO,Cl,, UO;', R,BiX, and R,SbX,. The structures of these polymers depend on the metal. Thus the tendency for internal carbonyl bridging increases for group-IVA polyesters in the order Pb, Sn, Ge, Si. This has been extended to include reactions with carboxylatecontaining polymers such as poly(rodium acrylate):

non bridging

bridging

Additional reviews have appeared'0'". (C.E. CARRAHER, JR.)

1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11.

J. Economy, J. H. Mason, L. C. 'Wohrer, Polym. Preprints, 7, 586 (1966).

L. Holliday, Inorg. Macromol. Rev., I , 3 (1970). S. Migdal, D. Gertner, A. Kilkha, J . Organomet. Chem., 11, 441 (1968). M. Iskenderor, K. Plekhanova, N. Adigezalora, Uch. Zap. Azerb. Gas. Unfr., Ser. Khim. Nauk, 4, 71 (1965). C. Carraher, R. L. Dammeier, Makromol. Chem., 141,251 (1971). C. Carraher, J. T. Reimer, J. Polym. Sci., Polymer Chem. Ed., 10, 3367 (1972). C. Carraher, Inorg. Macromol. Rev., I , 271 (1972). C. Carraher, Interfacial Synthesis, Vol. ZI, F. Millich, C. Carraher, eds., Marcel Dekker, New York, 1972. C. Carraher, J. Sheats, C. Pittman, Organometallic Polymers, Academic Press. New York, 1978. C. Carraher, L. Tisinger, W. Tisinger, Polyrn. Mater. 52, 177 (1985). J. Sheats, C. Carraher, C. Pittman, Metal-Containing Polymeric Systems, Plenum Press, New York, 1985.

15.2.13.5. M-0-R

(Polyether)-Containing Polymers.

Organometallic polyethers are formed utilizing low T condensation and addition systems'-6. Group-IVA and -B polymetalethers are synthesized utilizing nonaqueous interfacial systems and group-IVB polymetalethers are synthesized employing classical and inverse interfacial systems and aqueous solution systems: R,MX,

+ HOROH

-

R +MORO+"

R

(a)

193 15.2. Ring-Ring and Ring-Polymer Interconversions 15.2.13. Polymers Containing Metal-Oxygen and Metal-Carbon Bonds Polymers (Polyamidoximes and Polyoximes). 15.2.13.6. M-0-N-Containing

The polymers are linear, are soluble in dipolar aprotic solvents and exhibit good weight retention (80 % plus) in air and N, to 1000°C. The modification of various celluloses, including dextran and cotton, has been accomplished through condensation with RMX, and R,MX, those with M = Sn show reduced flammability, and show widespread antifungal and antibacterial activity and are hydrophobic. Mixed polymetaletheresters are synthesized utilizing xanthene and phenylsulfonphthalein dyes9-I4:

where M = Ti, Zr, Hf, Sn, etc.; X = C1, Br, I; R = aliphatic, aromatic. These polydyes show activity against bacteria and fungi. (C.E CARRAHER, JR.)

1. C. Carraher, Inorg. Macromol. Rev., 1, 271 (1972). 2. C. Carraher, Interfacial Synthesis, Vol. II, F. Millich, C. Carraher, eds., Marcel Dekker, New York, 1977, Ch. 21. 3. C. Carraher, J. Polym. Sci., A - I , 7, 2351, 2359 (1969). 4. C. Carraher, Makromol. Chem., 160, 259 (1972). 5. C. Carraher, L. M. Jambaya, Angew. Makromol. Chem., 52, 111 (1976). 6. C. Carraher, S. T. Bajah, Polymer, 15, 9 (1974). 7. C. Carraher, T. Gehrke, D. Giron, R. Cervents, unreported results. 8. C. Carraher, J. A. Schroeder, C. McNeely, D. J. Giron, J. H. Workman, Organ. Coatings Plastics Chem. 40, 560 (1979). 9. C. Carraher, R. Schwarz, J. Schroeder, M. Schwarz, M. Molloy, Interfacial Synthesis, Vol. III Recent Advances, C. Carraher, J. Preston, eds., Marcel Dekker, New York, 1981. 10. C. Carraher, R. Schwarz, J. Schroeder, M. Schwarz, H. M. Molloy, Organ. Coatings Plastics Chem., 43, 798 (1979). 11. C. Carraher, L. Tisinger, G. Solimine, M. Williams, S. Carraher, R. Strother, Polym. Mater., 55, 469 (1986). 12. C. Carraher, V. Foster, R. Linville, D. Stevison, Polym. Mater., 56, 401 (1987). 13. C. Carraher, R. Linville, D. Stevison, V. Foster, M. Williams, M. Aloi, Polym. Mater., 58, 8 5 (1988). 14. C. Carraher, M. Williams, L. Tisinger, I. Lopez, Polym. Mater., 58, 239 (1988). 15. Y. Naoshima, C. Carraher, H. Shido, M. Uenishi, Polym. Mater., 58, 553 (1988). 16. C. Carraher, V. Foster, R. Linville, D. Stevison, R. Ventatachalam, Polymers in the Hostel Environment. 17. L. H. Lee, eds, Plenum Press, New York, 1988, Ch. 19. 18. C. Carraher, Y. Naoshima, C. Butler, V. Foster, D. Gill, M. Williams, D. Giron, P. Mykytiuk, Polym. Mater., 57, 186 (1987). 19. C. Carraher, L. Speiling, Renewable Resource Materials, Plenum Press, New York, 1986.

15.2.13.6. M-0-N-Containing Polyoximes).

Polymers (Polyamidoximes and

Metal-containing polyamidoximes and polyoximes are synthesized from organometallic dihalides and divalent metal-containing moieties with diamidoximes and dioximes. These reactions are conducted using the aqueous interfacial and solution reaction

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc. 193 15.2. Ring-Ring and Ring-Polymer Interconversions 15.2.13. Polymers Containing Metal-Oxygen and Metal-Carbon Bonds Polymers (Polyamidoximes and Polyoximes). 15.2.13.6. M-0-N-Containing

The polymers are linear, are soluble in dipolar aprotic solvents and exhibit good weight retention (80 % plus) in air and N, to 1000°C. The modification of various celluloses, including dextran and cotton, has been accomplished through condensation with RMX, and R,MX, those with M = Sn show reduced flammability, and show widespread antifungal and antibacterial activity and are hydrophobic. Mixed polymetaletheresters are synthesized utilizing xanthene and phenylsulfonphthalein dyes9-I4:

where M = Ti, Zr, Hf, Sn, etc.; X = C1, Br, I; R = aliphatic, aromatic. These polydyes show activity against bacteria and fungi. (C.E CARRAHER, JR.)

1. C. Carraher, Inorg. Macromol. Rev., 1, 271 (1972). 2. C. Carraher, Interfacial Synthesis, Vol. II, F. Millich, C. Carraher, eds., Marcel Dekker, New York, 1977, Ch. 21. 3. C. Carraher, J. Polym. Sci., A - I , 7, 2351, 2359 (1969). 4. C. Carraher, Makromol. Chem., 160, 259 (1972). 5. C. Carraher, L. M. Jambaya, Angew. Makromol. Chem., 52, 111 (1976). 6. C. Carraher, S. T. Bajah, Polymer, 15, 9 (1974). 7. C. Carraher, T. Gehrke, D. Giron, R. Cervents, unreported results. 8. C. Carraher, J. A. Schroeder, C. McNeely, D. J. Giron, J. H. Workman, Organ. Coatings Plastics Chem. 40, 560 (1979). 9. C. Carraher, R. Schwarz, J. Schroeder, M. Schwarz, M. Molloy, Interfacial Synthesis, Vol. III Recent Advances, C. Carraher, J. Preston, eds., Marcel Dekker, New York, 1981. 10. C. Carraher, R. Schwarz, J. Schroeder, M. Schwarz, H. M. Molloy, Organ. Coatings Plastics Chem., 43, 798 (1979). 11. C. Carraher, L. Tisinger, G. Solimine, M. Williams, S. Carraher, R. Strother, Polym. Mater., 55, 469 (1986). 12. C. Carraher, V. Foster, R. Linville, D. Stevison, Polym. Mater., 56, 401 (1987). 13. C. Carraher, R. Linville, D. Stevison, V. Foster, M. Williams, M. Aloi, Polym. Mater., 58, 8 5 (1988). 14. C. Carraher, M. Williams, L. Tisinger, I. Lopez, Polym. Mater., 58, 239 (1988). 15. Y. Naoshima, C. Carraher, H. Shido, M. Uenishi, Polym. Mater., 58, 553 (1988). 16. C. Carraher, V. Foster, R. Linville, D. Stevison, R. Ventatachalam, Polymers in the Hostel Environment. 17. L. H. Lee, eds, Plenum Press, New York, 1988, Ch. 19. 18. C. Carraher, Y. Naoshima, C. Butler, V. Foster, D. Gill, M. Williams, D. Giron, P. Mykytiuk, Polym. Mater., 57, 186 (1987). 19. C. Carraher, L. Speiling, Renewable Resource Materials, Plenum Press, New York, 1986.

15.2.13.6. M-0-N-Containing Polyoximes).

Polymers (Polyamidoximes and

Metal-containing polyamidoximes and polyoximes are synthesized from organometallic dihalides and divalent metal-containing moieties with diamidoximes and dioximes. These reactions are conducted using the aqueous interfacial and solution reaction

15.2. Ring-Ring and Ring-Polymer lnterconversions 15.2.14. Coordination Polymers 15.2.14.2. General Synthetic Routes.

194

systems’-3. The synthesis of Sn- and Ti- containing polyoximes, where the oxime is derived from vitamin K, steroids and other biologically active compounds, give products with wide-spectrum biological activity:

N R,SnX,

/OH

+

R

€0-N

N-O-Snf”

I I

R ‘OH vitamin K

tin polyoxime

where X = C1, Br, I; R = aliphatic, aromatic. (C.E. CARRAHER JR.)

1. C. Carraher, L. P. Torre, Organ. Coatings Plastics Chem., 42, 18 (1979) 2. C. Carraher, M. Trembley, L. P. Torre, unpublished results. 3. C. Carraher, K. S . Wang, Makromol. Chem., 152,43 (1972).

15.2.14. Coordination Polymers 15.2.14.1. introduction.

Coordination polymers have been utilized since before recorded history, although not recognized as such until recently. For instance, the tanning of leather depends on the coordination of metal ions with the proteins that make up the hide. These complexes resist the bacterial attack, wear and weathering that typically befall nontanned animal skins. Metals bound to natural polymers, including proteins, affect numerous enzymatic and membrane interactions. Metals can be coordinated with organic ligands to form a polymer or to form a coordination complex within an already formed polymer. The reactions forming coordination polymers can be like those that form polymers through condensation routes. For ready cooperation of product structure with property, the coordination polymer should contain one metal ion per repeat unit. The exact structure of many coordination polymers are unknown and probably vary within a single polymer chain. (C.E. CARRAHER, JR.)

15.2.14.2. General synthetic Routes.

Coordination polymers can be prepared by preformed metal complexes polymerized through functional groups, where the actual polymer-forming step may be a condensation or addition reaction:

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc. 15.2. Ring-Ring and Ring-Polymer lnterconversions 15.2.14. Coordination Polymers 15.2.14.2. General Synthetic Routes.

194

systems’-3. The synthesis of Sn- and Ti- containing polyoximes, where the oxime is derived from vitamin K, steroids and other biologically active compounds, give products with wide-spectrum biological activity:

N R,SnX,

/OH

+

R

€0-N

N-O-Snf”

I I

R ‘OH vitamin K

tin polyoxime

where X = C1, Br, I; R = aliphatic, aromatic. (C.E. CARRAHER JR.)

1. C. Carraher, L. P. Torre, Organ. Coatings Plastics Chem., 42, 18 (1979) 2. C. Carraher, M. Trembley, L. P. Torre, unpublished results. 3. C. Carraher, K. S . Wang, Makromol. Chem., 152,43 (1972).

15.2.14. Coordination Polymers 15.2.14.1. introduction.

Coordination polymers have been utilized since before recorded history, although not recognized as such until recently. For instance, the tanning of leather depends on the coordination of metal ions with the proteins that make up the hide. These complexes resist the bacterial attack, wear and weathering that typically befall nontanned animal skins. Metals bound to natural polymers, including proteins, affect numerous enzymatic and membrane interactions. Metals can be coordinated with organic ligands to form a polymer or to form a coordination complex within an already formed polymer. The reactions forming coordination polymers can be like those that form polymers through condensation routes. For ready cooperation of product structure with property, the coordination polymer should contain one metal ion per repeat unit. The exact structure of many coordination polymers are unknown and probably vary within a single polymer chain. (C.E. CARRAHER, JR.)

15.2.14.2. General synthetic Routes.

Coordination polymers can be prepared by preformed metal complexes polymerized through functional groups, where the actual polymer-forming step may be a condensation or addition reaction:

15.2. Ring-Ring and Ring-Polymer Interconversions 15.2.14. Coordination Polymers 15.2.14.2. General Synthetic Routes.

0

195

0

or coordination of a metal ion by a polymer containing chelating groups:

0

I1

M(O-C-CH,),

and polymer formation through reaction of metal donor-atom coordination:

A listing of polymers utilized to form coordination polymers by route (b) appears in Table 1. Table 2 contains a listing of families of chelating compounds utilized to form coordination polymers by route (c). Polymers formed by routes (a) and (c) typically contain definable repeat units, whereas structures of polymers formed by route (b) vary depending on the conditions, including solvent, T, nature of reactants and concentration ratio of the two reactants; the structure will vary within a given chain for such polymers.

fCH,

I

+CH2-CHfn

I

fCH,-CHf,

I

fCH,-CHf,

I

fCH,-CHf,

CHELATES UTILIZED IN THE SYNTHESIS OF COORDlNATlON POLYMERS

15.2. Ring-Ring and Ring-Polymer Interconversions 15.2.14. Coordination Polymers 15.2.14.2. General Synthetic Routes.

T A ~ L1.L E POLYMER

196

I

d4-

197

15.2. Ring-Ring and Ring-Polymer lnterconversions 15.2.14. Coordination Polymers 15.2.14.2. General Synthetic Routes. TABLE2.

CHELATES UTILIZED IN THE SYNTHESIS OF COORDINATION POLYMERS

THROUGH CHELATION WITH

Chelate group

METALIONS Representative structure

Refs.

Bis-1,2-amino acids

Bis-0-aminophenols

Bisdiamines

1-9

N

N

II

10, 11

II

c-c I

)

R

R

\R’

H Bisdithiocarbamates

H’

S

I

12

\c-N-R-)I

II

S

H 0

II

Bis-a-hydroxyacids

H

0’

I,

P-c/c\ R R \R/

Bis-0-hydroxyazos

H 13, 14

15.2. Ring-Ring and Ring-Polymer lnterconversions 15.2.14. Coordination Polymers 15.2.14.2. General Synthetic Routes.

198

TABLE 2 (Cant.) Chelate group

Representative structure

Refs.

H 0

‘0

c

c

R

R

I/

II

I

3, 15, 16

t

kR1

3;

Bishydroxy nitrogen heteroaromatics

16

Bis-0-hydroxy bases

17, 18

H-0

Bisim ides

H-(’J-~-H H

Bisthiooxamides

Bisthiopicolinamides

I

S

-N-C-C-N-

Bisxanthanes

H

I

22

@C=h’-R%I

H Bisthiosemicarbazones

S

I1 II

/

S

23.24

S

H

/I I

H,N-C-N-N=C-R+j S

II

H\

18

s/ c , o2%

H

I

25

3

Cyclophosphazenes (with chelating groups)

26,27

Dicarbox ylates

13,14

199

15.2. Ring-Ring and Ring-Polymer Interconversions

15.2.14. Coordination Polymers 15.2.14.2. General Synthetic Routes.

TABLE2 (Cont.) Chelate group ~

Representative structure

Refs

~~

Dimercaptodiethers

0

Phosphinous anions

Bis-o-nitrites and 1,2-dinitriles

II

-o‘I’R

P

3

R’

NZC-CfR\ NZC-C

I

3

-R

(C.E. CARRAHER, JR.)

1. J. Gabe, D. Leonta, Rev. Chim. (Bucharest), 14, 1535 (1969). 2. J. C. Bailar, K. V. Martin, M. L. Judd, J. A. McLean, WADC Tech. Rept. No. 57-391 (Part 11), 1958. 3. A. H. Gerber, E. F. McInerney, Survey of Inorganic Polymers, Natl. Aeronaut. Space Admin. Cr-159563 HRI-396, Contract No. NAS3-21369, 1979; Chem. Abstr., 92, 94,921 (1980). 4. T. R. Musgrave, C. E. Mattson, Znorg. Chem., I , 1433 (1968). 5. G. P. Brown, S. Aftergut, J. Polym. Sci., A2, 1839 (1964). 6. M. Inone, M. Kishita, K. Kubo, Znorg. Chem., 4,626 (1965). 7. R. G. Charles, J. Polym. Sci.,AI, 267 (1963). 8. V. V. Korshak, A. M. Sladkov, L. K. Lunera, I. A. Bulgakova, Polym. Sci. U.S.S.R. (Engl. Transl.), 5, 363 (1964). 9. R. M. Klein, J. C. Bailar, Znorg. Chem., 2, 1190 (1963). 10. V. V. Korshak, M. S. Mirkamilova, N. I. Bekasova, Vysokomol.Soedin., Ser. B9, 748 (1967). 11. M. E. B. Jones, D. A. Thornton, R. F. Webb, Makromol. Chem., 49, 69 (1961). 12. A. P. Terent’ev, V. V. Rode, E. G. Rukhadze, Vysokomol.Soedin., 4, 1005 (1962); 4,821 (1962); Chem. Abstr., 59, 772 (1963). Kaaaku Zasshi, 73, 2539 (1970). 13 N. Hoio. H. Shirai. K. Fukatsu, A. Suzuki, KOQVO ~ . sci., 9, 374s (1965). 14 E. A. iomic, J. ~ p pPolym. 15, V. V. Korshak, S. V. Vinogradova, V. S. Artemova, Vysokomol.Soedin., 2,492 (1960). 16. S. Kanda, S. Yoshihiko, Bull. Chem. SOC.Jpn., 30, 192 (1957). 17. E. Horowitz, Adu. Chem. Ser., 85, 82 (1968). 18. I. B. Johns, H. R. DiPetro, US.Pat. 3,169,943 (1965), 3,211,698 (1965); Chem.Abstr., 62,10,62Og (1965), 64, 6845 (1966). 19. N. HOJO,H. Shirai, A. Suzuki, Kogyo Kagaku Zasshi, 73, 1438 (1970); 73, 2425 (1970); Chem. Abstr., 73, 131,428 (1970). 20. C. S. Marvel, M. M. Martin, J. Am. Chem. SOC.,80, 6600 (1958). 21. A. A. Berlin, A. I. Sherle, horg. Macromol. Rev., I, 235 (1971). 22. J. Xavier, P. Ray, J. Indian Chem. SOC.,589 (1958); Chem. Abstr., 53, 16,815 (1959). 23. K. V. Martin, J . Am. Chem. SOC.,80, 233 (1958). 24. C. S. March, S. A. Aspey, E. A. Dudley, J. Am. Chem. Soc., 78,4905 (1956). 25. M. Marcu, M. Dima, Rev. Chm. (Bucharest), 13, 359 (1968). 26. M. Kajiwara, M. Hashimoto, H. Sairo, Polymer, 14, 488 (1973). 27. H. R. Allcock, Phosphorus-Nitrogen Compounds, Academic Press, New York, 1972. I

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc. 200

15.2. Ring-Ring and Ring-Polymer Interconversions 15 2.14. Coordination Polymers 15.2.14.4. Pro pe rty-St r uct ure Pr inc i p Ies .

15.2.14.3. lonomers.

A major class of industrially important polymers, ionomers are ion-containing polymers in which up to 20 mol % of the repeat units contain ionic groups. Typically they are organic polymers with salt sites distributed throughout. Crosslinked polyelectrolyte resins, as beads, powders, membranes and coatings, are useful in ion exchange, electrodialysis and membrane separations; they are also used in elastomers. They can be considered reformable thermosetting materials. The ionic sites form clusters in which small numbers of ion dipoles associate and/or chemically bond, forming “ionic a ion-like crosslinks.” Commercially, the ions most used are Na’, K’, Ca2+ and Mg2’. The clusters increase the glass transition temperature within these regions. They can often be disrupted by input of a large amount of energy; thus, they act as hand blocks at low T, preventing flow. Since the “crosslinking” is random, no large crystalline regions form and most ionomers are transparent. The addition of these crosslinks has a dramatic effect on the physical properties, e.g., increases in glass transition T of 5OO0C,increases in modulus by > 3 orders of magnitude and increases in viscosity of > 4 orders of magnitude. Since ionomers contain ionic clusters, they adhere well to most polar surfaces. They are used for coatings of golf ball covers. Their thermoplasticity permits pressing in conventional molding machines. The addition of foaming agents produces flexible foams used in shoe soles and weatherstripping. Reviews of ionomer chemistry have appeared’-6. (C E CARRAHER, C. V. PITTMAN)

1. L. Holiday, ed., Ionic Polymers, Halstead Press, Wiley, New York, 1975. 2. A. Eisenberg, M. King, Ion Containing Polymers, Academic Press, New York, 1977. 3. A. Eisenberg, ed., Ions in Polymers, Advances in Chemistry Series No. 187, Am. Chem. SOC., Washington, DC, 1980. 4. W. J. MacKnight, T. R. Earnest, Jr., J . Polym. Sci., Macromol. Rev., 16, 41 (1981). 5 . C . G. Bazuin, A. Eisenberg, Ind. Eng. Chem. Prod. Res. D e n , 20, 271 (1981). 6. W. J. MacKnight, R. D. Lundberg,, Rub. Chem. Technol., 5713, (1984) 652.

15.2.14.4. Property-Structure Principles.

The synthesis and characterization of coordination polymers was supported by the US. Air Force in a search for materials that exhibited high thermal stabilities. However, attempts to prepare stable, tractable coordination polymers that simulate the exceptional thermal and/or chemical stability of model monomeric coordination compounds such as copper(I1) ethylenediaminobisacetylacetonate or phthalocyanine have been disa~pointing’-~.Typically, only short chains are formed, and the thermally stable monomers lose most of their stability when linked by the metals into polymeric units. The principles in designing coordination polymers4 are: a. Little flexibility is imparted by the metal ion or its immediate environment; thus, flexibility must arise from the organic moiety. b. Metal ions only stabilize ligands in their immediate vicinity; thus the chelates should be strong and close to the metal atom. c. Thermal, oxidative and hydrolytic stabilities are not directly related; thus polymers must be designed specifically for the properties desired.

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc. 200

15.2. Ring-Ring and Ring-Polymer Interconversions 15 2.14. Coordination Polymers 15.2.14.4. Pro pe rty-St r uct ure Pr inc i p Ies .

15.2.14.3. lonomers.

A major class of industrially important polymers, ionomers are ion-containing polymers in which up to 20 mol % of the repeat units contain ionic groups. Typically they are organic polymers with salt sites distributed throughout. Crosslinked polyelectrolyte resins, as beads, powders, membranes and coatings, are useful in ion exchange, electrodialysis and membrane separations; they are also used in elastomers. They can be considered reformable thermosetting materials. The ionic sites form clusters in which small numbers of ion dipoles associate and/or chemically bond, forming “ionic a ion-like crosslinks.” Commercially, the ions most used are Na’, K’, Ca2+ and Mg2’. The clusters increase the glass transition temperature within these regions. They can often be disrupted by input of a large amount of energy; thus, they act as hand blocks at low T, preventing flow. Since the “crosslinking” is random, no large crystalline regions form and most ionomers are transparent. The addition of these crosslinks has a dramatic effect on the physical properties, e.g., increases in glass transition T of 5OO0C,increases in modulus by > 3 orders of magnitude and increases in viscosity of > 4 orders of magnitude. Since ionomers contain ionic clusters, they adhere well to most polar surfaces. They are used for coatings of golf ball covers. Their thermoplasticity permits pressing in conventional molding machines. The addition of foaming agents produces flexible foams used in shoe soles and weatherstripping. Reviews of ionomer chemistry have appeared’-6. (C E CARRAHER, C. V. PITTMAN)

1. L. Holiday, ed., Ionic Polymers, Halstead Press, Wiley, New York, 1975. 2. A. Eisenberg, M. King, Ion Containing Polymers, Academic Press, New York, 1977. 3. A. Eisenberg, ed., Ions in Polymers, Advances in Chemistry Series No. 187, Am. Chem. SOC., Washington, DC, 1980. 4. W. J. MacKnight, T. R. Earnest, Jr., J . Polym. Sci., Macromol. Rev., 16, 41 (1981). 5 . C . G. Bazuin, A. Eisenberg, Ind. Eng. Chem. Prod. Res. D e n , 20, 271 (1981). 6. W. J. MacKnight, R. D. Lundberg,, Rub. Chem. Technol., 5713, (1984) 652.

15.2.14.4. Property-Structure Principles.

The synthesis and characterization of coordination polymers was supported by the US. Air Force in a search for materials that exhibited high thermal stabilities. However, attempts to prepare stable, tractable coordination polymers that simulate the exceptional thermal and/or chemical stability of model monomeric coordination compounds such as copper(I1) ethylenediaminobisacetylacetonate or phthalocyanine have been disa~pointing’-~.Typically, only short chains are formed, and the thermally stable monomers lose most of their stability when linked by the metals into polymeric units. The principles in designing coordination polymers4 are: a. Little flexibility is imparted by the metal ion or its immediate environment; thus, flexibility must arise from the organic moiety. b. Metal ions only stabilize ligands in their immediate vicinity; thus the chelates should be strong and close to the metal atom. c. Thermal, oxidative and hydrolytic stabilities are not directly related; thus polymers must be designed specifically for the properties desired.

201

15.2.14. Coordination Polymers 15.2.14.5. Selected Polymers 15.2.14.5.1. Metal Phosphinates. ~~

~

~~~~~~~

~~

d. Metal-ligand bonds rearrange more readily than organic bonds. e. Flexibility increases as the covalent nature of the metal-ligand bond increases. f. Coordination number and stereochemistry of the metal ion determines polymer structure about the metal atom. g. Complex formation is favored through use of pure reactants in stoichiometric amounts. h. If a solvent is used for the polymerization, it must not form a strong complex with either the metal ion or the chelating agent. ( C E CARRAHER, J R ) 1. A. H. Gerber, E. F. McInerney, Survey of Inorganic Polymers, NASA Cr-159563 HRI-396, Contract No. NAS3-21369, 1979; Chem. Abstr., 92, 94,921 (1980). 2. G. T. Morgan, N. J. Smith, J. Chem. Soc., 912 (1926). 3. J. C. Bailar, J. Chem. SOC.,51 (1961). 4. J. C. Bailar, Coordination polymers, in Preparative Inorganic Reactions, Vol. 1, W. L. Jolly, ed., Wiley-Interscience, New York, 1964, pp. 1-27.

15.2.14.5. Selected Polymers 15.2.14.5.1. Metal Phosphlnates.

Single-, double- and triple-bridged phosphinate polymers have been produced, containing such metals as Al, Be, Co, Cr, Ni, Ti and Zn Some of the products give films with tensile strengths of several thousand psi, and thermal stabilities at 450°C. Thermal degradation commences with loss of the organic groups, which in turn attack other phosphinate groups. Film-forming characteristics are enhanced by the use of plasticizers.

'.

R R \P/

P \p/R

single bridged

triple bridged

double bridged

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc. 201

15.2.14. Coordination Polymers 15.2.14.5. Selected Polymers 15.2.14.5.1. Metal Phosphinates. ~~

~

~~~~~~~

~~

d. Metal-ligand bonds rearrange more readily than organic bonds. e. Flexibility increases as the covalent nature of the metal-ligand bond increases. f. Coordination number and stereochemistry of the metal ion determines polymer structure about the metal atom. g. Complex formation is favored through use of pure reactants in stoichiometric amounts. h. If a solvent is used for the polymerization, it must not form a strong complex with either the metal ion or the chelating agent. ( C E CARRAHER, J R ) 1. A. H. Gerber, E. F. McInerney, Survey of Inorganic Polymers, NASA Cr-159563 HRI-396, Contract No. NAS3-21369, 1979; Chem. Abstr., 92, 94,921 (1980). 2. G. T. Morgan, N. J. Smith, J. Chem. Soc., 912 (1926). 3. J. C. Bailar, J. Chem. SOC.,51 (1961). 4. J. C. Bailar, Coordination polymers, in Preparative Inorganic Reactions, Vol. 1, W. L. Jolly, ed., Wiley-Interscience, New York, 1964, pp. 1-27.

15.2.14.5. Selected Polymers 15.2.14.5.1. Metal Phosphlnates.

Single-, double- and triple-bridged phosphinate polymers have been produced, containing such metals as Al, Be, Co, Cr, Ni, Ti and Zn Some of the products give films with tensile strengths of several thousand psi, and thermal stabilities at 450°C. Thermal degradation commences with loss of the organic groups, which in turn attack other phosphinate groups. Film-forming characteristics are enhanced by the use of plasticizers.

'.

R R \P/

P \p/R

single bridged

triple bridged

double bridged

202

15.2. Ring-Ring and Ring-Polymer Interconversions 15.2.14. Coordination Polymers 15.2.14.5. Selected Polymers

The metal phosphinates are prepared from metal salts and dialkyl or diarylphosphosphinic acids utilizing melt or solution systems. The metal polyphosphinates are utilized as additives; e.g., Cr(II1) polyphosphinates thicken silicones to greases and improve their high-pressure physical properties. Chromium and titanium polyphosphinates impart antistatic properties to a wide range of plastics. (C.E. CARRAHER, J R )

1. B. P. Block, Inorg. Macromol. Chem., I , 115 (1970).

15.2.14.5.2. Platinum and Pd Coordlnation Polymers.

Bacteria failed to divide in the presence of platinum electrodes but continued to grow, giving filamentous cells. A major cause of this inhibition to cell division is cisdichlorodiamineplatinum(I1) (c-DDP), which has been licensed as an antineoplast drug. The use of c-DDP is complicated due to a wide number of negative side effects. A number of these side effects may be overcome if the Pt is present in a polymer that both can act as a long-acting controlled-release agent and be prevented from entering catastrophically into the human circulatory and excretory systems. The c-DDP is attached to a preformed, water-soluble polymer, polybis(methy1amino)phosphazene, I, that bears coordination sites on both the side-group and chain nitrogen atoms'-3. Compound I reacts with K,[PtCl,] in organic media with 18-crown6-ether yielding a coordination complex, 11, containing c-DDP 3, which shows tumor inhibitory activity3.

I

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc. 202

15.2. Ring-Ring and Ring-Polymer Interconversions 15.2.14. Coordination Polymers 15.2.14.5. Selected Polymers

The metal phosphinates are prepared from metal salts and dialkyl or diarylphosphosphinic acids utilizing melt or solution systems. The metal polyphosphinates are utilized as additives; e.g., Cr(II1) polyphosphinates thicken silicones to greases and improve their high-pressure physical properties. Chromium and titanium polyphosphinates impart antistatic properties to a wide range of plastics. (C.E. CARRAHER, J R )

1. B. P. Block, Inorg. Macromol. Chem., I , 115 (1970).

15.2.14.5.2. Platinum and Pd Coordlnation Polymers.

Bacteria failed to divide in the presence of platinum electrodes but continued to grow, giving filamentous cells. A major cause of this inhibition to cell division is cisdichlorodiamineplatinum(I1) (c-DDP), which has been licensed as an antineoplast drug. The use of c-DDP is complicated due to a wide number of negative side effects. A number of these side effects may be overcome if the Pt is present in a polymer that both can act as a long-acting controlled-release agent and be prevented from entering catastrophically into the human circulatory and excretory systems. The c-DDP is attached to a preformed, water-soluble polymer, polybis(methy1amino)phosphazene, I, that bears coordination sites on both the side-group and chain nitrogen atoms'-3. Compound I reacts with K,[PtCl,] in organic media with 18-crown6-ether yielding a coordination complex, 11, containing c-DDP 3, which shows tumor inhibitory activity3.

I

15.2. Ring-Ring and Ring-Polymer lnterconversions 15.2.14. Coordination Polymers 15.2.14.5. Selected Polymers

203

cis-Dihaloplatinum diamine has been incorporated into polymers using diaminecontaining reactants including aromatics, aliphatics, pyrimides, pyridines and purines4q5: K,PtX,

+ H,N-R-NH,

TI1

X \ /x fPt-NH,-R--NH,+,

(b)

where X = C1, Br, I. These polymers show biological activity. At concentrations of 10-20 pg mL-' they can suppress, enhance or have no effect on the replication of viruses; but none of the polymers shows activity toward any of the tested cell lines (WISH, L929, HeLa) at these concentrations. At concentrations of 30 pg mL-' and above most of the platinum polyamines inhibit tumoral cell growth. Mice can tolerate dosages in excess of 20 pg g-' of weight with no apparent ill effects6-14. A number of analogous Pd-containing polymers have also been synthesized' 5*16. (C.E. CARRAHER, JR.)

1. H. Allcock, W. Cook, D. Mack, Inorg. Chem., 4, 2584 (1972). 2. H. Allcock, R. Allen, J. OBrien, J. Am. Chem. Soc., 97, 3914 (1977). 3. H. Allcock, in Organometallic Polymers, C. Carraher, J. Sheats and C. Pittman, eds., Academic Press, New York, 1978, Ch. 28. 4. C. Carraher, D. Giron, W. Scott, J. Schroeder, J. Macromol. SciXhem., A15, 125 (1981). 5. C. Carraher, D. Giron, I. Lopez, D. R. Cerutis, W. Scott, Organ. Coatings Plastics Chem., 44,221 (1981). 6. C. Carraher, C. Ademu-John, J. Fortman, D. Giron, C. Turner, J . Polym. Mater., 1(2), 116 (1984). 7. C. Carraher, D. Siegmann, D. Brenner, Polym. Muter., 59, 535 (1988). 8. C. Carraher, D. Siegman, J . Polym. Mater., 4, 29 (1987). 9. D. Siegmann, C. Carraher, A. Friend, J . Polym. Mater., 4, 19 (1987). 10. C. Carraher, N. Bigley, M. Trombley, D. Giron, Polym. Mater., 57, 177 (1987). 11. C. Carraher, R. Strother, D. Brenner, Polym. Muter., 57, 173 (1987). 12. C. Gebelin, C. Carraher, Adounces in Biomedical Polymers, Plenum Press, New York, 1987. 13. C. Gebelin, C . Carraher, Polymeric Materials in Medication, Plenum Press, New York, 1986. 14. C. Gebelin, C. Carraher, Bioactiue Polymeric Systems, Plenum Press, New York, 1985. 15. C. Carraher, W. Chen, G. Hess, D. Giron, Polym. Mater., 59, 530 (1988). 16. C. Carraher, G. Hess, W. Chen, Polym. Mater., 59, 744 (1988). 15.2.14.5.3. Uranyl Ion Complexation.

Polymers can complex the uranyl ion, the most common naturally occurring, watersoluble form of uranium, for both environmental and industrial purposes. Select salts of diacids, dioximes, etc., salts of polyacrylic acids, polyacrylic acid itself and a wide variety of carboxylic acid-, sulfonate- and sulfate-containing resins are capable of removing the uranyl ion to lo-, to lo-' mol L - l '-3. These complexed uranyl compounds have greatly reduced toxicities to a wide range of bacteria and fungi compared to 13:

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc. 15.2. Ring-Ring and Ring-Polymer lnterconversions 15.2.14. Coordination Polymers 15.2.14.5. Selected Polymers

203

cis-Dihaloplatinum diamine has been incorporated into polymers using diaminecontaining reactants including aromatics, aliphatics, pyrimides, pyridines and purines4q5: K,PtX,

+ H,N-R-NH,

TI1

X \ /x fPt-NH,-R--NH,+,

(b)

where X = C1, Br, I. These polymers show biological activity. At concentrations of 10-20 pg mL-' they can suppress, enhance or have no effect on the replication of viruses; but none of the polymers shows activity toward any of the tested cell lines (WISH, L929, HeLa) at these concentrations. At concentrations of 30 pg mL-' and above most of the platinum polyamines inhibit tumoral cell growth. Mice can tolerate dosages in excess of 20 pg g-' of weight with no apparent ill effects6-14. A number of analogous Pd-containing polymers have also been synthesized' 5*16. (C.E. CARRAHER, JR.)

1. H. Allcock, W. Cook, D. Mack, Inorg. Chem., 4, 2584 (1972). 2. H. Allcock, R. Allen, J. OBrien, J. Am. Chem. Soc., 97, 3914 (1977). 3. H. Allcock, in Organometallic Polymers, C. Carraher, J. Sheats and C. Pittman, eds., Academic Press, New York, 1978, Ch. 28. 4. C. Carraher, D. Giron, W. Scott, J. Schroeder, J. Macromol. SciXhem., A15, 125 (1981). 5. C. Carraher, D. Giron, I. Lopez, D. R. Cerutis, W. Scott, Organ. Coatings Plastics Chem., 44,221 (1981). 6. C. Carraher, C. Ademu-John, J. Fortman, D. Giron, C. Turner, J . Polym. Mater., 1(2), 116 (1984). 7. C. Carraher, D. Siegmann, D. Brenner, Polym. Muter., 59, 535 (1988). 8. C. Carraher, D. Siegman, J . Polym. Mater., 4, 29 (1987). 9. D. Siegmann, C. Carraher, A. Friend, J . Polym. Mater., 4, 19 (1987). 10. C. Carraher, N. Bigley, M. Trombley, D. Giron, Polym. Mater., 57, 177 (1987). 11. C. Carraher, R. Strother, D. Brenner, Polym. Muter., 57, 173 (1987). 12. C. Gebelin, C. Carraher, Adounces in Biomedical Polymers, Plenum Press, New York, 1987. 13. C. Gebelin, C . Carraher, Polymeric Materials in Medication, Plenum Press, New York, 1986. 14. C. Gebelin, C. Carraher, Bioactiue Polymeric Systems, Plenum Press, New York, 1985. 15. C. Carraher, W. Chen, G. Hess, D. Giron, Polym. Mater., 59, 530 (1988). 16. C. Carraher, G. Hess, W. Chen, Polym. Mater., 59, 744 (1988). 15.2.14.5.3. Uranyl Ion Complexation.

Polymers can complex the uranyl ion, the most common naturally occurring, watersoluble form of uranium, for both environmental and industrial purposes. Select salts of diacids, dioximes, etc., salts of polyacrylic acids, polyacrylic acid itself and a wide variety of carboxylic acid-, sulfonate- and sulfate-containing resins are capable of removing the uranyl ion to lo-, to lo-' mol L - l '-3. These complexed uranyl compounds have greatly reduced toxicities to a wide range of bacteria and fungi compared to 13:

204

15.2. Ring-Ring and Ring-Polymer Interconversions 15.2.14. Coordination Polymers 15.2.14.5. Selected Polymers

Usings resins to remove and detect metal ions, including the uranyl ion, has been practiced for some time4. Reactions in which metal-containing coordination polymers are formed are critical analytical and chromatographic reactions but are often not recognized as involving coordination polymer chemistry. Plants can chelate various e.g., Sphagnum Jimbriatum and Spagnum recurrum*. Thus future efforts at identification, separation and isolation of metalcontaining moieties may include use of naturally occurring reagents. (C.E. CARRAHER, JR.)

1. C. Carraher, J. Schroeder, Polym. Lett., 13, 215 (1975). 2. C. Carraher, in Interfacial Synthesis,Vol. 11, Technologyand Applications, F. Millich, C. Carraher, eds., Marcel Dekker, New York, 1978. 3. C. Carraher, S . Tsuji, D. S. Cerutis, W. Feld, J. DiNunzio, unpublished results. 4. J. Korkish, Modern Methods for the Separation of Rarer Metal Ions, Pergamon Press, Oxford, 1969. 5. A. Knight, W. Crooke, R. Inkson, Nature (London), 192, 142 (1961). 6. R. Clymo, Ann. Botany, New Ser., 27, 309 (1963). 7. J. Craigie, W. Maass, Ann. Botany, New Ser., 30, 153 (1966). 8. C. Carraher, B. Huntsman, J. Solch, T. Tiernan, M. Taylor, unreported results. 15.2.14.5.4. Bis-P-dikelone-Based Polymers.

The most widely studied coordination polymer family is that derived from bis-fi-diketones'. The diketones utilized are usually of two structural types, with the alkylene bridge occurring either between the two sets of carbonyls (I) or to one side of the carbonyls (11):

CH,

CH,

O

I

I

c=o

o=c I

I HC-R-CH I I o=c c=o I I CH,

R 0 \ / \ /

c

c

CH,

CH,

I

I

o=c I

CH,

i

I

c=o 1

CH, CH, I1

I

Such polymers are formed through the polymerization of the bis-fi-diketone with metal acetates or metal a c e t y l a ~ e t o n a t e s ~Thermal * ~ ~ ~ . decomposition of polymers of form 111occurs in the range of 225-350°C.

I11

where M = Mg, Co, Zn, Be, Cd, Ni, Cu. (C.E. CARRAHER, JR.)

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc. 204

15.2. Ring-Ring and Ring-Polymer Interconversions 15.2.14. Coordination Polymers 15.2.14.5. Selected Polymers

Usings resins to remove and detect metal ions, including the uranyl ion, has been practiced for some time4. Reactions in which metal-containing coordination polymers are formed are critical analytical and chromatographic reactions but are often not recognized as involving coordination polymer chemistry. Plants can chelate various e.g., Sphagnum Jimbriatum and Spagnum recurrum*. Thus future efforts at identification, separation and isolation of metalcontaining moieties may include use of naturally occurring reagents. (C.E. CARRAHER, JR.)

1. C. Carraher, J. Schroeder, Polym. Lett., 13, 215 (1975). 2. C. Carraher, in Interfacial Synthesis,Vol. 11, Technologyand Applications, F. Millich, C. Carraher, eds., Marcel Dekker, New York, 1978. 3. C. Carraher, S . Tsuji, D. S. Cerutis, W. Feld, J. DiNunzio, unpublished results. 4. J. Korkish, Modern Methods for the Separation of Rarer Metal Ions, Pergamon Press, Oxford, 1969. 5. A. Knight, W. Crooke, R. Inkson, Nature (London), 192, 142 (1961). 6. R. Clymo, Ann. Botany, New Ser., 27, 309 (1963). 7. J. Craigie, W. Maass, Ann. Botany, New Ser., 30, 153 (1966). 8. C. Carraher, B. Huntsman, J. Solch, T. Tiernan, M. Taylor, unreported results. 15.2.14.5.4. Bis-P-dikelone-Based Polymers.

The most widely studied coordination polymer family is that derived from bis-fi-diketones'. The diketones utilized are usually of two structural types, with the alkylene bridge occurring either between the two sets of carbonyls (I) or to one side of the carbonyls (11):

CH,

CH,

O

I

I

c=o

o=c I

I HC-R-CH I I o=c c=o I I CH,

R 0 \ / \ /

c

c

CH,

CH,

I

I

o=c I

CH,

i

I

c=o 1

CH, CH, I1

I

Such polymers are formed through the polymerization of the bis-fi-diketone with metal acetates or metal a c e t y l a ~ e t o n a t e s ~Thermal * ~ ~ ~ . decomposition of polymers of form 111occurs in the range of 225-350°C.

I11

where M = Mg, Co, Zn, Be, Cd, Ni, Cu. (C.E. CARRAHER, JR.)

15.2. Ring-Ring and Ring-Polymer Interconversions 15.2.14. Coordination Polymers 15.2.14.5. Selected Polymers

205

~

1. A. H. Gerber, E. F. McInerney, Survey of Inorganic Polymers, NASA Cr-159563 NRI-396, Contract No. NAS3-21369, 1979; Chem. Abstr., 92, 94,921 (1980). 2. R. G. Charles, J. Polym. Sci., A I , 267 (1963). 3. R. Charles, J. Phys. Chem., 64, 1747 (1960). 4. V. Korshak, S. Vinogradora, Dokl. Akad. Nauk SSSR, 138, 1353 (1961); Chem. Abstr., 55,21,646 (1961). 15.2.14.5.5. Bisthiopiocolinamide-Based Polymers.

Coordination polymers of Zn(I1) derived from bisthiopicolinamides show the best thermal stabilities and can be heated at 300°C for 6 h without change'. The Zn(I1) derivatives of bissalicaldimines also show the best thermal stability. The often found superior thermal stability of Zn(I1) derivatives arise from its single oxidation state'. Transition metals can be oxidized to higher states, catalyzing the decomposition of the polymer chains. (C E CARRAHER, JR.) 1. C. S. Marvel, S. Aspey, E. Dudley, J. Am. Chem. Soc., 78, 4905 (1956). 2. K. V. Martin, J. Am. Chem. Soc., 80, 233 (1958).

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc. 15.2. Ring-Ring and Ring-Polymer Interconversions 15.2.14. Coordination Polymers 15.2.14.5. Selected Polymers

205

~

1. A. H. Gerber, E. F. McInerney, Survey of Inorganic Polymers, NASA Cr-159563 NRI-396, Contract No. NAS3-21369, 1979; Chem. Abstr., 92, 94,921 (1980). 2. R. G. Charles, J. Polym. Sci., A I , 267 (1963). 3. R. Charles, J. Phys. Chem., 64, 1747 (1960). 4. V. Korshak, S. Vinogradora, Dokl. Akad. Nauk SSSR, 138, 1353 (1961); Chem. Abstr., 55,21,646 (1961). 15.2.14.5.5. Bisthiopiocolinamide-Based Polymers.

Coordination polymers of Zn(I1) derived from bisthiopicolinamides show the best thermal stabilities and can be heated at 300°C for 6 h without change'. The Zn(I1) derivatives of bissalicaldimines also show the best thermal stability. The often found superior thermal stability of Zn(I1) derivatives arise from its single oxidation state'. Transition metals can be oxidized to higher states, catalyzing the decomposition of the polymer chains. (C E CARRAHER, JR.) 1. C. S. Marvel, S. Aspey, E. Dudley, J. Am. Chem. Soc., 78, 4905 (1956). 2. K. V. Martin, J. Am. Chem. Soc., 80, 233 (1958).

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc.

16. Formation of Intercalation

Compounds

16.1. Introduction This chapter describes the formation of intercalation compounds as clathrates, as tunnel structures and as sheets.

208

16.2. The Formation of Clathrates 16.2.1. Definitions Clathrates are a special class of intercalation compounds. They are solids that result from the association, rather than from the combination, of two chemical species having no particular affinity. One species forms a three-dimensional open lattice exhibiting large cavities in which atoms or molecules of the second species are caged'-3. The atoms of the host lattice are linked together by strong classical bonds4. In the absence of guest species it is normally unstable and adopts another stucture. Thermodynamic stability, however, does not require occupancy of all of the available voids. Clathrates are therefore nonstoichiometric. The guest atoms or molecules exchange with the host lattice only weak intermolecular forces and no specific chemical bonds. Clathrates are known with noble gas atoms as guests. The stability of the host-guest association is governed by steric criteria. The structure formed, the corresponding chemical formula and its stability depend upon the size difference between the guest atoms or molecules and the available voids. The smaller this difference, the more stable the clathrate and the closer to stoichiometry the formula. O n the other hand, the size of the guest species does not influence the values of the unit cell parameters. A clathrate is thus an inclusion compound which is a solid solution of the guest atoms or molecules in a metastable crystalline form characterized by isolated large spherical voids of the host species. These structural criteria differentiate clathrates from the other intercalation compounds subsequently described (see 516.3 and 16.4), in which the available voids exhibit a uni- or two-dimensional arrangement (tunnel and sheet structures). Clathrates are known in which host and guest species are inorganic or organic compounds. Only those having inorganic host lattices are considered here. Among these, the formation of two kinds of clathrate compounds is reported: 1. the formation of clathrate hydrates 2. the formation of clathrate-type silicon, germanium and tin compounds Clathrate hydrates have been known since the early 19th Century, but their formation and nature were only understood a few decades ago5-'. The closely related clathrates of silicon, germanium and tin are more recent"-12. (C CROS)

1. D. E. Palin, H. M. Powell, J. Chem. Soc., 815 (1948). 2. H. M. Powell, in Non-stoichiornetric Compounds, L. Mandelcorn, ed., Academic Press, New York, 1964. 3. R. M. Barrer, in Non-stoichiometric Compounds, L. Mandelcorn, ed., Academic Press, New York, 1964. 4. L. A. K. Staveley, in Non-stoichiornetric Compounds,R. Ward, ed., Ado. Chem. Ser., 39, American Chemical Society, Washington, DC, 1963. 5. H. Davy, Phil. Trans. Roy. Soc. London, 101, 30 (1811).

209

210 6. 7. 8. 9. 10. 11. 12.

16.2. The Formation of Clathrates 16.2.2. The Formation of Clathrates Having a Water Host Lattice

M. Faraday, Quart. J. Sci. Lit.Arts, 15, 71 (1823). W. F. Claussen, J. Chem. Phys., 19,259 (1951). W. F. Clausen, J. Chem. Phys., 19, 1425 (1951). L. Pauling, R. E. Marsh, Proc. Natl. Acad. Sci. USA,38, (1952). J. S. Kasper, P. Hagenmuller, M. Pouchard, C. Cros, Science, 250, 1713 (1965). C. Cros, M. Pouchard, P. Hagenmuller, J. S. Kasper, Bull. SOC.Chim. Fr., 7,2737 (1968). J. Gallmeier, H. Schafer, A. Weiss, 2. Naturforsch., Teil B, 24, 665 (1969).

16.2.2. The Formation of Clathrates Having a Water Host Lattice A large number of substances form clathrate-type hydrates with water host lattices. The guest species are organic or inorganic molecules, or more rarely atoms which are insoluble or slightly soluble in water'-6. Four kinds of clathrate hydrates are known. The water networks are formed by close packing of large polyhedra whose vertices are

a

C

b

d

Figure 1. The polyhedra found in clathrate hydrates: (a) the pentagonal dodecahedra, (b) the tetrakaidecahedra, (c) the pentakaidecahedra, (d) the hexakaidecahedra (from ref. 7).

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc. 210 6. 7. 8. 9. 10. 11. 12.

16.2. The Formation of Clathrates 16.2.2. The Formation of Clathrates Having a Water Host Lattice

M. Faraday, Quart. J. Sci. Lit.Arts, 15, 71 (1823). W. F. Claussen, J. Chem. Phys., 19,259 (1951). W. F. Clausen, J. Chem. Phys., 19, 1425 (1951). L. Pauling, R. E. Marsh, Proc. Natl. Acad. Sci. USA,38, (1952). J. S. Kasper, P. Hagenmuller, M. Pouchard, C. Cros, Science, 250, 1713 (1965). C. Cros, M. Pouchard, P. Hagenmuller, J. S. Kasper, Bull. SOC.Chim. Fr., 7,2737 (1968). J. Gallmeier, H. Schafer, A. Weiss, 2. Naturforsch., Teil B, 24, 665 (1969).

16.2.2. The Formation of Clathrates Having a Water Host Lattice A large number of substances form clathrate-type hydrates with water host lattices. The guest species are organic or inorganic molecules, or more rarely atoms which are insoluble or slightly soluble in water'-6. Four kinds of clathrate hydrates are known. The water networks are formed by close packing of large polyhedra whose vertices are

a

C

b

d

Figure 1. The polyhedra found in clathrate hydrates: (a) the pentagonal dodecahedra, (b) the tetrakaidecahedra, (c) the pentakaidecahedra, (d) the hexakaidecahedra (from ref. 7).

16.2. The Formation of Clathrates 16.2.2. The Formation of Clathrates Having a Water Host Lattice 16.2.2.1. Clathrates of the Gas Hydrate Type.

211

occupied by oxygen atoms, the hydrogen atoms being located along the edges. These frameworks thus exhibit covalent and hydrogen bonding. The guest species occupy the various polyhedra with coordination numbers of at least 20 The most common polyhedron is the pentagonal dodecahedron. It is formed by 20 water molecules and has 12 pentagonal faces, 20 vertices and 30 edges. With a hydrogen bond distance of -280 pm, its volume is 170 x lo6 pm3 (Fig. la). Because of its fivefold symmetry, the close packing of such polyhedra cannot fill space; the pentagonal dodecahedron is therefore associated with slightly different polyhedra:

’.

1. The tetrakaidecahedron (24 water molecules), with 12 pentagonal faces, 2 hexagonal faces, 24 vertices and 36 edges (Fig. lb) 2. The pentakaidecahedron (26 water molecules), with 12 pentagonal faces, 3 hexagonal faces, 26 vertices and 39 edges (Fig. lc) 3. The hexakaidecahedron (28 water molecules), with 12 pentagonal faces, 4 hexagonal faces, 28 vertices and 42 edges (Fig. Id). Assuming the same 280-pm edge length, the volumes of these polyhedra are larger than that of pentagonal dodecahedra: 220 x lo6, 250 x lo6 and 260 x lo6 pm3, respectively. ( C . CROS)

1. 2. 3. 4. 5. 6. I.

W. F. Claussen, J. Chem. Phys., 19, 259 (1951). W. F. Claussen, J. Chem. Phys., 19, 1425 (1951). M. von Stackelberg, H. R. Miiller, J. Chem. Phys., 19, 1319 (1951). M. von Stackelberg, H. R. Miiller, Naturwissenschaften, 38, 456 (1951). L. Pauling, R. E. Marsh, Proc. Natl. Acad. Sci. USA, 3E (i952). M. von Stackelberg, H. R. Miiller, Z. Electrochem., 58, 25 (1954). G . A. Jeffrey, Structural Factors in the Formation of Clathrate Hydrates, Decherna Monograph, 47, 849 (1962).

16.2.2.1. Clathrates of the Gas Hydrate Type. The term “gas hydrate” has been adopted because most of the concerned guest species are gaseous at STP. The open ice host lattice is formed by the association of two pentagonal dodecahedra located at the corners and the center of a cubic unit cell with the six tetrakaidecahedra that fill the space between them. The unit cell parameter is 1200 pm (space group Pm3n) (Fig. 1). The total number of water molecules is 46. If all the available voids are occupied by a guest molecule, the corresponding formula is 8 M . 46 H,O (or M .5.75 H,O); if the smallest voids remain empty, the fomula becomes 6 M 46 H,O (or M . 7.67 H,O). This structural type is designated type I. Gas hydrate clathrates are formed by cooling water in presence of the guest species. The conditions of existence of the hydrates are conveniently represented by means of a pressure-temperature diagram such as Fig. 2,. The domain of stability of the hydrate is located on the left part of curves I1 and 111. Three data must be known in order to construct this diagram:

-

1. The critical point, above which it is impossible to obtain the hydrate independent of pressure, T,

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc. 16.2. The Formation of Clathrates 16.2.2. The Formation of Clathrates Having a Water Host Lattice 16.2.2.1. Clathrates of the Gas Hydrate Type.

211

occupied by oxygen atoms, the hydrogen atoms being located along the edges. These frameworks thus exhibit covalent and hydrogen bonding. The guest species occupy the various polyhedra with coordination numbers of at least 20 The most common polyhedron is the pentagonal dodecahedron. It is formed by 20 water molecules and has 12 pentagonal faces, 20 vertices and 30 edges. With a hydrogen bond distance of -280 pm, its volume is 170 x lo6 pm3 (Fig. la). Because of its fivefold symmetry, the close packing of such polyhedra cannot fill space; the pentagonal dodecahedron is therefore associated with slightly different polyhedra:

’.

1. The tetrakaidecahedron (24 water molecules), with 12 pentagonal faces, 2 hexagonal faces, 24 vertices and 36 edges (Fig. lb) 2. The pentakaidecahedron (26 water molecules), with 12 pentagonal faces, 3 hexagonal faces, 26 vertices and 39 edges (Fig. lc) 3. The hexakaidecahedron (28 water molecules), with 12 pentagonal faces, 4 hexagonal faces, 28 vertices and 42 edges (Fig. Id). Assuming the same 280-pm edge length, the volumes of these polyhedra are larger than that of pentagonal dodecahedra: 220 x lo6, 250 x lo6 and 260 x lo6 pm3, respectively. ( C . CROS)

1. 2. 3. 4. 5. 6. I.

W. F. Claussen, J. Chem. Phys., 19, 259 (1951). W. F. Claussen, J. Chem. Phys., 19, 1425 (1951). M. von Stackelberg, H. R. Miiller, J. Chem. Phys., 19, 1319 (1951). M. von Stackelberg, H. R. Miiller, Naturwissenschaften, 38, 456 (1951). L. Pauling, R. E. Marsh, Proc. Natl. Acad. Sci. USA, 3E (i952). M. von Stackelberg, H. R. Miiller, Z. Electrochem., 58, 25 (1954). G . A. Jeffrey, Structural Factors in the Formation of Clathrate Hydrates, Decherna Monograph, 47, 849 (1962).

16.2.2.1. Clathrates of the Gas Hydrate Type. The term “gas hydrate” has been adopted because most of the concerned guest species are gaseous at STP. The open ice host lattice is formed by the association of two pentagonal dodecahedra located at the corners and the center of a cubic unit cell with the six tetrakaidecahedra that fill the space between them. The unit cell parameter is 1200 pm (space group Pm3n) (Fig. 1). The total number of water molecules is 46. If all the available voids are occupied by a guest molecule, the corresponding formula is 8 M . 46 H,O (or M .5.75 H,O); if the smallest voids remain empty, the fomula becomes 6 M 46 H,O (or M . 7.67 H,O). This structural type is designated type I. Gas hydrate clathrates are formed by cooling water in presence of the guest species. The conditions of existence of the hydrates are conveniently represented by means of a pressure-temperature diagram such as Fig. 2,. The domain of stability of the hydrate is located on the left part of curves I1 and 111. Three data must be known in order to construct this diagram:

-

1. The critical point, above which it is impossible to obtain the hydrate independent of pressure, T,

Figure 1. The stacking of polyhedra in hydrates of type I (from ref. 1).

h

E

7

H20 Ilq

v

+

w 4

cc

3

m

v)

2cc a

'

02 iiq

3

/ /

2

1

0

-

-8

-4

0

8 12 TEMPERATURE

4

16 ("C)

Figure 2. Pressure vs. temperature diagramm for SO, hydrate (from ref. 2).

212

16.2. The Formation of Clathrates 16.2.2. The Formation of Clathrates Having a Water Host Lattice 16.2.2.1. Clathrates of the Gas Hydrate Type.

213

TABLE 1. PHYSICOCHEMICAL DATAON GAS HYDRATES (FROMREF. 3) Guest mol. M

Boiling point of M ("C)

Ar CH, Kr Xe C2H4 CZH,

- 190

NZO

PH, CZHZ COZ CH,F HZS ASH, H,Se C1Z GH5F CH,C1

so,

c10,

- 161 - 152 - 107 - 102

-93 - 89 - 87 - 84 - 79 - 78 - 60 -55 - 42 - 34 - 32 - 24 - 10 10

BrCl CH,Br CH,SH Br, a

lo5 N m-' = 1 bar

4 6 59 N

P(0"C) (lo5 N m-' 106 26.3 14.7 1.5 5.6 5.3 10.1 1.6 5.8 12.4 2.1 0.976 0.817 0.461 0.335 0.707 0.414 0.397 (0.213) (0.166) 0.249 0.318 0.059

1a

T(l atm) ("C) -42.8 -29.0 -27.8 - 3.4 - 13.4 - 15.8 - 19.3 - 6.4 -15.4 - 24 0.35 18 8.0 9.6 3.7 7.5 7.0 15.0 14 11.1 10

Tc ("C;105 N m-'

Lattice param. (Pm)

1197 14.5 12 28 16 10;45 18 29.5;23 28.3 30 28.7;6 22.8 21 12.1;2.3 18.2 14.5;1.5 12 6.2;0.12

1203 1204 1200 1206 1203 1200 1194 1202 1209 1212 1201

1 atm.

2. The temperature at which the decomposition pressure reaches one atmosphere, T I atm 3. The pressure of decomposition at O'C, These data are given in Table 1. Crystallization takes place at the phase boundaries within the water. The process is more rapid if the reactive surface is renewed by agitation. The formation of gas hydrate clathrates is only observed when the dimensions of the guest molecules or atoms are as close as possible to the free diameter of the two kinds of available voids in the type I host lattice: dI,iand dI,, (Fig. 3, left part). With guest species having a diameter less than dI,i (520 pm), both the dodecahedra1 and tetrakaidecahedral voids can be occupied and the corresponding stoichiometric hydrate is 8 M . 46 H,O; for molecules whose diameter is greater than dI,lbut less than dI,, (590 pm), only the largest voids are occupied and the corresponding hydrate formula is 6 M .46 H,O. For guest species having a larger diameter than dI,,, no gas hydrate-type structure is formed. (C. CROS)

1. R. K. McMullan, G. A. Jeffrey, J. Chem. Phys., 42, 2729 (1965). 2. H. W. B. Roozeboom, Reel Trau. Chzm., Pays-Bas, 3, 28 (1884). 3. M. von Stackelberg, H. R. Mullen, Z . Electrochem., 58,25 (1954).

214

16.2. The Formation of Clathrates 16.2.2 The Formation of Clathrates Having a Water Host Lattice 16.2.2.2. Clathrates of the Liquid Hydrate Type.

h2

‘1 2

‘1 I

‘11

Figure 3. Comparison of the dimensions of the guest molecules with the free diameter of the available voids in gas and liquid hydrates (from ref. 3).

16.2.2.2. Clathrates of the Liquid Hydrate Type The term “liquid” has been adopted because most of the concerned guest species are liquids at STP. The host lattice unit cell is cubic: a N 1740 pm, space group Fd 3m. It results from the association of 16 pentagonal dodecahedra and 8 hexakaidecahedra and is referred to as type I1 (Fig. 1). The total number of water molecules is 136. If all the polyhedra are filled by guest atoms or molecules, the corresponding hydrate formula is 24 M . 136 H,O (or M ’5.67 H,O), but this situation has never been observed. Liquid hydrates exist when the eight largest voids of the structure are occupied by large M molecules: 8 M ,136 H,O. In addition to these normal liquid hydrates, double hydrates are known, in which the 16 smallest voids are filled by a second guest species such as hydrogen sulfide or selenide. These species have been called help gases because they enhance the stability of the corresponding hydrate. The less volatile the help gas the greater the fraction of smaller voids it occupies, and the greater its stabilizing It should be noted however, that help gases taken alone tend to form a type I structure (gas hydrates). Liquid hydrate clathrates are formed in conditions similar to those for gas hydrate^^.^. The corresponding physicochemical data are given in Table 1. Liquid

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc. 214

16.2. The Formation of Clathrates 16.2.2 The Formation of Clathrates Having a Water Host Lattice 16.2.2.2. Clathrates of the Liquid Hydrate Type.

h2

‘1 2

‘1 I

‘11

Figure 3. Comparison of the dimensions of the guest molecules with the free diameter of the available voids in gas and liquid hydrates (from ref. 3).

16.2.2.2. Clathrates of the Liquid Hydrate Type The term “liquid” has been adopted because most of the concerned guest species are liquids at STP. The host lattice unit cell is cubic: a N 1740 pm, space group Fd 3m. It results from the association of 16 pentagonal dodecahedra and 8 hexakaidecahedra and is referred to as type I1 (Fig. 1). The total number of water molecules is 136. If all the polyhedra are filled by guest atoms or molecules, the corresponding hydrate formula is 24 M . 136 H,O (or M ’5.67 H,O), but this situation has never been observed. Liquid hydrates exist when the eight largest voids of the structure are occupied by large M molecules: 8 M ,136 H,O. In addition to these normal liquid hydrates, double hydrates are known, in which the 16 smallest voids are filled by a second guest species such as hydrogen sulfide or selenide. These species have been called help gases because they enhance the stability of the corresponding hydrate. The less volatile the help gas the greater the fraction of smaller voids it occupies, and the greater its stabilizing It should be noted however, that help gases taken alone tend to form a type I structure (gas hydrates). Liquid hydrate clathrates are formed in conditions similar to those for gas hydrate^^.^. The corresponding physicochemical data are given in Table 1. Liquid

16 2. The Formation of Clathrates 16 2.2. The Formation of Clathrates Having a Water Host Lattice 16.2.2.2. Clathrates of the Liquid Hydrate Type.

215

Figure 1. The stacking of polyhedra in hydrates of type I1 (from ref. 1). TABLE 1. PHYSICOCHEMICAL DATAON LIQUIDHYDRATES (FROMREF. 1) Guest mol., M

Boiling point of M ("C)

P(0"C) (lo5 N m-'

1a

T(l atm) ("C)

Tc ("C; lo5 N

Lattice param. (Pm)

Hydrates of the formula 8 M ' 136 H,O - 24 - 10

13 42 61

0.268 0.155 (0.067)

4.8; 0.787 1.7; 0.213 1.6; 0.080

1744 1753 1730 1731 1730

Double hydrates of the formula 8 M 16 M' ,136 H,O With M'=H,S

cos C3H8

CH,Br C,H5C1 (CH3)2S C,H,Br CH,CI, CH,I CS, CHC1, n-C,H,Br CCI, C6H6

CH,CI-CH,Cl CC1,Br CCI,NO, with M'=H,Se C,H,Cl CHCI,

- 50 -45 5 13 38 38 42 43 46 61 71 77 80 82 104 112 13 61 71

cc1,

A

lo5 N m -

=

1 bar z 1 atm

0 8

11.3 13

1730 1740 1731 1726 1739 1726 1728 1737 1730 1729 1742 1746 1748 1751 1757 1760

10 15 19

1741 1745 1760

7.2 14 13 14 14.2 8 16.3 17.3 3.5

216

16.2 The Formation of Clathrates 16.2.2. The Formation of Clathrates Having a Water Host Lattice 16.2.2.3. Other Clathrate Hydrates.

hydrates exist for species having a boiling point < 60°C, but in the presence of help gases this limit is significantly raised and the number of possible double hydrates is much greater. Steric criteria are also valuable to explain the formation of liquid hydrate clathrates (Fig. 3, $16.2.2.1). They appear for molecules having dimensions between the free diameters of the largest voids of types I and I1 structures, d,,2 and dII,2.The situation is the same for double liquid hydrates, but here a type I structure is theoretically possible in some cases, e.g., CH,Br, COS and CH,I. For molecules with dimensions > 690 pm, no hydrates are formed. This selectivity in encaging certain molecules but not others has been used for fractionation of natural gas by clathration and for desalination of sea water6-'. (C

ChUS)

M. von Stackelberg, H. R. Miiller, Z. Electrochem., 58,25 (1954). M. von Stackelberg, H. Friihbuss, Z. Electrochem., 58, 99 (1954). M. von Stackelberg, W Meinhold, Z. Electrochem., 58, 40 (1954). R. M. Barrer, W. I. Stuart, Puoc. Roy. Soc. London, Ser. A , 242, 172 (1957). J. H. van der Waals, J. C. Platteeuw, in Advances in Chemical Physics, Vol. 2, I. Prigogine, ed., Interscience, New York, 1959. 6. E. Hammerschmidt, Am. Gas Assoc. Monthly 18, 273, (1936). 7. A. J. Barduhn, H. E. Towlson, Y. C. Hu, Am. Inst. Chem. Eng. J., 8, 176 (1962).

1. 2. 3. 4. 5.

16.2.2.3. Other Clathrate Hydrates.

Other clathrate-type hydrates are known with strongly soluble cations having specific dimensions as guest species, e.g., tri-n-butyl or triisoamyl ammonium, tetra-nbutyl or tetraisoamyl ammonium or phosphonium'. Water molecules and anions together form the open host lattice, but because of their large size, guest cations are included in more than one cavity, causing significant distortion of the host framework. The interactions between host and guest species are, therefore, much stronger than in the previously described gas and liquid hydrates, since they involve ionic bonding. For these reasons such compounds cannot be considered pure clathrates, but only pseudoclathrates. Their structures, however, are closely related to type I and I1 hydrates. They also exhibit a polyhedral framework in which the pentagonal dodecahedron is the most common. Several structural types are known; data concerning the three predominent ones are given in Table 12-6. In isoamyl ammonium salts hydrates (type I11 structure) the host lattice is based on the two-dimensional packing of pentagonal dodecahedra sharing faces. Between these layers of polyhedra are larger tetrakaidecahedra and pentakaidecahedra. The tetraisoamyl ammonium cation lies in a large cavity formed by two tetrakaidecahedra and two pentakaidecahedra, as shown on Fig. 1. In tetra-n-butyl ammonium salts hydrates (type IV structure), the host lattice is formed by the association of slightly distorted dodecahedra sharing faces forming groups

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc. 216

16.2 The Formation of Clathrates 16.2.2. The Formation of Clathrates Having a Water Host Lattice 16.2.2.3. Other Clathrate Hydrates.

hydrates exist for species having a boiling point < 60°C, but in the presence of help gases this limit is significantly raised and the number of possible double hydrates is much greater. Steric criteria are also valuable to explain the formation of liquid hydrate clathrates (Fig. 3, $16.2.2.1). They appear for molecules having dimensions between the free diameters of the largest voids of types I and I1 structures, d,,2 and dII,2.The situation is the same for double liquid hydrates, but here a type I structure is theoretically possible in some cases, e.g., CH,Br, COS and CH,I. For molecules with dimensions > 690 pm, no hydrates are formed. This selectivity in encaging certain molecules but not others has been used for fractionation of natural gas by clathration and for desalination of sea water6-'. (C

ChUS)

M. von Stackelberg, H. R. Miiller, Z. Electrochem., 58,25 (1954). M. von Stackelberg, H. Friihbuss, Z. Electrochem., 58, 99 (1954). M. von Stackelberg, W Meinhold, Z. Electrochem., 58, 40 (1954). R. M. Barrer, W. I. Stuart, Puoc. Roy. Soc. London, Ser. A , 242, 172 (1957). J. H. van der Waals, J. C. Platteeuw, in Advances in Chemical Physics, Vol. 2, I. Prigogine, ed., Interscience, New York, 1959. 6. E. Hammerschmidt, Am. Gas Assoc. Monthly 18, 273, (1936). 7. A. J. Barduhn, H. E. Towlson, Y. C. Hu, Am. Inst. Chem. Eng. J., 8, 176 (1962).

1. 2. 3. 4. 5.

16.2.2.3. Other Clathrate Hydrates.

Other clathrate-type hydrates are known with strongly soluble cations having specific dimensions as guest species, e.g., tri-n-butyl or triisoamyl ammonium, tetra-nbutyl or tetraisoamyl ammonium or phosphonium'. Water molecules and anions together form the open host lattice, but because of their large size, guest cations are included in more than one cavity, causing significant distortion of the host framework. The interactions between host and guest species are, therefore, much stronger than in the previously described gas and liquid hydrates, since they involve ionic bonding. For these reasons such compounds cannot be considered pure clathrates, but only pseudoclathrates. Their structures, however, are closely related to type I and I1 hydrates. They also exhibit a polyhedral framework in which the pentagonal dodecahedron is the most common. Several structural types are known; data concerning the three predominent ones are given in Table 12-6. In isoamyl ammonium salts hydrates (type I11 structure) the host lattice is based on the two-dimensional packing of pentagonal dodecahedra sharing faces. Between these layers of polyhedra are larger tetrakaidecahedra and pentakaidecahedra. The tetraisoamyl ammonium cation lies in a large cavity formed by two tetrakaidecahedra and two pentakaidecahedra, as shown on Fig. 1. In tetra-n-butyl ammonium salts hydrates (type IV structure), the host lattice is formed by the association of slightly distorted dodecahedra sharing faces forming groups

CLATHRATE

HYDRATE STRUCTURES

Structural type, parameters and ideal composition

TABLE 1.

LARGECATIONIC GUESTS

Cationic guest species

WITH

(FROMREFS. 5,6) Anionic species

16.2. The Formation of Clathrates 16.2.2. The Formation of Clathrates Having a Water Host Lattice 16.2.2.3. Other Clathrate Hydrates.

217

218

16.2. The Formation of Clathrates 16.2.2. The Formation of Clathrates Having a Water Host Lattice 16.2.2.3. Other Clathrate Hydrates.

Figure 1. Tetra-isoamyl ammonium cation in the clathrate hydrate (iso-C,H,,),NF. 38 H,O (from ref. 3).

of five. The residual space is filled by tetrakaidecahedra and hexakaidecahedra. Cubic type V structures are closely related to type I hydrates as previously described. Single crystals of these pseudo-clathrates are conveniently prepared as follows. The peralkyl salts are usually obtained by metathesis reactions between the peralkyl ammonium or sulfonium iodides and the slightly soluble corresponding silver salt taken in excess. The mixture of insoluble silver salts is removed by filtration and the supernatant (1-2 mol L- I ) cooled until crystals are produced. For tetraisoamylammonium salts cooling to RT is sufficient, whereas for the tetra-n-butyl ammonium salts cooling to 5-10°C is necessary2. Because of the relatively strong host-guest interactions, these pseudo-clathrates are more stable than gas and liquid hydrates and exhibit higher melting points. (C CROS)

1. 2. 3. 4. 5.

D. L. Fowler, W. V. Loebenstein, D. B. Pall, C . A. Krauss, J. Am. Chem. SOC.,62, 1140 (1940). R. K. McMullan, G. A. Jeffrey, J. Chem. Phys., 31, 1231 (1959). D. Feil, G. A. Jeffrey, J. Chem. Phys., 35, 1863 (1961). G. A. Jeffrey, R. K. McMullan, J . Chem. Phys., 37, 2231 (1962).

G. A. Jeffrey, Structural Factors in the Formation of Clathrate Hydrates, Dechema Monograph, 47,

849 (1962) 6. G. Beurskene, G. A. Jeffrey, R. K. McMullan, J. Chem. Phys., 39, 3311 (1963). 7. R. K. McMullan, M. Bonamico, G. A. Jeffrey,J. Chem. Phys., 39, 3295 (1963). 8. R. K. McMullan, G. A. Jeffery, J. Chem. Phys., 42, 2725 (1965).

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc.

219

16.2. The Formation of Clathrates 16.2.3. The Formation of Clathrates

16.2.3. The Formation of Clathrates Having a Silicon, Germanium and Tin Host Lattice Silicon and germanium have been found to form clathrate-type host lattices in which guest species are alkali atoms. The host lattices are exactly the same as those of type I and I1 hydrates’-4 (see $16.2.2). They are formed by atoms of only one kind, which are bonded together by strongly covalent forces, as in the Si or Ge diamond structures. The Si-Si or Ge-Ge bond lengths are of the same order of magnitude as in classical Si or Ge with the bond angles 109’28’ which characterizes the tetrahedral sp3 hybridization of the carbon family. In Si or Ge type I and I1 clathrates the free diameters of the voids are less than in the hydrates because of shorter Si-Si or Ge-Ge distances. Using the same notation as in $16.2.2 the free diameters of the available voids are d,,,, = 396 pm, d, = 410 pm, d, = 460 pm and d,,,, = 534 pm for silicon, and d,,,, = 406 pm, d,,l = 420 pm, dI,2= 464 pm and d,,,, = 560 pm for germanium. As in gas and liquid hydrates, the formation of Si and Ge clathrates is mostly governed by steric Assuming for the alkali-metal atoms a radius close to the covalent radii6, the analytical and crystallographic data listed in Table 1 can be explained by means of a diagram such as Fig. 1. Sodium, with a covalent radius of 157 pm, can occupy all the available voids of both structures, but the type I1 host lattice is the more stable. Experimental data agree with this argument: type I clathrates are formed in milder conditions than type 11. This type I1 structure formed with small sodium atoms is unique and has no hydrate equivalent: both voids are occupied stoichiometrically and a large range of composition is observed : 3 < x < 22.

-

-

’

TABLE1. ANALYTICALAND CRYSTALLOGRAPHIC DATAON SILICONAND GERMANIUM CLATHRATES (FROMREF. 3) Silicon framework Guest atom M Type I structure: Na K Rb

cs

Type I1 structure: Na

K Rb

cs

Exp. x -8 -7

Param. (pm) MxSi46

1019 1026 -5 1027 Not observed MxSi136 - 3 1462 z 11 1462 N 22 1471 Not observed Not observed -7

Obtained in a mixture. Rb,Ge,, + (GeH),. Structural type uncertain. x-ray pattern too diffuse ‘Obtained in a mixture Na,Ge,,, + NaGe,.

a

Germanium framework

Exp. x

Param. (pm)

MxGe4, Not observed -8 1066 a 1070 b

b

MxGel

36

1540

b

Not observed Not observed b

220

"I

16.2. The Formation of Clathrates 16.2.3. The Formation of Clathrates

Sodium

Potassium

Rubidium

cesium

Figure 1. Comparison of the covalent radii of alkali-metal atoms with the free radius of the available voids in clathrate-type silicon host lattices (from ref. 3).

Potassium, with a covalent radius of -203 pm, is too large to occupy the smallest rII,l voids. The clathrate obtained has a type I host lattice with the eight available voids occupied by an alkali-metal atom. Rubidium, for which the covalent radius is between and rI 2 , also forms a type I framework, but only the six largest voids are filled. Cesium atoms, with a covalent radius of 235 pm, can only occupy the eight largest voids of a type I1 structure. Experimental data are in good agreement with the expected ones. The same arguments are also valuable for the formation of germanium clathrates, but these experimental data are more scarce (Fig. 2).

-

(C CROS)

1. J. S. Kasper, P. Hagenmuller, M. Pouchard, C. Cros, Science, 250, 1713 (1965). 2. C. Cros, M. Pouchard, P. Hagenmuller, J. S. Kasper, Bull. Soc. Chim.Fr., 7, 2737 (1968). 3. C. Cros, Thesis of Doctorat, University of Bordeaux, France, 1970.

16.2. The Formation of Clathrates 16 2.3. The Formation of Clathrates 16.2.3.1. The Formation of Silicon and Germanium Clathrates.

221

Figure 2. Comparison of the covalent radii of alkali-metal atoms with the free radius of the available voids in clathrate-type germanium host lattices (from ref. 3).

4. C. Cros, M. Pouchard, P. Hagenmuller, Bull. Soc. Chirn. Fr., 2, 379 (1971). 5. C. Cros, M. Pouchard, P. Hagenmuller, J. Solid State Chem., 2, 570 (1970). 6 . L. H. Ahrens, Geochem. Cosmochim. Acta, 2, 155 (1952).

16.2.3.1. The Formation of Silicon and Germanium Clathrates by Thermal Decomposition of Silicides and Germanides. Clathrate Si and G e compounds can be prepared by thermal decomposition of alkali-metal silicides and germanides of the general formula MSi or MGe ( M = alkali metal)'-'. Thermal treatment is performed under a slow argon flow (3 nL/mn) or under a high vacuum n/m '). The required experimental conditions are listed in Table 1. The thermal decomposition under an Ar flow leads to the formation of the same clathrate structure as by means of the high-vacuum treatment. This is true for KSi, but not for NaSi, for which the two kinds of networks are formed, according to the conditions of decomposition.

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc. 16.2. The Formation of Clathrates 16 2.3. The Formation of Clathrates 16.2.3.1. The Formation of Silicon and Germanium Clathrates.

221

Figure 2. Comparison of the covalent radii of alkali-metal atoms with the free radius of the available voids in clathrate-type germanium host lattices (from ref. 3).

4. C. Cros, M. Pouchard, P. Hagenmuller, Bull. Soc. Chirn. Fr., 2, 379 (1971). 5. C. Cros, M. Pouchard, P. Hagenmuller, J. Solid State Chem., 2, 570 (1970). 6 . L. H. Ahrens, Geochem. Cosmochim. Acta, 2, 155 (1952).

16.2.3.1. The Formation of Silicon and Germanium Clathrates by Thermal Decomposition of Silicides and Germanides. Clathrate Si and G e compounds can be prepared by thermal decomposition of alkali-metal silicides and germanides of the general formula MSi or MGe ( M = alkali metal)'-'. Thermal treatment is performed under a slow argon flow (3 nL/mn) or under a high vacuum n/m '). The required experimental conditions are listed in Table 1. The thermal decomposition under an Ar flow leads to the formation of the same clathrate structure as by means of the high-vacuum treatment. This is true for KSi, but not for NaSi, for which the two kinds of networks are formed, according to the conditions of decomposition.

16.2. The Formation of Clathrates 16.2.3. The Formation of Clathrates 16.2.3.2. The Formation of Silicon, Germanium and Tin Clathrates.

222

TABLE1. CONDITIONS OF FORMATIONS OF SILICON AND GERMANIUM CLATHRATES BY THERMAL DECOMPOSITION OF ALKALI SILICIDES AND GERMANIDES MSi AND MGe (FROMREF. 6)

Clathrate

Rb,Si4, K8Ge46

RbxGe4, NaxSi

13 6

x=3 x = 11 x = 22 Cs ,si 13 6 NaxGe136

Starting material

Condition of decomposition

Temperature range ("C)

NaSi KSi KSi RbSi KGe RbGe

argon flow argon flow vacuum vacuum vacuum vacuum

410--430 410-480 320-480 320-480 320-380 320-370

NaSi NaSi Na,Si, 36 CsSi NaGe

vacuum vacuum Na vapor pressure vacuum vacuum

445 340 280-320 320-470 330-400

Reaction time (days)

(C CROS)

I. 2. 3. 4. 5. 6. 7.

E. Hohmann, Z. Anorg. Allg. Chem., 257, 113 (1948). R. Schafer, W. Klemm, Z. Anorg. Allg. Chem., 312, 214 (1961). C. Cros, M. Pouchard, P. Hagenmuller, C. R. Hebd. Seances Acad. SCL,4764 (1965) C. Cros, M. Pouchard, P. Hagenmuller, Bull. SOC.Chzm. Fr., 2, 379 (1971). C. Cros, M. Pouchard, P. Hagenmuller, J. Solid State Chern., 2, 570 (1970). C. Cros, Thesis of Doctoral, Uniljersity of Bordeaux, France, 1970. C. Cros, J. C. Benejat, Bull. SOC.Chim. Fr., 5, 1736 (1972).

16.2.3.2. The Formation of Silicon, Germanium and Tin Clathrates from the Elements. Type I clathrate structures are synthesized from finely divided Si or Ge powder with potassium vapor at 700°C under low pressure',2. The homologous Sn compound is formed by melting together stoichiometric amounts of potassium and tin2. These conditions are easier to realize than the previous ones, since they do not require the intermediate, dangerous preparation of the highly reactive silicides and germanides, MSi and MGe. The only difference in result is that the value of the insertion rate, x, is exactly 8, which can be explained by the tendancy of the high-vacuum treatment to give nonstoichiometric compounds, whereas increased pressure favors complete occupancy of the available voids. The unit cell parameters are 1030 pm for K,Si,,, 1971 pm for K, Ge,, and 1203 pm for K,Sn,,, in agreement with those previously reported in Table 1, $16.2.3.

1. J. Gallmeier, H. Schafer, W. Weiss, 2. Naturforsch., Teil B, 22, 1080 (1967). 2. J. Gallmeier, H. Schafer, A. Weiss, Z . Naturforsch., Teil B, 24, 665 (1969).

( C . CROS)

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc. 16.2. The Formation of Clathrates 16.2.3. The Formation of Clathrates 16.2.3.2. The Formation of Silicon, Germanium and Tin Clathrates.

222

TABLE1. CONDITIONS OF FORMATIONS OF SILICON AND GERMANIUM CLATHRATES BY THERMAL DECOMPOSITION OF ALKALI SILICIDES AND GERMANIDES MSi AND MGe (FROMREF. 6)

Clathrate

Rb,Si4, K8Ge46

RbxGe4, NaxSi

13 6

x=3 x = 11 x = 22 Cs ,si 13 6 NaxGe136

Starting material

Condition of decomposition

Temperature range ("C)

NaSi KSi KSi RbSi KGe RbGe

argon flow argon flow vacuum vacuum vacuum vacuum

410--430 410-480 320-480 320-480 320-380 320-370

NaSi NaSi Na,Si, 36 CsSi NaGe

vacuum vacuum Na vapor pressure vacuum vacuum

445 340 280-320 320-470 330-400

Reaction time (days)

(C CROS)

I. 2. 3. 4. 5. 6. 7.

E. Hohmann, Z. Anorg. Allg. Chem., 257, 113 (1948). R. Schafer, W. Klemm, Z. Anorg. Allg. Chem., 312, 214 (1961). C. Cros, M. Pouchard, P. Hagenmuller, C. R. Hebd. Seances Acad. SCL,4764 (1965) C. Cros, M. Pouchard, P. Hagenmuller, Bull. SOC.Chzm. Fr., 2, 379 (1971). C. Cros, M. Pouchard, P. Hagenmuller, J. Solid State Chern., 2, 570 (1970). C. Cros, Thesis of Doctoral, Uniljersity of Bordeaux, France, 1970. C. Cros, J. C. Benejat, Bull. SOC.Chim. Fr., 5, 1736 (1972).

16.2.3.2. The Formation of Silicon, Germanium and Tin Clathrates from the Elements. Type I clathrate structures are synthesized from finely divided Si or Ge powder with potassium vapor at 700°C under low pressure',2. The homologous Sn compound is formed by melting together stoichiometric amounts of potassium and tin2. These conditions are easier to realize than the previous ones, since they do not require the intermediate, dangerous preparation of the highly reactive silicides and germanides, MSi and MGe. The only difference in result is that the value of the insertion rate, x, is exactly 8, which can be explained by the tendancy of the high-vacuum treatment to give nonstoichiometric compounds, whereas increased pressure favors complete occupancy of the available voids. The unit cell parameters are 1030 pm for K,Si,,, 1971 pm for K, Ge,, and 1203 pm for K,Sn,,, in agreement with those previously reported in Table 1, $16.2.3.

1. J. Gallmeier, H. Schafer, W. Weiss, 2. Naturforsch., Teil B, 22, 1080 (1967). 2. J. Gallmeier, H. Schafer, A. Weiss, Z . Naturforsch., Teil B, 24, 665 (1969).

( C . CROS)

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc.

16.3. The Formation of Tunnel Structures 16.3.1. Ion Exchange Compounds with a tunnel structure can be formed by cation-exchange reactions, on condition that their host lattice, which is generally a tridimensional infinite anion, allows the possibility of migration of the cations through the framework. The mobility of the intercalated cations is favored when the cavities are interconnected. For this reason, most of the ion exchangers of this class are characterized by an intersecting tunnel structure. Structures with parallel tunnels do not generally exhibit ion-exchange properties unless the tunnel diameter is much greater than that of the migration cation. The most common ion exchangers with a tunnel structure known at the present time are oxides. Among the intersecting tunnel structures, five classes must be distinguished: zeolites, pyrochlores, A,M,O,, compounds and their intergrowths, alkali antimonates and alkali molybdophosphates. Crancrinite and Na,YSi,O, 2-type silicates are the only known structures with parallel tunnels that show ion-exchange properties. (B RAVEAU)

16.3.1.l. Zeolites with Intersecting Tunnel Frameworks. The zeolites form a very important family of aluminosilicates and have been studied for their selective ion-exchange properties','. The host lattice of these compounds is built up of corner-sharing A10, and SiO, tetrahedra delimiting large cavities where the exchangeable cations are inserted, as for example, in the mineral faujasite, (Na,, Ca)[Ai,Si,O,,] .9 H,O, which is characterized by cubooctahedral cavities whose packing forms intersecting tunnels. Among the zeolites, the minerals have an important place (Table 1).The exchangeabie cations appearing in these compounds are predominantly Na', K', Ca2' and Ba". Synthetic zeolites have been obtained in recent years by hydrothermal preparation from solutions containing silica, alumina and alkali (Table l), but the use of the ionexchange properties of these materials remains an important method for their synthesis with various exchangeable cations i.e., Li', Ag', C s c , Tl', Rb', Sr2+, Cd", Pb", NH:, H,O' and alkylammonium ions such as [NH3CH3]+, [NH,(CH,),]' and [NH(CH,),] +.In the same way, feldspachoids, which can be considered zeolites in which the guest molecules are salts or bases, exhibit cation-exchange properties. This is true for the nosean-sodalite family, which includes, sodalite, (NaAlSiO,),(NaCl), , basic sodalite, (NaAlSiO,),(NaOH.H,O), nosean, (NaAlSiO,),(Na,SO,), ultramarine, (NaAlSiO,),Na,Sx, helvite, [(Mn, Fe, Zn)(BeSiO,) (Mn, Fe, Zn)S, hauynite, (NaCa, ,A1Si0,),(Na2, Ca)SO, and lazurite (NaCa, ,AlSiO,),(Na,, Ca)(SO,, S, 2 Cl). An example of such an exchange is provided by the action of hydrogen chloride on the basic sodalite which allows the migration of the sodium ions outside of the structure but not normally of anions, while water is formed in the structure according to: (OH-)sodahte

+ (Na')sodalite + HC1 gas

-

Na'C1-

+ (HZo)sodalite

(a)

outside

223

224

16.3. The Formation of Tunnel Structures 16.3.1. Ion Exchange 16.3.1.1. Zeolites with Intersecting Tunnel Frameworks.

In the same way the synthetic ultramarine exchanges its Na' ion for such ions as Li', K', Ag', T l t , Pb2+,Zn2' and Cd2+ with modification of color. Owing to their rigid framework, natural and synthetic zeolites are generally characterized by selective ion-exchange properties. The selectivity of the zeolites is determined by lattice forces and steric effects. The ionic size of the cation plays an important part in the selectivity of the exchange reaction. For example, analcite can exchange its Na' for Li ', K' and Ag but not for Cs'. For the same reason, no more than 60% of Na' can be exchanged for K' in basic soladite. In a similar way the exchange of Na' for [N(CH,),] 'in faujasite cannot exceed 20% because of the small volume of the tunnels3. The ion-exchange properties of certain zeolites are characterized by very simple behavior; for example, basic sodalite for which the exchanges of Na for KfLit or Ag TABLE1. SOME EXAMPLES OF NATURALLY OCCURRING ZEOLITES AND OF THE CORRESPONDIP~G SYNTHETIC ZEOLITES OBTAINED BY HYDROTHERMAL SYNTHESIS Natural zeolites Analcite Na,[AlSi,O& 2 H,O Wairakite Ca[AlSi,O,], 2 H,O Chabazite (Ca, Na,)[AlSi,O,], 6 H,O Gmelinite (Ca, Na,)[AlSi,O,], 6 H,O Levynite Ca[Al,Si,O,,] 5 H,O Erionite (Ca, Mg, Na,, K,)(A1Si3O,),~6 H,O Fauj asite (Na,, Ca)[A1,Sij0,,]~9 H,O Heulandite Ca[AlSi,O,], ' 5 H,O Clinoptilolite (Ca, Na,, K,)[AlSi,O,,],~X H,O Stilbite (Na,; Ca)[AISi,O,], 6 H,O Brewsterite (Sr, Ca, Ba)[AlSi,O,], ' 5 H,O Epistilbite Ca[AlSi,O,], ' 5 H,O Ferrierite (Na,, Mg)CAISijO,,l,~ 6 H,O Mordenite (Ca, K,, Na,)(AlSi,O,,),~ 6.7 H,O Ptilolite (Ca, K,Na,)[AlSi,O,,], '4H,O Dachiardite (Ca, K,, Na,)[AI,Si,O,,]~ 7 H,O

Synthetic zeolites Na; Ca; Sr; [NH,]; K; Rb; Cs; T1

(Ca, Na), Na, Sr Na form-Na aluminogermanate Na gallogermanate N(CH,),form

Ca

Na; Ca; Sr

16.3. The Formation of Tunnel Structures 16.3.1. Ion Exchange Zeolites with Intersecting Tunnel Frameworks. 16.3.1.l.

225

TABLE1. (Continued) Natural zeolites

Synthetic zeolites

Natrolite Na,A1,Si30,,~2 H,O Scolecite Ca[Al,Si,O,,] 3 H,O Mesolite Between Natrolite and Scolecite Edingtonite Ba[A1,Si,01,].3 H,O Thomsonite (Na,, Ca)[AlSiO,], ' 3 H,O Gonnardite (Na,, Ca)[Al,Si,O,,] .3.5 H,O Phillipsite (K,, Ca)[AlSi,O,], 4.5 H,O Harmotome (K,, Ba)lA1,Si501,1 5 H,O Wellsite (Ca, Ba, Na,, K,)[AlSi,O,], 3.5 H,O Laumonite Ca[A1SiZ0,],.4 H,O Gismondite Ca[AlSiO,],.4 H,O

Ca thomsonite

Na gallosilicate, Na gallogermanate

(Na, K), K Ca; Sr; Ba; (K, Na); NH,; N

obey the mass action law, leading to the sequence of selectivity Ag' % Na' > Li+ > K'. Similarly, the selectivity of all zeolites for Ag' and T1' are worthy of note. The high selectivity of ultramarine for N a + and of chabazite for Kf is also well known. All these observations show that the ionic size is not the only parameter which governs the ionexchange properties, but that the lattice forces are predominant to explain the behavior of these compounds. The selectivity rules are modified when the zeolites contain cavities of different sorts. The synthetic zeolite, Linde Sieves 4A, Na,,[A102],,[Si0,]12~x H20,shows such behavior: all 13 Na' can be exchanged for Ag', while only 12 Na' can be replaced by Tl', because one exchangeable site is structurally different from the 12 other sites. The ion-exchange isotherms of this compound show that this zeolite behaves like a mixture of two ion exchangers with different selectivity coefficients, in agreement with the presence of two sorts of cavities. Most zeolites do not show such regular behavior. These are characterized by strong variation of their selectivity coefficient with ionic composition. This phenomenon is observed when ion exchange is accompanied by a phase transition. This is particularly true for the exchange of N a + for K' in analcite which involves a transformation to leucite: Na'

-

analcite

+ K:q

K+

-

leucite

+ Na:q

(b)

Analcite can accommodate up to 20% K' ions without modifying its structure, and similarly leucite can accommodate up to 20 % N a + ion without modifying its framework,

226

16.3. The Formation of Tunnel Structures 16.3.1. Ion Exchange 16.3.1.2. Pyrochlores A, .M,O,

so that during the exchange reaction a lattice rearrangement occurs between these critical values. Between these limits of exchange both phases coexist. Furthermore, the ion exchange is reversible: leucite can be converted to analcite by exchanging K + for Na'. However, the conversions analcite -+ leucite and leucite -+ analcite do not occur for the same composition of solution. This hysteresis likely results from the existence of a potential energy barrier which delays the rearrangement of the structure, so that a metastable equilibrium is obtained. Sometimes the rearrangement allows the penetration of larger ions which are forbidden owing to their size in both final and initial compounds: this is the case in the introduction of Cs' during the transition N a + - analcite K + leucite. An irregular behavior can also be observed when there is no phase transition. It often occurs for exchange of ions of different valences, for example, replacement of Na' by Ca2+ on the 12 identical sites of Linde Sieves 4A. A quantitative theory explaining the behavior of these irregular systems has been proposed3. (B RAVEAU)

1. R. M. Barrer, in Non-stoichrometrzc compounds, L. Mandelcorn, ed., Academic Press, New York, 1964, pp. 309-399; excellent review. 2. F. Halferlich, Ion exchange, McGraw-Hill Book Company, New York, 1962; excellent review. 3. R. M. Barrer, J D. Falloner, Proc. Roy. Soc. London, Sev.A , 236, 227 (1956).

16.3.1.2. Pyrochlores A,

xM,O,

Besides the stoichiometric pyrochores A,M,X,, a family of oxides, fluorides and oxyfluorides, A, .,M,X, (Table l), with the same octahedral framework has been isolated, with A = K, Rb, Cs, T1. The host lattice, M,X,, of these compounds is built up of corner-sharing octahedra forming distorted hexagonal tunnels running along the (1 10) direction of the cubic cell. Most of the references concerning the structure and the synthesis by ion exchange of these compounds can be obtained from ref. 1. The 0,, cages that share their faces form intersecting tunnels, allowing good mobility of the cations through the framework. The ion-exchange properties of these compounds, and particularly those of the thallium compounds, have allowed the preparation of the nonstoichiometric hydronium pyrochlores using an acid solution according to the equation:

where M = Ta, Nb and 0 < x < 1. Hydrated pyrochlores of smaller ions can also be synthesized by exchange reaction between a solution of the corresponding salt and the hydronium pyrochlore HTaWO,.H,O according to the equation: HMWO,.H,O

+ A , ' , ~ A T a W O , . H , O + H&

(b)

whereA+ = N a + , L i ' , A g + ; M = Ta5+,Nb5',Sb5'.Theexchangeofamonovalention for a bivalent ion is more difficult: a quantitative replacement of the hydronium ion by a bivalent ion, however, is possible for the limiting tantalate, H,Ta,O,~H,O: H,Ta,O,. where A

=

Ca, Cd, Pb.

H,O

+ A::

ATa,O,. H,O

+ 2 HA

(c)

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc.

226

16.3. The Formation of Tunnel Structures 16.3.1. Ion Exchange 16.3.1.2. Pyrochlores A, .M,O,

so that during the exchange reaction a lattice rearrangement occurs between these critical values. Between these limits of exchange both phases coexist. Furthermore, the ion exchange is reversible: leucite can be converted to analcite by exchanging K + for Na'. However, the conversions analcite -+ leucite and leucite -+ analcite do not occur for the same composition of solution. This hysteresis likely results from the existence of a potential energy barrier which delays the rearrangement of the structure, so that a metastable equilibrium is obtained. Sometimes the rearrangement allows the penetration of larger ions which are forbidden owing to their size in both final and initial compounds: this is the case in the introduction of Cs' during the transition N a + - analcite K + leucite. An irregular behavior can also be observed when there is no phase transition. It often occurs for exchange of ions of different valences, for example, replacement of Na' by Ca2+ on the 12 identical sites of Linde Sieves 4A. A quantitative theory explaining the behavior of these irregular systems has been proposed3. (B RAVEAU)

1. R. M. Barrer, in Non-stoichrometrzc compounds, L. Mandelcorn, ed., Academic Press, New York, 1964, pp. 309-399; excellent review. 2. F. Halferlich, Ion exchange, McGraw-Hill Book Company, New York, 1962; excellent review. 3. R. M. Barrer, J D. Falloner, Proc. Roy. Soc. London, Sev.A , 236, 227 (1956).

16.3.1.2. Pyrochlores A,

xM,O,

Besides the stoichiometric pyrochores A,M,X,, a family of oxides, fluorides and oxyfluorides, A, .,M,X, (Table l), with the same octahedral framework has been isolated, with A = K, Rb, Cs, T1. The host lattice, M,X,, of these compounds is built up of corner-sharing octahedra forming distorted hexagonal tunnels running along the (1 10) direction of the cubic cell. Most of the references concerning the structure and the synthesis by ion exchange of these compounds can be obtained from ref. 1. The 0,, cages that share their faces form intersecting tunnels, allowing good mobility of the cations through the framework. The ion-exchange properties of these compounds, and particularly those of the thallium compounds, have allowed the preparation of the nonstoichiometric hydronium pyrochlores using an acid solution according to the equation:

where M = Ta, Nb and 0 < x < 1. Hydrated pyrochlores of smaller ions can also be synthesized by exchange reaction between a solution of the corresponding salt and the hydronium pyrochlore HTaWO,.H,O according to the equation: HMWO,.H,O

+ A , ' , ~ A T a W O , . H , O + H&

(b)

whereA+ = N a + , L i ' , A g + ; M = Ta5+,Nb5',Sb5'.Theexchangeofamonovalention for a bivalent ion is more difficult: a quantitative replacement of the hydronium ion by a bivalent ion, however, is possible for the limiting tantalate, H,Ta,O,~H,O: H,Ta,O,. where A

=

Ca, Cd, Pb.

H,O

+ A::

ATa,O,. H,O

+ 2 HA

(c)

16.3. The Formation of Tunnel Structures 16.3.1. Ion Exchange 16.3.1.2. Pyrochlores A, xM,O,

227

TABLE1. SOMEEXAMPLES OF PYROCHLORES AM,X,, AND A1+xMZX6THAT EXHIBIT ION EXCHANGE PROPERTIES DUE TO THEIRINTERSECTING TUNNEL STRUCTURE AMWO, AMTeO, A(M1/3W5,3)06 A(M 1/4w7/4)06 AL10 ZWl.80, T1l + X ( ~ I +- xX l 0 6~ I T1, -xNb,O, -%F, AM,O,F CsMM’O,F, CsMM’O,F, CsMTiOF, AMM’F, CsMM’F, AMgAIF, CsMCrF,

A = K,, or K, Rb, T1, Cs; M = Ta, Nb, Sb A = K, Rb, TI, Cs; M = Nb, Ta A = K, Rb, TI, Cs; M = Cr, Fe A = K, Rb, CS; M = A1 A = Rb, TI, Cs; M = Ga A = Rb, TI; M = Mg, CO A = Rb, TI M=Ta,Nb O < x < l

O<X
=

Co, Fe, Mn

In the case of strontium and barium, only a partial exchange has been observed. Molten and solid salts such as silver and lead nitrates’ undergo partial exchange of thallium TINbO,. All the exchange reactions in aqueous medium are reversible, but they d o not allow preparation of pyrochlores with large amounts of A ion: the exchange of HZq in H , + , M 2 0 6 ~ H 2 for 0 monovalent cations such as Na’ or K’ is not quantitative for x > 0. A solid-state exchange reaction between the alkali chlorides and the thallium pyrochlores has been developed3, based on the volatility of thallium chloride, which allows the synthesis of sodium- and potassium-rich pyrochlores: Tl1+.(Tal+,W1-.)O6

+ (1 + x) ACl-

500°C

(1

+ x) TlCl + A l + x ( T a l + x W l ~ , . ) 0 6 1H20

+ ~ ( ~ ~ l- x+ )xO~ ~l ’ ~ ~ O where A’ = Na’, K + and 0 < x < 1. The hydroxonium pyrochlores behave like protonic acids: [NH,][TaWO,] and [N,H,][TaWO,] can be synthesized in NH, and N,H, solutions. Unlike the zeolites, alkylammonium ions cannot penetrate in the framework because of the smaller size of the tunnels. 1. B. Raveau, Rev. Inorg. Chem., 1, 81 (1979). 2. J. L. Fourquet, F. Plet, R. de Pape, Mat. Res. Bull., 10, 933 (1975). 3. C. Michel, These de Doctorat d’Etat, Universite de Cam, 1974.

(8.RAVEAU)

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc. 228 16.3. The Formation of Tunnel Structures

16.3.1. Ion Exchange 16.3.15 . Ammonium and Potassium 12 Molybdophosphates A,PO,Mo,,O,,.

16.3.1-3. A,M,018 Structure and lntergrowths with Pyrochlores.

The octahedral host lattice of the hexagonal A2M,TiOl, (M = Ta, Nb; A = Rb, T1, Cs) oxides, which is built up of blocks of 2 x 3 edge-sharing octahedra, forms OZ1 cavities very similar to those observed in pyrochlores. It results in tunnels that intersect in a plane. The similarity of this structure with that of pyrochlores and also with the hexagonal tungsten bronze allows the synthesis of intergrowths of these structures' : A4MloM'030(A = Rb, Cs, T1; M = Ta, Nb; M' = Mo, W) and ANb29,2078(A = Rb, Cs, Tl). The host lattice of the latter oxides form intersecting tunnels similar to those observed in pyrochlores. All these structures, like pyrochlores, are characterized by good mobility of the A ions and exhibit similar exchange properties: AzM,Ti0,,.2 H,O, 1 OH,O have thus been synthesized by A4Ml0M'O30 . 4 H 2 0 and AiOM29 20,8. exchange reactions for A' = H', Li', Na', K + and Ag' 1-3. (B. RAVEAU)

1. B. Raveau, Rev. Inorg. Chem., I , 81 (1979). 2. C. Michel, A. Guymarc'h, B. Raveau, J. Solid State Chem., 22, 393 (1977). 3. C. Robert, G. Desgardin, B. Raveau, J. Inorg. Nucl. Chem., 41, 893 (1979).

16.3.1.4. Alkali Antimonates K,Sb,O,,

and K,Sb,O,,.

Monoclinic K2Sb40, and orthorhombic K3Sb5OI4are characterized by octahedral host lattices, in which the SbO, octahedra share corners and edges'. These frameworks form intersecting tunnels in both structures running along the b- and c-axes. The interconnection of the large cavities produces ion-exchange properties similar to those observed for other intersecting tunnel structures: K + can be exchanged for Na', Ag', Rb', TI+ and Cs' in molten salts'-'. Hydrated oxides have been synthesized by ion exchange in aqueous medium2, e.g., [H30]2Sb40,1 [H,0],,,~A2Sb,0,,~x H,O (A = Na, T1, K-Ag, K-Li) and Li3Sb5Ol4[HZO],,,. (8. RAVEAU)

1. H. Y. P. Hong, Acta Crystallogr., B30, 945 (1974). 2. Y. Piffard, M. Dion, M. Tournoux, C.R. Hebd. Seances Acad. Sci.,290, 437 (1980).

16.3.1.5. Ammonium and Potassium 12 Molybdophosphates A3P0,Mo,,0,e.

These compounds are built up of anionic units formed of a PO4 tetrahedron and 12 MO, octahedra, whose packing delimits intersecting tunnels where the A ions are located'. The ion-exchange properties of the ammonium compounds2, which can be partially replaced by alkali ions, are easily explained by the interconnection of the tunnels. The incompleteness of the exchange results of a phase transition during the exchange. (B. RAVEAU)

1. J. C. A. Boeyens, G. J. McDoual, J. Van R. Smit, J. Solid State Chem., 18, 191 (1976). 2. C. J. Coetzee, E. F. C. H. Rohwer, J. Inorg. Nucl. Chem., 32, 1711 (1970).

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc. 228 16.3. The Formation of Tunnel Structures

16.3.1. Ion Exchange 16.3.15 . Ammonium and Potassium 12 Molybdophosphates A,PO,Mo,,O,,.

16.3.1-3. A,M,018 Structure and lntergrowths with Pyrochlores.

The octahedral host lattice of the hexagonal A2M,TiOl, (M = Ta, Nb; A = Rb, T1, Cs) oxides, which is built up of blocks of 2 x 3 edge-sharing octahedra, forms OZ1 cavities very similar to those observed in pyrochlores. It results in tunnels that intersect in a plane. The similarity of this structure with that of pyrochlores and also with the hexagonal tungsten bronze allows the synthesis of intergrowths of these structures' : A4MloM'030(A = Rb, Cs, T1; M = Ta, Nb; M' = Mo, W) and ANb29,2078(A = Rb, Cs, Tl). The host lattice of the latter oxides form intersecting tunnels similar to those observed in pyrochlores. All these structures, like pyrochlores, are characterized by good mobility of the A ions and exhibit similar exchange properties: AzM,Ti0,,.2 H,O, 1 OH,O have thus been synthesized by A4Ml0M'O30 . 4 H 2 0 and AiOM29 20,8. exchange reactions for A' = H', Li', Na', K + and Ag' 1-3. (B. RAVEAU)

1. B. Raveau, Rev. Inorg. Chem., I , 81 (1979). 2. C. Michel, A. Guymarc'h, B. Raveau, J. Solid State Chem., 22, 393 (1977). 3. C. Robert, G. Desgardin, B. Raveau, J. Inorg. Nucl. Chem., 41, 893 (1979).

16.3.1.4. Alkali Antimonates K,Sb,O,,

and K,Sb,O,,.

Monoclinic K2Sb40, and orthorhombic K3Sb5OI4are characterized by octahedral host lattices, in which the SbO, octahedra share corners and edges'. These frameworks form intersecting tunnels in both structures running along the b- and c-axes. The interconnection of the large cavities produces ion-exchange properties similar to those observed for other intersecting tunnel structures: K + can be exchanged for Na', Ag', Rb', TI+ and Cs' in molten salts'-'. Hydrated oxides have been synthesized by ion exchange in aqueous medium2, e.g., [H30]2Sb40,1 [H,0],,,~A2Sb,0,,~x H,O (A = Na, T1, K-Ag, K-Li) and Li3Sb5Ol4[HZO],,,. (8. RAVEAU)

1. H. Y. P. Hong, Acta Crystallogr., B30, 945 (1974). 2. Y. Piffard, M. Dion, M. Tournoux, C.R. Hebd. Seances Acad. Sci.,290, 437 (1980).

16.3.1.5. Ammonium and Potassium 12 Molybdophosphates A3P0,Mo,,0,e.

These compounds are built up of anionic units formed of a PO4 tetrahedron and 12 MO, octahedra, whose packing delimits intersecting tunnels where the A ions are located'. The ion-exchange properties of the ammonium compounds2, which can be partially replaced by alkali ions, are easily explained by the interconnection of the tunnels. The incompleteness of the exchange results of a phase transition during the exchange. (B. RAVEAU)

1. J. C. A. Boeyens, G. J. McDoual, J. Van R. Smit, J. Solid State Chem., 18, 191 (1976). 2. C. J. Coetzee, E. F. C. H. Rohwer, J. Inorg. Nucl. Chem., 32, 1711 (1970).

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc. 228 16.3. The Formation of Tunnel Structures

16.3.1. Ion Exchange 16.3.15 . Ammonium and Potassium 12 Molybdophosphates A,PO,Mo,,O,,.

16.3.1-3. A,M,018 Structure and lntergrowths with Pyrochlores.

The octahedral host lattice of the hexagonal A2M,TiOl, (M = Ta, Nb; A = Rb, T1, Cs) oxides, which is built up of blocks of 2 x 3 edge-sharing octahedra, forms OZ1 cavities very similar to those observed in pyrochlores. It results in tunnels that intersect in a plane. The similarity of this structure with that of pyrochlores and also with the hexagonal tungsten bronze allows the synthesis of intergrowths of these structures' : A4MloM'030(A = Rb, Cs, T1; M = Ta, Nb; M' = Mo, W) and ANb29,2078(A = Rb, Cs, Tl). The host lattice of the latter oxides form intersecting tunnels similar to those observed in pyrochlores. All these structures, like pyrochlores, are characterized by good mobility of the A ions and exhibit similar exchange properties: AzM,Ti0,,.2 H,O, 1 OH,O have thus been synthesized by A4Ml0M'O30 . 4 H 2 0 and AiOM29 20,8. exchange reactions for A' = H', Li', Na', K + and Ag' 1-3. (B. RAVEAU)

1. B. Raveau, Rev. Inorg. Chem., I , 81 (1979). 2. C. Michel, A. Guymarc'h, B. Raveau, J. Solid State Chem., 22, 393 (1977). 3. C. Robert, G. Desgardin, B. Raveau, J. Inorg. Nucl. Chem., 41, 893 (1979).

16.3.1.4. Alkali Antimonates K,Sb,O,,

and K,Sb,O,,.

Monoclinic K2Sb40, and orthorhombic K3Sb5OI4are characterized by octahedral host lattices, in which the SbO, octahedra share corners and edges'. These frameworks form intersecting tunnels in both structures running along the b- and c-axes. The interconnection of the large cavities produces ion-exchange properties similar to those observed for other intersecting tunnel structures: K + can be exchanged for Na', Ag', Rb', TI+ and Cs' in molten salts'-'. Hydrated oxides have been synthesized by ion exchange in aqueous medium2, e.g., [H30]2Sb40,1 [H,0],,,~A2Sb,0,,~x H,O (A = Na, T1, K-Ag, K-Li) and Li3Sb5Ol4[HZO],,,. (8. RAVEAU)

1. H. Y. P. Hong, Acta Crystallogr., B30, 945 (1974). 2. Y. Piffard, M. Dion, M. Tournoux, C.R. Hebd. Seances Acad. Sci.,290, 437 (1980).

16.3.1.5. Ammonium and Potassium 12 Molybdophosphates A3P0,Mo,,0,e.

These compounds are built up of anionic units formed of a PO4 tetrahedron and 12 MO, octahedra, whose packing delimits intersecting tunnels where the A ions are located'. The ion-exchange properties of the ammonium compounds2, which can be partially replaced by alkali ions, are easily explained by the interconnection of the tunnels. The incompleteness of the exchange results of a phase transition during the exchange. (B. RAVEAU)

1. J. C. A. Boeyens, G. J. McDoual, J. Van R. Smit, J. Solid State Chem., 18, 191 (1976). 2. C. J. Coetzee, E. F. C. H. Rohwer, J. Inorg. Nucl. Chem., 32, 1711 (1970).

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc. 16.3. The Formation of Tunnel Structures 16.3.2. by Electrolysis

229

16.3.1-6. Structures with Parallel Tunnels. Cancrinite, [(Na, Ca,,,)AlSiO,],, (Na,, Ca)[C03] is one of the rate structures built up of parallel tunnels that exhibits ion-exchange properties of Na* and Ca" for all alkali cations up to Rb' ' - 3 . This is probably due the great dimension of its tunnels, which have free diameters of about 6 A. It is also the case with basic cancrinite, for which ion exchanges of Ag' for Na' and Li' have been observed with a hysteresis loop similar to that observed for structure transitions. It is, however, worthy of note that the exchange for Cs' is not possible for the latter structure despite the size of Cs' which is much smaller than that of a tunnel. It seems, therefore, that the ion-exchange reactions can only take place in a parallel tunnel structure, and only if the free diameter of the tunnel is considerably greater than that of the exchanged ions. This would allow the incoming and outgoing ions to move freely in opposite directions in the tunnels without mutual repulsion. A series of hexagonal silicates, Na,[MSi,O,,] isostructural with Na,[YSi,O,,], have been synthesized for M = Fe, In, Se, Y and the lanthanides Lu-Sm4. The host lattice of these compounds is built up of Si,,03, rings whose stacking forms columns that are connected through MO, octahedra to form tunnels parallel to the c axis. However, this structure does not contain purely parallel tunnels; some are interconnected, allowing Na' ions to move easily through the structure. The ion-exchange properties of these oxides are thus not surprising: Na' can be exchanged for Li', Ag', K + in molten nitrates. In conclusion, the ion-exchange properties of these materials, which reflect a certain mobility of the cations in the framework, can play an important part in the synthesis of solid electrolytes. (6 RAVEAU)

1. R. M. Barrer, in Non-stoichiometric Compounds, L. Mandelcorn, ed., Academic Press, New York, 1964, pp, 309-399; excellent review. 2. F. Helferiich, Zon Exchange, McGraw-Hill Book Company, New York, 1962; excellent review. 3. R. M. Barrer, J. D. Falloner, Pvoc. Roy. SOC.London, Ser. A , 236, 227 (1956). 4. R. D. Shannon, B. E. Taylor, T. E. Gier, H. Y. Chen, T. Berzins, Znorg. Chem., 17, 959 (1978).

16.3.2. by Electrolysis Electrolysis of molten salts is widely applicable to the synthesis of transition-metal compounds', including those with tunnel structures. The electrochemical reaction leads to a more reduced or more oxidized state in the product than in the molten salt. Only electroreduction has so far been used to prepare compounds with tunnel structures. In such structures the tunnels contain ions, e.g., alkali-metal ions in a transition-metal oxide host lattice. Ionization of the inserted alkalimetal gives electrons to the oxide host, reducing the transition-metal cations. Several conditions of electroreduction are needed for deposition to occur on the cathode. The product must be insoluble when polarized, for it to be separated from the melt; the melting point of the product must be higher than that of the reactants; the product must be a good electrical conductor at the temperature of reaction; the metal belonging to the host structure must be easily reducible.

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc. 16.3. The Formation of Tunnel Structures 16.3.2. by Electrolysis

229

16.3.1-6. Structures with Parallel Tunnels. Cancrinite, [(Na, Ca,,,)AlSiO,],, (Na,, Ca)[C03] is one of the rate structures built up of parallel tunnels that exhibits ion-exchange properties of Na* and Ca" for all alkali cations up to Rb' ' - 3 . This is probably due the great dimension of its tunnels, which have free diameters of about 6 A. It is also the case with basic cancrinite, for which ion exchanges of Ag' for Na' and Li' have been observed with a hysteresis loop similar to that observed for structure transitions. It is, however, worthy of note that the exchange for Cs' is not possible for the latter structure despite the size of Cs' which is much smaller than that of a tunnel. It seems, therefore, that the ion-exchange reactions can only take place in a parallel tunnel structure, and only if the free diameter of the tunnel is considerably greater than that of the exchanged ions. This would allow the incoming and outgoing ions to move freely in opposite directions in the tunnels without mutual repulsion. A series of hexagonal silicates, Na,[MSi,O,,] isostructural with Na,[YSi,O,,], have been synthesized for M = Fe, In, Se, Y and the lanthanides Lu-Sm4. The host lattice of these compounds is built up of Si,,03, rings whose stacking forms columns that are connected through MO, octahedra to form tunnels parallel to the c axis. However, this structure does not contain purely parallel tunnels; some are interconnected, allowing Na' ions to move easily through the structure. The ion-exchange properties of these oxides are thus not surprising: Na' can be exchanged for Li', Ag', K + in molten nitrates. In conclusion, the ion-exchange properties of these materials, which reflect a certain mobility of the cations in the framework, can play an important part in the synthesis of solid electrolytes. (6 RAVEAU)

1. R. M. Barrer, in Non-stoichiometric Compounds, L. Mandelcorn, ed., Academic Press, New York, 1964, pp, 309-399; excellent review. 2. F. Helferiich, Zon Exchange, McGraw-Hill Book Company, New York, 1962; excellent review. 3. R. M. Barrer, J. D. Falloner, Pvoc. Roy. SOC.London, Ser. A , 236, 227 (1956). 4. R. D. Shannon, B. E. Taylor, T. E. Gier, H. Y. Chen, T. Berzins, Znorg. Chem., 17, 959 (1978).

16.3.2. by Electrolysis Electrolysis of molten salts is widely applicable to the synthesis of transition-metal compounds', including those with tunnel structures. The electrochemical reaction leads to a more reduced or more oxidized state in the product than in the molten salt. Only electroreduction has so far been used to prepare compounds with tunnel structures. In such structures the tunnels contain ions, e.g., alkali-metal ions in a transition-metal oxide host lattice. Ionization of the inserted alkalimetal gives electrons to the oxide host, reducing the transition-metal cations. Several conditions of electroreduction are needed for deposition to occur on the cathode. The product must be insoluble when polarized, for it to be separated from the melt; the melting point of the product must be higher than that of the reactants; the product must be a good electrical conductor at the temperature of reaction; the metal belonging to the host structure must be easily reducible.

230 16.3. The Formation of Tunnel Structures 16.3.2. by Electrolysis 16.3.2.2. Mechanism of Electrochemical Preparation of Tungsten Bronzes

There are two main advantages in the use of electrolytic reactions as preparative methods:

1. It is possible to synthesize compounds that are not obtainable by other techniques. 2. Electroreduction often permits the growth of especially large crystals. (J.P. DOUMERC)

1. A. Wold, D. Bellavance, in Preparatiue Methods in Solid State Chemistry, P. Hagenmuller, ed., Academic Press, New York, 1972.

16.3.2.1. Rate of Growth of Deposit. The weight deposited, dm, during time, dt, is given by: dm - MEI dt xF where x is the number of Faraday equivalents (F) necessary for the deposition of 1 mol of a compound of atomic weight M ; I is the current and c the current efficiency. For a cubic crystal of volume v, the current density and the rate of growth are given by:

.

I=--

I 6 v2I3

dv - 6&Miv2I3 dt xFd where d is the density. Integration of Eq. (c) for the initial condition v = vo,, vo being the volume of the seed crystal, gives: 2~Mit v1'3

=

(=)

+ vy3

Substitution of Eq. (b) in Eq. (d) gives:

(

)

4.90~Mi~I't + 2.45i'/2vA/3 ''I2 = xdF The linearity of the plots of 1''' against time, t, shows that i, E and x remain constant during the electrolysis'. The yield corresponds to a value of E between 0.85 and 1'. (J.P DOUMERC)

1. R. A. Fredlein, A. Damjanovic, J. Solid State Chem., 4, 94 (1972).

16.3.2.2. Mechanism of Electrochemical Preparation of Tungsten Bronzes 16.3.2.2.1. Nature of the Reduced Species.

The melt used for electrochemical synthesis of tungsten bronzes M,WO,, is a mixture of the tungstate of the metal M and the oxide WO,.

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc. 230 16.3. The Formation of Tunnel Structures 16.3.2. by Electrolysis 16.3.2.2. Mechanism of Electrochemical Preparation of Tungsten Bronzes

There are two main advantages in the use of electrolytic reactions as preparative methods:

1. It is possible to synthesize compounds that are not obtainable by other techniques. 2. Electroreduction often permits the growth of especially large crystals. (J.P. DOUMERC)

1. A. Wold, D. Bellavance, in Preparatiue Methods in Solid State Chemistry, P. Hagenmuller, ed., Academic Press, New York, 1972.

16.3.2.1. Rate of Growth of Deposit. The weight deposited, dm, during time, dt, is given by: dm - MEI dt xF where x is the number of Faraday equivalents (F) necessary for the deposition of 1 mol of a compound of atomic weight M ; I is the current and c the current efficiency. For a cubic crystal of volume v, the current density and the rate of growth are given by:

.

I=--

I 6 v2I3

dv - 6&Miv2I3 dt xFd where d is the density. Integration of Eq. (c) for the initial condition v = vo,, vo being the volume of the seed crystal, gives: 2~Mit v1'3

=

(=)

+ vy3

Substitution of Eq. (b) in Eq. (d) gives: ''I2

=

(

)

4.90~Mi~I't + 2.45i'/2vA/3 xdF

The linearity of the plots of 1''' against time, t, shows that i, E and x remain constant during the electrolysis'. The yield corresponds to a value of E between 0.85 and 1'. (J.P DOUMERC)

1. R. A. Fredlein, A. Damjanovic, J. Solid State Chem., 4, 94 (1972).

16.3.2.2. Mechanism of Electrochemical Preparation of Tungsten Bronzes 16.3.2.2.1. Nature of the Reduced Species.

The melt used for electrochemical synthesis of tungsten bronzes M,WO,, is a mixture of the tungstate of the metal M and the oxide WO,.

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc. 230 16.3. The Formation of Tunnel Structures 16.3.2. by Electrolysis 16.3.2.2. Mechanism of Electrochemical Preparation of Tungsten Bronzes

There are two main advantages in the use of electrolytic reactions as preparative methods:

1. It is possible to synthesize compounds that are not obtainable by other techniques. 2. Electroreduction often permits the growth of especially large crystals. (J.P. DOUMERC)

1. A. Wold, D. Bellavance, in Preparatiue Methods in Solid State Chemistry, P. Hagenmuller, ed., Academic Press, New York, 1972.

16.3.2.1. Rate of Growth of Deposit. The weight deposited, dm, during time, dt, is given by: dm - MEI dt xF where x is the number of Faraday equivalents (F) necessary for the deposition of 1 mol of a compound of atomic weight M ; I is the current and c the current efficiency. For a cubic crystal of volume v, the current density and the rate of growth are given by:

.

I=--

I 6 v2I3

dv - 6&Miv2I3 dt xFd where d is the density. Integration of Eq. (c) for the initial condition v = vo,, vo being the volume of the seed crystal, gives: 2~Mit v1'3

=

(=)

+ vy3

Substitution of Eq. (b) in Eq. (d) gives: ''I2

=

(

)

4.90~Mi~I't + 2.45i'/2vA/3 xdF

The linearity of the plots of 1''' against time, t, shows that i, E and x remain constant during the electrolysis'. The yield corresponds to a value of E between 0.85 and 1'. (J.P DOUMERC)

1. R. A. Fredlein, A. Damjanovic, J. Solid State Chem., 4, 94 (1972).

16.3.2.2. Mechanism of Electrochemical Preparation of Tungsten Bronzes 16.3.2.2.1. Nature of the Reduced Species.

The melt used for electrochemical synthesis of tungsten bronzes M,WO,, is a mixture of the tungstate of the metal M and the oxide WO,.

16.3.2. by Electrolysis 231 16.3.2.2. Mechanism of Electrochemical Preparation of Tungsten Bronzes 16.3.2.2.2. Reactions at the Electrodes.

TABLE 1. NATUREOF REDUCEDSPECIESAS FOR TUNGSTEN BRONZES(AFTERREF 1)

A

FUNCTION OF MELT COMPOSITION

~

Melt composition

WO, concentration (mol %)

Nature of reduced species

Depending upon the nature of the metal M and the concentration of WO, in the melt the reduced species given by one electron transfer could either be a monomer WO, or a polymer (WO,), with n < 18 ' (Table 1). ( J P DOUMERC)

1. E. Banks, C. W. Fleischmann, L. Meites, J . Solid State Chem., 1, 372 (1970).

16.3.2.2.2. Reactions at the Electrodes.

Using a melt of composition 50% Na2W04-50% WO,, corresponding to Na,W,O,, there is evolution of oxygen at the anode': [W,0,12-

-

2 WO,

+ 2 e- +

0,

(a)

The two possible cathodic reactions are':

-

and, following the results of Table 1 (§16.3.4.2.1),for the given melt composition:

(wod4 + e-

[w0314

(c)

From the measured cathodic oxygen pressures and decomposition potentials, metallic sodium cannot be formed'. Consequently, the primary product is Na[WO,], . However, for the melt composition considered here the bronze Na,WO, obtained corresponds to an x value of 0.5 (see Fig. 1,816.3.2.3.2).Although kinetic studies do not permit a complete interpretation2s3 subsequent loss of WO, from Na[WO,], is the probable mechanism. ( J P DOUMERC)

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc. 16.3.2. by Electrolysis 231 16.3.2.2. Mechanism of Electrochemical Preparation of Tungsten Bronzes 16.3.2.2.2. Reactions at the Electrodes.

TABLE 1. NATUREOF REDUCEDSPECIESAS FOR TUNGSTEN BRONZES(AFTERREF 1)

A

FUNCTION OF MELT COMPOSITION

~

Melt composition

WO, concentration (mol %)

Nature of reduced species

Depending upon the nature of the metal M and the concentration of WO, in the melt the reduced species given by one electron transfer could either be a monomer WO, or a polymer (WO,), with n < 18 ' (Table 1). ( J P DOUMERC)

1. E. Banks, C. W. Fleischmann, L. Meites, J . Solid State Chem., 1, 372 (1970).

16.3.2.2.2. Reactions at the Electrodes.

Using a melt of composition 50% Na2W04-50% WO,, corresponding to Na,W,O,, there is evolution of oxygen at the anode': [W,0,12-

-

2 WO,

+ 2 e- +

0,

(a)

The two possible cathodic reactions are':

-

and, following the results of Table 1 (§16.3.4.2.1),for the given melt composition:

(wod4 + e-

[w0314

(c)

From the measured cathodic oxygen pressures and decomposition potentials, metallic sodium cannot be formed'. Consequently, the primary product is Na[WO,], . However, for the melt composition considered here the bronze Na,WO, obtained corresponds to an x value of 0.5 (see Fig. 1,816.3.2.3.2).Although kinetic studies do not permit a complete interpretation2s3 subsequent loss of WO, from Na[WO,], is the probable mechanism. ( J P DOUMERC)

232

16.3. The Formation of Tunnel Structures 16.3.2. by Electrolysis 16.3.2.3. Experimental Procedure and Data

1. E. Banks, C. W. Fleischmann, L. Meites, J. Solid State Chem., 1, 372 (1970). 2. R. A. Fredlein, A. Damjanovic, J. Solid Stare Chem., 4, 94 (1972). 3. J. P. Randin, Electrochzm. Acta, 19, 745 (1974).

16.3.2.3. Experimental Procedure and Data 16.3.2.3.1, the Electrolytic Cell.

The electrolysis is performed either in a platinum or an alumina cell, this being placed in a temperature-controlled furnace. Electrodes are usually either gold or platinum beads or plates soldered to the lead wires. A carbon anode can also be used'. The electrolysis can be carried out either at constant current or at constant potential, the latter being theoretically more reliable for the control of the composition of the products2. A double cell may also be used, consisting of two concentric crucibles, the inner one having holes at the b ~ t t o r n ~ Such - ~ . a setup maintains a separation between anodic and cathodic regions and thus prevents the oxygen effervescing at the anode from affecting the growth of the crystal. Sodium tungsten bronze6 crystals have been grown by attaching a crystal seed to the bottom of a rotating cathode, which can be lowered in order to keep the entire growing crystal at least 1 mm below the melt surface. Use of a variable-speed rotating crucible reduces the inhomogeneity in crystals of tantalum-substituted tungsten bronzes'. (J.P DOUMERC)

1. H. R. Shanks, J. Crystal Growth, 13/14, 433 (1972). 2. M. S. Whittingham, R. A. Huggins, Proceedings of 5th Materials Research Symposium,Natl. Bur. Stand. (US.), Spec. Publ. No 364, 51 (3972). 3. W. Kunnmann, A. Ferretti, Rev Sci. Instrum., 35,465 (1964). 4. D. B. Rogers, A. Ferrett, W. Kunnmann, J. Phys. Chem. Solids, 27, 1445 (1966). 5. A. F. Reids, J. A. Watts, J. Solid State Chem., 1, 310 (1970). 6. P. F. Weller, D. M. Granditz, J. Crystal Growth, 12, 63 (1972). 7. J. Marcus, Doctoral Thesis, Chemistry and Physics of Materials, Bordeaux, 1978.

16.3.2.3.2. Effect of the Melt Composition and Temperature.

The dependence of the insertion ratio, x, in tungsten bronzes, Na,WO,, upon the melt composition when the temperature of the charge is held just above its melting point is shown in. Fig. 1. This result is in relatively good agreement with the empirical relationship: x = a(M/W), where M/W = (atoms of inserted metal M)/(atoms of tungsten). This behavior is also observed for copper tungsten bronzes, Cu,WO, Substituted sodium tungsten bronzes where tungsten is partially replaced by another transition element are also prepared by adding an oxide of the corresponding cation to the For tantalum, the ratio Ta/W in the bronze crystals as a function of melt composition is given in Fig. 2.

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc. 232

16.3. The Formation of Tunnel Structures 16.3.2. by Electrolysis 16.3.2.3. Experimental Procedure and Data

1. E. Banks, C. W. Fleischmann, L. Meites, J. Solid State Chem., 1, 372 (1970). 2. R. A. Fredlein, A. Damjanovic, J. Solid Stare Chem., 4, 94 (1972). 3. J. P. Randin, Electrochzm. Acta, 19, 745 (1974).

16.3.2.3. Experimental Procedure and Data 16.3.2.3.1, the Electrolytic Cell.

The electrolysis is performed either in a platinum or an alumina cell, this being placed in a temperature-controlled furnace. Electrodes are usually either gold or platinum beads or plates soldered to the lead wires. A carbon anode can also be used'. The electrolysis can be carried out either at constant current or at constant potential, the latter being theoretically more reliable for the control of the composition of the products2. A double cell may also be used, consisting of two concentric crucibles, the inner one having holes at the b ~ t t o r n ~ Such - ~ . a setup maintains a separation between anodic and cathodic regions and thus prevents the oxygen effervescing at the anode from affecting the growth of the crystal. Sodium tungsten bronze6 crystals have been grown by attaching a crystal seed to the bottom of a rotating cathode, which can be lowered in order to keep the entire growing crystal at least 1 mm below the melt surface. Use of a variable-speed rotating crucible reduces the inhomogeneity in crystals of tantalum-substituted tungsten bronzes'. (J.P DOUMERC)

1. H. R. Shanks, J. Crystal Growth, 13/14, 433 (1972). 2. M. S. Whittingham, R. A. Huggins, Proceedings of 5th Materials Research Symposium,Natl. Bur. Stand. (US.), Spec. Publ. No 364, 51 (3972). 3. W. Kunnmann, A. Ferretti, Rev Sci. Instrum., 35,465 (1964). 4. D. B. Rogers, A. Ferrett, W. Kunnmann, J. Phys. Chem. Solids, 27, 1445 (1966). 5. A. F. Reids, J. A. Watts, J. Solid State Chem., 1, 310 (1970). 6. P. F. Weller, D. M. Granditz, J. Crystal Growth, 12, 63 (1972). 7. J. Marcus, Doctoral Thesis, Chemistry and Physics of Materials, Bordeaux, 1978.

16.3.2.3.2. Effect of the Melt Composition and Temperature.

The dependence of the insertion ratio, x, in tungsten bronzes, Na,WO,, upon the melt composition when the temperature of the charge is held just above its melting point is shown in. Fig. 1. This result is in relatively good agreement with the empirical relationship: x = a(M/W), where M/W = (atoms of inserted metal M)/(atoms of tungsten). This behavior is also observed for copper tungsten bronzes, Cu,WO, Substituted sodium tungsten bronzes where tungsten is partially replaced by another transition element are also prepared by adding an oxide of the corresponding cation to the For tantalum, the ratio Ta/W in the bronze crystals as a function of melt composition is given in Fig. 2.

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc. 232

16.3. The Formation of Tunnel Structures 16.3.2. by Electrolysis 16.3.2.3. Experimental Procedure and Data

1. E. Banks, C. W. Fleischmann, L. Meites, J. Solid State Chem., 1, 372 (1970). 2. R. A. Fredlein, A. Damjanovic, J. Solid Stare Chem., 4, 94 (1972). 3. J. P. Randin, Electrochzm. Acta, 19, 745 (1974).

16.3.2.3. Experimental Procedure and Data 16.3.2.3.1, the Electrolytic Cell.

The electrolysis is performed either in a platinum or an alumina cell, this being placed in a temperature-controlled furnace. Electrodes are usually either gold or platinum beads or plates soldered to the lead wires. A carbon anode can also be used'. The electrolysis can be carried out either at constant current or at constant potential, the latter being theoretically more reliable for the control of the composition of the products2. A double cell may also be used, consisting of two concentric crucibles, the inner one having holes at the b ~ t t o r n ~ Such - ~ . a setup maintains a separation between anodic and cathodic regions and thus prevents the oxygen effervescing at the anode from affecting the growth of the crystal. Sodium tungsten bronze6 crystals have been grown by attaching a crystal seed to the bottom of a rotating cathode, which can be lowered in order to keep the entire growing crystal at least 1 mm below the melt surface. Use of a variable-speed rotating crucible reduces the inhomogeneity in crystals of tantalum-substituted tungsten bronzes'. (J.P DOUMERC)

1. H. R. Shanks, J. Crystal Growth, 13/14, 433 (1972). 2. M. S. Whittingham, R. A. Huggins, Proceedings of 5th Materials Research Symposium,Natl. Bur. Stand. (US.), Spec. Publ. No 364, 51 (3972). 3. W. Kunnmann, A. Ferretti, Rev Sci. Instrum., 35,465 (1964). 4. D. B. Rogers, A. Ferrett, W. Kunnmann, J. Phys. Chem. Solids, 27, 1445 (1966). 5. A. F. Reids, J. A. Watts, J. Solid State Chem., 1, 310 (1970). 6. P. F. Weller, D. M. Granditz, J. Crystal Growth, 12, 63 (1972). 7. J. Marcus, Doctoral Thesis, Chemistry and Physics of Materials, Bordeaux, 1978.

16.3.2.3.2. Effect of the Melt Composition and Temperature.

The dependence of the insertion ratio, x, in tungsten bronzes, Na,WO,, upon the melt composition when the temperature of the charge is held just above its melting point is shown in. Fig. 1. This result is in relatively good agreement with the empirical relationship: x = a(M/W), where M/W = (atoms of inserted metal M)/(atoms of tungsten). This behavior is also observed for copper tungsten bronzes, Cu,WO, Substituted sodium tungsten bronzes where tungsten is partially replaced by another transition element are also prepared by adding an oxide of the corresponding cation to the For tantalum, the ratio Ta/W in the bronze crystals as a function of melt composition is given in Fig. 2.

16.3.2. by Electrolysis 16.3.2.3. Experimental Procedure and Data 16.3.2.3.2. Effect of the Melt Composition and Temperature.

233

1.o

-

A

0.8-

* 0.6-

.-

0

4

-

rn

L

.-

4 L

f 0.4-

o A f t e r [51

C31

o After A After

121

H c

0.2

-

I

0

I

1

20

1

I

40

1

1

60

Figure 1. Variation of composition of sodium tungsten bronzes with melt composition when the temperature of the charge is held just above its melting point.

crystal

0.300o.200; 0.100

/"-

~

Figure 2. Ratio Ta/W in tantalum-substituted sodium tungsten bronzes as a function of Ta/W ratio in the melt at 925°C.

234

16.3. The Formation of Tunnel Structures 16.3.2. by Electrolysis 16 3.2.3. Experimental Procedure and Data

1000 SODIUM TUNGSTEN BRONZES

900

-

V e

2

BOC

n

TETRAGONAL

/

-

/

K W

k

/

w

/

I-

/

’

--Y

TUNGSTEN METAL

W K

ORTHORHOMBIC

/

/

/

/

I

/

/

/

/

CUBIC BRONZE

/

70C

60C

Na2W04

10

20

30 40 50 MOLE % WO3-

60

7C

Figure 3. Different phases of sodium tungsten bronzes obtained as a function of melt composition and temperature (after 5).

The change in the composition and structure of crystals with the temperature of the melt is shown in Fig. 3 for sodium tungsten bronzes. (J P. DOUMERC)

E. Banks, C. W. Fleischmann, L. Meites, J. Solid State Chern., I , 372 (1970). R. A. Fredlein, A. Damjanovic, J. Solid State Chem., 4, 94 (1972). J. Marcus, Thise de Doctorat en Chimie et Physique des Materiaux, Bordeaux, 1978. P. F. Weller, B. E. Taylor, R. L. Mohler, Mater. Res. Bull., 5, 465 (1970). 5 . H. R. Shanks, J. Crystal Growth, 13/14, 433 (1972).

1. 2. 3. 4.

16.3.2.3.3. Examples of Oxides with Tunnel Structures Obtained by Molten Salt Electrolysis.

In Table 1 are gathered data concerning the preparation of selected oxides having

tunnel structures.

(J.P. DOUMERC)

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc.

234

16.3. The Formation of Tunnel Structures 16.3.2. by Electrolysis 16 3.2.3. Experimental Procedure and Data

1000 SODIUM TUNGSTEN BRONZES

900

-

n

TETRAGONAL

TUNGSTEN METAL

W K

/

2

BOC

-

/

K W

k w

/

/

I-

/

’

--Y

-

V

e

ORTHORHOMBIC

/

/

/

/

I

/

/

/

/

CUBIC BRONZE

/

70C

60C

Na2W04

10

20

30 40 50 MOLE % WO3-

60

7C

Figure 3. Different phases of sodium tungsten bronzes obtained as a function of melt composition and temperature (after 5).

The change in the composition and structure of crystals with the temperature of the melt is shown in Fig. 3 for sodium tungsten bronzes. (J P. DOUMERC)

E. Banks, C. W. Fleischmann, L. Meites, J. Solid State Chern., I , 372 (1970). R. A. Fredlein, A. Damjanovic, J. Solid State Chem., 4, 94 (1972). J. Marcus, Thise de Doctorat en Chimie et Physique des Materiaux, Bordeaux, 1978. P. F. Weller, B. E. Taylor, R. L. Mohler, Mater. Res. Bull., 5, 465 (1970). 5 . H. R. Shanks, J. Crystal Growth, 13/14, 433 (1972).

1. 2. 3. 4.

16.3.2.3.3. Examples of Oxides with Tunnel Structures Obtained by Molten Salt Electrolysis.

In Table 1 are gathered data concerning the preparation of selected oxides having

tunnel structures.

(J.P. DOUMERC)

< 0.9)

< 0.33)g

KxW03 (0.20 < x

Blue prisms

Black

Blue black prisms, 8 X l X 1 . Black small needles Length < 0.5. Dark lustrous prisms. 4 x 1 x 1 Yellow, orange, red”. Cubes up to 70 x 70 x 70. Blue, purple, red”

Color and crystal size (mm)

-

..Color as a function of the x valuela. Tetragonal potassium tungstcn bronze Hexagonal potassium tungsten brome. See Fie. 3. ’Sce Fig. 1. ‘The currcnt varies linearly with the potential6. Analogous tungsten bronrzs exist with Rb, Cs, TI’.

< 0.1)

(0.05 < x

(0.25 < x < 0.5)

Cs,Ti,O, ( x < 1) Na,WO, (0.5 < x

K3Ti8017

K,Ti,O, (x < 1)

Product composition

TKWBb

HKWB“

Distorted Pcrovskite

TKWBb

Perovskite

Hollandite

K3Ti8017

Hollanditc

Structurc 5(

100)

K,W0,(35-50) W0,(65 SO) K,W0,(50-60) WO, (50-40)

Na,W0,(35-50) W0,(65-50)’ Na,W0,(35-30) W0,(65-70)”

K,T1,0,(98.5) Nb,05 (1.5) Cs,Ti,0,(82) TiO,(lS) Na,W0,(50 95) W03(50-5)e

K,Ti,O

Melt compositions (mo) %I

f

d

f

f

750

f

d

750

10 100

8-900

d

10 f100

10-100

7 800 7-800

100

925

1.1

> 0.8

>0.8

>1

>1

0.8-0.9%

0.8-0.9X

3, 6, 8

3, 6, 8

6

3,4,6,8, 12

3-13

1

2

60

1010

16 h

1

Reh.

60

Time (h) or yield (%)

1020

cv)

V

I (mA)

T (“C)

TARLE 1. SELECTED 0 x 1 1 ) ~WITH s TUNNELSTRUCTIJRES OBTAINED BY MOLTENSALTELECTROLYSIS

236

16.3 The Formation of Tunnel Structures 16.3.2. by Electrolysis 16.3.2.3. Experimental Procedure and Data

1. A. R. Reids, J. A. Watts, J. Solid State Chem., 1, 310 (1970). 2. J. A. Watts, J. Solid State Chem., 1, 319 (1970). 3. A. Wold, D. Bellavance, in Preparative Methods in Solid State Chemistry, P. Hagenmulier, ed., Academic Press, New York, 1972. 4. R. A. Fredlein, A. Damjanovic, J. Solid State Chem., 4, 94 (1972). 5 . E. Banks, C. W. Fleischmann, L. Meites, J. Solid State Chem., 1, 372 (1970). 6. M. S. Whittingham, R. A. Huggins, Proceedings o f j t h Materials Research Symposium, Natl. Spec. Publ. No 364, 51 (1972). Bur. Stand. (U.S.), 7. J. P. Randin, Electrochim. Acta, 19, 745 (1974). 8. H. R. Shanks, J. Crystal Growth, 13/14, 433 (1972). 9. P. F. Weller, D. M. Grandits, J. Crystal Growth, 12, 63 (1972). 10. J. Marcus, These de Doctorat en Chimie et Physique des Materiaux, Bordeaux, 1978. 11. P. F. Weller, B. E. Taylor, R. L. Molher, Mat. Res. Bull., 5, 465 (1970). 12. E. Banks, A. Wold, Prep. Znorg. React., 4, 237 (1968). 13. C . Scheibler, J. Pratkt. Chem., 83, 821 (1861). 14. P. Hagenmuller, Prog. Solid Stare Chem., 5, 71 (1971).

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc.

16.4. The Formation of Sheet Structures 16.4.1. Introduction. Since 1975, graphite intercalation compounds (GIC) have attracted a great deal of attention because of reports of high electronic conductivity in certain graphite acceptor compo~nds'-~,although the magnitude of this conductivity has been disputed. Much research on graphite acceptors, excluding graphite-alkali metals, has occurred since 1980; only a small fraction of this is discussed here. The reader is directed to pertinent review articles for more complete coverage6-'. Although graphite and boron nitride are discussed separately, there are interesting new C/BN composites, dispersed on both atomic" and longer1L"2 length scales. "Buckminsterfullerene", a C, carbon cluster that can "intercalate" potassium and lanthanum, has been proposed13, although its existence is still debated14-16. Intercalation has been proposed for graphite-like carbonaceous materialsL7.Scanning tunneling microscopy18 and topographic imaging'y~*O are being applied to intercalation science.

(L 8.EBERT)

1. F. L. Vogel, J. Mater. Sci., 12, 982 (1977). 2. J. E. Fischer, T. E. Thompson, Phys. Today, 31, 36 (1978). 3. F. L. Vogel, R. Wachnik, L. A. Pendrys, in Physics of Intercalation Compounds, Springer Ser. Solid State Sci., 38, 288 (1981). 4. E. McRae, J. F. Mareche, M. Lelaurain, G. Furdin, A. Herold, J . Phys. Chem. Solids, 48, 957 (1987). 5. H. Kamimura, Phys. Today, 40, 64 (1987). 6. L. B. Ebert, J. Mol. Catal., 15, 275 (1982). 7. M. S. Dresselhaus, Phys. Today, 37, 60 (1984). 8. R. Setton, Synth. Met., 23, 519 (1988). 9. M. S. Dresselhaus, Muter. Sci. Eng., BI, 259 (1988). 10. J. Kouvetakis, R. B. Kaner, M. L. Sattler, N. Bartlett, J. Chem. Soc., Chem. Commun., 1758 (1986). 11. A. W. Moore, S. L. Strong, Graphite Intercalation Compounds: Science and Applications, M. Endo, M. S. Dresselhaus, G. Dresselhaus, eds., Materials Research Society, Pittsburgh, 1988, pp. 141-144. 12. A. W. Moore, S. L. Strong, J. Appl. Phys., in press. 13. Q. L. Zhang, S. C. O'Brien, J. R. Heath, Y. Liu, R. F. Curl, H. W. Kroto, R. E. Smalley, J. Phys. Chem., 90, 525 (1986). 14. D. M. Cox, D. J. Trevor, K. C. Reichmann, A. Kaldor, J. Am. Chem. Soc., 108,2457 (1986). 15. M. Frenklach, L. B. Ebert, J. Phys. Chem., 92, 561 (1988). 16. L. B. Ebert, J. C. Scanlon, C. A. Clausen, Energy Fuels, 2, 438 (1988). 17. L. B. Ebert, J. C. Scanlon, in Polynuclear Aromatic Compounds, L. B. Ebert, ed., American Chemical Society, Washington, D C 1988, pp. 367-382. 18. S. Gauthier, S. Rousset, J. Klein, W. Sacks, M. Belin, J. Vacuum Sci. Technol., A , 6, 360 (1988). 19. G. C. Chingas, J. Milliken, H. A. Resing, T. Tsang, Synth. Met., 12, 131 (1985). 20. R. Nishitani, Y Sasaki, Y. Nishina, J. Phys. Soc. Jpn., 56, 1051 (1987).

237

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc. 238

16.4 The Formation of Sheet Structures 16.4.2. Graphite and Boron Nitride Intercalation Compounds 16.4.2.1. Electron Acceptors

16.4.2. Graphite and Boron Nitride Intercalation Compounds 16.4.2.1, Electron Acceptors 16.4.2.1.1. Halides.

The interaction of graphite with electron acceptors to form intercalation compounds is analogous chemically to the reaction of polycyclic aromatic hydrocarbons with oxidants to form radical cations or dications, as is schematically depicted in Fig. 1. Thus, the requirements for the synthesis of graphite-acceptor compounds, whether with halides, oxyhalides, oxides, acids or halogens are the presence of an oxidant and the presence of a suitable counteranion. There are many synthetically distinct routes to achieve these ends. This section presents general approaches to the formation of graphite/ halide compounds, approaches that are also useful to synthesise compounds of graphite with oxyhalides, oxides, acids, or halogens. For listings of the numerous known graphite/ halide compounds, several recent review articles are available'-,. The simplest synthesis is by direct combination of graphite with halide in a sealed tube, with control of pressure and temperature. In this manner, such oxidizing halides as FeCl,, AsF, and UF, react with graphite to form intercalation compounds. For nonoxidizing halides such as AlCl,, TiCl,, PF, and WF, intercalation is achieved by only adding a suitable oxidant, such as CI,, CIF, or Br,; in fact many producers now add some halogen routinely in all graphite-halide reactions. In these cases of direct combination, the counteranion arises from the initially neutral halide in reactions as the following:

AROMATIC RADICAL CATION

GRAPHITE/ACCEPTOR COMPOUND

I"

Figure 1. In the manner that polycyclic aromatic hydrocarbons react with oxidants to form aromatic radical cations (top), graphite reacts with oxidants to form graphite-acceptor compounds (bottom).

16.4.2. Graphite and Boron Nitride Intercalation Compounds 16.4.2.1. Electron Acceptors 16.4.2.1.1. Halides.

for O < x

< 4;

10.83 C

+ FeCl, 28 C

-300"C, Clz

C ~ ~ ~ ~ ( F e , , , 7 C l , ) o+. '0.03 - FeCl,

+ PF, + C1F

25°C

+ 4.14 C

239

(b)

+ 0.5 C1,

C:,[PF,]-

Unfortunately, the direct combination method does not allow monitoring of halide activity during the reaction and necessitates disposing of xs unreacted halide after the reaction. To deal with these difficulties, the two-bulb, two-temperature method is used'. In one bulb, the graphite is maintained at a higher temperature than the halide in order to avoid condensation of unreacted, excess halide. Keeping either the graphite or the halide at constant temperature while varying the temperature differential allows the determination of intercalation or deintercalation isotherms or isobars, giving information on stage formation and hysteresis effects6.The two-bulb method is especially useful for formation of more dilute, higher stage, compounds7:

-

13.5 C (350°C) + FeC1, (300°C)

22.8 C (415°C) + FeC1, (300°C)

C,, ,FeCl,

C,,,,FeCl,

(stage 11, I,

=

1273 pm) (d)

(stage 111, I, = 1608 pm) (e)

where I, is the distance along c between halide layers, so co = nI, with n an integer. Instead of relying on the volatility of the halide, as in the direct combination and two-bulb approaches, the solubility of halides may be exploited to intercalate from solutions of nonaqueous solvents'. Although this technique allows reaction at relatively low temperatures, for which in situ monitoring of reaction progress is easy, the solvent may participate in the reaction through both co-intercalation and direct chemical reaction. Another way to intercalate nonvolatile halides is through complexation in the vapor phase. By forming a volatile adduct, a nominally nonvolatile halide can be transported to graphite a molecule at a time, thereby allowing intercalation. The first example of this technique was with the CoC1,-AlC1, system', and it has also been used for the trichlorides of rare earth elements'. The four above-mentioned methods all utilize chemical oxidants to remove charge from the graphite. Whether the oxidant is the halide itself, an added halogen or the solvent, a reduced chemical is produced. This possible complication can be eliminated by use of anodic oxidation, in which an external chemical circuit serves as an oxidant'. 24 C

+ [Bu,N]AsF,

SOzClF

C:,(AsF,)

+ [Bu,N] + Bu.

(f)

Once formed, the butyl radical can disproportionate to butene or dimerize to octane. This approach is discussed below in $1 6.4.2.6. The common theme in all five synthetic approaches is the reaction of graphite, in the presence of an oxidant, with small molecular or ionic fragments. These basic building blocks allow easy formation of a two-dimensional system within the interlamellar voids of the sheet like graphite structure. As indicated before, both a suitable oxidant and a suitable counteranion must be present before intercalation can occur. Thus, while neither C1F nor C1, reacts directly with graphite at RT, they may be used as oxidants to cause intercalation of acids electron-pair acceptor such as BF, or AlCl, '. However, these oxidants cannot facilitate the intercalation of a neutral molecule, which cannot accept an X - ( = F-, Cl-), as for example CCI,.

16.4. The Formation of Sheet Structures 16.4 2. Graphite and Boron Nitride Intercalation Compounds 16.4.2 1. Electron Acceptors

240

A conceptually distinct route to some intercalation compounds is through chemical reaction of preformed intercalation compounds. Thus, the comproportionation reaction, Eq. (g), is used to synthesize first-stage graphite-FeC1, from first-stage graphite-FeCI,, as shown in Eq. (h)". 2 FeC1, C,,,FeCl,

+ 0.5 Fe(CO),

+ Fe(0)

-

3 FeCl,

150atmCO,150"C

(g)

' C7.1(FeC12)1,5 + 2.5 co

(h)

In general, dilute halide compounds, as formed in Eq. (c), contain primarily the anionic form of the halide, but the more concentrated compounds such as C,AsF, or C,SbF, contain substantial quantities of the initial, neutral halide. Caution: Some care in handling these compounds is needed, because they hydrolyze in moist air". More specifically, the compound C,,AsF, reacts with N,F, in the same way as AsF, 1 2 : C,,AsF,,,

+ N,F,

-

C,,AsF,+,

+ trans-N,F,

(0

Similarly, C,SbF, isomerizes hydrocarbons in the same manner as pure SbF, 13. There have been few systematic studies of halide intercalation as a function of halogen. Intercalated bromides seem only slightly thicker in graphite than intercalated chlorides (stage-I1 C,,FeBr, has an I, of 1290 pm 14; stage-I1 C,, ,TlCI, has an I, of TABLE1. FLUORIDES THATINTERCALATE GRAPHITE~ Fluoride

Product composition

Stage

TiF, AsF, SbF, NbF, TaF, MoF, WF6 UF6

OsF, XeOF, XeF, XeF, XeF, KrF, PF, -F, i0,i CA~F,I -c GeF,-F, "021 '[BF4I in CH,NO, BF, -ClF" +

a

1, (pm) 1510 810 846 1176 1176 842 1170 1190 (2) 806 1130-1160 (2)

Refs.

-

16 17 18 18 18 19 19 20 21 22 23 24 25 26 27 21 28 29

720

4, 30

-

1140 -

807 800

For a given fluoride, only the most concentrated, lowest stage compound is entered. Although properly an oxyfluoride rather than a fluoride, XeOF, is included for reference to the other inert gas compounds

' This technique can be used to synthesize other fluoride compounds

16.4.2. Graphite and Boron Nitride Intercalation Compounds 16.4.2.1. Electron Acceptors 16.4.2.1.1. Halides.

241

GUEST-HOST AND GUEST-GUEST INTERACTIONS

Figure 2.

1312 pm, while stage-I1 C,,,,TlBr, has an I, of 1340 pm '). Intercalated iodides have not been reported. Many of the chlorides and bromides that intercalate graphite are solids that do not possess strong, oriented bonds in all three dimensions. Thus, on intercalation, they attempt to re-form their own lower dimensional lattices within graphite. Since central atom vacancies, rather than chemically reduced central atoms [see Eq. (b)], are considered to be the counteranions to the positive charge on graphite, a multivalent central atom is not a prerequisite for s y n t h e ~ i s ' ~ For ~ ' ~ .inserted species not showing oneor two-dimensional lattices, the concept of vacancies is less clear, and simple redox chemistry interpretations are preferred. Many of the fluorides (see Table 1) and some of the chlorides (as SbCl,), however, are molecular or oligomeric species at RT and, in the free state solidify to make molecular crystals. Thus, although similarities in the chargetransfer chemistry are expected among all halides, the molecular dynamics of fluorides within graphite3' are expected to be different from the rigid lattice behavior of most chlorides and bromides. As such, the fluoride intercalation compounds have much in common with the compounds with oxyhalides and inorganic acids discussed in the following sections. The concept of complex anions such as [As,F,,]- or [B,F,]- in graphite is of relevance to halide intercalation, as schematically depicted in Fig. 2. The large size of the graphite macrocation can be expected to stabilize such complex anion^^'-^^. In the concentrated AsF, and SbF, intercalation compounds, the presence of only a single l9F NMR signal at RT and the absence of 75Asor '"Sb signals argue for an interaction between MF, and [MF,]- species36;such an inference is also consistent with infrared reflectivity studies3'. Other work is available on these system^^'-^^. Biintercalation, an interleaving of different inserted species, has been studied in the case of inserted halide specie^^^-^,. Halogen-containing solvents, with metal chlorides, can also lead to ternary phases, such as graphite-FeC13-C,H4C12 47. A different approach to inserted halides is the intercalation compound with trimethyltin chloride4'. Finally, the intercalation of halides into carbon fibers is an area of increasing intere~t~',~~. (L.B. EBERT)

242

16.4. The Formation of Sheet Structures 16.4.2. Graphite and Boron Nitride Intercalation Compounds 16.4.2.1. Electron Acceptors

1. L. B. Ebert, Annu. Rev. Muter. Sci., 6, 181 (1976). 2. A. Herold, Muter. Sci. W g . , 31, l(1977). 3. A. Herold, in Physics and Chemistry ofMaterials with LayeredStructures, Vol. 6, F. Levy, ed., D. Reidel, Dordrecht, 1979, pp. 323-421. 4. H. Selig, L. B. Ebert, Adz. Inorg. Chem. Radiochem., 23, 281 (1980). Recommended for its treatment of fluorides. 5. A. Herold, Bull. Soc. CJzim. Fr., 999 (1955). 6. J. G. Hooley, Carbon, 18, 83 (1980). C. E. Pettinos Award Lecture. 7. D. Hohlwein, F. D. Grigutch, A. Knappwost, Angew. Chem., 81, 333 (1969); Angew. Chem., Int. Ed. Engl., 8, 382 (1969). 8. E. Stumpp, Muter. Sci.Eng., 31, 53 (1977). Thorough review of halides. 9. J. 0. Besenhard, H. Mohlwald, J. J. Nickl, Synth. Met., 3, 187 (1981). 10. B. Pritzlaff, H. Strahl, Carbon, 15, 399 (1977). 11. L. V. Interrante, R. S. Markiewicz, D. W. McKee, Synth. Met., I , 287 (1980). 12. V. Munch, H. Selig, Synth. Met., 1, 407 (1980). 13. K. Laali, M. Muller, J Sommer, J. Chem. Soc., Chem. Commun.,p. 1088 (1980). 14. H. Stahl, Z. Anorg. Allg. Chem., 428, 269 (1977). 15. G. K. Wertheim, P. M. Th. M. van Attekum, H. S. Guggenheim, K. E. Clements, Solid State. Commun.;33, 809 (1980). 16. E. Buscarlet, P. Touzain, L. Bonnelain, Carbon, 14, 75 (1976). 17. E. R. Falardeau, L. R. Hanlon, T. E. Thompson, horg. Chem., 17, 301 (1978). 18. J. Melin, A. Herold, Carbon, 13, 357 (1975). 19. A. Hamwi, P. Touzain, L. Bonnetain, Muter. Sci. Eng., 31, 95 (1977). 20. L. B. Ebert, J. P. DeLuca, A. H. Thompson, J. C. Scanlon, Muter. Res. Bull., 12, 1135 (1977). 21. N. Bartlett, R. N. Biagoni, B. W. McQuillan, A. S. Robertson, A. C. Thompson, J. Chem. Soc., Chem. Commun., p. 200 (1978). 22. H. Selig, 0. Gani, Inorg. Nucl. Chem. Lett., 11, 75 (1975). 23. H. Selig, M. Rabinovitz, I. Agranat, C.-H. Lin, L. B. Ebert, J. Am. Chem. Soc., 98, 1601 (1976). 24. H. Selig, M. Rabinovitz, I. Agranat, C.-H. Lin, L. B. Ebert, J. Am. Chem. Soc., 99, 953 (1977). 25. M. Rabinovitz, I. Agranat, H. Selig, C.-H. Lin, L. B. Ebert, J. Chem. Res. (3,216; (M), 2353 (1977). 26. H. Selig, P. K. Gallagher, L. B. Ebert, Znorg. Nucl. Chem. Lett., 13, 427 (1977). 27. A. D. Cohen, US.Pat. 4,128,499 (Dec. 5, 1978); Chem. Abstr., YO, 9 3 , 1 5 2 ~(1979). 28. E. M. McCarron, Y. J Grannec, N. B. Bartlett, J. Chem. Soc., Chem. Commun., p. 890 (1980). 29. D. Billaud, A. Pron, F. L. Vogel, A. Herold, Muter. Res. Bull., 15, 1627 (1980). 30. L. B. Ebert, H. Selig, Synth. Met., 3, 53 (1981). 31. L. B. Ebert, D. R. Mills, J. C. Scanlon, Muter. Res. Bull., 14, 1369 (1979). 32. L. B. Ebert, D. R. Mills, J. C. Scanlon, H. Selig, Muter. Res. Bull., 16, 831 (1981). 33. L. B. Ebert, J. Mol. Catal., 15, 275 (1982). 34. L. B. Ebert, D. R. Mills, J. C. Scanlon, Muter. Res. Bull., 18, 1505 (1983). 35. I. Stang, G. Roth, K. Lueders, H. J. Guentherodt, Synth. Met., 12, 85 (1985). 36. L. B. Ebert, A. R. Garcia, H. Selig, Rev. Chim. Mineral, 23, 543 (1986). 37. F. Jost, Y. Yacoby, D Heitmann, S. Roth, Phys. Rev. Ser., B, 39, 5444 (1989). 38. D. Vaknin, D. Davidov, H. Selig, J. Fluorine Chem., 32, 345 (1986). 39. B. Sundqvist, B. Lundberg, J. Phys. C, Solid State Phys., 19, 6915 (1986). 40. I. Ohana, Y. Yacoby, NATO ASZ Series B, 148, 345 (1986). 41. W. Wang, A. Philipp, K. Seeger, J. Muter. Sci., 22, 223 (1987). 42. M Lelaurain, J. F.Mareche, E. McRae, G. Furdin, A. Herold, J. Muter. Res., 3, 87 (1988) 43. A. Herold, G . Furdin, D. Guerard, L. Hachim, M. Lelaurain, N. E. Nadi, R. Vangelisti, Synth. Met., 12, 11 (1985). 44. G. Furdin, L. Hachim, N. E. Nadi, M. Lelaurain, R. Vangelisti, A. Herold, C. R. Hebd. Seances Acad Sci., Ser. 2, 301, 87 (1985). 45. N. Nadi, E. McRae, J. F. Mareche, M. Lelaurain, A. Herold, Carbon, 24, 695 (1986). 46. P. Scharff, E. Stumpp, C. Ehrhardt, Synth. Met., 23, 415 (1988). 47. M. Inagaki, T. Mitsuhashi, Y. Soneda, J. Chini Phys., Phys.-Chim. Biol., 84, 1439 (1987); Chem. Abst., 109, 15,4592 (1988). 48. R. Schogl, H. P. Boehm, Z . Nuturforsch., Tell B, 39 788 (1984). 49. C. Manini, E. McRae, J. F. Mareche, A. Herold, Rev. Chim. Mineral, 22, 161 (1985). 50. S. Luski, I. Ohana, H. Selig, Carbon, 25, 799 (1987).

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc. 16 4.2. Graphite and Boron Nitride Intercalation Compounds 16.4.2.1. Electron Acceptors 16.4.2.1 2 Oxyhalides.

243

_ _ _ _ _ _ _ _ ~ ~ ~ ~

16.4.2.1.2. Oxyhalides.

In 1956, CrO,F,, CrO,Cl, and UO,Cl, were reported to intercalate graphite. Intercalated CrO,Cl, shows the behavior, so far unique, of becoming more resistant to removal the longer time spent in the graphite lattice'. Graphite reacts with XeOF, by direct combination at RT to yield C, ,XeOF, '. Unlike XeF, , which is reduced to Xe(1V) on intercalation, the intercalated species here is primarily XeOF, The reaction of IOF, with graphite at 75°C is accompanied by oxygenation of the graphite skeleton to yield C, ,O, JF, '. Similarly, graphite reacts with ReOF, to form C,, ,20,(ReF,)(ReF,)o '. Side reactions involving formation of covalent bonds to graphite have been noted for products of graphite with CrO,,, KrF,' and IF,'. Intercalation compounds of graphite with MoOC1, l o and VOF, '' have been claimed. The oxyfluoride S,O,F, yields first-stage intercalation compounds with both graphite and boron nitride'2p13. With graphite, at RT a blue first-stage compound C,f_,,[SO,F]- (I, = 787 pm) is obtained, which decomposes to give a second-stage compound of I, = 1131 pm. With white boron nitride, a blue first-stage compound, [BN]:[SO,F](I, = 803 pm), is obtained, which is stated to be an electrical conductor". The result with boron nitride is exciting and unexpected. Although having structural similarities to graphite',, boron nitride is an electrical insulator of band gap 3.85 eVI5. The p, electrons are localized in BN, as they are in the molecular analog b ~ r a z i n e ' ~ - 'Previous ~. efforts to intercalate metal halides into boron nitride did not achieve loadings as high as (BN),X, although problems with the synthesis and interpretation may have Although five different samples of BN, have been used, a conducting, blue sample of [BN],[SO,F] has not been recovered with any BN specimen". There may be materials problems with BN, therefore, and the existence of a graphite, types structure for BN is difficult to establish". However, the reaction of graphite with S,O,F, has been confirmed, and additional studies utilizing both S,O,F, and HSO,F have been made". Separately, CsS0,F has been shown to react with graphite to yield an intercalation compound that does not contain the initial CsSO,F, a situation analogous to that found with XeF, and IF, 2 3 .

,

(L B. EBERT)

1. R. C. Croft, Aust. J. Chem., 9, 184 (1956). 2. J. C. Hooley, Can. J. Chem., 40, 745 (1962). 3 H. Selig, 0. Gani, Inorg. Nucl. Chem. Lett., 11, 75 (1975). 4. L. B. Ebert, H. Selig, Mater. Sci. Eng., 31, 177 (1977). 5. H. Selig, L. B. Ebert, A h . Inorg. Chem. Radiochem., 23, 281 (1980). 6. V. Munch, H. Selig, L. B. Ebert, J. Fluorine Chem., 15, 223 (1980). 7. L. B. Ebert, R. A. Huggins, J. I. Brauman, Carbon, Z2, 199 (1974). 8. H. Selig, P. K. Gallagher, L. B. Ebert, Inorg. Nucl. Chem. Lett., 13, 427 (1977). 9. H. Selig, M. J. Vasilie, F. A. Stevie, W A. Sunder, J . Fluorine Chem., 10, 299 (1977). 10. A. Boeck, W. Rudorff, Z. Anorg. Allg. Chem., 397, 179 (1973). 11. A. Herold, in Physics and Chemistry of Materials with Layered Structures, Vol. 6, Intercalated Layered Materials, F. Levy, ed., D. Reidel, Dordrecht, 1979, pp. 323-421; Chem. Abstr., 92, 156,283 (1979). 12. N Bartlett, R. N. Biagoni, B. W McQuillan, A. S. Robertson, A. C. Thompson, J. Chem. Soc., Chem. Commun., p. 200 (1978). 13. N. Bartlett, R. N. Biagoni, G . McCarron, B. McQuillan, F. Tanzella, in Molecular Metals, W. E. Hatfield, ed , Plenum Press, New York, 1979, pp. 293-299. 14. F. Hulliger, in Structural Chemistry of'Layer Type Phases, F. Levy, ed., D. Reidel, Dordrecht, 1976, pp. 50-51.

244

16.4. The Formation of Sheet Structures 16.4.2. Graphite and Boron Nitride Intercalation Compounds 16.4.2.1. Electron Acceptors

15. M. B. Khusidman, Sou. Phys.-Sol. St., 14, 2791 (1973). 16. E. L. Muetterties, The Chemistry of Boron and Its Compounds, New York, Wiley, 1967, pp. 410-423. 17. J. E. Huheey, Inorganic Chemistry: Principles of Structure and Reactivity, New York, Harper and Row, 1972, pp. 516-518. 18. K. Ohhashi, T. Shinjo, Bull. Inst. Chem. Res. (Kyoto),55, 441 (1977). 19. J. G. Hooley, in Preparation and Crytal Growth of Materials with Layered Structures, R. M. A. Lieth, ed., D. Reidel, Dordrecht, 1977, pp. 1-33. 20. J. G. Hooley, Carbon, 21, 181 (1983). 21. T. S. Bartnitskaya, J. Less-Common Met., 117, 253 (1986). 22. S. Karunanithy, F. Aubke, Mater. Sci. Eng., 62, 241 (1984). 23. L. B. Ebert, E. H. Appelman, Phys. Rev. Ser. B, 28, 1637 (1983). 16.4.2.1.3. Oxides.

The oxides CrO,, MOO,, Sb,O,, Tl,S(+S), Cu(+S), WS,, Fe(+S), Sb,S5 and PdS( + S) are reported' to intercalate graphite. Some disagreement has arisen over the and nature of the two most concentrated intercalation compounds, C,,,CrO, Cg9Mo0, ,, with evidence indicating these to be merely mixtures of graphite and reduced simple metal oxides. Thus, in CrO,, direct combination with either graphite or boron nitride under the previous conditions leads to the formation of Cr,O, and/or C r 2 0 5without a change in the lamellar lattices of graphite or boron nitride2. Only with a synthetic method involving CrO, in refluxing CH,COOH is intercalation to a compound of formula C::.46([Cr0,] -)o,46(Cr03)o,54 o b s e r ~ e d ~Assuming ,~. one C r + 5for every Cr+6,an oligomeric dimer can be envisioned as illustrated in Fig. 3. A refinement of this synthetic method involves the passage of electrolytic current through a solution of CrO, in CH,COOH at RT, which allows direct formation of a stage-I11compound5.In a related synthesis6, chromyl trifluoroacetate in trifluoroacetic anhydride intercalates graphite to give a first-stage compound, C,,CrO2[O2CCF3],. The nature of the graphite can influence the intercalation reaction7-,. Thus, the direct combination method generates' C,,CrO, for particles 20 pm x 3 pm in size, but only C,,,CrO, for particles 1100 pm x 32 pm in size. Unfortunately, the reported intercalation product was treated with 6 mol HCl at 100°C before characterization, so it

GRAPHITE MACRO-CATION STABILIZES INTERCALATED OLlGOMERlC ANIONS

Figure 3.

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc. 244

16.4. The Formation of Sheet Structures 16.4.2. Graphite and Boron Nitride Intercalation Compounds 16.4.2.1. Electron Acceptors

15. M. B. Khusidman, Sou. Phys.-Sol. St., 14, 2791 (1973). 16. E. L. Muetterties, The Chemistry of Boron and Its Compounds, New York, Wiley, 1967, pp. 410-423. 17. J. E. Huheey, Inorganic Chemistry: Principles of Structure and Reactivity, New York, Harper and Row, 1972, pp. 516-518. 18. K. Ohhashi, T. Shinjo, Bull. Inst. Chem. Res. (Kyoto),55, 441 (1977). 19. J. G. Hooley, in Preparation and Crytal Growth of Materials with Layered Structures, R. M. A. Lieth, ed., D. Reidel, Dordrecht, 1977, pp. 1-33. 20. J. G. Hooley, Carbon, 21, 181 (1983). 21. T. S. Bartnitskaya, J. Less-Common Met., 117, 253 (1986). 22. S. Karunanithy, F. Aubke, Mater. Sci. Eng., 62, 241 (1984). 23. L. B. Ebert, E. H. Appelman, Phys. Rev. Ser. B, 28, 1637 (1983). 16.4.2.1.3. Oxides.

The oxides CrO,, MOO,, Sb,O,, Tl,S(+S), Cu(+S), WS,, Fe(+S), Sb,S5 and PdS( + S) are reported' to intercalate graphite. Some disagreement has arisen over the and nature of the two most concentrated intercalation compounds, C,,,CrO, Cg9Mo0, ,, with evidence indicating these to be merely mixtures of graphite and reduced simple metal oxides. Thus, in CrO,, direct combination with either graphite or boron nitride under the previous conditions leads to the formation of Cr,O, and/or C r 2 0 5without a change in the lamellar lattices of graphite or boron nitride2. Only with a synthetic method involving CrO, in refluxing CH,COOH is intercalation to a compound of formula C::.46([Cr0,] -)o,46(Cr03)o,54 o b s e r ~ e d ~Assuming ,~. one C r + 5for every Cr+6,an oligomeric dimer can be envisioned as illustrated in Fig. 3. A refinement of this synthetic method involves the passage of electrolytic current through a solution of CrO, in CH,COOH at RT, which allows direct formation of a stage-I11compound5.In a related synthesis6, chromyl trifluoroacetate in trifluoroacetic anhydride intercalates graphite to give a first-stage compound, C,,CrO2[O2CCF3],. The nature of the graphite can influence the intercalation reaction7-,. Thus, the direct combination method generates' C,,CrO, for particles 20 pm x 3 pm in size, but only C,,,CrO, for particles 1100 pm x 32 pm in size. Unfortunately, the reported intercalation product was treated with 6 mol HCl at 100°C before characterization, so it

GRAPHITE MACRO-CATION STABILIZES INTERCALATED OLlGOMERlC ANIONS

Figure 3.

16.4.2. Graphite and Boron Nitride Intercalation Compounds 16.4.2.1. Electron Acceptors 16.4.2.1.4. Acids.

245

was exposed to molecular chlorine?, a situation not present in earlier work'. The only other example of metal oxide intercalation is with Re,O, '. Nonmetallic oxides that intercalate graphite include N,O,, SO,, Cl,O, and SeO,, corresponding to anhydrides of inorganic acids''-' '. The intercalation compounds of these anhydrides have much in common with those of graphite with inorganic acids, discussed in 616.4.2.1.4. With use of a new aqueous impregnation method, a C,,CrO, intercalation compound has been claimed", although neither diffra~tion'~ nor ESR14 supports the presence of intercalation. (L.B. EBERT)

1. R. C. Croft, Aust. J. Chem., 9, 201 (1956). 2. L. B. Ebert, R. A. Huggins, J. I. Brauman, Carbon, 12, 199 (1974). 3. V. P. Elutin, Dokl. Acad. Nauk SSSR, 191, 73 (1970). 4. L. B. Ebert, L. Matty, Synth. Met., 4, 345 (1982). 5. W. Metz, H. Meyer-Spasche, Synth. Met., 1, 63 (1979). 6. P. Touzain, E. Buscarlet, L. Bonnetain, Ann. Chim. Fr., 3, 193 (1978). 7. J. G. Hooley, M. Reimer, Carbon, 13, 401 (1975). 8. U'.Metz, H. Meyer-Spasche, Synth. Met., 1, 53 (1979). 9. H. Fuzellier, A. Herold, Proc. 4th London Con$ on Carbon and Graphite, 322 (1974). 10. L. B. Ebert, Annu. Rev. Muter. Sci., 6, 181 (1976). 11. A. Herold, in Physics and Chemistry of Materials with Layered Structures, Vol. 6, Intercalated Layered Materials, F. Levy, ed., D. Reidel, Dordrecht, 1979,323-421; Chem. Abstr., 92, 156,283 (1979) 12. J. M. Skowronski, Carbon, 24, 185 (1986). 13. L. B. Ebert, J. C. Scanlon, Carbon, 25, 437 (1987). 14. L. B. Ebert, J. C. Scanlon, L. A. Gebhard, Carbon, 26, 906 (1988). 16.4.2.1.4. Acids.

The details of intercalation of graphite by inorganic acids have been long known'. The presence of an oxidation agent, as HNO,, CrO,, K[Mn04], [NH4]2[SzOs], PbO,, HIO,, HIO,, Mn3+, Mn4+ or an electrochemical anode, in the proportion 1 equiv: 27molcarbon,isnecessaryfortheformation ofbluegraphitesalts withsuchacids asH,S04, HSeO, and HClO,. Anodic oxidation is used to synthesize compounds of graphite with sulfuric, chloro- and fluorosulfonic, selenic, perchloric, and nitric acids, from which compounds containing trifluoroacetic, arsenic, periodic, phosphoric, and pyrophosphoric acids are generated by double decomposition'. Significantly, electrochemical experiments, suggest such acid salts in stage I to have the approximate formula C14A-(HA),, consistent with the 1938 work. The formation of such salt structures is not the only possible reaction, however; covalent carbon-oxygen bond formation and ultimate evolution of carbon dioxide are always possible side reactions4. The relative importance of this oxygenation chemistry depends on the graphite employed'. A discussion of the origin of the conductivity increase when graphite is intercalated has been presented6. The degree of charge transfer is apparently not the predominant factor determining the high conductivity?. As expected, the formation of covalent carbon-oxygen bonds has a deleterious effect on electronic conductivity'. Furthermore, acids within graphite may have properties different from free acids9-''. Intercalation compounds of graphite with perfluoroalkanesulfonic acids and alkane sulfonic acids occur' '. Phase transitions occur in graphite-nitric acid compounds1*-14. (L.B EBERT)

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc. 16.4.2. Graphite and Boron Nitride Intercalation Compounds 16.4.2.1. Electron Acceptors 16.4.2.1.4. Acids.

245

was exposed to molecular chlorine?, a situation not present in earlier work'. The only other example of metal oxide intercalation is with Re,O, '. Nonmetallic oxides that intercalate graphite include N,O,, SO,, Cl,O, and SeO,, corresponding to anhydrides of inorganic acids''-' '. The intercalation compounds of these anhydrides have much in common with those of graphite with inorganic acids, discussed in 616.4.2.1.4. With use of a new aqueous impregnation method, a C,,CrO, intercalation compound has been claimed", although neither diffra~tion'~ nor ESR14 supports the presence of intercalation. (L.B. EBERT)

1. R. C. Croft, Aust. J. Chem., 9, 201 (1956). 2. L. B. Ebert, R. A. Huggins, J. I. Brauman, Carbon, 12, 199 (1974). 3. V. P. Elutin, Dokl. Acad. Nauk SSSR, 191, 73 (1970). 4. L. B. Ebert, L. Matty, Synth. Met., 4, 345 (1982). 5. W. Metz, H. Meyer-Spasche, Synth. Met., 1, 63 (1979). 6. P. Touzain, E. Buscarlet, L. Bonnetain, Ann. Chim. Fr., 3, 193 (1978). 7. J. G. Hooley, M. Reimer, Carbon, 13, 401 (1975). 8. U'.Metz, H. Meyer-Spasche, Synth. Met., 1, 53 (1979). 9. H. Fuzellier, A. Herold, Proc. 4th London Con$ on Carbon and Graphite, 322 (1974). 10. L. B. Ebert, Annu. Rev. Muter. Sci., 6, 181 (1976). 11. A. Herold, in Physics and Chemistry of Materials with Layered Structures, Vol. 6, Intercalated Layered Materials, F. Levy, ed., D. Reidel, Dordrecht, 1979,323-421; Chem. Abstr., 92, 156,283 (1979) 12. J. M. Skowronski, Carbon, 24, 185 (1986). 13. L. B. Ebert, J. C. Scanlon, Carbon, 25, 437 (1987). 14. L. B. Ebert, J. C. Scanlon, L. A. Gebhard, Carbon, 26, 906 (1988). 16.4.2.1.4. Acids.

The details of intercalation of graphite by inorganic acids have been long known'. The presence of an oxidation agent, as HNO,, CrO,, K[Mn04], [NH4]2[SzOs], PbO,, HIO,, HIO,, Mn3+, Mn4+ or an electrochemical anode, in the proportion 1 equiv: 27molcarbon,isnecessaryfortheformation ofbluegraphitesalts withsuchacids asH,S04, HSeO, and HClO,. Anodic oxidation is used to synthesize compounds of graphite with sulfuric, chloro- and fluorosulfonic, selenic, perchloric, and nitric acids, from which compounds containing trifluoroacetic, arsenic, periodic, phosphoric, and pyrophosphoric acids are generated by double decomposition'. Significantly, electrochemical experiments, suggest such acid salts in stage I to have the approximate formula C14A-(HA),, consistent with the 1938 work. The formation of such salt structures is not the only possible reaction, however; covalent carbon-oxygen bond formation and ultimate evolution of carbon dioxide are always possible side reactions4. The relative importance of this oxygenation chemistry depends on the graphite employed'. A discussion of the origin of the conductivity increase when graphite is intercalated has been presented6. The degree of charge transfer is apparently not the predominant factor determining the high conductivity?. As expected, the formation of covalent carbon-oxygen bonds has a deleterious effect on electronic conductivity'. Furthermore, acids within graphite may have properties different from free acids9-''. Intercalation compounds of graphite with perfluoroalkanesulfonic acids and alkane sulfonic acids occur' '. Phase transitions occur in graphite-nitric acid compounds1*-14. (L.B EBERT)

246

1. 2. 3. 4.

16.4. The Formation of Sheet Structures 16.4.2 Graphite and Boron Nitride Intercalation Compounds 16.4.2.1. Electron Acceptors

W. Rudorff, U. Hofmann, Z. Anorg. Allg. Chem., 238, l(1938). M. J. Bottomley, G. S. Parry, A. R. Ubbelohde, D. A. Young, J. Chem. Soc., p. 5674 (1963).

A. R. Ubbelohde, Carbon, 2, 23 (1968). J. G. Hooley, in Preparation and Growth of Materials with Layered Structures, R. M. A. Lieth, ed., D. Reidel, Dordrecht, 1977, pp. 1-33. 5. D. Horn, H. P. Boehm, Z. Anorg. Allg. Chem., 456, 117 (1979). 6. L. B. Ebert, J. C. Scanlon, Ind. Eng. Chem., Prod. Res. Dev., 19, 103 (1980). 7. L. Pietronero, S. Strassler, H. R. Zeller, M. J. Rice, Phys. Rev. B, 22, 904 (1980). 8. E. McRae, A. Metrot, P. Willmann, A. Herold, Physzca, 99B, 541 (1980). 9. A. Avogadro, M. Villa, J. Chem. Phys., 70, 109 (1979). 10. B. Iskander, P. Vast, A. Lorriaux-Rubbens, M. L. Dele-Dubois, P. Touzain, Mater. Sci.Eng., 43, 59 (1980). 11. H. P. Boehm, W. Helle, B. Ruisinger, Synth. Met., 23, 395 (1988). 12. I. Rosenman, F. Batallan, A. Magerl, H. Fuzellier, Synth. Met., 12. 117 (1985). 13. I. Rosenman, C. Simon. F. Batallan, A. Magerl, Europhys. Lett., 3, 1013 (1987). 14. F. Batallan, I. Rosenman, C. Simon, A. Magerl, Synth. Met., 23, 49 (1988). 16.4.2.1.5. Halogens and Interhalogens.

Each of the molecular halogens reacts with graphite in a different way': Fluorine forms covalent bonds with carbon (516.4.2.4). Chlorine forms only dilute compounds, stable below -20°C '. Bromine reacts easily with graphite at RT to form a second-stage compound, C,,Br, with I, = 1030 pm '. Iodine does not intercalate into graphite at all', although it is adsorbed on graphite4. Interhalogens that intercalate into graphite include ICl, IBr, ClF,, ClF,, BrF,, IF, and IF, Despite the seeming simplicity of these intercalation compounds, there is much uncertainty about their nature. Thus, the existence of the graphite-chlorine compound is may be incondisputed'. Similarly, the claims for Br; 8-9 in C,Br, compounds"-" sistent with the magnetic properties of the graphite-bromine system". Current evidence suggests that the predominant interaction of bromine with graphite is weakl3-I4, more analogous to that of the aromatic charge-transfer complexes' than to that of discrete aromatic ions',. However, covalent carbon-bromine bond formation at defects has been noted", and the interhalogen compound IF, strongly fluorinates graphite to yield CF,(IF,) 5-6.

,-'.

(L.B. EBERT)

L. B. Ebert, Annu. Rev. Mater. Sci., 6, 181 (1976). G. Furdin, M. Lelaurin. E. McRae, J. F. Marcehe, A. Herold, Carbon, 17, 329 (1979). W. T. Eeles, J. A. Turnbull, Nature (London), 198, 877 (1963). M. Fleischmann, P. J. Hendra, J. Robinson, Nature (London),288, 152 (1980). H. Selig, W. A. Sunder, M. J. Vasilie, F. A. Stevie P. K. Gallagher, L. B. Ebert, J. Fluorine Chem., 12, 397 (1978). 6. H. Selig, L. B. Ebert, Ado. Inorg. Chem. Radiochem., 23, 281 (1980). 7. J. G. Hooley, Carbon, 8, 333 (1970). 8. I. Marov, M. C. R. Symons, J. Chem. Soc., A , p. 201 (1971). 9. L. D. Kispert, J. Pearson, J. Phys. Chem., 76, 133 (1972). 10. D. A. Young, Carbon, 15, 373 (1977). 11. J. D Hibbs, D. A. Young, Chem. Phys. Lett., 53, 361 (1978). 12. G. R. Hennig, B. Smaller, E. L. Yasaitas, Phys. Rev., 95, 1088 (1954). 13. J. J. Song, D. D. L. Chung, P. C. Eklund, M. S. Dresselhaus, Soltd. State Commun., 20, 1111 (1976). 14. S. M. Heald, E. A. Stern, Phys. Rev. B, 17, 4069 (1978). 15. W. B. Person, C. F. Cook, H. B. Friedhch, J . Chem. Phys., 46,2521 (1967). 16. R. Foster, Organic Charge Transfer Complexes,Academic Press, New York, 1969. 1. 2. 3. 4. 5.

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc. 246

1. 2. 3. 4.

16.4. The Formation of Sheet Structures 16.4.2 Graphite and Boron Nitride Intercalation Compounds 16.4.2.1. Electron Acceptors

W. Rudorff, U. Hofmann, Z. Anorg. Allg. Chem., 238, l(1938). M. J. Bottomley, G. S. Parry, A. R. Ubbelohde, D. A. Young, J. Chem. Soc., p. 5674 (1963).

A. R. Ubbelohde, Carbon, 2, 23 (1968). J. G. Hooley, in Preparation and Growth of Materials with Layered Structures, R. M. A. Lieth, ed., D. Reidel, Dordrecht, 1977, pp. 1-33. 5. D. Horn, H. P. Boehm, Z. Anorg. Allg. Chem., 456, 117 (1979). 6. L. B. Ebert, J. C. Scanlon, Ind. Eng. Chem., Prod. Res. Dev., 19, 103 (1980). 7. L. Pietronero, S. Strassler, H. R. Zeller, M. J. Rice, Phys. Rev. B, 22, 904 (1980). 8. E. McRae, A. Metrot, P. Willmann, A. Herold, Physzca, 99B, 541 (1980). 9. A. Avogadro, M. Villa, J. Chem. Phys., 70, 109 (1979). 10. B. Iskander, P. Vast, A. Lorriaux-Rubbens, M. L. Dele-Dubois, P. Touzain, Mater. Sci.Eng., 43, 59 (1980). 11. H. P. Boehm, W. Helle, B. Ruisinger, Synth. Met., 23, 395 (1988). 12. I. Rosenman, F. Batallan, A. Magerl, H. Fuzellier, Synth. Met., 12. 117 (1985). 13. I. Rosenman, C. Simon. F. Batallan, A. Magerl, Europhys. Lett., 3, 1013 (1987). 14. F. Batallan, I. Rosenman, C. Simon, A. Magerl, Synth. Met., 23, 49 (1988). 16.4.2.1.5. Halogens and Interhalogens.

Each of the molecular halogens reacts with graphite in a different way': Fluorine forms covalent bonds with carbon (516.4.2.4). Chlorine forms only dilute compounds, stable below -20°C '. Bromine reacts easily with graphite at RT to form a second-stage compound, C,,Br, with I, = 1030 pm '. Iodine does not intercalate into graphite at all', although it is adsorbed on graphite4. Interhalogens that intercalate into graphite include ICl, IBr, ClF,, ClF,, BrF,, IF, and IF, Despite the seeming simplicity of these intercalation compounds, there is much uncertainty about their nature. Thus, the existence of the graphite-chlorine compound is may be incondisputed'. Similarly, the claims for Br; 8-9 in C,Br, compounds"-" sistent with the magnetic properties of the graphite-bromine system". Current evidence suggests that the predominant interaction of bromine with graphite is weakl3-I4, more analogous to that of the aromatic charge-transfer complexes' than to that of discrete aromatic ions',. However, covalent carbon-bromine bond formation at defects has been noted", and the interhalogen compound IF, strongly fluorinates graphite to yield CF,(IF,) 5-6.

,-'.

(L.B. EBERT)

L. B. Ebert, Annu. Rev. Mater. Sci., 6, 181 (1976). G. Furdin, M. Lelaurin. E. McRae, J. F. Marcehe, A. Herold, Carbon, 17, 329 (1979). W. T. Eeles, J. A. Turnbull, Nature (London), 198, 877 (1963). M. Fleischmann, P. J. Hendra, J. Robinson, Nature (London),288, 152 (1980). H. Selig, W. A. Sunder, M. J. Vasilie, F. A. Stevie P. K. Gallagher, L. B. Ebert, J. Fluorine Chem., 12, 397 (1978). 6. H. Selig, L. B. Ebert, Ado. Inorg. Chem. Radiochem., 23, 281 (1980). 7. J. G. Hooley, Carbon, 8, 333 (1970). 8. I. Marov, M. C. R. Symons, J. Chem. Soc., A , p. 201 (1971). 9. L. D. Kispert, J. Pearson, J. Phys. Chem., 76, 133 (1972). 10. D. A. Young, Carbon, 15, 373 (1977). 11. J. D Hibbs, D. A. Young, Chem. Phys. Lett., 53, 361 (1978). 12. G. R. Hennig, B. Smaller, E. L. Yasaitas, Phys. Rev., 95, 1088 (1954). 13. J. J. Song, D. D. L. Chung, P. C. Eklund, M. S. Dresselhaus, Soltd. State Commun., 20, 1111 (1976). 14. S. M. Heald, E. A. Stern, Phys. Rev. B, 17, 4069 (1978). 15. W. B. Person, C. F. Cook, H. B. Friedhch, J . Chem. Phys., 46,2521 (1967). 16. R. Foster, Organic Charge Transfer Complexes,Academic Press, New York, 1969. 1. 2. 3. 4. 5.

16.4.2. Graphite and Boron Nitride Intercalation Compounds 16.4.2.2 Electron Donors 16.4.2.2.1. Graphite and Carbon Intercalation Compounds.

247

17. L. B. Ebert, L. Matty, D. R. Mills, J. C. Scanlon, Mater. Rex Bull., 15, 251 (1980). 18. C. Mazieres, G. Colin, J. Jegoudez, R., Setton, Carbon, 14, 176 (1976).

16.4.2.2. Electron Donors 16.4.2.2.1. Synthesis of Graphite and Carbon Intercalation Compounds Containing One metal'+.

(i) Heavy alkali metals. General. Because of the partial ionization of the intercalated metal, the metal-graphite compounds can be called graphitides. Potassium, Rb and Cs directly intercalate in the liquid and vapor phases. The yellow-bronze MC, compounds of stage I are stoichiometric: the in-plane spacing of the intercalated layers is twice that of the graphite sheets. The difference between the experimental and theoretical M:C ratio is < 1-2 %. The steel-blue compounds of stage 11, which contain liquid-like metallic layers, are not stoichiometric: the MC,, formula is only approximate. This is also true for the formulas MC,,, MC,,, . . . , MC,,, generally used for the dark blue compound of stage I11 and the black compounds of stages IV, . . . , n. (ii) Synthesis in the liquid phase. The molten metal in excess reacts strongly with all kinds of graphite, leading to the MC, first-stage compounds6,'. Because of the large expansion of the elemental crystallites along the axis, the natural and industrial polycrystalline graphites are pulverized during intercalation. The reaction is exothermic (-AH in J per mole of metal is about 40, 46.5 and 64, respectively, for K, Rb and Cs). Excess metal is separated by distillation or centrifugation (for this technique see $16.4.2.2.2). Because the liquid-phase reaction is fast, it produces many defects in the solid. (iii) Synthesis in the vapor phase. Vapor-phase intercalation is slower (requiring several hours) and leads to better materials. It can be carried out in a two-bulb reactor (Fig. l.a)738. When the difference T, - T, between the temperature, T,, of the bulb containing the graphite and the temperature, T,, of the bulb containing the molten metal is low, the MC, first-stage compound is formed. For higher temperature differences the secondstage MC,,, and the nth stages MC,,, are formed. Figure 1.b shows an isobaric curve of the graphite-potassium system where T, = T, is constant at 250°C. When TG - T, increases from 0 to 360°C, the K: C ratio decreases from 1/8 to zero. First-stage KC, and second-stage KC,, are in equilibrium under the saturated vapor pressure of the metal at a temperature, O,, corresponding to the almost vertical line that separates the long plateau of the two compounds; 0, is related to the temperature, T,, of the metal by:

c

-

1

P

--=I$.TM QG

The experimental values of c1 and fl corresponding to the transitions first stage F? second stage are given in Table 1. The change of color upon passing from the first to the second stage and vice versa, which has been used to study these equilibria3. Because of the larger domain of stability, this color change can be used during synthesis to monitor the firststage to second-stage transition.

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc. 16.4.2. Graphite and Boron Nitride Intercalation Compounds 16.4.2.2 Electron Donors 16.4.2.2.1. Graphite and Carbon Intercalation Compounds.

247

17. L. B. Ebert, L. Matty, D. R. Mills, J. C. Scanlon, Mater. Rex Bull., 15, 251 (1980). 18. C. Mazieres, G. Colin, J. Jegoudez, R., Setton, Carbon, 14, 176 (1976).

16.4.2.2. Electron Donors 16.4.2.2.1. Synthesis of Graphite and Carbon Intercalation Compounds Containing One metal'+.

(i) Heavy alkali metals. General. Because of the partial ionization of the intercalated metal, the metal-graphite compounds can be called graphitides. Potassium, Rb and Cs directly intercalate in the liquid and vapor phases. The yellow-bronze MC, compounds of stage I are stoichiometric: the in-plane spacing of the intercalated layers is twice that of the graphite sheets. The difference between the experimental and theoretical M:C ratio is < 1-2 %. The steel-blue compounds of stage 11, which contain liquid-like metallic layers, are not stoichiometric: the MC,, formula is only approximate. This is also true for the formulas MC,,, MC,,, . . . , MC,,, generally used for the dark blue compound of stage I11 and the black compounds of stages IV, . . . , n. (ii) Synthesis in the liquid phase. The molten metal in excess reacts strongly with all kinds of graphite, leading to the MC, first-stage compounds6,'. Because of the large expansion of the elemental crystallites along the axis, the natural and industrial polycrystalline graphites are pulverized during intercalation. The reaction is exothermic (-AH in J per mole of metal is about 40, 46.5 and 64, respectively, for K, Rb and Cs). Excess metal is separated by distillation or centrifugation (for this technique see $16.4.2.2.2). Because the liquid-phase reaction is fast, it produces many defects in the solid. (iii) Synthesis in the vapor phase. Vapor-phase intercalation is slower (requiring several hours) and leads to better materials. It can be carried out in a two-bulb reactor (Fig. l.a)738. When the difference T, - T, between the temperature, T,, of the bulb containing the graphite and the temperature, T,, of the bulb containing the molten metal is low, the MC, first-stage compound is formed. For higher temperature differences the secondstage MC,,, and the nth stages MC,,, are formed. Figure 1.b shows an isobaric curve of the graphite-potassium system where T, = T, is constant at 250°C. When TG - T, increases from 0 to 360°C, the K: C ratio decreases from 1/8 to zero. First-stage KC, and second-stage KC,, are in equilibrium under the saturated vapor pressure of the metal at a temperature, O,, corresponding to the almost vertical line that separates the long plateau of the two compounds; 0, is related to the temperature, T,, of the metal by:

c

-

1

P

--=I$.TM QG

The experimental values of c1 and fl corresponding to the transitions first stage F? second stage are given in Table 1. The change of color upon passing from the first to the second stage and vice versa, which has been used to study these equilibria3. Because of the larger domain of stability, this color change can be used during synthesis to monitor the firststage to second-stage transition.

248

16.4. The Formation of Sheet Structures 16.4.2. Graphite and Boron Nitride Intercalation Compounds 16.4.2 2. Electron Donors

'G

(b) Figure 1.

Some convenient values of T, and T, are given in Table 2. Experimental diagrams with the coordinates I/T and log p (p = vapor pressure of the metal) make the domains of stability of the different stages beyond RT11p13precise. The purple phase, called MC,,, is a disordered nonstoichiometric first-stage compound" that instantaneously decomposes into MC, and MC,, below a critical temperature (375°C in the case of potassium) and cannot be kept at RT TABLE 1. CONSTANT CONCERNING THE FIRST STAGE@ SECONDSTAGEEQUILIBRIA. Metals c(

B

K

Rb

Cs

0.26 x 10-5 1.19

0.83 x 10-4 1.19

1.66 x 10-4 1.51

TABLE 2. TEMPERATURES PREPARING FOR HEAVYALKALI METALS-GRAPHITE COMPOUNDSO

Stage I Stage I1

TJC) TJC) TJC)

K

Rb

cs

250 225-320 350-400

208 215-330 375-430

194 200-425 475-530

16.4.2. Graphite and Boron Nitride Intercalation Compounds 16.4.2.2. Electron Donors 16.4.2.2.1. Graphite and Carbon Intercalation Compounds.

249

The compounds MC,,, of stages n > 2 have a small domain of stability, and their synthesis in the two-bulb reactor is difficult, requiring successive tests to find the convenient T, and T, temperatures. Another method consists in heating graphite in a sealed tube or in an air-tight vessel with a calculated amount of metal or of first-stage c~mpound’~,’~. The stage can be characterized by x - r a y ~ ~up~ to , ~stage ’ 10 and can be evaluated up to stage 15. Theoretical calculations have led to models of phase diagrams with coordinates of composition and temperature’ that have received partial experimental verifi~ation’~. (iv) Intercalation into ungraphitized carbons. Heavy alkali metals are able to intercalate into all kinds of carbons, leading to nonstoichiometric compounds”. The in-plane density of the metallic layers is higher in the corresponding compounds than in the graphite ones. Consequently, in the soft carbons the ratio M:C is higher than in graphite. In the hard carbons, the intercalations remains incomplete because of the stiffness of the texture and the M:C ratio is lower than in graphite. (v) Intercalation by reaction with organometallic compounds. Graphite-potassium compounds of different stages are synthetized by the reaction of graphite with K[(C,H,)Co(PMe,),] dissolved in pentane”. (vi) Compounds of sodium. The stability of the intercalation compounds decreases in passing from Cs to K. Liquid or vapor phase Na does not intercalate in pure graphite”, or hardly intercalates at 400°C leading to an VIIIth stage compound, NaC,,23. Compounds of stages VII and VI are prepared in the vapor phase at lower temperatures or in the liquid phase by introducing calcium into the molten sodiumz4. On the other hand, sodium intercalates in the soft carbons with increasing facility as the maximal temperature of treatment of carbon (HTT) is lowered; e.g., a petroleum coke at 1250°C leads to a third stage NaC,,, and a coke heated at 800°C can lead to a first stage with a composition near NaC,. Hard carbons are also able to accept 20 % of their weight or more of sodium, partly by intercalation, and more by adsorption and condensation in the porosity”. (vii) Intercalation of lithium. General. The direct synthesis of well-defined lithium graphite can be carried out in the vapor, liquid or solid phases. A metallic vessel is needed because lithium strongly attacks glasses. The temperature must not exceed 400°C to avoid the formation of the acetylide, Li,C,. The yellow-brass LiC, first stage is a stoichiometric compound. The blue second stage can also be stoichiometric, with the formula LiC,, especially in the presence of the first stage. However, the experimental C:Li ratio can vary from 12 to > 25. Nonstoichiometry is also a general characteristic of the compounds of higher stages. (viii) Synthesis in the vapor phase. The first-stage compound, LiC,, can be prepared from lithium vapor at 400°C with previously outgassed pyrographite or single crystals in a sealed copper or stainless steel tube”. Complete intercalation requires several hours or days, varying with the dimensions of the sample. The synthesis of second-stage compoundsz6 by using a convenient difference between the temperatures T, and T, could not be confirmed by x-ray. (ix) Synthesis in the liquid phase. Complete intercalation of pyrographite leading to an MC, compound can be carried out in several hours by immersion in molten lithium in a stainless steel crucible at 350°C under argon. The excess metal that covers the intercalated sample must be mechanically separated”.

250

16.4. The Formation of Sheet Structures 16.4.2. Graphite and Boron Nitride Intercalation Compounds 16.4.2.2. Electron Donors

The use of dilute solutions of lithium in liquid sodium offers greater possibilities: pure LiC, can be prepared by immersing pyrographite samples in molten sodium containing a few percent lithium at 350"C2' or at 15o"C3O in a stainless steel apparatus under argon pressure. The product can be separated from the bath by centrifugation. Compounds of stages > I can be prepared by using convenient concentrations of Li in the bath. Stages 11, 111, IV and V have been obtained by successively introducing several pyrographite samples in the same sodium lithium alloy bath3'. (x) Synthesis in the solid phase. Partial intercalation occurs by compression under several kilobars of a mixture of graphite and lithium powder. The reaction goes to completion by heating the pellet under vacuum or an argon atmosphere at -200°C. Compounds of the first or higher stages, especially the second, can be prepared using convenient Li: C ratios. This method is useful in preparing large quantities". (xi) Formation of new metastable phases under high pressure. By Le Chatelier's principle, intercalation compounds submitted to very high pressure transform into new phases containing metal layers of higher in-plane density and a higher proportion of empty galleries. For instance, the first-stage KC, transforms into a stage II/III compound in which the metal layer corresponds to a C:K ratio of 6; the second stage KC,, becomes a third stage containing metal layers with a C:K of 832-34. On the other hand, by compressing mixtures of alkai metal and graphite of an appropriate composition, new, rich first-stage compounds such as KC, and KC435, NaC, and NaC,,, and LiC, and LiC,35337are formed. The different phases obtained under high pressure are metastable at RT and alm pressure. (xii) Alkaline-earth metal-graphite compounds. Clear yellow MC, first-stage compounds3*are formed in the vapor phase by heating stainless steel tubes containing the metal, M, and single crystals of graphite or pyrographite samples. For Sr and Ba, complete intercalation can be carried out in several weeks at 470 and 500"C, respectively. The intercalation of calcium remains superficial. By heating at higher temperatures the carbides MC, are formed. SrC, and BaC, can also be prepared by heating pellets of compressed metal and graphite powders. Mixtures of higher stage compounds of formula MC,,, can appear transiently during the synthesis of the first stages in the vapor phase. (xiii) Lanthanide-graphite compounds. Clear yellow LiC, first-stage compounds are synthesized for Ln = Sm, Eu, Tm, Yb3'. Their synthesis is at the limits of possibility, because of the low vapor pressure of the metals and the parallel reactions leading to nonstoichiometrtic interstitial carbides, LnC,. The best results are obtained in the vapor phase, using sealed metallic or borosilicate glass tubes at 450-500°C corresponding to pressures of to torr. Pure EuC, and YbC, can be prepared by intercalation in graphite single crystals, but only superficial reaction has been obtained for samarium and thulium. The use of pyrographite samples also leads to superficial reactions. ( A . HEROLD)

1. A. Herold, R. Setlon, A. Platzer, in Les Carbones, Vol. 11, A. Pacault, ed. Masson, Paris, 1965, p. 499. 2. D. Berger, B. Carton, A. Metrot, A. Herold, Chem. Phys. Carbon, 12, l(1975). 3. A. Herold, in Physics and Chemistry of Materials with Layered Structures, Vol. 6, Intercalated Layered Materials, F. A. Levy, ed., D. Reidel, Dordecht, 1979, p. 323; Chem. Abstract, 92 156-283 (1979). 4. M. S. Dresselhaus, G. Dresselhaus, Adv. Phys., 30 (2), 139 (1981). 5 . A. P. Legrand, S. Flandrois, NATO AS1 Series, Series B, Physics, Vol. 172 (1987)

16.4.2 Graphite and Boron Nitride Intercalation Compounds 16.4.2.2. Electron Donors 16.4.2.2 2 Graphite Intercalation Compounds. 6. 7. 8. 9. 10. 11.

12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39.

251

E Weinlraub, U.S. Pat. 992,645 (1909). K. Fredenhagen, G. Cadenbach, Z. Anorg., ANg. Chem., 158, 249 (1926). A. Herold, Bull. Soc. Chim. Fr., , 999 (1955) A. Herold, D. Saehr, Bull. SOC.Chim. F r , 1287 (1964). D. E. Nixon, G. S. Parry, Br. Appl. Phys., Ser. 2, 1, 291, (1968) S. Aronson, F. J. Salzano, J. Chem. Phys., 42, 1323 (1965); 45, 2221 (1966); 45, 4551 (1966). B. Carton, Bull. Soc Chim. Fr., 4624 (1970). N. Nishitani, Y. Uno, H. Suematsu, Phys. Rev., B, 27, 6573 (1983). W. Rudorff, E. Schulze, Angew. Chem., 66, 305 (1954). A. Maaroufi, S. Flandrois, D. Guerard, J. Chim. Phys. (Fr). 84, 11/12 (1987). D. Billaud, J. F. Madehe, E. McRae, A. Herold, Synth, Met., 2, 37 (1980). C. Underhill, T. Krapchev, M. S. Dresselhaus, Synth. Met., 2, 47 (1980). S. Safran, Synth. Met., 2, 1, (1980); ref. 5, p. 47. H. Suematsu, K. Suda, N. Metoki, Synth. Met., 23, 7 (1988). B. Carton, A. Herold, Bull. Soc. Chim. Fr. 521 (1972); ref. 2. H. F. Klein, J. Gross, J. 0. Besenhard, Angew. Chem., Int. Ed. Engl., 19, 491 (1980). A. Metrot, A. Herold, J. Chim. Phys., 71, 71 (1959); ref. 2. R. C. Asher, J. Inorg. Nucl. Chem., 10, 238 (1959). A. Metrot, D. Guerard, D. Billaud, Synth. Met., I , 363 (1979/1980). D. Guerard, A. Herold, Carbon, 13, 337 (1975). P. Pfluger, V. Geiser, S. Stolz, H. J. Guntherodt, Synth. Met. 3, 27 (1989). D. Guerard, unpublished experiments. M. Zanini, S. Basu, J E. Fischer, Carbon, 16, 211 (1978). S. Basu, C. Zeller, P. Flanders, C. J. Fuerst, W. D. Johnson, J. Fischer, Mater. Sci. Eng., 38, 27 (1 979). D. Billaud, A. Herold, Carbon, 17, 183 (1979). D. Billaud, E. McRae, J. F. Markhe, A. Herold, Synth. Met., 3, 21 (1981). J. E. Fischer, C. D. Fuerst, K. C. Woo, Synth. Met., 7, l(1983). R. Clarke, N. Wada, S. A. Solin, Phys. Rev. Lett., 44, 1616 (1980). J. E. Fischer, In ref. 5, p 59. K. N. Semenenko, V. V. Avdeev, V. Z. Mordkovich, Proc. Acad. Sci. Chem. SOC.(Engl. Transl.) 271, 273 (1983); Russ. J. Inorg. Chem. (Engl. Transl.), 29, 1277 (1984). I. T. Belash, A. D. Bronnikov, 0. V. Zharikov, A. V. Pal’nichenko, Solid State Commun.,64, 1445 (1987). I. T. Belasn, A. D. Bronnikov, 0.V. Sharikov, A. V. Pal’nichenko, Solid State Commun.,59, 921 (1989). D. Guerard, M. Chaabouni, P. Lagrange, M. El Makrini, A. Herold, Carbon, 18, 257 (1980). M. El Makrini, D. Guerard, P. Lagrange, A. Herold, Carbon, 18, 203 (1980).

16.4.2.2.2. Graphite Intercalation Compounds Containing Two Metals (Ternary Compounds).

(i) General. Two kinds of ternaries containing two metals can be distinguished, the solid solutions and the true ternary compounds. A solid solution, M I -,MkC,, is derived from an MC, compound by partial substitution of the metal M by the metal M‘ in the intercalated layers. In the ternary compounds, the structures of the metallic layers differ from those of the binaries, and the M/M‘ ratio is well defined. (ii) Solid solutions of two heavy alkali metals. Two binary first stage compounds, MC, and M’C, (in which M and M’ are heavy alkali metals), are able to form a continuous range of solid solutions, M,-,M~C, (0 < x < l), all of the same yellowbronze color. These ternaries can be synthesized in the liquid or vapor phase using the same methods as for the binaries, with M,-,ML alloys instead of a pure metal M (see $16.4.2.2.1, heavy alkali metals). In the liquid phase the excess of molten alloy can be separated by volatilization or centrif~gationl-~.

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc. 16.4.2 Graphite and Boron Nitride Intercalation Compounds 16.4.2.2. Electron Donors 16.4.2.2 2 Graphite Intercalation Compounds. 6. 7. 8. 9. 10. 11.

12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39.

251

E Weinlraub, U.S. Pat. 992,645 (1909). K. Fredenhagen, G. Cadenbach, Z. Anorg., ANg. Chem., 158, 249 (1926). A. Herold, Bull. Soc. Chim. Fr., , 999 (1955) A. Herold, D. Saehr, Bull. SOC.Chim. F r , 1287 (1964). D. E. Nixon, G. S. Parry, Br. Appl. Phys., Ser. 2, 1, 291, (1968) S. Aronson, F. J. Salzano, J. Chem. Phys., 42, 1323 (1965); 45, 2221 (1966); 45, 4551 (1966). B. Carton, Bull. Soc Chim. Fr., 4624 (1970). N. Nishitani, Y. Uno, H. Suematsu, Phys. Rev., B, 27, 6573 (1983). W. Rudorff, E. Schulze, Angew. Chem., 66, 305 (1954). A. Maaroufi, S. Flandrois, D. Guerard, J. Chim. Phys. (Fr). 84, 11/12 (1987). D. Billaud, J. F. Madehe, E. McRae, A. Herold, Synth, Met., 2, 37 (1980). C. Underhill, T. Krapchev, M. S. Dresselhaus, Synth. Met., 2, 47 (1980). S. Safran, Synth. Met., 2, 1, (1980); ref. 5, p. 47. H. Suematsu, K. Suda, N. Metoki, Synth. Met., 23, 7 (1988). B. Carton, A. Herold, Bull. Soc. Chim. Fr. 521 (1972); ref. 2. H. F. Klein, J. Gross, J. 0. Besenhard, Angew. Chem., Int. Ed. Engl., 19, 491 (1980). A. Metrot, A. Herold, J. Chim. Phys., 71, 71 (1959); ref. 2. R. C. Asher, J. Inorg. Nucl. Chem., 10, 238 (1959). A. Metrot, D. Guerard, D. Billaud, Synth. Met., I , 363 (1979/1980). D. Guerard, A. Herold, Carbon, 13, 337 (1975). P. Pfluger, V. Geiser, S. Stolz, H. J. Guntherodt, Synth. Met. 3, 27 (1989). D. Guerard, unpublished experiments. M. Zanini, S. Basu, J E. Fischer, Carbon, 16, 211 (1978). S. Basu, C. Zeller, P. Flanders, C. J. Fuerst, W. D. Johnson, J. Fischer, Mater. Sci. Eng., 38, 27 (1 979). D. Billaud, A. Herold, Carbon, 17, 183 (1979). D. Billaud, E. McRae, J. F. Markhe, A. Herold, Synth. Met., 3, 21 (1981). J. E. Fischer, C. D. Fuerst, K. C. Woo, Synth. Met., 7, l(1983). R. Clarke, N. Wada, S. A. Solin, Phys. Rev. Lett., 44, 1616 (1980). J. E. Fischer, In ref. 5, p 59. K. N. Semenenko, V. V. Avdeev, V. Z. Mordkovich, Proc. Acad. Sci. Chem. SOC.(Engl. Transl.) 271, 273 (1983); Russ. J. Inorg. Chem. (Engl. Transl.), 29, 1277 (1984). I. T. Belash, A. D. Bronnikov, 0. V. Zharikov, A. V. Pal’nichenko, Solid State Commun.,64, 1445 (1987). I. T. Belasn, A. D. Bronnikov, 0.V. Sharikov, A. V. Pal’nichenko, Solid State Commun.,59, 921 (1989). D. Guerard, M. Chaabouni, P. Lagrange, M. El Makrini, A. Herold, Carbon, 18, 257 (1980). M. El Makrini, D. Guerard, P. Lagrange, A. Herold, Carbon, 18, 203 (1980).

16.4.2.2.2. Graphite Intercalation Compounds Containing Two Metals (Ternary Compounds).

(i) General. Two kinds of ternaries containing two metals can be distinguished, the solid solutions and the true ternary compounds. A solid solution, M I -,MkC,, is derived from an MC, compound by partial substitution of the metal M by the metal M‘ in the intercalated layers. In the ternary compounds, the structures of the metallic layers differ from those of the binaries, and the M/M‘ ratio is well defined. (ii) Solid solutions of two heavy alkali metals. Two binary first stage compounds, MC, and M’C, (in which M and M’ are heavy alkali metals), are able to form a continuous range of solid solutions, M,-,M~C, (0 < x < l), all of the same yellowbronze color. These ternaries can be synthesized in the liquid or vapor phase using the same methods as for the binaries, with M,-,ML alloys instead of a pure metal M (see $16.4.2.2.1, heavy alkali metals). In the liquid phase the excess of molten alloy can be separated by volatilization or centrif~gationl-~.

252

16.4. The Formation of Sheet Structures 16.4.2. Graphite and Boron Nitride Intercalation Compounds 16.4.2.2. Electron Donors

At the beginning of the synthesis, the intercalated alloy has approximately the composition of the vapor or liquid in contact with the graphite; somewhat later, an equilibrium is established between the intercalated alloy, M, -xM;, and the molten alloy in excess, M,-,MI. The former is always richer in the heavier alkali metal. At a given temperature, the ratio: x 1-y c=x1-x y

-

is constant, but it decreases with temperature (c 370 at 165"C, and 78 at 288°C for M = Rb and M' = Cs). The time to attain the equilibrium is 1-3 weeks for powdered natural graphite and much greater for pyrographite. (iii) Heterostructures. (a) Two heavy alkali metals. Yellow first-stage ternary compounds are prepared by action of another alkali metal M', on second-stage MC,, '. The successful preparation of such a compound has been performed on CsC,, (prepared from HOPG in the standard way) by transferring the platelet to a glass tube containing metallic K, sealing under vacuum and heating to 70°C with the platelet immersed in the molten K 43. At the end of the reaction (12 days), xs K is eliminated by spinning the glass tube in a centrifuge maintained at 70°C. The product is a mixture of two phases, KC, and KCsC,,. This ternary is called a heterostructure or a biintercalation compound; all its interlayer spaces are filled alternately with single-layer sheets of pure K or pure Cs according to the sequence , . . C K C Cs C K C Cs C . . . along the E axis. Using similar experimental conditions, but a longer time of reaction, it is possible to prepare KRbC,, (in admixture with KC,), corresponding to the sequence.. . C K C Rb C K C Rb C . . . '. (iv) One heavy alkali metal and sodium or thallium. Solid solutions of stage-I, M, -.Na,, in which M = K or Cs, have been prepared in the liquid phase; the index x can attain 0.25, but the ternaries are metastable with respect to the M1-,Na, alloys. Higher stage ternaries can also be obtained. As the stage n increases, so does the index, x, which can attain 0.50 ,. Thallium can also be partly substituted for the heavy alkali metals, but at lower concentrations than sodium7. (iv) (a) Ternary compounds. Sodium barium. Blue second-stage compounds can be prepared by reacting molten Na-Ba alloys with pyrographite samples in a stainless steel apparatus separated into two chambers by a metallic cloth filter. After intercalation the reactor is centrifuged to separate the excess molten alloy from the pyrographite. The intercalated layers have a hexagonal structure, but their in-plane spacing seems to be incompatible with that of graphite; consequently, the C:Na + Ba ratio cannot be an integer but lies between 7 and 8. On the other hand, the Na:Ba ratio is often 2 but can vary between 1.6 and 6, probably by partial substitution of one metal for another'. (b) Mercury and potassium or rubidium. The ternary compounds containing mercury and a heavy alkali metal are mercurographitides because of the charge transfer from the alkali metal to the mercury and graphite sheets. The intercalated layers possess a hexagonal structure, and their in-plane parameter is twice that of graphite. The mercurographitides are therefore stoichiometric and exist in four well-defined specie^^-'^: 1. The pink KHgC, and RbHgC,, which are stage 1; 2. The blue KHgC, and RbHgC,, which are stage 2

16.4.2. Graphite and Boron Nitride Intercalation Compounds 16.4.2.2. Electron Donors 16.4.2.2.2. Graphite Intercalation Compounds.

253

3. The first stage MHgC, compounds, which are prepared from graphite with the compounds, MHg. These latter are obtained by heating stoichiometric quantities of mercury and potassium or rubidium in sealed glass tubes at 200°C, and they can be easily. pulverized under an argon atmosphere.

-

-

The intercalation of KHg or RbHg into graphite can be carried out in three ways: (1) the reaction of graphite with a molten MHg amalgam in a sealed glass tube at 200°C under vacuum, which is fast but requires separation of the excess amalgam, making it more convenient for pyrographite than for graphite; (2) the same reaction in the vapor phase at -200°C in a two-bulb tube (see graphite compounds with the heavy alkali metals in 816.4.2.2.1) under vacuum (the reaction is often incomplete when powdered graphite is used); (3) the stoichiometric reaction of a powdered amalgam and powdered graphite in a sealed glass tube at 200°C. This method allows large amounts of MHgC, to be prepared in a powder form. The MHgC, second-stage compounds can also be prepared by three methods: (1) heating a mixture of graphite and MHgC, powder (at a C to MHg ratio of 4 : l ) at -200°C in a sealed tube under vacuum; (2) heating a mixture of graphite and MHg powders (8 C for 1 MHg) under the same conditions; and (3) reaction of an MC, compound (M = K, Rb) with a stoichiometric amount of mercury in the vapor phase in a sealed tube under vacuum at 100-150°C. The first two methods are fast (several hours); the third is slower but allows intercalation into pyrographite samples. (c) Thallium and potassium or rubidium. The intercalated thallium-heavy alkalimetal alloys result in thick layers, without long range in-plane ordering. Therefore, there is no theoretical support for stoichiometric formulas. Nevertheless, the composition of the ternaries is quite reproducible. Four compounds have been d e ~ c r i b e d ~ ~ the ' ~ ~ pale ' ~ : yellow KTl,,,C, and RbTl,,,C, are of first stage, and the pink KTl,,,C, and RbTl,,,C, are second stage. Because of the high boiling point of thallium, the synthesis of these species must be carried out in the liquid or solid phases. Different techniques are used for the compounds containing potassium and rubidium. Both KTl,,, and KTll,,C, are prepared (1) by the reaction of graphite with molten alloys; MTl,,, (which is not a defined compound) leads to the first stage, KTl, ,C4, at -340°C; KTl, leads to the second stage, KTl,,,C,, at 400°C. This method can be used for pyrographite samples. Another means (2) by compressing a mixture of graphite and solid KTl,,, powder under 10-20 kilobars then heating the pellet at 340°C. The first or second stage is synthesized from initial mixtures of suitable composition. Complete intercalation of the alloy needs successive compressions and heating, leading to powders. The product RbTl,,,C, exists in two forms, CI and p; the interplanar distance is lower in the former than in the latter. (3) The molten RbTl,,, alloy intercalates into pyrographite, leading to the CI form near 340"C, to the /3 form >450°C and to a mixture between these temperatures. (4)Compressing the heating a mixture of graphite and the alloy powders leads to RbTl, ,C4 or RbTl,,,C,, depending on the ratio C to RbTl,,,. (d) Bismuth, antimony or arsenic. The three semimetals Bi, Sb and As can intercalate into graphite, alloyed with a heavy alkali metal, in the same way as mercury and thallium' ,-18. However, with these elements, only one synthesis method is possible: allowing the liquid binary alloy to react on pyrolytic graphite platelet (HOPG), varying the concentration of the alloy, the reaction temperature and the reaction time. This reaction is realized in a glass tube sealed under vacuum.

-

16.4. The Formation of Sheet Structures 16.4.2. Graphite and Boron Nitride Intercalation Compounds 16.4.2.2. Electron Donors

254

TABLE 1. ALKALI-METAL BISMUTHO-, ANTIMONO-AND ARSENIOGRAPHITIDES

K-Bi

Rb-Bi Cs-Bi Cs-Sb

CS-AS

Stage

Type

2 3

c!

4

1

5

c!

1

2

Formula

Color

KB10 6%

Blue Grey blue Grey Grey Violet Blue Grey blue Violet Blue green Blue green Grey blue Grey Grey Grey Silver blue Dark blue Grey Blue

KBio 6C12 -

2 3 1

c!

RbBi, $2, RbBi, 7C,

a

-

1 2

P P

1 2 3 1

Y 6

1

1 2 2

c!

CsBi, 53C4 CsBiC, CsBiC, CsSb, &, CsSb, J4

c!

a

P 2

1/'

P Y

csO

7SSb0 7SC8

-

CsAso 7 4 c 4 CsAsC, Cso ,AsC, CsAsC,

Reaction temperature Reaction ("C) time 450 400 500 5 50

500 420 600 450 580 600

530

580 590 600 600 590 640 620

2d 2d

2d 2d 10 d 10 d 10 d 24 h 10 d 10 d

4h

20 min lh 1 mon

30 min Id lw

36 h

Composition of the reactive alloy 40 at. % Bi 50 at. % Bi 54 at. % Bi 57 at. % Bi 40 at. % Bi 50 at. % Bi 35 at. % Bi 45 at. % Bi 45 at. % Bi 45 at. % Bi 38 at. % Bi 38 at. %Sb 50 at. % Sb 50 at. % Sb 38 at. % As 48 at. % As 40 at. % As 50 at. % As

Bismuth can i n t e r ~ a l a t e with ' ~ K, Rb and Cs. This extension to cesium, which is the more electropositive alkali metal, is probably allowed by the electronegativity of bismuth, which is a little higher than that of mercury and thallium. It seems the more electronegative is the element the easier is its intercalation into graphite with a highly electropositive alkali metal. Finally, in order to build a ternary compound, a n equilibrium between the electronegativities of carbon, alkali metal and a second metal (or semimetal) is necessary. This evolution continues since with Sb and As", which are even more electronegative than Bi, only the intercalation with Cs is possible. All the ternaries containing Bi, Sb and As, that have been synthesized and isolated are collected in Table 1. Numerous other phases have been observed, but always in admixture with other compounds, so these phases are not indicated in this table. It is important to emphasize that in every system the intercalated metallic sheets exhibit often various compositions and orderings. These are distinguished by greek letters in order of their respective thicknesses: the thicker the intercalated sheet, the later the corresponding letter in the alphabet. In all these systems, the 2D and C -axis orderings of the intercalated sheets are generally c ~ m p l i c a t e d ' ~Last, . in the same table, the best synthesis conditions are indicated for every phase. (P. LAGRANGE, A HEROLD)

1. D. Billaud, Thesis, Nancy 1973 CNRS no A09043. 2. D. Billaud, A. Herold, Bull. SOC.Chim. Fr., 103 (1972). 3. D. Billaud, D. Balesdent, A. Herold, Bull. SOC.Chim. Fr., 2402 (1974). 4. B. R. York, S. K. Hark, S. A. Solin, Phys. Rev. Lett., 50, 1470 (1983). 5. B. R. York, S. K. Hark, S. A. Solin, Synth. Met., 7, 25(1983).

16.4 The Formation of Sheet Structures 16.4.2. Graphite and Boron Nitride Intercalation Compounds 16.4.2.3. Mixed Metal-Molecule-Graphite Tenaries. ~~~~~~~~~~~

255

~

6. D. Billaud, A. Herold, F. L. Vogel, Muter. Sci.Eng. 45, 55 (1980). 7. D. Billaud, A. Herold, F. L. Vogel, Synth. Met., 3, 279 (1981). 8. D. Billaud, A. Herold, Bull. Soc. Chim. Fr., 2715 (1974). 9. M. El Makrini, Thesis, Nancy (1980). 10. M. El Makrini, P. Lagrange, D. Guerard, A. Herold, C.R. Hebd. Seances Acad. Scz. Ser. C, 288, 303 (1979). 11. P. Lagrange, M. El Makrini, D. Guerard, A. Herold, Physica. Ser. B + C , 99, 473 (1980). 12. M. El Makrini, P. Lagrange, D. Guerard, A. Herold, Carbon, 18, 211 (1980). 13. P. Lagrange, M. El Makrini, A. Herold, C.R. Hebd. Seances, Acad. Sci.Ser. C, 290,203 (1980). 14. M. El Makrini, P. Lagrange, A. Herold, Carbon, 18, 374 (1980). 15. P. Lagrange, A. Bendriss-Rerhrhaye, Carbon, 26, 283 (1988). 16. A. Essaddek, P. Lagrange, C.R. Hebd. Seances, Acad. Sci.Ser. 11, 306, 1077 (1988). 17. J. Assouik, P. Lagrange, C.R. Hebd. Seances, Acad. Sci. Ser. II, 307, 493 (1988). 18. A. Essaddek, J. Assouik, P. Lagrange, J. Muter. Res., 4, 244 (1989). 19. A. Bendriss-Rerhrhaye, P. Lagrange, F. Rousseaux, Synth. Met., 23, 89 (1988).

16.4.2.3. Mixed Metal-Molecule-Graphite Ternaries.

Ternary graphite intercalation compounds (TGIC), MS,C,, contain, besides the metal M, molecules S coordinated through their p- or n-electrons. Five methods are available for their preparation, which must be carried out in absence of air or moisture: (i) M is dissolved in S and allowed to react with graphite. The product, where M can be an alkali (Li to Rb) or an alkaline earth (Ca to Ba) with S = NH, or other amines, is then filtered, washed and dried at the lowest possible temperature. Excess solution yields a first-stage compound with ideal formula M(NH,),C,,, while excess graphite leads to second- and higher stage derivatives. With Li and Na, NH, replacing by OP[N(CH3)J3 (HMPT) yields Li(HMPT)C,, or Na(HMPT)C,, , both first stage3. (ii) The metal M, dissolved in a solution of naphthalene4, benzonitrile' or benzophenone in THF4p6 or DME', reacts with graphite, yielding, after workup as above, first-stage GIC of varying compositions, such as Na(THF),C,, or K(DME)C,,, accompanied by free graphite if the latter is in excess during the preparation. The use of the cobalt complex [L3LCo], (with L = a phosphine, such as P(CH,),, and L' = an olefin, such as C,H, dissolved in an aromatic solvent, e.g., such as benzene or toluene, leads to a TGIC containing the aromatic molecule7. (iii) A salt of M (Al, Be or Eu7 or an alkali') in an anhydrous solvent (NH, or DMSO') is electrolyzed with a graphite cathode. For the alkali metals, controlling the conditions of electrolysis allows the choice of s (the stage), and the production of K(DMSO)C,,, with s < 4, whereas a single compound, M(CH,)C,,, is obtained with s = 4 for M = A1 and s = 3 for M = Be. (iv) Part of the metal in the binary MC, is removed with simultaneous intercalation of S. This can be done by dropping the binary in liq NH, ', or an aromatic polynuclear hydrocarbong or other unsaturated molecule', THF9 or DME". In the latter, the quantity of metal removed increases with the electron affinity of the aromatic complexforming molecule in the solution'' and also depends on the stage of MC,. Thus, KC' yields the first-stage K(THF)C,, with phenanthrene, but a third-stage K(THF)C,, with perylene, both in THF". (v) Substance S (as a liquid or vapor) reacts with a second- or higher stage binary MC, to form MS,C, 13, where S may be benzene', (with M = K ; y = 1 , 2 or 3; x = 24) or THF" (M = K, y = 1 or 2, x = 24). This is the method of choice for the best defined TGIC. '3'

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc. 16.4 The Formation of Sheet Structures 16.4.2. Graphite and Boron Nitride Intercalation Compounds 16.4.2.3. Mixed Metal-Molecule-Graphite Tenaries. ~~~~~~~~~~~

255

~

6. D. Billaud, A. Herold, F. L. Vogel, Muter. Sci.Eng. 45, 55 (1980). 7. D. Billaud, A. Herold, F. L. Vogel, Synth. Met., 3, 279 (1981). 8. D. Billaud, A. Herold, Bull. Soc. Chim. Fr., 2715 (1974). 9. M. El Makrini, Thesis, Nancy (1980). 10. M. El Makrini, P. Lagrange, D. Guerard, A. Herold, C.R. Hebd. Seances Acad. Scz. Ser. C, 288, 303 (1979). 11. P. Lagrange, M. El Makrini, D. Guerard, A. Herold, Physica. Ser. B + C , 99, 473 (1980). 12. M. El Makrini, P. Lagrange, D. Guerard, A. Herold, Carbon, 18, 211 (1980). 13. P. Lagrange, M. El Makrini, A. Herold, C.R. Hebd. Seances, Acad. Sci.Ser. C, 290,203 (1980). 14. M. El Makrini, P. Lagrange, A. Herold, Carbon, 18, 374 (1980). 15. P. Lagrange, A. Bendriss-Rerhrhaye, Carbon, 26, 283 (1988). 16. A. Essaddek, P. Lagrange, C.R. Hebd. Seances, Acad. Sci.Ser. 11, 306, 1077 (1988). 17. J. Assouik, P. Lagrange, C.R. Hebd. Seances, Acad. Sci. Ser. II, 307, 493 (1988). 18. A. Essaddek, J. Assouik, P. Lagrange, J. Muter. Res., 4, 244 (1989). 19. A. Bendriss-Rerhrhaye, P. Lagrange, F. Rousseaux, Synth. Met., 23, 89 (1988).

16.4.2.3. Mixed Metal-Molecule-Graphite Ternaries.

Ternary graphite intercalation compounds (TGIC), MS,C,, contain, besides the metal M, molecules S coordinated through their p- or n-electrons. Five methods are available for their preparation, which must be carried out in absence of air or moisture: (i) M is dissolved in S and allowed to react with graphite. The product, where M can be an alkali (Li to Rb) or an alkaline earth (Ca to Ba) with S = NH, or other amines, is then filtered, washed and dried at the lowest possible temperature. Excess solution yields a first-stage compound with ideal formula M(NH,),C,,, while excess graphite leads to second- and higher stage derivatives. With Li and Na, NH, replacing by OP[N(CH3)J3 (HMPT) yields Li(HMPT)C,, or Na(HMPT)C,, , both first stage3. (ii) The metal M, dissolved in a solution of naphthalene4, benzonitrile' or benzophenone in THF4p6 or DME', reacts with graphite, yielding, after workup as above, first-stage GIC of varying compositions, such as Na(THF),C,, or K(DME)C,,, accompanied by free graphite if the latter is in excess during the preparation. The use of the cobalt complex [L3LCo], (with L = a phosphine, such as P(CH,),, and L' = an olefin, such as C,H, dissolved in an aromatic solvent, e.g., such as benzene or toluene, leads to a TGIC containing the aromatic molecule7. (iii) A salt of M (Al, Be or Eu7 or an alkali') in an anhydrous solvent (NH, or DMSO') is electrolyzed with a graphite cathode. For the alkali metals, controlling the conditions of electrolysis allows the choice of s (the stage), and the production of K(DMSO)C,,, with s < 4, whereas a single compound, M(CH,)C,,, is obtained with s = 4 for M = A1 and s = 3 for M = Be. (iv) Part of the metal in the binary MC, is removed with simultaneous intercalation of S. This can be done by dropping the binary in liq NH, ', or an aromatic polynuclear hydrocarbong or other unsaturated molecule', THF9 or DME". In the latter, the quantity of metal removed increases with the electron affinity of the aromatic complexforming molecule in the solution'' and also depends on the stage of MC,. Thus, KC' yields the first-stage K(THF)C,, with phenanthrene, but a third-stage K(THF)C,, with perylene, both in THF". (v) Substance S (as a liquid or vapor) reacts with a second- or higher stage binary MC, to form MS,C, 13, where S may be benzene', (with M = K ; y = 1 , 2 or 3; x = 24) or THF" (M = K, y = 1 or 2, x = 24). This is the method of choice for the best defined TGIC. '3'

256

16.4. The Formation of Sheet Structures 16.4.2. Graphite and Boron Nitride Intercalation Compounds 16 4.2.3. Mixed Metal-Molecule-Graphite Tenaries.

In the last two methods, the intercalation of S usually modifies the stage of the binary GIC since the metal ions, which occupy only l/s of the interlayer space (as implied by the pleated-layer model14), must separate to accommodate the molecules, S, and thus revert to a lower stage derivative. An exception is for first-stage binary GIC, in which the change of structure is an increase of the crystal parameter C163'7.Whatever the stage of the starting binary FGIC, this is well accounted for by the van der Waals dimensions of S and the need to coordinate the p or 71 electrons with the metal ion. As a result of this coordination, the chemical properties of the TGIC are toned-down versions of those of the binary MC,. The physical properties, however, are often quite different. Although the first-stage TGIC are blue or violet, second or higher stages are black. Because of the electrons donated by M, they are paramagnetic, with an EPR signal often much narrower than that of the parent binary, indicative of a decreased spin-orbit coupling even in the case of heavy M atoms'5s17.Also, coordination of the Mf ion enhances the electrical conductivity15,but this may be adversely affected by a rotation of the organic molecule at higher temperatures, as shown by NMR' 8,19, and by the capture of some of the charge carriers by the organic m o l e c ~ l e ' ~ . Even if the first intercalation of S into the binary is not accompanied by immediate chemical changes, these may ultimately occur, especially at higher temperatures. Thus, benzene, which will not intercalate by itself in KC,, forms a quaternary GIC in the presence of THF18; attempts to expel the benzene show that it is transformed into biphenyl and partly hydrogenated derivatives. Even THF, believed to give reversible inter~alation'~, is modified when the TGIC is heatedz0.

(A HEROLD, R SETTON)

1. W. Riidorff, E. Schulze, Angew. Chem., 66, 305 (1954). 2. W. Riidorff, E. Schulze, D. Rubisch, Z. Anorg. Allg. Chem., 282, 232 (1955).

3. 4. 5. 6. 7.

8. 9. 10. 11. 12. 13. 14.

15. 16. 17. 18. 19. 20. 21. 22. 23.

D. Ginderow, R. Setton, Carbon, 6, 81 (1968). M. Nomine, L. Bonnetain, C. R. Hebd. Seances Acad. Sci., Ser. C, 264, 2084 (1967). D. Ginderow, R. Setton, C. R. Hebd. Seances Acad. Sci., Ser. C, 270, 153 (1970). D. Ginderow, R. Setton, C. R. Hebd. Seances Acad. Sci., Ser. C, 266, 1515 (1968). J. 0. Besenhard, I. Kain, H. F. Klein, H. Mourald, H. Witt, Mater. Rex. Soc., Fall Mtg. Proc., 1983, p. 221. W. Ostertag, Diplomasarbeit, Tiibingen, 1956. J. 0. Besenhard, Carbon, 14, 111 (1976). F. Beguin, R. Setton, Carbon, 13, 273 (1975). I. B. Rashkov, I. M. Panayotov, V. C. Shishkova, Carbon, 17, 103 (1979). I. B. Rashkov, V. Shishkova, I. M. Panayotov, G. Merle, J. M. Letoffe, 3rdInternational Carbon Conference, Baden-Baden, 1980, Preprints p. 96. F. Beguin, Ph.D. Thesis Orleans, 1980. A. Hamwi, P. Touzain, L. Bonnetain, 2nd International Carbon Conference, Baden-Baden, 1976, Preprints, p. 143. N. Daumas, A. Herold, C. R. Hebd. Seances Acad. Sci., Ser. C, 268, 273 (1969). F. Beguin, R. Setton, J. Chem. Soc., Chem. Commun., 611 (1976). J. Jegoudez, C. Mazitres, R. Setton, Carbon, 24, 747 (1986). L. Facchini, M. F. Quinton, A. P Legrand, F. Beguin, R. Setton, Physzca, 9 9 4 525 (1980). F. BCguin, R. Setton, L. Facchini, A. P. Legrand, G. Merle, C. Mai, Synth. Met., 2, 161 (1980). J. Amiell, P. Delhaes, F. Beguin, R. Setton, Mater. Sci. Eng., 31, 243 (1977). S. Malsuyaki, M. Tanaguchi, M. Sano, Synth. M e t . , 16, 343 (1986). F. Beguin, R. Setton, A. Hamwi, P. Touzain, Mater. Scz. Eng., 40, 187 (1979). F. Beguin, J. Jegoudez, C. Mayieres, R. Setton, C.R. Hebd. Seances Acad. Sci., 293, 969 (1986).

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc. 16.4.2. Graphite and Boron Nitride Intercalation Compounds 16.4.2.4. Alkali-Metal-Hydrogen-Graphite Ternaries. 16.4.2.4.2. Intercalation of Alkali-Metal Hydrides into Graphite.

257

16.4.2.4. Alkali-Metal-Hydrogen-Graphite Ternaries. 16.4.2.4.1. Hydrogenation of the binary phases MC, and MC,, (M = K, Rb, Cs).

Hydrogen reacts with the yellow-bronze first-stage MC, compounds (M = K, Rb), leading to MC,H, ternaries, with 0 < x < 0.67. For x < 0.1, the initial structure and stage remain the same. For x > 0.6, a new homogeneous phase is formed of ideal formula KC,II,/3. For 0.1 < x < 0.6, the two phases The blue KC8H2,, belongs to the second stage3. The KC, slowly absorbs hydrogen (or deuterium) under 1 atm at RT, but it is preferable to heat at 120°C. For MC, compounds, prepared from natural powdered graphite, the value x = 0.6 is attained in several days and the limit x = 0.67, corresponding to the formula KC,H2/3, after two weeks. Samples prepared from single crystals or from pyrographite cannot be hydrogenated.

-

KC,

+ 2 H, X

-

KCSH,

(a)

-

is reversible; between x = 0.15 and 0.6 the dissociation pressure attains 1 atm for T 300°C. The pressure of KC,D, is more than twice that of KC8H,1’2. The RbC, ternary absorbs small amounts of hydrogen under 1 atm, but ternaries RbC,H, can be synthesized under pressure. Blue homogeneous RbC8H,,3 is obtained under 140 atm, but can be stored under vacuum at RT’. The KC,, ternary also forms hydrogen or deuterium ternaries, KC,,H, or KC,,D, (the limiting values are x = 0.36, y = 0.3 under 1 atm at 60°C; x = 0.55 under 13 atm)’. These ternaries consist of a mixture of KC,H,,, or KC,D,,, and higher stage binaries, MC12n3. The binaries CsC,, RbC,, and CsC,, do not react at all with hydrogen, even at pressures of 200 atm.

-

(D. GUERARD, A HEROLD)

1. D. Saehr, A. Herold, Bull. Soc. Chim. Fr., 3130 (1965) 2. M. Colin, A. Herold, Bull. Soc. Chim., Fr., 90 (1971). 3. P. Lagrange, A. Herold, Carbon, 16, 235 (1978). 16.4.2.4.2 Intercalation of Alkali-Metal Hydrides into Graphite.

The general synthesis method consists in heating a sample of graphite immersed in a large excess of a powder of the hydride. The reaction occurs generally in two steps: 1. fast intercalation (a few minutes to a few hours) of the alkali metal alone, which leads to a binary first-stage (yellow) or a second-stage (steel-blue) compound 2. Slower cointercalation (a few hours to two weeks) of both elements, which leads to the ternary compound’. The change of colors can be used to follow the progress of the reactions. The values corresponding to each system are given in the Table 1.There are two exceptions with this process. The first is sodium hydride, which intercalates directly’. This accounts for a H:Na ratio close to 1 and for a very weak intercalation of sodium into pure graphite6. Second is lithium hydride, which is unable to intercalate by itself and needs the presence

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc. 16.4.2. Graphite and Boron Nitride Intercalation Compounds 16.4.2.4. Alkali-Metal-Hydrogen-Graphite Ternaries. 16.4.2.4.2. Intercalation of Alkali-Metal Hydrides into Graphite.

257

16.4.2.4. Alkali-Metal-Hydrogen-Graphite Ternaries. 16.4.2.4.1. Hydrogenation of the binary phases MC, and MC,, (M = K, Rb, Cs).

Hydrogen reacts with the yellow-bronze first-stage MC, compounds (M = K, Rb), leading to MC,H, ternaries, with 0 < x < 0.67. For x < 0.1, the initial structure and stage remain the same. For x > 0.6, a new homogeneous phase is formed of ideal formula KC,II,/3. For 0.1 < x < 0.6, the two phases The blue KC8H2,, belongs to the second stage3. The KC, slowly absorbs hydrogen (or deuterium) under 1 atm at RT, but it is preferable to heat at 120°C. For MC, compounds, prepared from natural powdered graphite, the value x = 0.6 is attained in several days and the limit x = 0.67, corresponding to the formula KC,H2/3, after two weeks. Samples prepared from single crystals or from pyrographite cannot be hydrogenated.

-

KC,

+ 2 H, X

-

KCSH,

(a)

-

is reversible; between x = 0.15 and 0.6 the dissociation pressure attains 1 atm for T 300°C. The pressure of KC,D, is more than twice that of KC8H,1’2. The RbC, ternary absorbs small amounts of hydrogen under 1 atm, but ternaries RbC,H, can be synthesized under pressure. Blue homogeneous RbC8H,,3 is obtained under 140 atm, but can be stored under vacuum at RT’. The KC,, ternary also forms hydrogen or deuterium ternaries, KC,,H, or KC,,D, (the limiting values are x = 0.36, y = 0.3 under 1 atm at 60°C; x = 0.55 under 13 atm)’. These ternaries consist of a mixture of KC,H,,, or KC,D,,, and higher stage binaries, MC12n3. The binaries CsC,, RbC,, and CsC,, do not react at all with hydrogen, even at pressures of 200 atm.

-

(D. GUERARD, A HEROLD)

1. D. Saehr, A. Herold, Bull. Soc. Chim. Fr., 3130 (1965) 2. M. Colin, A. Herold, Bull. Soc. Chim., Fr., 90 (1971). 3. P. Lagrange, A. Herold, Carbon, 16, 235 (1978). 16.4.2.4.2 Intercalation of Alkali-Metal Hydrides into Graphite.

The general synthesis method consists in heating a sample of graphite immersed in a large excess of a powder of the hydride. The reaction occurs generally in two steps: 1. fast intercalation (a few minutes to a few hours) of the alkali metal alone, which leads to a binary first-stage (yellow) or a second-stage (steel-blue) compound 2. Slower cointercalation (a few hours to two weeks) of both elements, which leads to the ternary compound’. The change of colors can be used to follow the progress of the reactions. The values corresponding to each system are given in the Table 1.There are two exceptions with this process. The first is sodium hydride, which intercalates directly’. This accounts for a H:Na ratio close to 1 and for a very weak intercalation of sodium into pure graphite6. Second is lithium hydride, which is unable to intercalate by itself and needs the presence

16.4. The Formation of Sheet Structures 16.4.2. Graphite and Boron Nitride Intercalation Compounds 16.4.2.4. Alkali-Metal-Hydrogen-Graphite Ternaries.

258

TABLE1. TERNARY AND QUATERNARY OF ALKALI-METALHYDRIDES WITH GRAPHITE'

Temperature ("C)

Length of reaction (days)

Stage,

Intercalate

S

Formula

NaH

400-500

1-8

1-11

KH

400 300 450 450 450 450 430-500

1 2-3 2" 15 8 8 1-4

1 2 1% l?

NaHC, 5 s to NaH4,5, KHO.SC4

RbH

CsH LiH + K a

2P 1

1-4

KH,

&ti

RbH0.8C3.6 RbH0.8C3,6

RbHO8CSl CsHiJ.& K,LiHC

under 100 x lo5 Pa of H,.

of 25-40% in weight of K metal4. In this case, the first step is the intercalation of K alone, which leads to KC,. During the second step, there is cointercalation of the three elements; K, Li and H, leading to the quaternary phases K'LiHC,, (s = 1-4). The formulas of these compounds show a large variation from NaHC,,,, to NaHC,.,,, s being the stage. This variation is probably due to inclusions of free hydride between the graphene layers, as shown in most of the x-ray diagrams. The color varies from dark blue (mixture of stages 1 and 2 ) to steel blue (stage 2) and grey (higher stages)'. (i) Potassium hydride. The composition of the phases obtained by reacting KH with graphite is slightly different from that of the compound obtained by hydrogenation of KC,: the H:K ratio is 0.8 instead of 2:3. This is probably due to an incomplete hydrogenation of KC,: the sorption of hydrogen slows down as the ratio reaches 2 : 3 and is probably not achieved, even after a few weeks of reaction. The first-stage compound is violet, the second is steel blue. (ii) Rubidium and Cesium hydride. These are not commercialized and must be prepared in the laboratory'. Rubidium hydride compounds exhibit three different interplanar distances (di), called x , p and y as the value of di increases. This is a phenomenon often observed in the ternary compounds of graphite with metal alloys6. (D. GUERARD, A HEROLD)

1. D. Guerard, P. Lagrange, Chemical Physics oflntercalation A. P. Legrand, S . Flandrois eds., Nato AS1 Series 172 B, 1988 p. 341. 2. D. Guerard, N. E. Elalem, C. Takoudjou, F. Rousseaux, Synth. Met., 12, 195 (1985). 3. A. Metrot, D. Guerard, D. Billaud, A. Herold, Synth. Met. 363, l(1980). 4. N. E. Elalem, L. Elansari, A. Oufkir, D. Guerard, Proc. ConJ Int. Composes Lamelluzres, D. Guerard, P Lagrange, eds., Pont-a-Mousson, 1988 p. 71. 5. D. Guerard, N. E. Elalem, S. E. Hadigui, L. Elansari, F. Rousseaux, H. Eslrade-Szwarckopf, J. Conard, P. Lauginie, J. Less-Common Met., 131, 173 (1987). 6. P. Lagrange, A. Bendriss-Rerhrhaye, Carbon, 26, 283 (1988).

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc. 16.4. The Formation of Sheet Structures 16.4.2. Graphite and Boron Nitride Intercalation Compounds 16.4.2.5. Graphite Oxides

259

16.4.2.4.3. Intercalation of Quaternary Compounds with Hydrogen and Two Alkali Metals into Graphite. (i) Lithium hydride. Lithium hydride unable to intercalate by itself, reacts when mixed with 25-40% by wt of K metal. The compounds of stages I-IV have the general formula K,LiHC,, (ii) Biintercalation Compounds. The second-stage ternary compounds with hydrogen and a heavy alkali metal react with calculated amounts of a heavy alkali metal when’ the second stage compound is in powder form and the temperature is just above the melting point of the heavy alkali metal. The reaction is

’.

where M = K, Rb; M’ = K, Rb or Cs. The intervals between graphene layers are occupied alternately by three-layered planes, M-H-M ’, and single layers of M. (D GUERARD, A. HEROLD)

1. N. E. Elalem, L. Elansari, A. Oufkir, D. Guerard, Proc. Con$ Int. ComposPe Lamellaires,

D. Guerard, P. Lagrange, eds., Pont-A-Mousson, 1988, p. 71. 2. P. Lagrange, A. Metrot, A. Herold, C.R. Acad. Sci. Paris, 278C, 701 (1974). 3. A. Bendriss-Rerhrhaye, P. Lagrange, F. Rousseaux, Synth. Met., 23, 89 (1988).

16.4.2.4.4. Physiosorption of Hydrogen and Deuterium in the Second-Stage Compounds MC,,(M = K, Rb or Cs).

When a binary second-stage compound MC,, is cooled down to 7 7 K , a rapid sorption of H, is observed. The combination is stable if the temperature remains below 100 K:

When the metal is K, the amount of stored hydrogen is around 1.2% by wt, less than in the LaNi, alloy, but the reaction is completely reversible in the first case. There is an isotopic effect in this reaction that can be used to separate hydrogen and deuterium. One interesting aspect in this case is that there is no dissociation of the molecule and the two phases are only H, and D,

’.

(D GUERARD, A. HEROLD)

1. P. Lagrange, A. Metrot, A. Herold, C.R. Acad. Sc. Paris. 275C, 765 (1972). 2. P. Lagrange, D. Guerard, J. F. Mareche, A. Herold, J. Less-Common Met. 131, 371 (1987).

16.4.2.5. Graphite Oxides

The literature up to 1966 is covered by ref. 1. For surface oxides of graphite see 516.4.2.7. (J 0. BESENHARD)

1. Gmelzns Handbuch der Anorganischen Chemze,Vol. 14, Part B-3, Verlag Chemie, Weinheim 1968. Review and data collection.

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc. 16.4. The Formation of Sheet Structures 16.4.2. Graphite and Boron Nitride Intercalation Compounds 16.4.2.5. Graphite Oxides

259

16.4.2.4.3. Intercalation of Quaternary Compounds with Hydrogen and Two Alkali Metals into Graphite. (i) Lithium hydride. Lithium hydride unable to intercalate by itself, reacts when mixed with 25-40% by wt of K metal. The compounds of stages I-IV have the general formula K,LiHC,, (ii) Biintercalation Compounds. The second-stage ternary compounds with hydrogen and a heavy alkali metal react with calculated amounts of a heavy alkali metal when’ the second stage compound is in powder form and the temperature is just above the melting point of the heavy alkali metal. The reaction is

’.

where M = K, Rb; M’ = K, Rb or Cs. The intervals between graphene layers are occupied alternately by three-layered planes, M-H-M ’, and single layers of M. (D GUERARD, A. HEROLD)

1. N. E. Elalem, L. Elansari, A. Oufkir, D. Guerard, Proc. Con$ Int. ComposPe Lamellaires,

D. Guerard, P. Lagrange, eds., Pont-A-Mousson, 1988, p. 71. 2. P. Lagrange, A. Metrot, A. Herold, C.R. Acad. Sci. Paris, 278C, 701 (1974). 3. A. Bendriss-Rerhrhaye, P. Lagrange, F. Rousseaux, Synth. Met., 23, 89 (1988).

16.4.2.4.4. Physiosorption of Hydrogen and Deuterium in the Second-Stage Compounds MC,,(M = K, Rb or Cs).

When a binary second-stage compound MC,, is cooled down to 7 7 K , a rapid sorption of H, is observed. The combination is stable if the temperature remains below 100 K:

When the metal is K, the amount of stored hydrogen is around 1.2% by wt, less than in the LaNi, alloy, but the reaction is completely reversible in the first case. There is an isotopic effect in this reaction that can be used to separate hydrogen and deuterium. One interesting aspect in this case is that there is no dissociation of the molecule and the two phases are only H, and D,

’.

(D GUERARD, A. HEROLD)

1. P. Lagrange, A. Metrot, A. Herold, C.R. Acad. Sc. Paris. 275C, 765 (1972). 2. P. Lagrange, D. Guerard, J. F. Mareche, A. Herold, J. Less-Common Met. 131, 371 (1987).

16.4.2.5. Graphite Oxides

The literature up to 1966 is covered by ref. 1. For surface oxides of graphite see 516.4.2.7. (J 0. BESENHARD)

1. Gmelzns Handbuch der Anorganischen Chemze,Vol. 14, Part B-3, Verlag Chemie, Weinheim 1968. Review and data collection.

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc. 16.4. The Formation of Sheet Structures 16.4.2. Graphite and Boron Nitride Intercalation Compounds 16.4.2.5. Graphite Oxides

259

16.4.2.4.3. Intercalation of Quaternary Compounds with Hydrogen and Two Alkali Metals into Graphite. (i) Lithium hydride. Lithium hydride unable to intercalate by itself, reacts when mixed with 25-40% by wt of K metal. The compounds of stages I-IV have the general formula K,LiHC,, (ii) Biintercalation Compounds. The second-stage ternary compounds with hydrogen and a heavy alkali metal react with calculated amounts of a heavy alkali metal when’ the second stage compound is in powder form and the temperature is just above the melting point of the heavy alkali metal. The reaction is

’.

where M = K, Rb; M’ = K, Rb or Cs. The intervals between graphene layers are occupied alternately by three-layered planes, M-H-M ’, and single layers of M. (D GUERARD, A. HEROLD)

1. N. E. Elalem, L. Elansari, A. Oufkir, D. Guerard, Proc. Con$ Int. ComposPe Lamellaires,

D. Guerard, P. Lagrange, eds., Pont-A-Mousson, 1988, p. 71. 2. P. Lagrange, A. Metrot, A. Herold, C.R. Acad. Sci. Paris, 278C, 701 (1974). 3. A. Bendriss-Rerhrhaye, P. Lagrange, F. Rousseaux, Synth. Met., 23, 89 (1988).

16.4.2.4.4. Physiosorption of Hydrogen and Deuterium in the Second-Stage Compounds MC,,(M = K, Rb or Cs).

When a binary second-stage compound MC,, is cooled down to 7 7 K , a rapid sorption of H, is observed. The combination is stable if the temperature remains below 100 K:

When the metal is K, the amount of stored hydrogen is around 1.2% by wt, less than in the LaNi, alloy, but the reaction is completely reversible in the first case. There is an isotopic effect in this reaction that can be used to separate hydrogen and deuterium. One interesting aspect in this case is that there is no dissociation of the molecule and the two phases are only H, and D,

’.

(D GUERARD, A. HEROLD)

1. P. Lagrange, A. Metrot, A. Herold, C.R. Acad. Sc. Paris. 275C, 765 (1972). 2. P. Lagrange, D. Guerard, J. F. Mareche, A. Herold, J. Less-Common Met. 131, 371 (1987).

16.4.2.5. Graphite Oxides

The literature up to 1966 is covered by ref. 1. For surface oxides of graphite see 516.4.2.7. (J 0. BESENHARD)

1. Gmelzns Handbuch der Anorganischen Chemze,Vol. 14, Part B-3, Verlag Chemie, Weinheim 1968. Review and data collection.

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc. 260

16.4. The Formation of Sheet Structures 16.4.2. Graphite and Boron Nitride Intercalation Compounds 16.4.2.5. Graphite Oxides

16.4.2.5.1. Definition and Nomenclature of Graphite Oxides.

Graphite oxide (GO) is a collective name for covalent layered C-0-H compounds made by oxidation of graphite in strong acids. First mentioned in 1859', the nomenclature varies-graphitic acid, graphite oxyhydroxide, carbon oxide and pyrographitic acid are outdated synonyms for GO. GO cannot be obtained as a reproducible substance of fixed stoichiometry. The composition of GO preparations depends on the pristine graphite and the conditions of oxidation. The degree of oxidation of GO is commonly characterized by the carbon : oxygen ratio, after taking into account the hydrogen content, i.e., by x in C,O. y H'O. There is controversy over the kind and position of the carbon-oxygen bonds in GO. Besides hydroxyl groups, either ether linkages3 or keto groups4 on puckered layers of carbon hexagons are possible, and keto-enol tautomerism allows the carbon layers to remain planar5s6. Carbon-oxygen IR bands of GO may be assigned to any of these

Figure 1. Schematic drawing of a GO structural model, on the left: plane view, on the right: profile (after 5).

16.4.2. Graphite and Boron Nitride Intercalation Compounds 16.4.2.5. Graphite Oxides 16.4.2.5.2. Preparation of Graphite Oxides.

261

p o ~ s i b i l i t i e sAdditional ~~~. carboxyl groups are situated at the plane edges of the carbon layers. Unfortunately, the structure of GO cannot be determined unequivocally by x-ray or by electron diffraction because only hkO and 001 but no hkl reflections are observed. The interplanar distance of vacuum-dried GO is in the range 590-670 pm5, and the C-C distance in the carbon planes is only slightly higher than in graphite. Thus, structural models for GO are based mainly on its reactivity and physical properties. GO has oxidizing and acidic character; reactions with reductants and bases of different strength reveal that neither the oxidizing nor the acidic functions are always equivalent. Pure GO preparations are light colored and nonconducting. Exposure to light or warming at 80°C causes darkening, but no significant change in composition4. A model that agrees with these properties corresponds to the ideal stoichiometry, C,O,(OH),, and has puckered carbon layers with C-C single and double bonds, hydroxyl groups and carbonyl groups in different surroundings as structural elements (Fig. l)5. (J.O. BESENHARD)

1. Gmelins Handbuch der Anerganischen Chemie, Vol. 14, B-3, Verlag Chemie, Weinheim 1968.

2. 3. 4. 5. 6. 7.

Review and data collection. B. C. Brodie, Phil. Trans. Roy. SOC.London, 149, 249 (1859). G. Ruess, Monatsh. Chem., 76, 381 (1947). A. Glauss, R. Plass, H. P. Boehm, U. Hofmann, 2. Anorg. Allg. Chem., 291, 205 (1957). W. Scholz, H. P. Boehm, Z . Anorg. Allg. Chem., 369, 327 (1969). K. E. Carr, Carbon, 8, 245 (1970). R. J. Lagow, R. B. Badachape, J. L. Wood, 3. L. Margrave, J. Chem. SOC.,Dalton Trans., 1268 (1974).

16.4.2.5.2. Preparation of Graphite Oxides.

There are three classical methods of preparing for GO1, all based on further oxidation of acid salts of graphite' (cf. §16.4.2.1.4), which yield products of varied composition and properties. Fuming nitric acid is added to a mixture of graphite and K[ClO,] which is then kept at 60°C for 4 d. The product is washed with a large volume of water. After drying, the oxidation procedure is repeated until yellowish GO is obtained3. Caution: Addition of fuming HNO, to graphite KClO, may be hazardous. In a modified preparation' fuming HNO, is added slowly (during 5 h) to the cooled (ice-salt bath) mixture. Alternatively, K[ClO,] is added slowly at RT to a suspension of graphite in conc HJO, HNO,. The reaction product is washed with water and dried4$5.Caution: Fast addition of K[CIO,] may cause explosions by decomposition of ClO,, which is formed from K[CIO,] in the presence of conc H,SO,. It is recommended that only small lots of GO be prepared by this procedure, and to spread the addition of chlorate over 10 d'. Just as in the above preparation, completeness of reaction decreases with increasing particle size-coarse-flaked graphite requires up to three oxidation processes. Finally, graphite is stirred into a mixture of Na[NO,] and conc H , S 0 4 which is cooled in an ice bath. Thereupon, KCMnO,] is added to the suspension6. Caution: During addition of KCMnO,] the temperature should not exceed 20°C to avoid explosive autodecomposition of Mn,O,. The reaction mixture is kept at 35°C for 30 min, then

+

+

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc. 16.4.2. Graphite and Boron Nitride Intercalation Compounds 16.4.2.5. Graphite Oxides 16.4.2.5.2. Preparation of Graphite Oxides.

261

p o ~ s i b i l i t i e sAdditional ~~~. carboxyl groups are situated at the plane edges of the carbon layers. Unfortunately, the structure of GO cannot be determined unequivocally by x-ray or by electron diffraction because only hkO and 001 but no hkl reflections are observed. The interplanar distance of vacuum-dried GO is in the range 590-670 pm5, and the C-C distance in the carbon planes is only slightly higher than in graphite. Thus, structural models for GO are based mainly on its reactivity and physical properties. GO has oxidizing and acidic character; reactions with reductants and bases of different strength reveal that neither the oxidizing nor the acidic functions are always equivalent. Pure GO preparations are light colored and nonconducting. Exposure to light or warming at 80°C causes darkening, but no significant change in composition4. A model that agrees with these properties corresponds to the ideal stoichiometry, C,O,(OH),, and has puckered carbon layers with C-C single and double bonds, hydroxyl groups and carbonyl groups in different surroundings as structural elements (Fig. l)5. (J.O. BESENHARD)

1. Gmelins Handbuch der Anerganischen Chemie, Vol. 14, B-3, Verlag Chemie, Weinheim 1968.

2. 3. 4. 5. 6. 7.

Review and data collection. B. C. Brodie, Phil. Trans. Roy. SOC.London, 149, 249 (1859). G. Ruess, Monatsh. Chem., 76, 381 (1947). A. Glauss, R. Plass, H. P. Boehm, U. Hofmann, 2. Anorg. Allg. Chem., 291, 205 (1957). W. Scholz, H. P. Boehm, Z . Anorg. Allg. Chem., 369, 327 (1969). K. E. Carr, Carbon, 8, 245 (1970). R. J. Lagow, R. B. Badachape, J. L. Wood, 3. L. Margrave, J. Chem. SOC.,Dalton Trans., 1268 (1974).

16.4.2.5.2. Preparation of Graphite Oxides.

There are three classical methods of preparing for GO1, all based on further oxidation of acid salts of graphite' (cf. §16.4.2.1.4), which yield products of varied composition and properties. Fuming nitric acid is added to a mixture of graphite and K[ClO,] which is then kept at 60°C for 4 d. The product is washed with a large volume of water. After drying, the oxidation procedure is repeated until yellowish GO is obtained3. Caution: Addition of fuming HNO, to graphite KClO, may be hazardous. In a modified preparation' fuming HNO, is added slowly (during 5 h) to the cooled (ice-salt bath) mixture. Alternatively, K[ClO,] is added slowly at RT to a suspension of graphite in conc HJO, HNO,. The reaction product is washed with water and dried4$5.Caution: Fast addition of K[CIO,] may cause explosions by decomposition of ClO,, which is formed from K[CIO,] in the presence of conc H,SO,. It is recommended that only small lots of GO be prepared by this procedure, and to spread the addition of chlorate over 10 d'. Just as in the above preparation, completeness of reaction decreases with increasing particle size-coarse-flaked graphite requires up to three oxidation processes. Finally, graphite is stirred into a mixture of Na[NO,] and conc H , S 0 4 which is cooled in an ice bath. Thereupon, KCMnO,] is added to the suspension6. Caution: During addition of KCMnO,] the temperature should not exceed 20°C to avoid explosive autodecomposition of Mn,O,. The reaction mixture is kept at 35°C for 30 min, then

+

+

262

16.4. The Formation of Sheet Structures 16.4.2. Graphite and Boron Nitride Intercalation Compounds 16.4.2.5. Graphite Oxides

TABLE1. COMPOSITION AND PROPERTIES OF GO DEPENDING ON THE METHOD OF PREPARATION

Method of preparation

Oxidation by K[C10,] + HNO,

Interplanar distance (pm) C: 0 ratio (x in C,O y H,O) -OH acidic (mequiv g- ') -OH total (mequiv g-') H total (mequiv g-') C=O (mequiv g-') Content of impurities Surface area

592-598 3.1-4.1 -

13.3-23.3 27.5

Low Low

Oxidation by K[C10,] + HNO,

Oxidation by NaNO, +

%504

+ H2S04

+ K[Mn04]

630-635 2.7-3 5.5-5.9 10.4 14.3-19.6 32.4-33

660-570 2.6-3.8 5.6 12.6 14.4-18.8 27-28.5

Medium High

Ref.

High Medium

washed with a large volume of warm water and dried over P,O,, in vacuo. Residual [MnO,]- and MnO, may be reduced with H,O,. Improvements in these classical preparations are listed in Table 1. GO with low contamination can be obtained by anodic oxidation of graphite in 70% HClO, and other strong acids containing some water2s839.The concentration of water is critical; too little results in incomplete GO formation, which may be understood as a hydrolysis of graphite salts: C;A-

+ H,O

C,-OH

+ HA

@)lo

In too dilute acids the dominant reaction at graphite anodes is 0, evolution. Anodically prepared GO is distinguished from chemically prepared GO by a significantly lower degree of oxidation and by electronic conductivity comparable to that of the pristine graphite''. The conducting GO can be used for positive electrodes in aqueous8 and nonaqueous batteriesl2.l3. The C:O ratio of washed and vacuum-dried samples is -4.614; the interplanar distance is in the range 610-630pm. A (C,F),-like structure (516.4.2.6.2) has been suggested15 for graphite oxides with a C:O ratio of 4. (J 0 BESENHARD)

1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15.

H. P. Boehm, W. Scholz, Justus Liebigs Ann. Chem., 691, 1 (1965). H. P. Boehm, M. Eckel, W. Scholz, Z. Anorg. Allg. Chem., 353, 236 (1967). B. C. Brodie, Justus Liebigs Ann. Chem., 114, 6 (1860). L. Staudenmaier, Ber. Dtsch. Chem. Ges., 31, 1481 (1898). V. Kohlschutter, P. Haenni, Z. Anorg. Allg. Chem., 105, 121 (1919). W. S. Hummers, R. E. Offeman, J. Am. Chem. SOC., 80, 1339 (1958). W. Scholz, H. P. Boehm, 2. Anorg. Allg. Chem., 369, 327 (1969). B K. Brown, 0. W. Storey, Trans. Am. Electrochem. Soc., 53, 129 (1928). J. 0. Besenhard, H. P. Fritz, Z . Anorg. ANg. Chem., 416, 106 (1975). M. J. Bottomley, G. S. Parry, A. R. Ubbelohde, D. A. Young, J. Chem. Soc., 5674 (1963). J. 0. Besenhard, H. Mohwald, J. J. Nickl, Synth. Met., 3, 187 (1981). J. 0. Basenhard, E. Theodoridou, H. Mohurald, J. J. Nickl, Synth. Met., 4, 211 (1982). Ph Tougain, R. Yagami, J. Power Sources, 14, 99 (1985). J. 0. Besenhard, Z. Naturforsch., Teil B, 32, 1210 (1977). T. Nakajima, in International Colloquium on Layered Compounds, D. Guerand, P. Lagrange, eds., Universitk de Nancy I, Nancy, 1988, p. 125.

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc. 16.4.2. Graphite and Boron Nitride Intercalation Compounds 16.4.2.5. Graphite Oxides 16.4.2.5.3. Reactivity of Graphite Oxides

263

16.4.2.5.3. Reactivity of Graphite Oxides.

Reactions of GO may be classified as (i) reactions of specific groups and (ii) intercalation reactions. (i) Etherification and Esterification of -OH and -COOH groups. The acidic -OH groups in GO can be methylated (with diazomethane) and acetylated (with acetic anhydride). These reactions cause the interlayer distance to increase by 300-400 pm’. Esterification and chlorination of -COOH groups at plane edges does not change the interlayer distance. (ii) Fluorination. Fluorine in the presence of H F in a nickel tube at 4 bar (4 x lo’ Pa) raises the C:O ratio from 2.2 to 4.2-4.4, with the fluorine content in the range 16.25-22.4%. These oxofluorides exhibit an IR carbon fluorine stretch at 1095 cm- indicating the formation of C-F bonds’. An oxofluoride with higher fluorine content is prepared by fluorination of GO with K,[MnF,] in fluorosulfuric acid3. It may be represented by the formula C,OF,. From IR evidence and the fact that C,0F3 is not wetted by water, fluorine is bound to carbon. (iii) Cation-Exchange Reactions. Protons in GO can be exchanged for other cations. The -COOH groups undergo cation exchange reactions (e.g., with Ca’+) in neutral media. The acidity of -OH groups is low, and strongly basic media are required for cation exchange4. Cation exchange is accompanied by solvent intercalation between the GO layers, and consequently by an increase in the interplanar distance which depends on the kind of solvent and cation and on the cation concentration. For a range of low concentrations of cations, i.e., low charge density and low electrostatic interaction between cations and negatively charged GO layers, the interplanar distance may become infinite (=sol formation)’. On both sides of this concentration range, the interplanar distances are finite. (iv) Solvates and Adducts. Polar agents are intercalated into GO and cause the interplanar distance to increase (see Table 1). Single- and multiple-layer solvation of GO as well as adduct formation is possible. The increase in interplanar distance ranges from a few pm for ammonia up to more than 4000 pm for long-chain amines and alcohol^^,^. Agents with low dielectric constants cause only little intracrystalline swelling of GO. Classification of molecular intercalates as solvates and adducts is arbitrary. Provided that the intercalants interact with specific groups of GO, the compounds are regarded as adducts. The final concentration of intercalants and the increase in interplanar distance is low in adducts. Typical adducts are formed between NH, and acidic groups of GO. Nonspecific, physical adsorption of the intercalants leads to solvates of GO. Typical solvates are formed with dioxane or sulfoxides. Frequently, the uptake of intercalants occurs in two stages: the formation of adducts is followed by solvation. (v) Thermal decomposition. Rapid heating to 150-200°C causes exothermic deflagration of GO. If the temperature is increased slowly, the thermal decomposition yields H’O, CO,, 0, and only little C O 14. (vi) Reduction. Reduction of GO by various agents never yields graphite or carbon as the final product”. By electrochemical oxidation-reduction cycles, the GO formation can be reversed with more than 90% coulomb efficiency. The difference from 100% is attributed to irreversible oxidation of carbon at defect sites’,. (vii) Reaction with H,S. Up to 90% of the oxygen in GO can be replaced by sulfur if an aqueous suspension of GO is allowed to stand for 2 wk at 50°C under 100 torr x

’,

264

16.4. The Formation of Sheet Structures 16.4.2. Graphite and Boron Nitride Intercalation Compounds 16.4.2.5. Graphite Oxides TABLE2. ONE-DIMENSIONAL TNTRA-CRYSTAL LINE SWELLING OF GO IN VARIOUS AGENTS Interplanar distance

Om)

1160 1240 932

840

910 1450 1413 1535 655 725 1260 708 1260 714

pressure of H,S

Agent

Ref.

Water 0.05 N aq NaOH Ethanol Diethyl ether Dioxane (single layer) Dioxane (double layer) Dipropyl sulfoxide Dibutyl sulfoxide Ammonia (adduct) Ammonia (solvate) Acetic acid Nitric acid (adduct) Nitric acid (solvate) Nitrogen dioxide

5 5 8 9 10 10 11 11 12 12 9 13 9 13

The resulting graphite sulfide is a n ill-defined sulfur analog of GO. and the dehydration of cyclohexanol'* are catalyzed by GO. Ion exchange with noble metal cations gives access to carbon-supported noble metal catalyst^'^^^^. 15.

(viii) Catalytic properties. The decomposition of H 2 0 2

(J 0.BESENHARD)

1. W. Rudorff, Ado. Inorg. Chem. Radiochem., 1, 223 (1959). 2. R. J. Lagow, R. B. Badachape, J. L. Wood, J. L. Margrave, J. Chem. Soc., Dalton Trans., 1268 (1974). 3. A. S. Nazarov, N. F. Yudanov, I. I. Yakovlev, Russ. J. Inorg. Chem. (Engl. Transl.), 22, 1128 (1977). 4. W. Scholz, H. P. Boehm, Z. Anorg. Allg. Chem., 369, 321 (1969). 5. A. Clauss, R. Plass, H. P. Boehm, U. Hofmann, 2.Anorg. Allg. Chem., 291, 205 (1957). 6. F. Aragon de la Cruz, J. Can0 Ruiz, D. M. C. MacEwan, Nature (London), 183, 740 (1959). 7. A. Ramirez Garcia, J. Can0 Ruiz, D. M. C. MacEwan, Nature (London), 203, 1063 (1964). 8. J. C. Derksen, J. R. Katz, Reel. Trau. Chim. Pays-Bas, 53, 652 (1934). 9. G. R. Hennig, Prog. Inorg. Chem., 1, 125 (1959). Classical review on graphite intercalation compounds. 10. G . Ruess, Monatsh. Chem., 76, 381 (1947). 11. A. Angoso Catalina, R. J. Ruano Casero, J. €3. Vericad Raga, An. Quim., 73, 512 (1977). 12. W. H. Slabaugh, B. C. Seiler, J. Phys. Chem., 66, 396 (1962). 13. A. S . Nazarov, V. V. Lisitsa, I. I. Yankovlev, Russ. J. Inorg. Chem. (Engl. Transl.), 23,995 (1978). 14. W. Scholz, H. P. Boehm, Z . Anorg. Allg. Chem., 331, 129 (1964). 15. U. Hofmann, A. Frenzel, E. Csalan, Justus Liebigs Ann. Chem., 510, 1 (1934). 16. E. Theodoridou, J. 0. Besenhard, H. P. Fritz, J. Electroanal. Chem., 122, 67 (1981). 17. I. P. Kolesnikova, A. G. Voloshin, Kinet. Katal. (Engl. Transl.), 16, 700 (1975). 18. A. Tomita, Carbon, 12, 92 (1974). 19. D. R. Lowde, J. 0.Williams, P. A. Attwood, R. J. Bird, B. D. McNicol, R.T. Short,J. Chem. Sac., Faraday Trans. I, 75, 2312 (1979). 20. E. Theodoridou, A. D. Jannakoudakis, J. 0. Basenhand, R. F. Sauter, Synth. Met., 14, 125 (1986).

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc. 16.4.2. Graphite and Boron Nitride Intercalation Compounds 16.4.2.6. Graphite Fluorides 16.4.2.6.1. Poly(monocarbon monofluoride), (CF), .

265

16.4.2.5.4. Graphite Oxide Membranes.

Semipermeable membranes with cation-exchange properties can be prepared by evaporating colloidal solutions of GO. Supported GO membranes are impermeable to 0, and N, and can be used to determine cation activities and to desalinate water',2. (J 0. BESENHARD)

1. H. P. Boehm, A. Clauss, U. Hofmann, J. Chim. Phys., 58, 141 (1961). Review on graphite oxide and membranes thereof. 2. Gmelins Handbuch der Anorganischen Chemie, Vol. 14, Part B-3, Verlag Chemie, Weinheim 1968. Review.

16.4.2.6. Graphite Fluorides

Fluorination of graphite at elevated temperature yields covalent layered C-F compounds with variable fluorine content. The compositions and properties of these preparations depend on particle size and orientation of the pristine carbon material and on the conditions of fluorination. Nevertheless, graphite fluorides are better defined than graphite oxides, and there are at least three different binary graphite fluorides: poly(monocarbon monofluoride), poly(dicarbon monofluoride) and poly(tetracarbon monofluoride). Covalent C-F bonds are also detectable in lamellar 7c charge-transfer graphite C,KrF,' and halogen fluorides3 (see $16.4.2.1.1 and compounds such as C,SbF, These mixed covalent-7c charge-transfer compounds arise from internal fluorination' by the intercalant.

',

(J.O. BESENHARD)

1. L. Streifinger, H. P. Boehm, R. Schlogl, R. Pentenrieder, Carbon, 17, 195 (1979). 2. H. Selig, P. K. Gallagher, L. B. Ebert, Inorg. Nucl. Chem. Lett., 13, 427 (1977). 3. H. Selig, L. B. Ebert, Adu. Znorg. Chem. Radiochem., 23, 281 (1980). Comprehensive review on graphite intercalation compounds. 4. A. W. Moore, Chem. Phys. Carbon, 17, 233 (1981). 5. N. Bartlett, B. W. McQuillan, in Intercalation Chemistry,M. S. Wittinghaus, B. W. Jacobson, eds., Academic Press, New York, 1982, p. 19. 16.4.2.6.1. Poly(monocarbon monofluoride), (CF),

.

The literature up to 1966 is covered by ref. 1, and the properties of poly(monocarbon monofluoride) have been s ~ r v e y e d ~There , ~ . are reviews on preparations and properties of (CF), in Japanese6 and Czech7. The (CF), compound is nonstoichiometric with considerable phase range and so is indicated by the formula (CF,),. The properties of (CF,), vary with x: electronic conductivity and heat of immersion (wetting agent, butanol)' decrease with increasing x, and the color changes from black for x I 0.715 grey to white for x 2 0.998'. On the other hand, the frequency of the C-F stretch is nearly independent of x '. In poly(monocarbon monofluoride) the fluorine atoms form layers above and below puckered layers of sp3-hybridized carbon atomszs3. (i) Preparation of Poly(monocarbon monofluoride). Preparation of (CF,), with x 1 is by action of F, (760 torr) on graphite at 420°C in a copper tubes; x increases with

-

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc. 16.4.2. Graphite and Boron Nitride Intercalation Compounds 16.4.2.6. Graphite Fluorides 16.4.2.6.1. Poly(monocarbon monofluoride), (CF), .

265

16.4.2.5.4. Graphite Oxide Membranes.

Semipermeable membranes with cation-exchange properties can be prepared by evaporating colloidal solutions of GO. Supported GO membranes are impermeable to 0, and N, and can be used to determine cation activities and to desalinate water',2. (J 0. BESENHARD)

1. H. P. Boehm, A. Clauss, U. Hofmann, J. Chim. Phys., 58, 141 (1961). Review on graphite oxide and membranes thereof. 2. Gmelins Handbuch der Anorganischen Chemie, Vol. 14, Part B-3, Verlag Chemie, Weinheim 1968. Review.

16.4.2.6. Graphite Fluorides

Fluorination of graphite at elevated temperature yields covalent layered C-F compounds with variable fluorine content. The compositions and properties of these preparations depend on particle size and orientation of the pristine carbon material and on the conditions of fluorination. Nevertheless, graphite fluorides are better defined than graphite oxides, and there are at least three different binary graphite fluorides: poly(monocarbon monofluoride), poly(dicarbon monofluoride) and poly(tetracarbon monofluoride). Covalent C-F bonds are also detectable in lamellar 7c charge-transfer graphite C,KrF,' and halogen fluorides3 (see $16.4.2.1.1 and compounds such as C,SbF, These mixed covalent-7c charge-transfer compounds arise from internal fluorination' by the intercalant.

',

(J.O. BESENHARD)

1. L. Streifinger, H. P. Boehm, R. Schlogl, R. Pentenrieder, Carbon, 17, 195 (1979). 2. H. Selig, P. K. Gallagher, L. B. Ebert, Inorg. Nucl. Chem. Lett., 13, 427 (1977). 3. H. Selig, L. B. Ebert, Adu. Znorg. Chem. Radiochem., 23, 281 (1980). Comprehensive review on graphite intercalation compounds. 4. A. W. Moore, Chem. Phys. Carbon, 17, 233 (1981). 5. N. Bartlett, B. W. McQuillan, in Intercalation Chemistry,M. S. Wittinghaus, B. W. Jacobson, eds., Academic Press, New York, 1982, p. 19. 16.4.2.6.1. Poly(monocarbon monofluoride), (CF),

.

The literature up to 1966 is covered by ref. 1, and the properties of poly(monocarbon monofluoride) have been s ~ r v e y e d ~There , ~ . are reviews on preparations and properties of (CF), in Japanese6 and Czech7. The (CF), compound is nonstoichiometric with considerable phase range and so is indicated by the formula (CF,),. The properties of (CF,), vary with x: electronic conductivity and heat of immersion (wetting agent, butanol)' decrease with increasing x, and the color changes from black for x I 0.715 grey to white for x 2 0.998'. On the other hand, the frequency of the C-F stretch is nearly independent of x '. In poly(monocarbon monofluoride) the fluorine atoms form layers above and below puckered layers of sp3-hybridized carbon atomszs3. (i) Preparation of Poly(monocarbon monofluoride). Preparation of (CF,), with x 1 is by action of F, (760 torr) on graphite at 420°C in a copper tubes; x increases with

-

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc. 16.4.2. Graphite and Boron Nitride Intercalation Compounds 16.4.2.6. Graphite Fluorides 16.4.2.6.1. Poly(monocarbon monofluoride), (CF), .

265

16.4.2.5.4. Graphite Oxide Membranes.

Semipermeable membranes with cation-exchange properties can be prepared by evaporating colloidal solutions of GO. Supported GO membranes are impermeable to 0, and N, and can be used to determine cation activities and to desalinate water',2. (J 0. BESENHARD)

1. H. P. Boehm, A. Clauss, U. Hofmann, J. Chim. Phys., 58, 141 (1961). Review on graphite oxide and membranes thereof. 2. Gmelins Handbuch der Anorganischen Chemie, Vol. 14, Part B-3, Verlag Chemie, Weinheim 1968. Review.

16.4.2.6. Graphite Fluorides

Fluorination of graphite at elevated temperature yields covalent layered C-F compounds with variable fluorine content. The compositions and properties of these preparations depend on particle size and orientation of the pristine carbon material and on the conditions of fluorination. Nevertheless, graphite fluorides are better defined than graphite oxides, and there are at least three different binary graphite fluorides: poly(monocarbon monofluoride), poly(dicarbon monofluoride) and poly(tetracarbon monofluoride). Covalent C-F bonds are also detectable in lamellar 7c charge-transfer graphite C,KrF,' and halogen fluorides3 (see $16.4.2.1.1 and compounds such as C,SbF, These mixed covalent-7c charge-transfer compounds arise from internal fluorination' by the intercalant.

',

(J.O. BESENHARD)

1. L. Streifinger, H. P. Boehm, R. Schlogl, R. Pentenrieder, Carbon, 17, 195 (1979). 2. H. Selig, P. K. Gallagher, L. B. Ebert, Inorg. Nucl. Chem. Lett., 13, 427 (1977). 3. H. Selig, L. B. Ebert, Adu. Znorg. Chem. Radiochem., 23, 281 (1980). Comprehensive review on graphite intercalation compounds. 4. A. W. Moore, Chem. Phys. Carbon, 17, 233 (1981). 5. N. Bartlett, B. W. McQuillan, in Intercalation Chemistry,M. S. Wittinghaus, B. W. Jacobson, eds., Academic Press, New York, 1982, p. 19. 16.4.2.6.1. Poly(monocarbon monofluoride), (CF),

.

The literature up to 1966 is covered by ref. 1, and the properties of poly(monocarbon monofluoride) have been s ~ r v e y e d ~There , ~ . are reviews on preparations and properties of (CF), in Japanese6 and Czech7. The (CF), compound is nonstoichiometric with considerable phase range and so is indicated by the formula (CF,),. The properties of (CF,), vary with x: electronic conductivity and heat of immersion (wetting agent, butanol)' decrease with increasing x, and the color changes from black for x I 0.715 grey to white for x 2 0.998'. On the other hand, the frequency of the C-F stretch is nearly independent of x '. In poly(monocarbon monofluoride) the fluorine atoms form layers above and below puckered layers of sp3-hybridized carbon atomszs3. (i) Preparation of Poly(monocarbon monofluoride). Preparation of (CF,), with x 1 is by action of F, (760 torr) on graphite at 420°C in a copper tubes; x increases with

-

266

16.4. The Formation of Sheet Structures 16.4.2. Graphite and Boron Nitride Intercalation Compounds 16.4.2.6. Graphite Fluorides TABLE1. RELATIONSHIPS BETWEENTHE REACTIONTEMPERATURES AND F:C RATIOSIN (CF,),, (AFTER REF. 13)

Temperature

a

("C)

Reaction time

Empirical formula

378 400 450 450 475 47 5 500 525 550 570 570 600 600 640

120 h 50 h 10 h 70 h 5h 50 h 150 rnin I00 min 50 rnin 40 rnin 120 h 20 min 140 h 5h

CFO 58 CFO

61

CFO

67

CFO

67

CFO

71

CF0.72

CFO 8 2 CFO 88 CFO 9 0 CFO 9 3 CFO 9 3 CFO

96

CFO 9 7 CFLOO

Natural graphite (200-250 mesh), 200 tom F,.

reaction time but is only -0.88 after 6.5 h 8 . Oxygen-free charcoal is fluorinated at 280°C. Oxygen-containing charcoal burns in F, at room temperature'. Caution: Higher temperature (500-700°C for graphite, 400-450°C for charcoal) can cause severe explosions' and formation of CF,, C,F, and other perfluorinated hydrocarbons. Thus fluorination of graphite should be regarded as a potentially hazardous experiment. Preparation of (CF,), with x up to 0.988 is by fluorination of graphite in a copper tube at 525-550°C 7 , 9 . The maximum fluorine content reached in preparations of (CF,), depends mainly on the reaction temperature (see Table 1). Prolonged reaction time, i.e., going beyond -0.5-2 h, causes only insignificant further increases of the F:C r a t i 0 ~ 9 ~At ~ ' ~>6OO0C . the formation of gaseous fluorides becomes considerable". The interlayer spacing of fluorinated carbon as a function of the fluorination temperature is shown in Fig. 1. Catalytic quantities of H F l 2 or FSO,H l 3 decrease the temperature required for fluorination of graphite; -250°C is sufficient at a H F partial pressure of 200 torr'. Catalytic fluorination of graphite proceeds via ionic or n charge-transfer intermediates; which are characterized by high mobility of the intercalant. In contrast, there is no evidence for mobile fluorine in (CF,), '. Silver fluorideI4, LiF and CuF, 1 s , 1 6h a w also been used as fluorination catalysts. As a rule, catalytically prepared (CF,), contains small quantities of the catalyst. At high pressures of fluorine (> 15 atm), (CF,), can be obtained even at RTI7-as the fluorination reaction may be violent, this procedure is extremely hazardous. Instead of F,, ClF, and BrF, are also used as fluorinating agents for graphite. At low T BrF, forms mainly .n charge-transfer compounds'8, but an increase in T yields the at least partly covalent products (CF,);y ClF, and (CF,);y BrF, ", and finally pure (CF,)"

16.4.2. Graphite and Boron Nitride Intercalation Compounds 16.4.2.6.Graphite Fluorides 16.4.2.6.1.Poly(monocarbon monofluoride), (CF), .

267

900 -

800-

700 -

600 -

L 400 500 600 Temperature

(OC)

Figure 1. Interlayer spacing do,, as a function of fluorination temperature (after ref. 13). In a fluorine plasma, (CF,), with x = 1.19 is synthesized at a gas temperature between 100 and 150°C ”. The synthesis of superstoichiometric (CF,), with x 2 1 is claimed (x 2 1.12 1 7 , x 2 1.14 22, x 2 1.23 - 1.25 1 3 , x 2 1.26 23). Excess fluorine is present in =CF, and -CF, groups at plane edges and therefore the F:C ratio increases with decreasing particle size ”. Adsorbed HF or aliphatic fluorocarbons may also raise the analytical F content. Small amounts of (CF,), are formed on carbon anodes in fluoride meltsz4. Some (CF,), preparations are commercially availablez5. (ii) Reactions of Poly(monocarbon monofluoride). The (CF,), compound is less reactive than any other graphite compound and has some poly(monocarbon trifluoride)like properties: it is not wetted by water, is insoluble in polar and nonpolar solvents, does not show swelling in n-donor or n-donor solvents, is thermaily stable up to 450°C and is not reduced by H,, even at 400°C On the other hand, (CF,), intercalates fluorineacceptor electron-pair acceptor acidsz6, decomposes into amorphous carbon and aliphatic fluorocarbons at 450°C and is an oxidant with commercial application for positive electrodes in batteries with lithium negative electrodes. (iii) Reactions with Electron-Pain Acceptor Acids. The electron-pair acceptor acid SbF, reacts with (CF,), to form a compound containing mobile fluorine species:

’.

C,F

+ SbF,

-

C,’(SbF,)F-

(a>

It is not possible to bring this reaction to the C+(SbF,)F- limit, but more dilute compounds, such as (CF, 06)1 ,SbF, (nominal stoichiometry), can be formedz5.

268

16.4. The Formation of Sheet Structures 16.4.2. Graphite and Boron Nitride Intercalation Compounds 16.4.2.6 Graphite Fluorides

Similar compounds [e.g., C: F(BF,); 26], have been prepared with other fluorineacceptor electron-pair acceptor acids, e.g., with PF,, BF, 2 7 3 2 8 and with halogen fluorides29. (iv) Thermal Decomposition. Thermal decomposition of (CF,), is an exothermic and topotactic r e a c t i ~ n ~ ~ due ~ ~ to , ~the ~ , thermal ~'; stability up to -4OO"C, (CF,), is a potential solid lubricant31 , 3 2 . The primary gaseous decomposition products, C,F4 33 and C,F, or CF, 29, form CF,, C2F6,C3F8 and other fluorocarbons by secondary reactions. After the fluorocarbons are released, the thickness of the residual amorphous graphite particles in the c direction is just one-half that of the parent graphite particles from which (CF,), is preparedz9. (v) Reduction. The (CF,), compound does not react with HI, is only slowly reduced to carbon and F- by Zn in acetic acid' but is readily reduced: (i) cathodically in dipolar aprotic solutions of lithium salts and (ii) by organolithium compounds such as nbutyllithium and lithium n a ~ h t h a l i d e ~ ~ . The (CF,), compound is applied in high-energy-density organic electrolyte lithium batteries. The irreversible discharge reaction is:

'

(CF,),

+ (nx) Li

-

C,

+ (nx) LiF

(b)

assuming finely divided carbon and LiF to be the discharge products35. There is evidence for a more complex discharge mechanism via lithium intercalation into (CF,),, yielding ternary compounds, (CLi,F,), 10,34,36. The interplanar distance in (CF,), increases during reduction34 and may reach 935 prn3,. Moreover, the open-cell voltage of the redox couple (CF,),/Li is only 3 V, whereas 4.57 V is calculated from thermodynamic data via Eq. (b)37. Ultraviolet radiation enhances the reductive decomposition of (CF,),. Alcohols are oxidized to the corresponding carbonyl compounds3':

-

2 (CF)

+ RR'CHOH

-

2 (C)+ 2 H F + RRCO

(c)

Fine particles of diamond can be synthesized from (CF,), by reduction with metals: (CF,),

+ nx Cu

nx CuF

+n C

(d)39

under high static pressure39 or in a shock wave4'. (J.O. BESENHARD)

1 . Gmefins Handbuch der Anorganischen Chemie, Vol. 14, Part B-3, Verlag Chemie, Weinheim, 1968. Review and data collection. 2. P. Kamarchik Jr., J. L. Margrave, Ace. Chem. Res., 1 1 , 296 (1978). 3. L. B. Ebert, Annu. Rev. Mazer. Sci., 6, 181 (1976). Comprehensive review on graphite intercalation compounds. 4. N. Watanabe, Kagaku Sosetsu, 27, 37 (1980); Chem. Abstr., 93, 49,478 (1980). 5. I. Peka, V. Petrzila, Chem. Listy, 74, 32 (1980); Chem. Abstr., 92, 149,231 (1980). 6. Y . Kita, N. Watanabe, Y. Fujii, J. Am. Chem. SOC.,101, 3832 (1979). 7. W. Rudorff, G. Rudorff, Z . Anorg. Allg. Chem., 253,281 (1947). 8. 0. Ruff, 0. Bretschneider, F. Ebert, 2. Anorg. Allg. Chem., 217, l(1934). 9. W. Rudorff, Adv. Inorg. Chern. Radiochem., I , 223 (1959). Classical review on graphite compounds.

16.4.2. Graphite and Boron Nitride Intercalation Compounds 16.4 2.6. Graphite Fluorides 16.4.2.6.2. Poly(dicarbon monofluoride), (C,F), .

269

N. Watanabe, Solid State Zonics, I, 87 (1980). A. K. Kuriakose, J. L. Margrave, J. Phys. Chem., 69, 2772 (1965). W. Rudorff, G. Rudorff, Chem. Ber., 80,413 (1947). A. S. Nazarov, A. M. Danilenko, I. I. Yakovlev, Zh. Neorg. Khim., 25, 350 (1980); Chem. Abstr., 92, 157,031 (1980). 14. T. Nakajima, T. Ino, K. Nakane, N. Watanabe, in International Colloquium on Layered Compounds, D. Guerard, P. Lagrange, eds., Universite de Nancy I, Nancy, 1988, p. 119. 15. T. Nakajima, K. Nakane, N. Watanabe, Nippon Kugaku Kaishi, 1790 (1985); Chem. Abstr., 104, 60,922 (1986). 16. T. Nakajima, N. Watanabe, I. Kameda, M. Endo, Carbon, 24,343 (1986). 17. R. J. Lagow, R. B. Badachhape, J. L. Wood, J. L. Margrave, J. Chem. Soc., Dalton Trans., 1268 (1974) 18. A. A. Opalovskii, A. S. Nazarov, A. A. Uminskii, Rum. J. Inorg. Chem. (Engl. Transl.), 17, 632 (1972). 19. H. Selig, W. A. Sunder, M. J. Vasile, F. A. Stevie, P. K. Gallagher, L. B. Ebert, J. Fluorine Chem., 12, 397 (1978). 20. Y. I. Nikonorov, E. F. Khairetdinov, Inorg. Muter. (Engl. Transl.), 15, 1254 (1979). 21. R. J. Lagow, L. A. Shimp, D. K. Lam, R. F. Baddour, Znorg. Chem., 11,2568 (1972). 22. N. Watanabe, S. Koyama, Bull. Chern. Soc. Jpn., 53, 3093 (1980). 23. D. T. Mcshri, Polym. News, 6, 200 (1980). 24. 0. Ruff, Ber. Dtsch. Chem. Ges., 69, 181A (1936). 25. E.g., from Ozark-Mahoning, Tulsa, OK, and from Alfa-Ventron, Beverly, MA. 26. L. B. Ebert, R. A. Huggins, J. I. Braurnan, Mater. Res. Bull., 11, 615 (1976). 27. Y. I. Nikonorov, Kinet. Katal. (Engl. Transl.), 20, 1322 (1979). 28. R. Hulme, US. Pat. 4, 119,655; Chem. Abstr., 90, 90,081 (1979). 29. N. Watanabe, S. Koyama, H. Imoto, Bull. Chem. Soc. Jpn., 53, 2731 (1980). 30. N. Watanabe, T. Kawamura, S. Koyarna, Bull. Chem. Soc. Jpn., 53, 3100 (1980). 31. P. Cadman, G. M. Gossedge, J. Mat. Sci., 14, 1465 (1979). 32. C . Martiu, J. Sulleu, M. Roussel, Wear, 34, 215 (1975); Chem. Abstr. 84, 19,994 (1975). 33. P. Kamarchik Jr., J. L. Margrave, J. T h e m . Anal., 11, 259 (1977). 34. M. S. Whittingham, J. Electrochein. Soc., 122, 526 (1975). 35. M. Fukuda, T. Iiiirna. in Power Sources, Vol. 5, D. H. Collins, ed., Academic Press, New York, 1975, p. 713. 36. N. Watanabe. M. Endo. K. Ueno. Solid State Zonics. 1. 501 (1980). 37. A. J. Valerga,’R. B. Badachhape, G. D. Parks, P. Kamarchik: J. L.’Wood, J. L. Margrave, U.S. Army Electronics Command, Technical Report, ECOM-0056-F (1974). 38. N. Watanabe, K. Ueno, Bull. Chem. Soc. Jpn., 53, 388 (1980). 39. J. L. Margrave, R. G. Bautista, P. J. Ficalora, R. B. Badachhape, U S . Pat. 3,711, 595; Chem. Abstr., 78, 99,841 (1973). 40. S. Yamaguchi, N. Setaka, J. Electrochem. Soc., 127, 245 (1980).

10. 11. 12. 13.

16.4.2.6.2. Poly(dicarbon monofluoride), (C,F),

.

The (C’F,), compound with x 1 is obtained by action of F, on graphite with the reaction temperature kept as low as possible (see 516.4.2.6.1, Table 1). This minimum temperature is determined by the crystallinity of the graphite-highly oriented graphite requires 30O-35OcC, whereas 300°C is sufficient for poorly oriented material’,2. The reaction time is affected by the particle size’.’. The (C,F,), compound is also prepared by action of ClF, and H F on graphite3. In a structural model of (C,F,), ‘J the fluorine atoms form layers above and below interconnected puckered double layers of sp3-hybridized carbon atoms, i.e., diamond double layers. Poly(dicarbon monofluoride) is a black nonconductor. Its chemistry is basically the same as that of poly(monocarbon monofluoride). Surprisingly, poly(dicarbon monofluoride) is less stable than poly(monocarbon monofluoride), as indicated by the lower N

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc. 16.4.2. Graphite and Boron Nitride Intercalation Compounds 16.4 2.6. Graphite Fluorides 16.4.2.6.2. Poly(dicarbon monofluoride), (C,F), .

269

N. Watanabe, Solid State Zonics, I, 87 (1980). A. K. Kuriakose, J. L. Margrave, J. Phys. Chem., 69, 2772 (1965). W. Rudorff, G. Rudorff, Chem. Ber., 80,413 (1947). A. S. Nazarov, A. M. Danilenko, I. I. Yakovlev, Zh. Neorg. Khim., 25, 350 (1980); Chem. Abstr., 92, 157,031 (1980). 14. T. Nakajima, T. Ino, K. Nakane, N. Watanabe, in International Colloquium on Layered Compounds, D. Guerard, P. Lagrange, eds., Universite de Nancy I, Nancy, 1988, p. 119. 15. T. Nakajima, K. Nakane, N. Watanabe, Nippon Kugaku Kaishi, 1790 (1985); Chem. Abstr., 104, 60,922 (1986). 16. T. Nakajima, N. Watanabe, I. Kameda, M. Endo, Carbon, 24,343 (1986). 17. R. J. Lagow, R. B. Badachhape, J. L. Wood, J. L. Margrave, J. Chem. Soc., Dalton Trans., 1268 (1974) 18. A. A. Opalovskii, A. S. Nazarov, A. A. Uminskii, Rum. J. Inorg. Chem. (Engl. Transl.), 17, 632 (1972). 19. H. Selig, W. A. Sunder, M. J. Vasile, F. A. Stevie, P. K. Gallagher, L. B. Ebert, J. Fluorine Chem., 12, 397 (1978). 20. Y. I. Nikonorov, E. F. Khairetdinov, Inorg. Muter. (Engl. Transl.), 15, 1254 (1979). 21. R. J. Lagow, L. A. Shimp, D. K. Lam, R. F. Baddour, Znorg. Chem., 11,2568 (1972). 22. N. Watanabe, S. Koyama, Bull. Chern. Soc. Jpn., 53, 3093 (1980). 23. D. T. Mcshri, Polym. News, 6, 200 (1980). 24. 0. Ruff, Ber. Dtsch. Chem. Ges., 69, 181A (1936). 25. E.g., from Ozark-Mahoning, Tulsa, OK, and from Alfa-Ventron, Beverly, MA. 26. L. B. Ebert, R. A. Huggins, J. I. Braurnan, Mater. Res. Bull., 11, 615 (1976). 27. Y. I. Nikonorov, Kinet. Katal. (Engl. Transl.), 20, 1322 (1979). 28. R. Hulme, US. Pat. 4, 119,655; Chem. Abstr., 90, 90,081 (1979). 29. N. Watanabe, S. Koyama, H. Imoto, Bull. Chem. Soc. Jpn., 53, 2731 (1980). 30. N. Watanabe, T. Kawamura, S. Koyarna, Bull. Chem. Soc. Jpn., 53, 3100 (1980). 31. P. Cadman, G. M. Gossedge, J. Mat. Sci., 14, 1465 (1979). 32. C . Martiu, J. Sulleu, M. Roussel, Wear, 34, 215 (1975); Chem. Abstr. 84, 19,994 (1975). 33. P. Kamarchik Jr., J. L. Margrave, J. T h e m . Anal., 11, 259 (1977). 34. M. S. Whittingham, J. Electrochein. Soc., 122, 526 (1975). 35. M. Fukuda, T. Iiiirna. in Power Sources, Vol. 5, D. H. Collins, ed., Academic Press, New York, 1975, p. 713. 36. N. Watanabe. M. Endo. K. Ueno. Solid State Zonics. 1. 501 (1980). 37. A. J. Valerga,’R. B. Badachhape, G. D. Parks, P. Kamarchik: J. L.’Wood, J. L. Margrave, U.S. Army Electronics Command, Technical Report, ECOM-0056-F (1974). 38. N. Watanabe, K. Ueno, Bull. Chem. Soc. Jpn., 53, 388 (1980). 39. J. L. Margrave, R. G. Bautista, P. J. Ficalora, R. B. Badachhape, U S . Pat. 3,711, 595; Chem. Abstr., 78, 99,841 (1973). 40. S. Yamaguchi, N. Setaka, J. Electrochem. Soc., 127, 245 (1980).

10. 11. 12. 13.

16.4.2.6.2. Poly(dicarbon monofluoride), (C,F),

.

The (C’F,), compound with x 1 is obtained by action of F, on graphite with the reaction temperature kept as low as possible (see 516.4.2.6.1, Table 1). This minimum temperature is determined by the crystallinity of the graphite-highly oriented graphite requires 30O-35OcC, whereas 300°C is sufficient for poorly oriented material’,2. The reaction time is affected by the particle size’.’. The (C,F,), compound is also prepared by action of ClF, and H F on graphite3. In a structural model of (C,F,), ‘J the fluorine atoms form layers above and below interconnected puckered double layers of sp3-hybridized carbon atoms, i.e., diamond double layers. Poly(dicarbon monofluoride) is a black nonconductor. Its chemistry is basically the same as that of poly(monocarbon monofluoride). Surprisingly, poly(dicarbon monofluoride) is less stable than poly(monocarbon monofluoride), as indicated by the lower N

270

16.4. The Formation of Sheet Structures

16.4.2, Graphite and Boron Nitride Intercalation Compounds 16.4.2.6 Graphite Fluorides

temperature of decomposition and the more positive redox potential'. The electrochemical reduction of (C,F,), in dipolar aprotic lithium salt solutions yields ternary lithium intercalation compounds, (C,Li,F,), '. (J.0 BESENHARD)

1. Y. Kita, N. Watanabe, Y. Fujii, J. Am. Chem. Soc., 101, 3832 (1979). 2. N. Watanabe, Solid State Ionics, I , 87 (1980). 3. B. M. L. Rao, P.A. Malachesky, U.S. Pat. 4,057,676; Chem. Abstr., 88, 107,898 (1978). 16.4.2.6.3. Poly(tetracarb0n monofluoride), (C,F),.

The (C4F,), compound with 1 is obtained by fluorination of graphite: (i) in the presence of H F and (ii) at T < 80-1CO"C There is always some H F associated with (C4F,), preparations that (i) Formation of Poly(tetracarb0n monofluoride).

x

I.

cannot be removed in vacuo132. The (C4F,), compound is covalent but characterized by a considerable electronic conductivity. The fluorine atoms form layers above and below carbon single layers which are assumed to be In the classical preparation a F,-HF mixture (HF partial pressure -200 torr) is passed over graphite at RT ( - 1.5 h, copper tube). The rate of reaction drops rapidl'y with decreasing H F pressure. This preparation yields (C4F,), with typical values for x in the range 1-1.1 '. Because the presence of H F is indispensable, the lamellar acid compound C:,[HF,] - . 4 H F (see 516.4.2.1.4) may be an intermediate product. Alternately, a static bomb reactor [nickel or Monel (an NiCuCFe alloy), prefluorinated] and a total pressure of the F,-HF mixture of -5 atm can be used. The H F is introduced first ( - 1 atm), and then the pressure is slowly raised to 5 atm by adding F,. The reaction is complete in 15 min '. Caution: The bomb may become hot owing to the exothermic reaction producing (C4FX)". Therefore, the use of thick-walled bombs is recommended'. (ii) Reactions of Poly(tetracarb0n monofluoride). Like poly(monocarbon monofluoride) and poly(dicarbon monofluoride), poly(tetracarbon monofluoride) is an inert substance. It is stable in air and it is not attacked by dilute alkalis or acids or by reducing agents such as FeCl, and HI or by oxidizing agents such as dichromate in concentrated sulfuric acid3. The (C4F,), compound intercalates ClF, to yield nominal stoichiometry C,F. 0.46 ClF, '. The thermal decomposition of this compound results in evolution of chlorine and further fluorination of the graphite matrix4. Action of NH, on C4F. 0.46 ClF, yields carbon fluorides containing NH, groups5. The (C4F,), compound is thermally less stable than carbon fluorides with a higher F:C ratio. Decomposition starts at > 100°C 3; thermodynamic data on (C,F,), are given in ref. 7. The (C4F,), compound can be cathodically reduced in dipolar aprotic solutions of lithium salts. The redox potential is about the same as that of (CF,),; however, the reduction is kinetically more hindered'.

-

(J 0.BESENHARD)

1. W. Rudorff, G. Rudorff, Chem. Ber., 80,417 (1947). 2. R. J. Lagow, R. B. Badachhape, P. Ficalora, J. L. Wood, J. L. Margrave, Synth. Inorg. Metal-Org. Chem., 2, 145 (1972).

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc. 270

16.4. The Formation of Sheet Structures

16.4.2, Graphite and Boron Nitride Intercalation Compounds 16.4.2.6 Graphite Fluorides

temperature of decomposition and the more positive redox potential'. The electrochemical reduction of (C,F,), in dipolar aprotic lithium salt solutions yields ternary lithium intercalation compounds, (C,Li,F,), '. (J.0 BESENHARD)

1. Y. Kita, N. Watanabe, Y. Fujii, J. Am. Chem. Soc., 101, 3832 (1979). 2. N. Watanabe, Solid State Ionics, I , 87 (1980). 3. B. M. L. Rao, P.A. Malachesky, U.S. Pat. 4,057,676; Chem. Abstr., 88, 107,898 (1978). 16.4.2.6.3. Poly(tetracarb0n monofluoride), (C,F),.

The (C4F,), compound with 1 is obtained by fluorination of graphite: (i) in the presence of H F and (ii) at T < 80-1CO"C There is always some H F associated with (C4F,), preparations that (i) Formation of Poly(tetracarb0n monofluoride).

x

I.

cannot be removed in vacuo132. The (C4F,), compound is covalent but characterized by a considerable electronic conductivity. The fluorine atoms form layers above and below carbon single layers which are assumed to be In the classical preparation a F,-HF mixture (HF partial pressure -200 torr) is passed over graphite at RT ( - 1.5 h, copper tube). The rate of reaction drops rapidl'y with decreasing H F pressure. This preparation yields (C4F,), with typical values for x in the range 1-1.1 '. Because the presence of H F is indispensable, the lamellar acid compound C:,[HF,] - . 4 H F (see 516.4.2.1.4) may be an intermediate product. Alternately, a static bomb reactor [nickel or Monel (an NiCuCFe alloy), prefluorinated] and a total pressure of the F,-HF mixture of -5 atm can be used. The H F is introduced first ( - 1 atm), and then the pressure is slowly raised to 5 atm by adding F,. The reaction is complete in 15 min '. Caution: The bomb may become hot owing to the exothermic reaction producing (C4FX)". Therefore, the use of thick-walled bombs is recommended'. (ii) Reactions of Poly(tetracarb0n monofluoride). Like poly(monocarbon monofluoride) and poly(dicarbon monofluoride), poly(tetracarbon monofluoride) is an inert substance. It is stable in air and it is not attacked by dilute alkalis or acids or by reducing agents such as FeCl, and HI or by oxidizing agents such as dichromate in concentrated sulfuric acid3. The (C4F,), compound intercalates ClF, to yield nominal stoichiometry C,F. 0.46 ClF, '. The thermal decomposition of this compound results in evolution of chlorine and further fluorination of the graphite matrix4. Action of NH, on C4F. 0.46 ClF, yields carbon fluorides containing NH, groups5. The (C4F,), compound is thermally less stable than carbon fluorides with a higher F:C ratio. Decomposition starts at > 100°C 3; thermodynamic data on (C,F,), are given in ref. 7. The (C4F,), compound can be cathodically reduced in dipolar aprotic solutions of lithium salts. The redox potential is about the same as that of (CF,),; however, the reduction is kinetically more hindered'.

-

(J 0.BESENHARD)

1. W. Rudorff, G. Rudorff, Chem. Ber., 80,417 (1947). 2. R. J. Lagow, R. B. Badachhape, P. Ficalora, J. L. Wood, J. L. Margrave, Synth. Inorg. Metal-Org. Chem., 2, 145 (1972).

16.4.2. Graphite and Boron Nitride Intercalation Compounds 16.4.2.7. Residue and Surface Compounds 16.4.2.7.1. Introduction and Definitions.

27 1

3. W. Riidorff, Adv. lnorg. Chem. Radiochem., I , 223 (1959). 4. A. S. Nazarov, I. I. Yakovlev, A. F. Antimonov, V. M. Grank~n,V. B. Durasov, P. P.

Samyannikov, Zh. Neorg. Khim, 25, 1506 (1980); Chem. Abstr., 93, 124,977 (1980). 5. A. S. Nazarov, A. F. Antimonov, Z h Neorg. Khim., 25, 2123 (1980); Chem. Abstr., 93, 178,724 (1980) 6. J. L. Wood, A. J. Valerga, R. B. Badachhape, J. L. Margrave, Report Ad-755 934 (1972); Chem. Abstr., 78, 165,105 (1973). 7. A. J. Valerga, R. B. Badachhape, G. D. Parks, P. Kamarchik, J. L. Wood, Report AD-776 990 (1974); Chem. Abstr., 81, 69,310 (1974). 8. H. F. Hunger, G. J. Heymach, J. Electrochem. SOC.,120, 1161 (1973).

16.4.2.7. Residue and Surface Compounds 16.4.2.7.1. Introduction and Definitions.

Lamellar graphite compounds cannot be decomposed completely under ordinary conditions. Although most of the intercalated material can be removed easily, a residue is more tenaciously retained’. These decomposition products are residlie compounds whose composition remains constant at RT in air, i.e., at zero partial pressure of the intercalant’. The structure and properties of a residue compound, C,X,, - m l , prepared by partial deintercalation of C,X: C,X

-

C,X,,

-m)

+mX

(a)

-

differ from those of the lamellar compound, C,X(, - m ) , prepared by combination of C, and X: C,

+ (1 - m) X

CnX(l-m)

(b)

The residual intercalant is retained at crystal imperfections such as crystallite boundaries or twin lines’, and x-ray diffraction may show graphite reflections only3. Residue compounds with large interplanar distances have residual intercalant between the carbon planes4-*. In bromine residue compounds ordered layers of intercalant are found by electron d i f f r a ~ t i o n ~Thus * ~ . there is no distinct borderline between dilute lamellar and residue compounds. The term “surface compound of carbon” is restricted to chemisorption, i.e., linkage of the adsorbate to the surface by covalent bonds. Surface groups on graphite are situated on edge carbon atoms or on defect sites; there is no attack of basal planesg. Carbonoxygen surface compounds are present on any black carbon in contact with air; there are also surface compounds with S, H, N or halogens bonded to carbong-”. (J.0 BESENHARD)

1. G. R. Hennig, Prog. Inorg. Chem., I , 125 (1959). Classical review on graphite intercalation

compounds. 2. J. G. Hooley, Chem. Phys. Carbon, 5, 321 (1969). Revlew on intercalation isotherms. 3. L. B. Ebert, Annu. Rev. Mater. Sci., 6, 181 (1976). 4. M. Inagaki, Carbon, 5, 317 (1967). 5. M. Inagaki, J. C. Rouillon, G . Fug, P. Delhaes, Carbon, 15, 181 (1977). 6. D. D. L. Chung, Rev. Sci. Znstrum., 51, 933 (1980). 7. D. D. L. Chung, J. Electron. Mater., 7, 189 (1978). 8. J. 0. Besenhard, H.-F. Klein, J. Gross, H. Mohwald, J. J. Nickl, Synth. Met., 4, 51 (1981).

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc. 16.4.2. Graphite and Boron Nitride Intercalation Compounds 16.4.2.7. Residue and Surface Compounds 16.4.2.7.1. Introduction and Definitions.

27 1

3. W. Riidorff, Adv. lnorg. Chem. Radiochem., I , 223 (1959). 4. A. S. Nazarov, I. I. Yakovlev, A. F. Antimonov, V. M. Grank~n,V. B. Durasov, P. P.

Samyannikov, Zh. Neorg. Khim, 25, 1506 (1980); Chem. Abstr., 93, 124,977 (1980). 5. A. S. Nazarov, A. F. Antimonov, Z h Neorg. Khim., 25, 2123 (1980); Chem. Abstr., 93, 178,724 (1980) 6. J. L. Wood, A. J. Valerga, R. B. Badachhape, J. L. Margrave, Report Ad-755 934 (1972); Chem. Abstr., 78, 165,105 (1973). 7. A. J. Valerga, R. B. Badachhape, G. D. Parks, P. Kamarchik, J. L. Wood, Report AD-776 990 (1974); Chem. Abstr., 81, 69,310 (1974). 8. H. F. Hunger, G. J. Heymach, J. Electrochem. SOC.,120, 1161 (1973).

16.4.2.7. Residue and Surface Compounds 16.4.2.7.1. Introduction and Definitions.

Lamellar graphite compounds cannot be decomposed completely under ordinary conditions. Although most of the intercalated material can be removed easily, a residue is more tenaciously retained’. These decomposition products are residlie compounds whose composition remains constant at RT in air, i.e., at zero partial pressure of the intercalant’. The structure and properties of a residue compound, C,X,, - m l , prepared by partial deintercalation of C,X: C,X

-

C,X,,

-m)

+mX

(a)

-

differ from those of the lamellar compound, C,X(, - m ) , prepared by combination of C, and X: C,

+ (1 - m) X

CnX(l-m)

(b)

The residual intercalant is retained at crystal imperfections such as crystallite boundaries or twin lines’, and x-ray diffraction may show graphite reflections only3. Residue compounds with large interplanar distances have residual intercalant between the carbon planes4-*. In bromine residue compounds ordered layers of intercalant are found by electron d i f f r a ~ t i o n ~Thus * ~ . there is no distinct borderline between dilute lamellar and residue compounds. The term “surface compound of carbon” is restricted to chemisorption, i.e., linkage of the adsorbate to the surface by covalent bonds. Surface groups on graphite are situated on edge carbon atoms or on defect sites; there is no attack of basal planesg. Carbonoxygen surface compounds are present on any black carbon in contact with air; there are also surface compounds with S, H, N or halogens bonded to carbong-”. (J.0 BESENHARD)

1. G. R. Hennig, Prog. Inorg. Chem., I , 125 (1959). Classical review on graphite intercalation

compounds. 2. J. G. Hooley, Chem. Phys. Carbon, 5, 321 (1969). Revlew on intercalation isotherms. 3. L. B. Ebert, Annu. Rev. Mater. Sci., 6, 181 (1976). 4. M. Inagaki, Carbon, 5, 317 (1967). 5. M. Inagaki, J. C. Rouillon, G . Fug, P. Delhaes, Carbon, 15, 181 (1977). 6. D. D. L. Chung, Rev. Sci. Znstrum., 51, 933 (1980). 7. D. D. L. Chung, J. Electron. Mater., 7, 189 (1978). 8. J. 0. Besenhard, H.-F. Klein, J. Gross, H. Mohwald, J. J. Nickl, Synth. Met., 4, 51 (1981).

272

16.4. The Formation of Sheet Structures 16.4.2. Graphite and Boron Nitride Intercalation Compounds 16.4.2.7. Residue and Surface Compounds

9. H. P. Boehm, Angew. Chem., Int. Ed. Engl., 5, 533 (1966). Brief review, recommended reading. 10. H. P. Boehm, A&. Cutal., 16, 179 (1966). Comprehensive review. 11. B. R. Puri, Chem. Phys. Curbon, 6, 191 (1970). Comprehensive review.

16.4.2.7.2. Preparation of Residue Compounds.

Preparations and properties of residue compounds are briefly summarized in several general reviews on graphite intercalation compounds'-5; their synthesis is more systematically dealt with in ref. 6. Residue compounds of graphite are prepared from the corresponding lamellar compounds by partial deintercalation. This process ultimately yields residue compounds; via more dilute lamellar compounds. Figure 1 shows a typical composition isotherm for an intercalation-deintercala tion cycle. Lamellar compounds that form spontaneously by action of the intercalant on graphite can be deintercalated simply by thermal or in vacuo treatment; deintercalation of graphite saltq requires a reductant. In residue compounds prepared from lamellar compounds based on artificial polycrystalline graphite, 1/3 of the original intercalant may be retained', depending on particle size and crystallinity of the material. Lamellar compounds prepared from large crystals decompose only slowly, not just because of longer diffusion pathways in the interlayer gap but because the amount of retained intercalant also increases with the thickness of the graphite flakes6,8-10.

-

120

--

CeBr

E"0

-5

0 0 0 0 F

a

L

Q

i0!

4-

b

ki

oljo o b Partial pressure of Br2

O; 'oh0

O L

I

0

Figure 1. Composition isotherm, Br, on naturally occurring graphite from the Malagasi Republic (Madagascar graphite), 30°C (after ref. 7).

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc. 272

16.4. The Formation of Sheet Structures 16.4.2. Graphite and Boron Nitride Intercalation Compounds 16.4.2.7. Residue and Surface Compounds

9. H. P. Boehm, Angew. Chem., Int. Ed. Engl., 5, 533 (1966). Brief review, recommended reading. 10. H. P. Boehm, A&. Cutal., 16, 179 (1966). Comprehensive review. 11. B. R. Puri, Chem. Phys. Curbon, 6, 191 (1970). Comprehensive review.

16.4.2.7.2. Preparation of Residue Compounds.

Preparations and properties of residue compounds are briefly summarized in several general reviews on graphite intercalation compounds'-5; their synthesis is more systematically dealt with in ref. 6. Residue compounds of graphite are prepared from the corresponding lamellar compounds by partial deintercalation. This process ultimately yields residue compounds; via more dilute lamellar compounds. Figure 1 shows a typical composition isotherm for an intercalation-deintercala tion cycle. Lamellar compounds that form spontaneously by action of the intercalant on graphite can be deintercalated simply by thermal or in vacuo treatment; deintercalation of graphite saltq requires a reductant. In residue compounds prepared from lamellar compounds based on artificial polycrystalline graphite, 1/3 of the original intercalant may be retained', depending on particle size and crystallinity of the material. Lamellar compounds prepared from large crystals decompose only slowly, not just because of longer diffusion pathways in the interlayer gap but because the amount of retained intercalant also increases with the thickness of the graphite flakes6,8-10.

-

120

--

CeBr

E"0

-5

0 0 0 0 F

a

L

Q

i0!

4-

b

ki

oljo o b Partial pressure of Br2

O; 'oh0

O L

I

0

Figure 1. Composition isotherm, Br, on naturally occurring graphite from the Malagasi Republic (Madagascar graphite), 30°C (after ref. 7).

16.4.2. Graphite and Boron Nitride Intercalation Compounds 16.4.2.7. Residue and Surface Compounds 16.4.2.7.2. Preparation of Residue Compounds.

273

Lamellar compounds based on highly oriented graphite can be deintercalated more completely. By electrochemical deintercalation of first-stage acid salts (cf. 516.4.2.1.4) based on highly oriented pyrolytic or natural graphite 95 % of the intercalant can be removed under quasi-equilibrium conditions”-’ 3; for much less crystalline carbon fibers, the corresponding value is 80 % 14. Because deintercalation is slow, it may be hard to distinguish between kinetically hindered deintercalation and a true residue compound in equilibrium. The preparations and properties of residue compounds of various intercalants are reported, e,g., of Br, 1,6,7,10,15-17 C1 15,18-20 ICI6,17,21, CrO,Cl, 5.6,17, FeCl 6 , 2 2 $ 2 3 1 HNO, 6315,23,24 and H,S04 1 3 1 5 3 2 5 , 2 6 .Conductivity measurements of decomposed samples of KC, indicate K-residue corn pound^^^^^^. The electronic conductivity of residue compounds is several times higher than that of the parent graphite, but still below the conductivity of either donor- or acceptor-type lamellar compounds. Thus conductivity is a sensitive probe for the reversibility of an intercalation reaction. Benzene-solvated K-graphite forms an air-stable second-stage residue compound in which the carbon planes adjacent to the residual intercalant are separated by 575 pm”.

-

-

(J.O. BESENHARD)

G. R. Hennig, Prog. Inorg. Chem., I , 125 (1959). L. B. Ebert, Annu. Rev. Mater. Sci., 6, 181 (1976). H. Selig, L. B. Ebert, Ado. Inorg. Chem. Radiochem., 23, 281 (1980). A. R. Ubbelohde, in Intercalated Layered Materials, Vol. 6, F. Levy, ed., D. Reidel Publishing Company, Dordrecht, 1979. 5. J. G. Hooley, in Preparation and Crystal Growth of Materials with Layered Structures, Vol. 1, R. M. A. Leith, ed., D. Reidel Publishing Company, Dordrecht, 1977. 6. J. G. Hooley, Chem. Phys. Carbon, 5, 321 (1969). 7. J. Maire, J. Mering, Proceedings of the 3rd Conference on Carbon, Buffalo, New York, 1957, Pergamon Press; Oxford, 1959, p. 337. 8. J. G. Hooley, Mater. Sci. Eng., 31, 17 (1977). 9. J. G. Hooley, Carbon, 18, 83 (1980). 10. G. A. Saunders, A R. Ubbelohde, D. A. Young, Proc. Roy. Soc. (London),,4271, 499 (1963). 11. J. 0. Besenhard, H. Mohwald, J. J. Nickl, Synth. Met., 3, 187 (1981). 12. W. Bibcrachen, A. Lerf, J. 0.Basenhard, H. Mohwald, T. Buty, Mater. Res. Bull. 17,1385 (1982). 13. J. 0. Basenhard, E. Wudy, H. Mohwald, J. J. Nickl, W. Bibcracker, W. Foug, Synth. Met., 7, 185 (1983). 14. J. 0. Besenhard, H. Mohwald, J. J. Nickl, Rel;. Chim. Mineral., 19, 588 (1982). 15. Grnelins Handbuch der Anorganrschen Chernie, Vol. 14, Part B-3, Verlag Chemie, Weinheim 1968. Review and data collection. 16. L. H. Reyerson, J. E. Wertz, W. Weltner, H. Whitehurst, J . Phys. Chem., 61, 1334 (1957). 17. J. G. Hooley, Can. J. Chem., 40, 745 (1962). 18. A. HCrold, Bull. Soc. Chim. Fr., 22, 999 (1955). 19. R. Juza, P. Jonck, A. Schmeckenbecher, 2. Anorg. Allg. Chem., 292, 34 (1957). 20. R. Juza, A. Schmeckenbecher, Z. Anorg. Allg. Chem., 292,46 (1957). 21. M. Saito, T. Tsuzuku, Carbon, 15, 347 (1977). 22. J. G. Hooley, M. Bartlett, Carbon, 5, 417 (1967). 23. M. Inagaki, Carbon, 5, 317 (1967). 24. M. Inagaki, J. C. Rouillon, G. Fug, P. Delhaes, Carbon, 15, 181 (1977). 25. G. Hennig, J. Chem. Phys., 19, 922 (1951). 26. M. L. Dzurus, G. R. Hennig, J. Chem. Phys., 27, 275 (1957). 27. A. S. Fialkov, T. N. Zhuikova, T. K. Kaz’mina, N. A. Savost’yanova, Yu. N. Novlkov, Inorg. Mater. (Engl. Transl.), 14, 1432 (1978). 28. J. 0. Besenhard, H.-F. Klein, J. Gross, H. Mohwald, J. J. Nickl, Synth. Met., 4, 51 (1981). 1. 2. 3. 4.

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc. 274

16.4. The Formation of Sheet Structures 16.4.2. Graphite and Boron Nitride Intercalation Compounds 16.4 2.7. Residue and Surface Compounds

16.4.2.7.3. Preparation of Surface Compounds.

The surface chemistry of black carbon has been reviewed c~mprehensivelyl-~; electrochemical surface reactions of carbon have been surveyed4x5. Since surface reactions of carbon are reactions of defect sites, molar amounts of adsorbate are taken up only by microcrystalline materials such as charcoal or carbon black with high concentrations of defects and large internal and external surface areas per weight unit. Surface compounds on carbon are of technological interest' in heterogeneous catalysis, rubber reinforcement, composites from carbon fiber-reinforced polymer^^-^, printing inks and in the development of surface-modified electrodeslO-lz. (i) Surface Oxides. Surface oxides of carbon are different from bulk graphite oxides (cf. 516.4.2.5). Two classes of surface oxides can be distinguished: basic oxides, likely consisting of pyrone-type structures13, are formed when carbon freed from surface compounds by heating in vacuo comes into contact with oxygen at low T; the irreversible uptake of oxygen starts at -40°C ' , I 4 . The basic oxides adsorb acids in the presence of oxygen and liberate H,O, when acid is adsorbed'. Acidic oxides are formed when carbon is treated with oxygen (in the presence of some H,O) near its ignition point15; acidic groups are also formed upon reaction with oxidizing solutions (e.g., and conc18 HNO,, [Mn04]-16, [Cr04]'-16, [OCl]-16, [S,0,]2-'6 and other^','^,'^) at RT. Some acidic functional groups that are present on oxidized carbon surfaces are presented in Fig. 1. The same functional groups, although in much lower concentration, are found on surface-oxidized graphite; in addition, there are hydroperoxidesl. Electrochemical oxidation of carbon in 1 mol L-'H,SO, is claimed to result in hydroquinone- or quinone-like groups in different e n v i r ~ n m e n t s ~ J -however ~ ~ ; there is not much support for these findings in later ~ o r k ' ~ , 'The ~ . anodically produced surface film differs from that chemically formed5.

-

Open forms

Lactone forms

Figure 1. Model of functional groups in acidic surface oxides of carbon (after ref. 20).

16.4.2. Graphite and Boron Nitride Intercalation Compounds 16.4.2.7. Residue and Surface Compounds 16.4.2.7.3. Preparation of Surface Compounds.

27 5

A pretreatment that makes even basal plane carbon atoms accessible to oxidation is by electrochemical oxidation-reduction-cycles in 97 % H,SO, 2 6 ; oxidation of basal plane carbon atoms is also possible by ozonez7. The maximum oxygen concentration in surface oxides depends on the specific surface area of the parent carbon material; values of 15 ' to 25 % oxygen are reported for sugar charcoal. An oxygen content in this range corresponds to an acid content (determined by neutralization with NaOEt) of 2 to 3 mequiv g- l . Surface oxides on graphite undergo the reactions typical of carboxylic acids, phenols and alcohols. Methylation with diazomethaneZ8or dimethylsulfate, silanization of -OH groups by reaction with halosilanes, formation of acyl chlorides by reaction with thionyl chloride, further reactions of these acyl chlorides, neutralization of acidic surface groups with different bases, ion exchange2' and more special reactions are used to characterize surface oxides on carbon and to modify their properties for technical application^'^',^.^^. High surface area carbons must be purified prior to further reaction, e.g., by extraction with xylene, followed by outgassing in vacuo at 1000-1200°C; at > 12OO0C, most carbons start recrystallizing'. (ii) Surface Sulfides. Reaction of charcoal with elementary sulfur at 600°C yields a solid product containing -20 wt. % sulfurlS2.At least part of this sulfur is chemically bonded to the surface3'; fixing of sulfur is also by capillary condensation and physical adsorption3'. Surface sulfides are also prepared by reacting of carbon with CS,, H,S and

- -

'

so, 'J.

(iii) Other Surface Groups. Besides oxygen, hydrogen is present in high surface area carbon in contact with air (see surface oxides). Hydrogen directly bonded to edge carbon is obtained by reaction of carbon with hydrogen at 1035-1375"C32. Ammonia and amines are adsorbed on acidic surface oxides',', but outgassed charcoal also fixes some nitrogen species on treatment with NH, at 750-900"C33. There are indications that nitrogen is present as nitrile groups'. The reactivity of halogens on carbon decreases C1 > Br > I '; however, the intercalation of halogens has to be considered also besides surface reactions34. On treatment of carbon black with chlorine the maximum chlorine uptake ( - 20 wt %) occurs at 450°C A microwave-induced fluorination of basal-plane carbon surfaces also occurs35. (iv) Radicals. ESR shows the existence of free radicals in carbons. The unpaired electrons are attributed to surface groups (e.g., aroxyls). Their concentration can be determined by combination with other free r a d i c a l s ' ~ ~ ~ .

'.

(J.O. BESENHARD)

1. H. P. Boehm, Adc. Catal., 16, 179 (1966). Recommended reading.

2. B. R. Puri, Chem. Phys. Carbon, 6, 191 (1970). Revieg. 3. H. P. Boehm, H. Knozinyer, Catalysis-Science and Technology, Vol. 4, J. R. Anderson, M.

Boudart, eds., Springer-Verlag, Berlin, 1983, p. 40. 4. R. E. Panzer, P. J. Elving, Electrochrm. Acta, 20, 635 (1975). Review. 5. J.-P. Randin, in Encyclopedia of Electrochemistry of the Elements, Vol. VII, A. J. Bard, ed., Marcel Dekker, Inc., New York, 1976. Review. 6. K. Wolf, R. E. Fornes, J. D. Memory, R. D. Gilbert, Chem. Phys. Carbon, 18, 33 (1982). 7. I. L. Kalnin, H. Jager, in Carbon Fibres and Their Composites, E. Fitzer, ed., Springer-Verlag, Berlin, 1985, p. 62 8. J. G. Morley, High Performance Fibre Composites, Academic Press, London, 1987. 9. E. Fitzer, M. Heine, in Fibre Reinforcement for Composite Materials, A. R. Burnsell, ed., Elsevier, Amsterdam, 1988, p. 73. 10. K. D. Snell, A. G . Keenan, Chem. SOC.Rev., 8, 259 (1979). Review.

16.4. The Formation of Sheet Structures 16.4.2. Graphite and Boron Nitride Intercalation Compounds 16.4.2.8. Electrochemical Formation

276

11. 12. 13. 14.

15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36.

R. W. Murray, Ace. Chem. Res., 13, 135 (1980). Review. J. Schreurs, E. Barendrecht, Reel. Trav. Chim. Pays-Bas, 103, 205 (1984). M. Voll, H. P. Boehm, Carbon 7,481 (1971). G. G. Fedorov, Ju. A. Zarif'yants, V. F. Kiselev, Russ. J . Phys. Chem. [Engl. Transl.), 37, 1267 (1963). V. A. Garten, D. E. Weiss, Austr. J. Chem., 10, 309 (1957). J. B. Donnet, F. Hueber, C. Reitzer, J. Oddous, G. Riess, Bull. Soc. Chim. Fr., 29, 1727 (1962). E. Fitzer, K.-H. Geigl, W. Huttner, R. Weiss, Carbon, 18, 389 (1980). U. Hofmann, G. Ohlerich, Angew. Chem., 62, 16 (1950). J. B. Donnet, P. Ehrburger, Carbon, 15, 143 (1977). H. P. Boehm, Angew. Chenz., In?. Ed. Engl., 5, 533 (1966). Review. B. D. Epstein, E. Dalle-Molle, J. S. Mattson, Carbon, 9, 609 (1971). K. F. Blurton, Electrochim. Acta, 18, 869 (1973). K. Kinoshita, J. A. S. Bett, Carbon, 11, 403 (1973). E. Theodorideu, A. D. Jannakoudakis, J. 0. Besenhard, R. F. Sauter, Synth. Metals, 14, 125 (1986). J. 0. Besenhard, H. P. Fritz, Angew. Chem., Int. Ed. Engl., 22, 950 (1983). E. Theodoridou, J. 0. Besenhard, H. P. Fritz, J. Electroanal. Chem., 122, 67 (1984). G. I. Emel'yanova, L. F. Atyaksheva, T. I. Treylazova, in Mater.-Vses. Mezhvuz. KonJ Ozonu, 2nd, Yu. A. Mal'tsev, ed., Moskovskii Gos. Univ., Moscow, 1977, p. 12; Chem.Abstr., 89,165,586 (1978). S. S. Barton, D. J. Gillespie, B. H. Harrison, W. Kemp, Carbon, 18, 363 (1978). P. A. Attwood, R. J. Bird, B. D. McNicol, R. T. Short, J. Chem. Soc., Faraday Trans. I, 75,2312 (1979). K. W. Sykes, P. White, Trans. Faraday Soc., 52, 660 (1956). R. Juza, W. Blanke, Z. Anorg. Allg. Chem., 210, 81 (1933). J. P. Redmond, P. L. Walker Jr., J. Phys. Chem., 64, 1093 (1960). P. H. Emmett, Chem. Rev., 43, 69 (1948). J. G. Hooley, Can. J. Chem., 37, 899 (1959). P. Cadman, J. D. Scott, J. M. Thomas, SurJ Interface Anal., I , 115 (1979); Chem. Abstr., 93, 159,090 (1980). J. P. Donnet, G. Henrich, Bull. Soc. Chim. Fr., 27, 1609 (1960).

16.4.2.8 Electrochemical Formation

For the electrochemical formation of surface compounds of graphite see 516.4.2.7. Electrochemical intercalation reactions of boron nitride, which is practically an insulator, have not yet been reported. The electrochemistry of black carbons has been reviewed

'.

(J.0 BESENHARD)

1. J. 0. Besenhard, H. P. Fritz, Angew. Chem., Int. Ed. Engl., 22, 950 (1983).

16.4.2.8.1. Strengths and Limitations of the Electrochemical Technique.

Cation- and anion-graphite intercalation compounds can be prepared by electrochemical reduction or oxidation of graphite in appropriate electrolytes. Such electrochemical solid-state reactions require mixed conductivity of the solid, i.e., existence of electronic and ionic conductivity. Electrochemical reduction or oxidation of graphite, in contrast to chemical reactions, makes it possible to control n in C,X compounds simply by controlling the current supply. Moreover, the stoichiometry' and thermodynamic constants' of graphite com-

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc. 16.4. The Formation of Sheet Structures 16.4.2. Graphite and Boron Nitride Intercalation Compounds 16.4.2.8. Electrochemical Formation

276

11. 12. 13. 14.

15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36.

R. W. Murray, Ace. Chem. Res., 13, 135 (1980). Review. J. Schreurs, E. Barendrecht, Reel. Trav. Chim. Pays-Bas, 103, 205 (1984). M. Voll, H. P. Boehm, Carbon 7,481 (1971). G. G. Fedorov, Ju. A. Zarif'yants, V. F. Kiselev, Russ. J . Phys. Chem. [Engl. Transl.), 37, 1267 (1963). V. A. Garten, D. E. Weiss, Austr. J. Chem., 10, 309 (1957). J. B. Donnet, F. Hueber, C. Reitzer, J. Oddous, G. Riess, Bull. Soc. Chim. Fr., 29, 1727 (1962). E. Fitzer, K.-H. Geigl, W. Huttner, R. Weiss, Carbon, 18, 389 (1980). U. Hofmann, G. Ohlerich, Angew. Chem., 62, 16 (1950). J. B. Donnet, P. Ehrburger, Carbon, 15, 143 (1977). H. P. Boehm, Angew. Chenz., In?. Ed. Engl., 5, 533 (1966). Review. B. D. Epstein, E. Dalle-Molle, J. S. Mattson, Carbon, 9, 609 (1971). K. F. Blurton, Electrochim. Acta, 18, 869 (1973). K. Kinoshita, J. A. S. Bett, Carbon, 11, 403 (1973). E. Theodorideu, A. D. Jannakoudakis, J. 0. Besenhard, R. F. Sauter, Synth. Metals, 14, 125 (1986). J. 0. Besenhard, H. P. Fritz, Angew. Chem., Int. Ed. Engl., 22, 950 (1983). E. Theodoridou, J. 0. Besenhard, H. P. Fritz, J. Electroanal. Chem., 122, 67 (1984). G. I. Emel'yanova, L. F. Atyaksheva, T. I. Treylazova, in Mater.-Vses. Mezhvuz. KonJ Ozonu, 2nd, Yu. A. Mal'tsev, ed., Moskovskii Gos. Univ., Moscow, 1977, p. 12; Chem.Abstr., 89,165,586 (1978). S. S. Barton, D. J. Gillespie, B. H. Harrison, W. Kemp, Carbon, 18, 363 (1978). P. A. Attwood, R. J. Bird, B. D. McNicol, R. T. Short, J. Chem. Soc., Faraday Trans. I, 75,2312 (1979). K. W. Sykes, P. White, Trans. Faraday Soc., 52, 660 (1956). R. Juza, W. Blanke, Z. Anorg. Allg. Chem., 210, 81 (1933). J. P. Redmond, P. L. Walker Jr., J. Phys. Chem., 64, 1093 (1960). P. H. Emmett, Chem. Rev., 43, 69 (1948). J. G. Hooley, Can. J. Chem., 37, 899 (1959). P. Cadman, J. D. Scott, J. M. Thomas, SurJ Interface Anal., I , 115 (1979); Chem. Abstr., 93, 159,090 (1980). J. P. Donnet, G. Henrich, Bull. Soc. Chim. Fr., 27, 1609 (1960).

16.4.2.8 Electrochemical Formation

For the electrochemical formation of surface compounds of graphite see 516.4.2.7. Electrochemical intercalation reactions of boron nitride, which is practically an insulator, have not yet been reported. The electrochemistry of black carbons has been reviewed

'.

(J.0 BESENHARD)

1. J. 0. Besenhard, H. P. Fritz, Angew. Chem., Int. Ed. Engl., 22, 950 (1983).

16.4.2.8.1. Strengths and Limitations of the Electrochemical Technique.

Cation- and anion-graphite intercalation compounds can be prepared by electrochemical reduction or oxidation of graphite in appropriate electrolytes. Such electrochemical solid-state reactions require mixed conductivity of the solid, i.e., existence of electronic and ionic conductivity. Electrochemical reduction or oxidation of graphite, in contrast to chemical reactions, makes it possible to control n in C,X compounds simply by controlling the current supply. Moreover, the stoichiometry' and thermodynamic constants' of graphite com-

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc. 16.4. The Formation of Sheet Structures 16.4.2. Graphite and Boron Nitride Intercalation Compounds 16.4.2.8. Electrochemical Formation

276

11. 12. 13. 14.

15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36.

R. W. Murray, Ace. Chem. Res., 13, 135 (1980). Review. J. Schreurs, E. Barendrecht, Reel. Trav. Chim. Pays-Bas, 103, 205 (1984). M. Voll, H. P. Boehm, Carbon 7,481 (1971). G. G. Fedorov, Ju. A. Zarif'yants, V. F. Kiselev, Russ. J . Phys. Chem. [Engl. Transl.), 37, 1267 (1963). V. A. Garten, D. E. Weiss, Austr. J. Chem., 10, 309 (1957). J. B. Donnet, F. Hueber, C. Reitzer, J. Oddous, G. Riess, Bull. Soc. Chim. Fr., 29, 1727 (1962). E. Fitzer, K.-H. Geigl, W. Huttner, R. Weiss, Carbon, 18, 389 (1980). U. Hofmann, G. Ohlerich, Angew. Chem., 62, 16 (1950). J. B. Donnet, P. Ehrburger, Carbon, 15, 143 (1977). H. P. Boehm, Angew. Chenz., In?. Ed. Engl., 5, 533 (1966). Review. B. D. Epstein, E. Dalle-Molle, J. S. Mattson, Carbon, 9, 609 (1971). K. F. Blurton, Electrochim. Acta, 18, 869 (1973). K. Kinoshita, J. A. S. Bett, Carbon, 11, 403 (1973). E. Theodorideu, A. D. Jannakoudakis, J. 0. Besenhard, R. F. Sauter, Synth. Metals, 14, 125 (1986). J. 0. Besenhard, H. P. Fritz, Angew. Chem., Int. Ed. Engl., 22, 950 (1983). E. Theodoridou, J. 0. Besenhard, H. P. Fritz, J. Electroanal. Chem., 122, 67 (1984). G. I. Emel'yanova, L. F. Atyaksheva, T. I. Treylazova, in Mater.-Vses. Mezhvuz. KonJ Ozonu, 2nd, Yu. A. Mal'tsev, ed., Moskovskii Gos. Univ., Moscow, 1977, p. 12; Chem.Abstr., 89,165,586 (1978). S. S. Barton, D. J. Gillespie, B. H. Harrison, W. Kemp, Carbon, 18, 363 (1978). P. A. Attwood, R. J. Bird, B. D. McNicol, R. T. Short, J. Chem. Soc., Faraday Trans. I, 75,2312 (1979). K. W. Sykes, P. White, Trans. Faraday Soc., 52, 660 (1956). R. Juza, W. Blanke, Z. Anorg. Allg. Chem., 210, 81 (1933). J. P. Redmond, P. L. Walker Jr., J. Phys. Chem., 64, 1093 (1960). P. H. Emmett, Chem. Rev., 43, 69 (1948). J. G. Hooley, Can. J. Chem., 37, 899 (1959). P. Cadman, J. D. Scott, J. M. Thomas, SurJ Interface Anal., I , 115 (1979); Chem. Abstr., 93, 159,090 (1980). J. P. Donnet, G. Henrich, Bull. Soc. Chim. Fr., 27, 1609 (1960).

16.4.2.8 Electrochemical Formation

For the electrochemical formation of surface compounds of graphite see 516.4.2.7. Electrochemical intercalation reactions of boron nitride, which is practically an insulator, have not yet been reported. The electrochemistry of black carbons has been reviewed

'.

(J.0 BESENHARD)

1. J. 0. Besenhard, H. P. Fritz, Angew. Chem., Int. Ed. Engl., 22, 950 (1983).

16.4.2.8.1. Strengths and Limitations of the Electrochemical Technique.

Cation- and anion-graphite intercalation compounds can be prepared by electrochemical reduction or oxidation of graphite in appropriate electrolytes. Such electrochemical solid-state reactions require mixed conductivity of the solid, i.e., existence of electronic and ionic conductivity. Electrochemical reduction or oxidation of graphite, in contrast to chemical reactions, makes it possible to control n in C,X compounds simply by controlling the current supply. Moreover, the stoichiometry' and thermodynamic constants' of graphite com-

16.4.2. Graphite and Boron Nitride Intercalation Compounds 16.4.2.8. Electrochemical Formation 16.4.2.8.1. Strengths and Limitations of the Electrochemical Technique.

277

pounds as well as some kinetic information can be obtained when current-potential-time relations are monitored in electrochemical intercalation or deintercalation reactions. The one main disadvantage of electrochemical as compared to chemical preparations is the difficulty in scaling up since, due to transport limitations, the size of the electrode cannot be increased as desired. This is especially troublesome with poorly oriented graphite, since the mobility of ions in graphite compounds, and consequently the rate of electrochemical intercalation and deintercalation reactions, decreases with decreasing order of the host lattice. The reactivity of poorly oriented material, however, can be increased considerably by prior electrochemical intercalation-deintercalation cycles3. A favorable parent material for electrochemical preparations of graphite compounds is graphite foil; owing to high porosity (- 50 %), the solid state ionic transport (i.e., the ionic transport through the interlayer gap) is enhanced by liquid-phase ionic transport in the pores of the electrode. A means of reacting small particle-size graphite is by a suspended electrode, where a graphite suspension in the electrolyte is stirred onto an inert substrate electrode. The most conventional method for electrochemically preparing graphite compounds is constant-current reduction or oxidation of weighed graphite samples for a To avoid material losses because of extensive exfoliation and definite period of degradation, graphite electrodes are usually sandwiched between porous glass, etc4. At low current density, i.e., close to equilibrium, the formation of stages can be followed on a continuous-potential recording (Fig. 1). Cyclic voltammetry is another powerful method for investigating electrode reactions of graphite, Figure 2 shows cyclic voltammograms of a Pt and a graphite electrode,

Equiv. of current

(C2;

= 1)

Figure 1. Potential of a graphite electrode during anodic oxidation in 100% H,SO, (after ref. 1; see also refs. 5 and 6).

16.4. The Formation of Sheet Structures 16.4.2. Graphite and Boron Nitride Intercalation Compounds 16.4.2.8. Electrochemical Formation

278

?

L

2(

30

1(

15

C

0

-1c

-1 5

-20

-3 0

-3 0

-45

3 0

-40

-2

-3

-1

0

Potential (V vs. SCE ) Figure 2. Cyclic voltammograms of Pt and graphite foil in saturated NMe,CI-DMSO; sweep rate SO mV s-

l,

SCE = saturated calomel electrode (after ref. 7).

16.4.2. Graphite and Boron Nitride Intercalation Compounds 16.4.2.8. Electrochemical Formation 16.4.2.8.1, Strengths and Limitations of the Electrochemical Technique.

A+:

cations stable vs. reduction, e.g., [alkali]+, [NR,]+

X-:

Anions stable vs. oxidation.

279

( s o h ) : solvent stable vs. reduction or oxidation, respectively Figure 3. Class&sof lamellar graphite compounds accessible by electrochemical reduction of oxidation of graphite. respectively, in a NMe,Cl-dimethylsulfoxide (DMSO) electrolyte. Graphite undergoes reduction in this medium in a reversible reaction, whereas at Pt[NMe,]+ is reduced irreversibly. The range of lamellar compounds that can be prepared by electrochemical reactions of graphite is shown schematically in Fig. 3. Reduction yields lamellar cation intercalation compounds that can be solvated. Solvated cation compounds are formed in donor electrolytes, i.e., solutions of alkali- or NR,-salts in solvents such as NH, or DMSO some binary (= unsolvated) cation compounds can be obtained in molten salt or solid electrolytes. Oxidation yields lamellar anion intercalation compounds that, again, depending on the electrolyte, may be solvated. Solvated anion compounds are formed, e.g., in acids, HA (HA = solvent), and in propylene carbonate (PC) electrolytes'0; binary anion compounds can be obtained, e.g., in S0,ClF electrolytes3. "Overoxidation" and solvolysis of graphite salts may yield covalent layered compounds with a fraction of sp3 carbon, e.g., graphite oxide (cf. §16.4.2.5), which is formed in H,O-containing acids"-13. Any ionic graphite compound may be formed electrochemically, but since most of these compounds are also accessible by other methods, the following examples are restricted to compounds that have not yet been prepared otherwise or that may be prepared electrochemically with advantage. Applications of electrochemical reactions of graphite and graphite compounds have been r e ~ i e w e d ' ~ ~ ' ~ . ,s9;

(J 0 BESENHARD)

1. M. J. Bottomley, G. S. Parry, A. R. Ubbelohde, D. A. Young, J. Chem. SOC.,5674 (1963).

Classical paper.

2. S. Aronson, C. Frishberg, G. Frankl, Carhon, 9, 715 (1971).

16.4 The Formation of Sheet Structures 16.4.2. Graphite and Boron Nitride Intercalation Compounds 16.4.2.8. Electrochemical Formation

280

3. 4. 5. 6. 7 8. 9. 10. 11. 12. 13. 14. 15. 16.

J. 0. Besenhard, H. Mohwald, J. J. Nickl, Synth. Met., 3, 187 (1981). J. 0. Besenhard, Carbon, 14, 111 (1976). W. Biberacher,A. Lerf, J. 0.Besenhard, H. Mohwald, T. Buty, Mater. Res. Bull., 17,1385 (1982). B. Bouyad, H. Fugellier, M. LeLaurain, A. Metrot, F. Rousseaux, Synth. Met. 7, 325 (1983). J. 0. Besenhard. H. P. Fritz, J. Electroanal. Chem., 53, 329 (1974). W. F. K. Wynne-Jones,Proc. Symp. Colston Rex Soc., 10, 35 (1958); Chem. Abstr., 53, 21,046 (1959). J. 0. Besenhard, H. Mohwald, J. J. Nickl, Carbon, 18, 399 (1980). J. 0. Besenhard, H. P. Fritz, H. Mohwald, J. J. Nickl, 3rd Int. Carbon Con$, Baden-Baden (1980), Preprints, p. 147. H. P. Boehm, M. Eckel, W. Scholz, Z. Anorg. Allg. Chem., 353, 236 (1967). A. Mktrot, J. E. Fischer, Synth. Met., 3, 201 (1981). J. 0. Besenhard, H. P. Fritz, Z. Anorg. Allg. Chem., 416, 106 (1975). H. Selig, L. B. Ebert, AdL. Inorg. Chem. Radiochem., 23, 281 (1980). J. 0. Besenhard, E. Theodoridou, H. Mohwald, J. J. Nickl, Synth. Met., 4, 211 (1982). S. Flandrois, J. Herran, Synth. Met., 14, 103 (1986).

16.4.2.8.2. Cathodic Reduction of Graphite.

Preparation of binary-alkali metal graphites by cathodic reduction of graphite in molten salt electrolytes is possible in principle's2 but has no advantage over conventional vapor-phase intercalation, particularly because of difficulties related to the separation of the product from the melt. Cathodic intercalation of alkali cations from solid3 or organic polymer4 electrolyte is also possible5. Cathodic reduction of graphite in dipolar aprotic solutions of alkali and [NR,] ' salts (e.g., in ethers, esters, amides) usually yields solvated cation intercalation compound^^-^; however, at high cation density the solvent may be squeezed out'. The corresponding phosphonium and sulfonium graphite intercalation compounds are unstable; their formation is accompanied by reductive decomposition of these cation^^,'^.

Alternative methods for preparing solvated alkali-metal graphites include either secondary solvation of the binary compounds or reduction of graphite by solutions of the alkali metal in electron-solvating media such as hexamethylphosphorus triamide (HMPT)'' or polycyclic aromatic systems in ether^",'^, leading, e.g.. to HMPT- or ether-solvated compounds. Early work on cathodic reduction of graphite in solutions of NR, and alkali salts in liq NH, verified the formation of unstable NH,-solvated cation intercalation compound^'^, but the DMSO-solvated compounds A+(DMSO),C; (A' = Li', Na-, K', Rb', Cs' or various quarternary ammonium cations)5b6are more stable15. These are prepared by constant current reduction (counterelectrode separated from the working electrode by a diaphragm, typical current density 0.1-0.5 mA cm-') of graphite foil or natural flake graphite (pressed in stainless steel gauze) in purified (trace water!) 0.5 mol L-' solutions of A' iodides, perchlorates or hexafluorophosphates in DMSO '. From their large interplanar distances (e.g., 1493 pm for first-stage K'(DMSO),Ci4, 1582 pm for [NMe4]+(DMSO),C&J6 it was concluded that the intercalated cations are three-dimensionally solvated with DMSO. In contrast, the interplanar distances of tetrahydrofuran (THF)- and 1,2-dimethoxyethane (DME)solvated compounds correspond just to the van der Waals diameter of the solvent molecule^'^^'^; i.e., the cations in the interlayer gap are only two-dimensionally solvated.

-

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc. 16.4 The Formation of Sheet Structures 16.4.2. Graphite and Boron Nitride Intercalation Compounds 16.4.2.8. Electrochemical Formation

280

3. 4. 5. 6. 7 8. 9. 10. 11. 12. 13. 14. 15. 16.

J. 0. Besenhard, H. Mohwald, J. J. Nickl, Synth. Met., 3, 187 (1981). J. 0. Besenhard, Carbon, 14, 111 (1976). W. Biberacher,A. Lerf, J. 0.Besenhard, H. Mohwald, T. Buty, Mater. Res. Bull., 17,1385 (1982). B. Bouyad, H. Fugellier, M. LeLaurain, A. Metrot, F. Rousseaux, Synth. Met. 7, 325 (1983). J. 0. Besenhard. H. P. Fritz, J. Electroanal. Chem., 53, 329 (1974). W. F. K. Wynne-Jones,Proc. Symp. Colston Rex Soc., 10, 35 (1958); Chem. Abstr., 53, 21,046 (1959). J. 0. Besenhard, H. Mohwald, J. J. Nickl, Carbon, 18, 399 (1980). J. 0. Besenhard, H. P. Fritz, H. Mohwald, J. J. Nickl, 3rd Int. Carbon Con$, Baden-Baden (1980), Preprints, p. 147. H. P. Boehm, M. Eckel, W. Scholz, Z. Anorg. Allg. Chem., 353, 236 (1967). A. Mktrot, J. E. Fischer, Synth. Met., 3, 201 (1981). J. 0. Besenhard, H. P. Fritz, Z. Anorg. Allg. Chem., 416, 106 (1975). H. Selig, L. B. Ebert, AdL. Inorg. Chem. Radiochem., 23, 281 (1980). J. 0. Besenhard, E. Theodoridou, H. Mohwald, J. J. Nickl, Synth. Met., 4, 211 (1982). S. Flandrois, J. Herran, Synth. Met., 14, 103 (1986).

16.4.2.8.2. Cathodic Reduction of Graphite.

Preparation of binary-alkali metal graphites by cathodic reduction of graphite in molten salt electrolytes is possible in principle's2 but has no advantage over conventional vapor-phase intercalation, particularly because of difficulties related to the separation of the product from the melt. Cathodic intercalation of alkali cations from solid3 or organic polymer4 electrolyte is also possible5. Cathodic reduction of graphite in dipolar aprotic solutions of alkali and [NR,] ' salts (e.g., in ethers, esters, amides) usually yields solvated cation intercalation compound^^-^; however, at high cation density the solvent may be squeezed out'. The corresponding phosphonium and sulfonium graphite intercalation compounds are unstable; their formation is accompanied by reductive decomposition of these cation^^,'^.

Alternative methods for preparing solvated alkali-metal graphites include either secondary solvation of the binary compounds or reduction of graphite by solutions of the alkali metal in electron-solvating media such as hexamethylphosphorus triamide (HMPT)'' or polycyclic aromatic systems in ether^",'^, leading, e.g.. to HMPT- or ether-solvated compounds. Early work on cathodic reduction of graphite in solutions of NR, and alkali salts in liq NH, verified the formation of unstable NH,-solvated cation intercalation compound^'^, but the DMSO-solvated compounds A+(DMSO),C; (A' = Li', Na-, K', Rb', Cs' or various quarternary ammonium cations)5b6are more stable15. These are prepared by constant current reduction (counterelectrode separated from the working electrode by a diaphragm, typical current density 0.1-0.5 mA cm-') of graphite foil or natural flake graphite (pressed in stainless steel gauze) in purified (trace water!) 0.5 mol L-' solutions of A' iodides, perchlorates or hexafluorophosphates in DMSO '. From their large interplanar distances (e.g., 1493 pm for first-stage K'(DMSO),Ci4, 1582 pm for [NMe4]+(DMSO),C&J6 it was concluded that the intercalated cations are three-dimensionally solvated with DMSO. In contrast, the interplanar distances of tetrahydrofuran (THF)- and 1,2-dimethoxyethane (DME)solvated compounds correspond just to the van der Waals diameter of the solvent molecule^'^^'^; i.e., the cations in the interlayer gap are only two-dimensionally solvated.

-

16 4.2. Graphite and Boron Nitride Intercalation Compounds 16.4.2 8 Electrochemical Formation 16 4.2.8.3. Anodic Oxidation of Graphite.

281

For DME-solvated Li compounds, Li+(DME),C, was found by constant current titration' to be the limiting stoichiometry. The standard potentials of solvated alkali-metal graphites are more positive vs. the free alkali metals than those of binary alkali-metal graphites, but there is a decrease in the positive shift from DMSO to DME5. Solvated alkali-metal graphites are more ionic in character than unsolvated ones; this is indicated by their electronic properties6 and also by the small H, evolution on reaction with water, which may be regarded as a probe for ionicity18. (J.0 BESENHARD)

A. Htrold, Bull. Soc. Chim. Fr., 999 (1955). Y. N. Novikov, M. E. Vol'pin, Russ. Chem. Rev. (Engl. Transl.), 40, 733 (1971). P. Pfluger, V. Geiser, S. Stolz, H.-J. Guntherodt, Synth. Met., 3, 27 (1981). R. Yoyami, in Chemical Physics oflnteiculaiion, A. P. Lagrand, S.Flanders, eds., Plenum Press, New York, 1987, p. 457. 5. J. 0. Besenhard, Carbon, 14>111 (1976). 6. J. 0. Besenhard, H Mohwald, J. J. Nickl, Carbon, 18, 399 (1980). 7. J. Simonet, H. Lund, J. Electroanal. Chem., 75, 719 (1977). 8. Ph. Touyain, B. Marcus, Y. Maeda; L. Bonnetain, in International Colloquium on Layered Compounds, D. Guerard, P Lagrange, eds., Universite de Nancy I, Nancy, 1988, p. 131 9 E. A. H. Hall, J. Simonet, H. Lund, J. Electroanal. Chem., 100, 197 (1979). 10. G. Bernard, J. Simonet, J Electroanal. Chern., 112, 117 (1980). 11. D. Ginderow, R. Setton, Carbon, 6, 5 (1971). 12. C. Stein, L. Bonnetain, J. Gole, Bull. Soc. Chim. Fr., 3166 (1966). 13. Co-Minh-Duc, J. Gole, J. Chim. Phys., 69, 986 (1972). 14. 0. Rubisch, Ph. D. Thesis; Univ. Tubingen (1957), H. H. Sick, Ph. D. Thesis, Univ. Tubingen (1959). 15. It is questionable whether any of the solvated cation intercalation compounds of graphite may be thermodynamically stable vs. self-decomposition by internal reduction of the intercalated polar solvent. 16. F. Btguin, R. Setton, Carbon, 13, 293 (1975). 17. J. Amiell, P. Belhaes, F. Btguin, R. Setton, Mater. Sci. E n g , 31, 243 (1977). 18. D. E. Bergbreitcr, J. M. Killough, J. Am. Chern. Soc., 100, 2126 (1978). 1. 2. 3. 4.

16.4.2.8.3. Anodic Oxidation of Graphite. (I) Formation of Acid Salts of Graphite. Anodic oxidation of graphite in strong protonic acids, HA, is the favored preparation of acid salts C,'(HA),A- (see 516.4.2.1.4). In contrast, chemical oxidation or chemical oxidation followed by anion exchange is less universal and yields contaminated products. Examples for well-characterized acid salts of graphite prepared by anodic oxidation are listed in Table 1; in first-stage compounds n is -24. There is some evidence for a first-stage salt with n 12 obtained by anodic oxidation in H,SO, Salt formation is also observed with graphite anodes in C1S0,H9, H,Se0,3, HI0,3, H , A s O , ~ and CF3COOH l o , perfluoroalkane sulfonic acids and alkane sulfonic acids1'. Since hydrolysis of graphite salts t o graphite oxide (see 516.4.2.3) is a common side reaction, the acid should be as concentrated as possible, but a first-stage salt may be prepared in 70 % HClO, and even in less concentrated acids the anodic intercalation of anions is possible'2,'3. Anodic oxidation of graphitic materials in H,SO, is also used as a test reaction to characterize their degree of orientation and defect concentration'4s15.Ether-solvated

-

'3'.

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc. 16 4.2. Graphite and Boron Nitride Intercalation Compounds 16.4.2 8 Electrochemical Formation 16 4.2.8.3. Anodic Oxidation of Graphite.

281

For DME-solvated Li compounds, Li+(DME),C, was found by constant current titration' to be the limiting stoichiometry. The standard potentials of solvated alkali-metal graphites are more positive vs. the free alkali metals than those of binary alkali-metal graphites, but there is a decrease in the positive shift from DMSO to DME5. Solvated alkali-metal graphites are more ionic in character than unsolvated ones; this is indicated by their electronic properties6 and also by the small H, evolution on reaction with water, which may be regarded as a probe for ionicity18. (J.0 BESENHARD)

A. Htrold, Bull. Soc. Chim. Fr., 999 (1955). Y. N. Novikov, M. E. Vol'pin, Russ. Chem. Rev. (Engl. Transl.), 40, 733 (1971). P. Pfluger, V. Geiser, S. Stolz, H.-J. Guntherodt, Synth. Met., 3, 27 (1981). R. Yoyami, in Chemical Physics oflnteiculaiion, A. P. Lagrand, S.Flanders, eds., Plenum Press, New York, 1987, p. 457. 5. J. 0. Besenhard, Carbon, 14>111 (1976). 6. J. 0. Besenhard, H Mohwald, J. J. Nickl, Carbon, 18, 399 (1980). 7. J. Simonet, H. Lund, J. Electroanal. Chem., 75, 719 (1977). 8. Ph. Touyain, B. Marcus, Y. Maeda; L. Bonnetain, in International Colloquium on Layered Compounds, D. Guerard, P Lagrange, eds., Universite de Nancy I, Nancy, 1988, p. 131 9 E. A. H. Hall, J. Simonet, H. Lund, J. Electroanal. Chem., 100, 197 (1979). 10. G. Bernard, J. Simonet, J Electroanal. Chern., 112, 117 (1980). 11. D. Ginderow, R. Setton, Carbon, 6, 5 (1971). 12. C. Stein, L. Bonnetain, J. Gole, Bull. Soc. Chim. Fr., 3166 (1966). 13. Co-Minh-Duc, J. Gole, J. Chim. Phys., 69, 986 (1972). 14. 0. Rubisch, Ph. D. Thesis; Univ. Tubingen (1957), H. H. Sick, Ph. D. Thesis, Univ. Tubingen (1959). 15. It is questionable whether any of the solvated cation intercalation compounds of graphite may be thermodynamically stable vs. self-decomposition by internal reduction of the intercalated polar solvent. 16. F. Btguin, R. Setton, Carbon, 13, 293 (1975). 17. J. Amiell, P. Belhaes, F. Btguin, R. Setton, Mater. Sci. E n g , 31, 243 (1977). 18. D. E. Bergbreitcr, J. M. Killough, J. Am. Chern. Soc., 100, 2126 (1978). 1. 2. 3. 4.

16.4.2.8.3. Anodic Oxidation of Graphite. (I) Formation of Acid Salts of Graphite. Anodic oxidation of graphite in strong protonic acids, HA, is the favored preparation of acid salts C,'(HA),A- (see 516.4.2.1.4). In contrast, chemical oxidation or chemical oxidation followed by anion exchange is less universal and yields contaminated products. Examples for well-characterized acid salts of graphite prepared by anodic oxidation are listed in Table 1; in first-stage compounds n is -24. There is some evidence for a first-stage salt with n 12 obtained by anodic oxidation in H,SO, Salt formation is also observed with graphite anodes in C1S0,H9, H,Se0,3, HI0,3, H , A s O , ~ and CF3COOH l o , perfluoroalkane sulfonic acids and alkane sulfonic acids1'. Since hydrolysis of graphite salts t o graphite oxide (see 516.4.2.3) is a common side reaction, the acid should be as concentrated as possible, but a first-stage salt may be prepared in 70 % HClO, and even in less concentrated acids the anodic intercalation of anions is possible'2,'3. Anodic oxidation of graphitic materials in H,SO, is also used as a test reaction to characterize their degree of orientation and defect concentration'4s15.Ether-solvated

-

'3'.

282

16.4. The Formation of Sheet Structures 16.4.2. Graphite and Boron Nitride Intercalation Compounds 16.4.2.8. Electrochemical Formation

TABLE1. ANODICALLY PREPARED ACID SALTS OF GRAPHITE Acid

Lowest stage obtained

HW, HC10, HNO, HF FS03H CF,SO,H

First First First First First First

Refs.

3-5 3, 4 3 6

I 8

acid salts, e.g., C18[BF,]- . 1.4 HBF4.0.9 Et,O (empirical formula), have been obtained by anodic oxidation of graphite in a solution of HBF, in Et,O 16. (ii) Formation of Neutral Salts of Graphite. In contrast to the acid salts, which always contain molar amounts of acid, HA, neutral salts of graphite are either binary compounds, C;A-, or solvated compounds, C;(solv),A-, (solv) being a polar aprotic solvent stable to oxidation. Neutral graphite salts are stronger oxidants than acid salts. With a few exceptions they have not been prepared otherwise than by anodic oxidation of graphite. An alternative preparation is graphite oxidation by [O,]' l 7 , I 8 and [NO]' l 9 or [NO,]' 1920 salts. Neutral salts of graphite can be prepared by anodic oxidation at RT, e.g., in nonaqueous [C104]-, [BF4]-, [PF,]-, [AsF,]-, [AsF,]- [SbFJ and [AICl,]solutions. Anodic oxidation of graphite in nonaqueous F- solutions does not yield ionic salts of graphite; however, formation of a nitromethane-solvated fluoride is claimed3'. Of course the anions have to be fairly oxidation-resistant, as the standard potential of firststage salts (in general corresponding to the stoichiometry C14) is + 2 V. Suitable solvents are sulfolane and 3-methylsulfolane2 dimethyl sulfite", propylene c a r b ~ n a t e ~m~i st r~~~m, e t h a n e ~ ~acetonitrileZ4, ,'~, S0,ClF 14,31, etcZ6;for solubility reasons Li' and [NR,] salts of the above-mentioned anions are most convenient. Caution: Prolonged oxidation of graphite in [CIO,] - electrolytes may cause explosions, presumably owing to side reactions of CIO,. radicals. Explosion hazards are also present3' for the cathodic deposition of alkali metals in nitromethane-this is the counterreaction to the anodic formation of graphite salts in A'X--nitromethane electrolytes. In nitr~methane,~ and S0,C1F14,31 unsolvated graphite salts C;X- are formed; vacuum-dried salts prepared in acetomitrile electrolytes are also solvent freez4. On the other hand, graphite salts prepared in propylene carbonate electrolytes are s o l ~ a t e d ~ ~ , ~ ~ .

'-"

',

-

+

(iv) Formation of Partly Ionic Electron-Pair A ~ c e p t o r ~Acid ~ , ~ ~Intercalation . Compounds. In contrast to oxidizing electron-pair acceptor acids such as FeC1, or SbF,,

which spontaneously form graphite intercalation compounds, nonoxidizing electron pair acceptor acids such as AlCl, can be intercalated only in the presence of oxidants such as chlorine (see $16.4.2.1.1). The required partial oxidation of graphite can also be performed anodically; AlCl, 35, BiCI, 3 6 and F3B.O(C,H,), 2 6 , 3 7 have been intercalated in this way. (v) Formation of Bi Intercalation Compounds. Compounds of the type MCl3-graphite-H,SO4-graphite-H,SO,-graphite-MCl3. . . can be prepared by anodic oxidation of second-stage MCI,-graphite (M = Bi, In) in sulfuric acid3*.

16.4.2. Graphite and Boron Nitride Intercalation Compounds 16.4.2.8. Electrochemical Formation 16.4.2.8.3. Anodic Oxidation of Graphite. ~

~~~~~

283

~~

(VI)Formation of Covalent Graphite Compounds.

Graphite fluoride (see 516.4.2.6) and graphite oxide (see 516.4.2.5) can be obtained by anodic oxidation of graphite in fluoride melts3’ and in strong protonic acids containing water4’, respectively. The anodic formation of graphite fluoride is not a bulk reaction but a surface process, as (CF,), is no F- conductor and consequently the reaction is blocked when a surface film of (CF,), has formed. Covalent fluorination of graphite is a common side reaction to the electrochemical formation of neutral graphite salts of such complex fluoride ions as [PFJ. An indirect RT synthesis of graphite fluoride is by anodic oxidation of graphite in nonaqueous solutions of F- plus [BFJ salts41; C,BF,, which is formed in the primary step, is converted to graphite fluoride by an exchange process. The anodic formation of graphite oxide (see $16.4.2.5.2) is a quantitative bulk reaction as acid HA, anion A- and H,O, which are involved in this reaction [see 516.4.2.5.2, Eq. (a)], are mobile in the gap between the graphite oxide layers. Anodically prepared graphite oxide is distinguished from chemically prepared graphite oxide by a lower degree of oxidation (see $16.4.2.5.2); therefore, it is still electronically conductive and black. With increasing degree of oxidation the electronic conductivity decreases and finally the further oxidation is blocked by an insulating film of green to yellowish graphite oxide. (J 0 BESENHARD)

1. A. Metrot, J. E. Fischer, Synth. Met., 3, 201 (1981). 2. R. Fujii, K. Matsuo, Tunso, 73, 44 (1973). 3. M. J. Bottomley, G. S. Parry, A. R. Ubbelohde, D. A. Young, J. Chem. Soc., 5674 (1963). Classical paper. 4. L. C. F. Blackman, J. F. Mathews, A. R. Ubbelohde, Proc. Soc., London, Sect. A , 258,329 (1960). 5. W. Riidorff, Z. Anorg. Allg. Chem., 238, 1 (1938). 6. W. Riidorff, Z. Anorg. Allg. Chem., 254, 319 (1947). 7. P. Touzain, E. Buscarlet, L. Bonnetain, Carbon, 16, 403 (1978). 8. D. Horn, H. P. Boehm, Muter. Sci. Eng., 31, 87 (1977). 9. J. Melin, G. Furdin,H. Fuzellier, R. Vasse, A. Herold, Muter. Sci.Eng., 31, 61 (1977). 10. W. Riidorff, W.-F. Siecke, Chem. Ber., 91, 1348 (1958). 11. H. P. Boehm, W. Helle, B. Ruisinger, Synth. Met. 23, 395 (1988). 12. F. Beck, H. Junge, H. Krohn, Electrochim. Acta, 26, 799 (1981). 13. F. Beck, H. Krohn, W. Kaiser, J. Appl. Electrochem., 12, 505 (1982). 14. J. 0. Besenhard, H. Mohwald, J. J. Nickl, Synth. Met., 3, 287 (1981). 15. D. Horn, H. P. Boehm, Z. Anorg. Allg. Chem., 456, 117 (1979). 16. A. Metrot, P. Willmann, A. Herold, Carbon, 19, 119 (1981). 17. N. Bartlett, B. McQuillan, A. S. Robertson, Muter. Res. Bull., 13, 1259 (1978). 18. N. Bartlett, E. M. McCarron, B. McQuillan, T. E. Thompson, Synth. Met., 1, 221 (1979j80). 19. D. Billaud, A. Pron, F. L. Vogel, A. Herold, Mazer. Res. Bull., 15, 1627 (1980). 20. W. C . Forsman, H. E. Mertway, Synth. Met., 2, 171 (1980). 21. A. Brenner, J. Electrochem. Soc., 118,461 (1971). 22. J. S . Dunning, W. H. Tiedemann, L. Hsueh, D. N. Bennion, J. Electrochem. Soc., 118, 1886 (1971). 23. J. 0. Besenhard, H. P. Fritz, Z . Nuturforsch., Teil B, 26, 1125 (1971). 24. J. 0. Besenhard, H. P. Fritz, Z. Naturforsch., Teil B, 27, 1294 (1972). 25. D. Billaud, A. Metrot, P. Willmann, A. Herold, 3rd Int. Carbon Conf, Baden-Baden (1980), Preprints, p. 140. 26. J. 0. Besenhard. Ph. D. Thesis. T. H. Miinchen (19731, =NASA Report N 74-30505. 27. Ph. Touyain, A.’Jobert, Electrochim. Actu, 26, 1133 (1981). 28. D. Billaud, A. Chenite, J. Power Sources, 13, 1 (1984). 29. J. Berthelot, J. Simonet, Electrochim. Actu, 29, 1189 (1984).

284

30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 41.

16.4. The Formation of Sheet Structures 16.4.3. Dichalcogenides 16.4.3.1. Layered Dichalcogenides.

A. Chenite, A. Metrot, D. Billaud, A. Herold, Synth. Met., 7, 201 (1983). J. 0. Besenhard, H. Mohwald, in International Colloquium on Layered Compounds, J. A. Titus, Chem. Eng. News, 49, 6 (1971). J. 0. Besenhard, H. P. Fritz, H. Mohwald, J. J. Nickl, 3rd Znt. Carbon Con$, Baden-Baden (1980), Preprints, p. 147. A. Jobert, Ph. Touyain, L. Bonnetain, Carbon, 13, 193 (1981). M. Fouletier, M. Armand, Carbon, 17, 427 (1979). E. Stumpp, K. Wloka, Synth. Met., 3, 209 (1981). A. Metrot, P. Willmann, A. Herold, Mater. Sci. Eng., 31, 83 (1977). P. Scharff, E. Stumpp, C. Erhardt, Synth. Met., 23, 415 (1988). 0. Ruff, Ber., 69, A181 (1936). J. 0. Besenhard, H. P. Fritz, 2.Anorg. Allg. Chem., 416, 106 (1975). M. Armand, Et. Sci. Thesis, Univ. Grenoble (1978).

16.4.3. Dichalcogenides 16.4.3.1. Layered Dichalcogenides.

Layered dichalcogenides are typically two-dimensional compounds. Their structure is built up with [XMX] slabs (X = S, Se, Te) structural units. Within these units are strong ionocovalent or metallic bonds, whereas they are separated by rather large distances (generally of the order of the radii of closest approach) in agreement with weak interslab bondings. Table 1 shows the elements that form layered dichalcogenides. In addition a few other elements such as Ni, Co, Ir, Rh, and Pd give only ditellurides derivatives. The structures can be classified with respect to the nature of the slabs and the way they are stacked. The slabs are made up of three atomic layers, two anionic layers sandwiching a metallic one (Fig. 1).The coordination of the metal in the slabs can be either octahedral or trigonal prismatic. The octahedral surroundings correspond to an anionic close packing that is generally hexagonal, leading to a Cd1,-type structure (in a few cases, such as NbTe,, it can be cubic, leading to a CdCl, arrangement). Representing the anionic TABLE1. FORMATION OF LAYERED DICHALCOGENIDES: WITH TRIGONAL PRISMATIC COORDINATION

"'

OCTAHEDRAL OK

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc. 284

30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 41.

16.4. The Formation of Sheet Structures 16.4.3. Dichalcogenides 16.4.3.1. Layered Dichalcogenides.

A. Chenite, A. Metrot, D. Billaud, A. Herold, Synth. Met., 7, 201 (1983). J. 0. Besenhard, H. Mohwald, in International Colloquium on Layered Compounds, J. A. Titus, Chem. Eng. News, 49, 6 (1971). J. 0. Besenhard, H. P. Fritz, H. Mohwald, J. J. Nickl, 3rd Znt. Carbon Con$, Baden-Baden (1980), Preprints, p. 147. A. Jobert, Ph. Touyain, L. Bonnetain, Carbon, 13, 193 (1981). M. Fouletier, M. Armand, Carbon, 17, 427 (1979). E. Stumpp, K. Wloka, Synth. Met., 3, 209 (1981). A. Metrot, P. Willmann, A. Herold, Mater. Sci. Eng., 31, 83 (1977). P. Scharff, E. Stumpp, C. Erhardt, Synth. Met., 23, 415 (1988). 0. Ruff, Ber., 69, A181 (1936). J. 0. Besenhard, H. P. Fritz, 2.Anorg. Allg. Chem., 416, 106 (1975). M. Armand, Et. Sci. Thesis, Univ. Grenoble (1978).

16.4.3. Dichalcogenides 16.4.3.1. Layered Dichalcogenides.

Layered dichalcogenides are typically two-dimensional compounds. Their structure is built up with [XMX] slabs (X = S, Se, Te) structural units. Within these units are strong ionocovalent or metallic bonds, whereas they are separated by rather large distances (generally of the order of the radii of closest approach) in agreement with weak interslab bondings. Table 1 shows the elements that form layered dichalcogenides. In addition a few other elements such as Ni, Co, Ir, Rh, and Pd give only ditellurides derivatives. The structures can be classified with respect to the nature of the slabs and the way they are stacked. The slabs are made up of three atomic layers, two anionic layers sandwiching a metallic one (Fig. 1).The coordination of the metal in the slabs can be either octahedral or trigonal prismatic. The octahedral surroundings correspond to an anionic close packing that is generally hexagonal, leading to a Cd1,-type structure (in a few cases, such as NbTe,, it can be cubic, leading to a CdCl, arrangement). Representing the anionic TABLE1. FORMATION OF LAYERED DICHALCOGENIDES: WITH TRIGONAL PRISMATIC COORDINATION

"'

OCTAHEDRAL OK

16.4. The Formation of Sheet Structures 16.4.3. Dichalcogenides 16.4.3.1. Layered Dichalcogenides.

-Van

der Waals gap

285

X

X

'T

T X

X Trigonal prismatic slab

Octahedral slab

Figure 1. Basic features for a representation of lamellar dichalcogenides.

layers by A, B, C and the cationic planes by a, b, c, the CdI, structure of TiS, is described by the . . . A b C A b C . . . sequence in relation with the (AC), hexagonal close packing of the anions. In the case of a structure built up with trigonal prismatic slabs, as in NbS, 2 H, the sequence is A b A C b C, with successive A A or C C blocks made of superposed chalcogen layers. In both cases (TiS, and NbS,) octahedral sites exist in the van der Waals gap (see Fig. 1) between the C and A (TiS,) or the A and C (NbS,) chalcogen layers (Fig. 2). This is important in the process of intercalation. These simple models can be elaborated in different ways, among which three appear to be very important. First, the weak interslab bonding allows them to slide one over the other, leading to numerous polytypes that are generally referred to in a notation' that takes into account the symmetry of the lattice (trigonal, hexagonal or rhombohedral), the number of slabs concerned and the stacking sequence (1 T, 3 R, 4 Ha, 4 Hc . . .48 R . . .). From this point of view 2 H MoS, and 2 H NbS, can be regarded as two different stacking modes of trigonal prismatic slabs. In 2 H MoS, the atomic sequence appears as A b A B a B with a B sulfur layer above the b cationic layer and a similar arrangement for a and A layers (Fig. 2). The second elaboration arises from the fact that slabs having different cation coordination may be interleaved in many ways; different possibilities are shown in Fig. 3 in the case of tantalum diselenide. The last point concerns folding or distortion of the slabs. This is generally related to the formation of metal-metal bonds

C

C b

B a

C

0

b

A

I .

I

C b

I

0

1

A b

A

A

A

b

A Ti S q

NbS22 H

MoS22H

Figure 2. Octahedral (TiS,) and trigonal prismatic (NbS2, MoS,) slabs in lamellar dichalcogenides.

16.4. T h e Formation of Sheet Structures 16.4.3. Dichalcogenides 16.4.3.1,Layered Dichalcogenides.

286

. toot

0 Se 0 Ta

0

0

0

0 IT TaSep

2H TaSe2

3 R TaSe2

0

4Ha TaSep

0 0 4Hb TaSep

0 6R TaSep

Figure 3. TaSe, polytypes

and can be observed in rhenium chalcogenides and in the high-temperature forms of MoTe, and WTe,. A partial occupancy of the van der Waals gap (Fig. 1) by extra metal atoms reflects a M i +xX, nonstoichiometry that is often encountered in dichalcogenides. The x extra metal atoms can lead to various types of ordered arrangements in the van der Waals gap’, as in the case of the structurally related metal-rich phases M,X, (Ml,5,,X2),M,X, (M,,25Xz)where an ordering of vacancies is observed in some planes3,,. However much more complicated arrangements in domains have been observed in the case of the Ti, +& polytypes5. This partial filling of this region by metal atoms destroys the low dimensional character of the structure. Bridge bondings are formed between the slabs and can have drastic effects upon the ability of intercalation and its kinetics.

Figure 4. Simplified band model for octahedral (left) and trigonal prismatic slabs (right).

16.4. The Formation of Sheet Structures 16.4.3. Dichalcogenides 16.4.3.1. Layered Dichalcogenides.

E 'S,P'

'd / ( PI"' 'd 2' 'p / (d I'

'S'

N(E) Figure 5. Density of state scheme for trigonal prismatic group. VA dichalcogenides (after ref. 8). Figure 4 gives a representation of the band structures of octahedral and trigonal prismatic dichalcogenides. The valence bond arises from the nonmetal s and p orbitals, while the cationic s and p levels form antibonding states. The metal d orbitals, split by the chalcogen ligand field, are situated between these groups and play an essential role concerning the electrical and optical properties. A t2g band with six fold electron occupancy is the lowest empty band in the case of an octahedral dichalcogenide, whereas an a; narrow band (dz') and an e' group (dxy, dx2 - yz) are successively found in the case of a trigonal prismatic lab^,^. With octahedral coordination and empty d bands, ZrS, is a large-gap semiconductor; NbS,, with a half-filled dz2 band, is a metal; MoS,, with a filled dzZ band, is a narrow-band semiconductor. Group-VII compounds (ReS,, for example) expected to be metals have in fact distorted structures with metal-metal pairing. This results in a full band splitting away and the formation of a narrow-gap semiconductor. In relation to its stoichiometry, TiS, is also a particular case. Photoemission studies (UPS and XPS) in conjunction with the optical absorption and reflectivity data and along with band structure calculation^^^^*^ lead to more sophisticated models. Owing to the covalency of the compounds (particularly the trigonal prismatic derivatives, which are less ionic), there is a mixing between metal d states and chalcogen p states to form the M-X covalent bond. The valence band is thus broadened while the metal d band acquires some p-like character resulting in an antibonding d/p* band. Figure 5 gives a representation of the resulting density of states for a group-VA element trigonal prismatic dichalcogenide. (J. ROUXEL)

288

16.4. The Formation of Sheet Structures 16.4.3. Dichalcogenides 16.4.3.2. Alkali-Metal Intercalates.

1. L. Ramsdell, Am. Mineral, 32, 64 (1967). 2. F. Jellinek, International Conference on Electron Diffraction and the Nature of Defects in Crystals, Pergamon Press, Oxford, 1966. 3. F. Jellinek, Acta Crystallogr., 10, 620 (1957). 4. M. Chevreton, F. Bertaut, C. R. Hebd. Seances Acad. Sci., 255, 1275 (1962). 5. J. S. Legendre, M. Huber, J. Appl. Crystallogr., 13, 193 (1980). 6 . J. A. Wilson, A. D. Yoffee, Adu. Phys, 18, 193 (1979). 7. R. Huisman, R. De Souge, C. Haas, F. Jellinek, J. Solid State Chem., 3, 56 (1971). 8. M. G. Bell, W. Y. Liang, Adu. Phys., 25, 53 (1976). 9. R. M. White, G. Lucovsky, Solid State Commun.,11, 1369 (1972).

16.4.3.2. Alkali-Metal Intercalates.

In two-dimensional structures the process of intercalation results in the pulling apart of the slabs of the structure. It should be reversible and allow the true intercalation compounds to be distinguished from the usual insertion derivatives. It means that no strong bond is broken in the host structure. All these features are well verified in the case of dichalcogenides. Owing to their variety (composition, structures . . .), the dichalcogenides provide some of the best examples for the study of intercalation chemistry, despite the fact that only electron donors have been intercalated up to now. Since the first preparation of alkali-metal intercalates by the liquid ammonia technique’, many preparation methods have been proposed. The most commonly used fall into four main groups: (i) use of alkali-metal solutions in liquid ammonia, (ii) use of organometallic reagents, (iii) solid-state techniques and (iv) electrochemical processes. The use of alkali-metal solutions in liq NH, leads to fast reactions and proves very useful, for it can be used for all alkali metals (and also for some alkaline-earth metals, Eu and Yb). On the other hand, it presents severe drawbacks: the working conditions are delicate and the method often leads to the fixation of NH,, which must be eliminated. A good procedure consists in using thick-walled sealed borosilicate glass tubes at RT z. The tubes possess several branches allowing, through chosen temperature gradients, separation of NH, from the intercalated product and preparation of the samples in situ for x-ray analysis. Pure and dried NH, (distilled over Na) is necessary to avoid side reactions and the formation of A,(H,O),MS,-hydrated products. Other solvents can be used instead of NH,, e.g., hexamethylphosphotriamide (HMPT) for Li ,. The use of organometallic reagents is well represented by the butyllithium technique. A solution of n-butyllithium in hexane is a mild and efficient reagent to form the lithium The host structure is simply brought into derivatives of transition-metal chalcogenide~~. contact with a diluted solution of n-BuLi in hexane (0.2-1.5 mol L - l ) under an inert atmosphere for several days at RT. Octane is formed as a byproduct of the reaction: n-C,H,Li

+ host

-

Li,host

+ X2 n-C,H,, ~

(a)

Some butane and butene may be found in certain cases. Other organometallics can be used, such as alkali naphthalide solutions in THF, which form the lithium, sodium and potassium intercalates. A direct reaction between alkali metals and the host structure can lead to intercalation compounds. Weighed mixtures of the alkali metal and the host are allowed

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc. 288

16.4. The Formation of Sheet Structures 16.4.3. Dichalcogenides 16.4.3.2. Alkali-Metal Intercalates.

1. L. Ramsdell, Am. Mineral, 32, 64 (1967). 2. F. Jellinek, International Conference on Electron Diffraction and the Nature of Defects in Crystals, Pergamon Press, Oxford, 1966. 3. F. Jellinek, Acta Crystallogr., 10, 620 (1957). 4. M. Chevreton, F. Bertaut, C. R. Hebd. Seances Acad. Sci., 255, 1275 (1962). 5. J. S. Legendre, M. Huber, J. Appl. Crystallogr., 13, 193 (1980). 6 . J. A. Wilson, A. D. Yoffee, Adu. Phys, 18, 193 (1979). 7. R. Huisman, R. De Souge, C. Haas, F. Jellinek, J. Solid State Chem., 3, 56 (1971). 8. M. G. Bell, W. Y. Liang, Adu. Phys., 25, 53 (1976). 9. R. M. White, G. Lucovsky, Solid State Commun.,11, 1369 (1972).

16.4.3.2. Alkali-Metal Intercalates.

In two-dimensional structures the process of intercalation results in the pulling apart of the slabs of the structure. It should be reversible and allow the true intercalation compounds to be distinguished from the usual insertion derivatives. It means that no strong bond is broken in the host structure. All these features are well verified in the case of dichalcogenides. Owing to their variety (composition, structures . . .), the dichalcogenides provide some of the best examples for the study of intercalation chemistry, despite the fact that only electron donors have been intercalated up to now. Since the first preparation of alkali-metal intercalates by the liquid ammonia technique’, many preparation methods have been proposed. The most commonly used fall into four main groups: (i) use of alkali-metal solutions in liquid ammonia, (ii) use of organometallic reagents, (iii) solid-state techniques and (iv) electrochemical processes. The use of alkali-metal solutions in liq NH, leads to fast reactions and proves very useful, for it can be used for all alkali metals (and also for some alkaline-earth metals, Eu and Yb). On the other hand, it presents severe drawbacks: the working conditions are delicate and the method often leads to the fixation of NH,, which must be eliminated. A good procedure consists in using thick-walled sealed borosilicate glass tubes at RT z. The tubes possess several branches allowing, through chosen temperature gradients, separation of NH, from the intercalated product and preparation of the samples in situ for x-ray analysis. Pure and dried NH, (distilled over Na) is necessary to avoid side reactions and the formation of A,(H,O),MS,-hydrated products. Other solvents can be used instead of NH,, e.g., hexamethylphosphotriamide (HMPT) for Li ,. The use of organometallic reagents is well represented by the butyllithium technique. A solution of n-butyllithium in hexane is a mild and efficient reagent to form the lithium The host structure is simply brought into derivatives of transition-metal chalcogenide~~. contact with a diluted solution of n-BuLi in hexane (0.2-1.5 mol L - l ) under an inert atmosphere for several days at RT. Octane is formed as a byproduct of the reaction: n-C,H,Li

+ host

-

Li,host

+ X2 n-C,H,, ~

(a)

Some butane and butene may be found in certain cases. Other organometallics can be used, such as alkali naphthalide solutions in THF, which form the lithium, sodium and potassium intercalates. A direct reaction between alkali metals and the host structure can lead to intercalation compounds. Weighed mixtures of the alkali metal and the host are allowed

16.4 The Formation of Sheet Structures 16.4 3. Dichalcogenides 16.4.3.2. Alkali-Metal Intercalates. ~~

289

~

to react in a sealed quartz tube at 600-800°C. This has been used in the case of AXMY, derivatives with A = Li, Na, K; M = Nb, Ta and Y = S, Se5. The Na,VS, compounds are also obtained from the elements in evacuated quartz tubes6. These methods present several drawbacks. A reduction of the host structure may be observed, homogeneous products are difficult to ascertain and the reactor can be severely attacked with the formation of byproducts that cannot be eliminated. However, solid-state technique are very useful in preparing well-crystallized 1 : 1 products; LiVS, is obtained by reacting mixtures of Li,CO, and V,O, in an H,S atmosphere at 500-700°C; LiCrS, is formed by heating a mixture of Li,CO, and Cr,S, at 800°C in a flow of CS, vapor and Ar. More generally, AMS, derivatives may be obtained by heating the A,S and M,S, sulfides at about 800-900°C. The preparation of NaVS, or LiVS, before VS, was known is a successful application of the solid-state techniques. The growth of single crystals of intercalation compounds is a major problem that has not yet been solved in a completely satisfying way. Single crystals of TiS,, ZrS, or NbS, react with alkali-metal solutions in liq NH,. Intercalation occurs, but the quality of the crystals obtained in that manner may be very bad, particularly in the case of Na (problems of two-phase domains) and in the case of the biggest alkali metals. The nbutyllithium technique can also be used to intercalate crystals. Reactions of layered transition-metal disulfides with H,S in alkali halide melts, or in a gas flow reactor at 800- lOOO"C, produce crystals in thin, hexagonal platelets, up to several mm in diameter'. The problem in that case is to separate the crystals from the alkali halide matrix. None of these methods is utterly satisfactory. The use of alkali-metal solutions in liq NH, allows intercalation from Li to Cs to be covered. It can readily be used to prepare nonstoichiometric phases but the method presents experimental difficulties.Ammonia is often cointercalated, which favors the formation of trigonal prismatic intercalates owing to the preference of the NH, for this type of site. On the other hand, the solvation of the A + ions by NH, may determine the composition of the final product in relation to the formation of stable complex species, with a definite formulation, between the slabs of the host. The thermal treatment necessary to remove the NH, may also lead to different structures than those formed at RT; it certainly plays a role concerning the phase limits. Alkali-metal solutions in liq NH, are powerful reducing solutions, and in the case of tellurides, or even for some selenides or sulfides, a reduction of the host structure can occur. The use of organometallic compounds is efficient in the case of Li. Solvent is not cointercalated which is an advantage compared to the liq NH, case, but a disadvantage concerning the diffusion of the alkali-metal ions, and a complete intercalation may not be achieved (the NH, molecule separates the sheets and favors A' ion diffusion). Solid-state high-T techniques are well adapted to preparing the stoichiometric samples, each time mixtures of sulfides or oxides are used. Well-crystallized products are obtained. The electrochemical processes are based on the use of the chalcogenide in an electrochemical cell with an A anode and a solution of A t ions as the electrolyte. They are discussed specifically 516.4.3.3. In the host structure the alkali metal can occupy all the van der Waals gaps (stage I compounds; see Fig. 1, §16.4.3.1),or only one out of two (stage I1 compounds), or one out of four (stage IV compounds). Other stages are very likely despite the fact that the experimental evidence is still very poor. The chalcogen surroundings of the alkali metal can be either octahedral or trigonal prismatic (Fig. 1). With an octahedral coordination, a nickel arsenide structure is

16.4. The Formation of Sheet Structures 16.4.3. Dichalcogenides 16.4 3.2. Alkali-Metal Intercalates.

290

0 s

0

Alkali metal

0

Ti

0

0 8

0

0 0

0

17 Ni A s

T i S2

D

I

8 0

0

0

0

0

0

0 0

pJj 1 a-

0 0

octahedra I coorznation

--I

I

I

1

t rigona I prisma t ic coord in a t io n

Figure 1. Structural models for TiS, intercalates (after ref. 9).

expected when starting from a Cd1,-like host structure. This should be the case of the octahedral intercalates of TiS,; in fact, however, only Li,TiS, with 0 < x I 1 exhibits this structural form', corresponding to the classical Cd1,-NiAs transition. A NaHF, arrangement with an ordering involving three MX, slabs and three van der Waals layers is generally observed. The stacking of the anionic layers has changed from an hexagonal ABAB... sequence to a distorted cubic face-centered one, ABCABC. . . . The 3 R octahedral structure (Ia type) that is observed corresponds to the filling of a CdCl, structure (CdC1,-NaHF, transition). A 3 R (Ib) structure is generally observed with the trigonal prismatic structure, which implies a shift of the slabs to give an AABBCC stacking of the sulfur layers. In any case the (STiS) slabs are unchanged, but they are shifted so that the unit cell now includes three. Both Na in TiS, and potassium in ZrS, are examples of the two types of first-stage compounds with the two kinds of coordination'. In the sulfides the coordination of the alkali metal depends on three factors': (i) the size and (ii) the amount of alkali metal and (iii) the nature of the bonding in the slabs of the host structure. The influence of these factors can be clearly understood by noting that an octahedron is the best equilibrium position of six negative charges and consequently can accommodate higher charges on the anions than a trigonal prism. With a larger alkali-metal atom the chalcogen layers are more distant, thus minimizing the repulsions, and the trigonal prismatic structure is favored. The general ionization scheme xA +, MS;- explains why for a given alkali metal the octahedral form may appear for the higher values of x (highest charges on the anions), whereas the trigonal-prismatic form is obtained for the lower values of x. The latter factor involves the ionicity of the bonding in the slabs of the host structure: ZrS,, more ionic, favors the formation of the octahedral species as compared to TiS,. It is clear that such considerations do not apply so easily to less ionic systems such as selenides and tellurides. In the case of sulfides they allow the

29 1

16.4. The Formation of Sheet Structures 16.4.3. Dichalcogenides 16.4.3.2. Alkali-Metal Intercalates.

building of a general diagram concerning the ionicity-structure relationship in the case of Cd1,-like host structures". By plotting the r l / r i - ratio of the ionic radii vs. a function related to the fractional ionicities of the M-S and A-S bonds and the stoichiometry, it is possible to draw an unambiguous line between the 3 R octahedral (more exactly, the antitrigonal prismatic ATP) and the 3 R prismatic domains. (Fig. 2). For each type of site, the increases Ac of the c parameters in the direction perpendicular to the slabs are found to depend on the composition. A comparison for the same composition and the same alkali metal, when possible, always shows much higher values in the trigonal prismatic sites. Typical values are 0.98, for Na,,,,ZrS, with octahedral coordination, and 1.27 8, for Na,,,,TiS, with trigonal prismatic sites (1.32 in Na,,, lNbS,), which is significant despite differing host structures having to be considered. TiS2 ZrSp

!A

rS2

-- -- -

1

cs

Rb

K

Na

Li

0

k

Ib

i'zxfifi:100

Figure 2. Ionicity-structure diagram for various intercalates of Cd1,-like host structures (after ref. 10).

16 4. T h e Formation of Sheet Structures 16.4.3. Dichalcogenides 16.4.3.2. Al kal i-Metal Intercalates.

292

For a given type of site and the same amount of intercalation, Ac is proportional to the radius of the concerned A t ion. The explanation of the formation of multistage phases remains an open question. The cost in energy is lower if fewer van der Waals gaps have to be opened at the beginning. Then the increase of the repulsive energy between A + ions when the population of an occupied van der Waals gap rises may lead to a situation where the opening of a new van der Waals gap becomes energetically more favorable. The repulsive energy between successive A t intercalated layers also can be considered. Possibly the A t ions occupy at first more distant van der Waals gaps, thus lowering the repulsion. The slab of the host screens this repulsion. From this point of view the A"Ta,S,C phases are of some importance. The fact that no second-stage phases are observed in that case" is probably related to the bigger screening effect (Fig. 3) owing to a five-layer slab (S-Ta-C-Ta-S). In general screening is better in MX, intercalates than for graphite, where stage compounds are much more prevalent. If the energetic factor of opening the van der Waals gaps were the only one to interfere, there should be more high-stage phases in the dichalcogenides in which the slabs are more rigid, are more difficult to distort and require a higher energy to separate. However, the real nature of the second-stage phases is not yet clearly understood. The Ac parameters observed fit very well with the occupancy of one over two van der Waals gap in stage I1 compounds, one over four in stage IV compounds, etc., and this has been taken as proof of the existence of multistage phases (for example, the c parameter of a second-stage phase A,MS, fits with 6c (host) 3Ac of the A,,MS, first stage phase). This agrees with a classical model of multistage phases but also with the model" for graphite intercalates shown in Fig. 4. In both cases we should have a 2x population of the occupied van der Waals gap, which agrees with the type of coordination suggested by ionicity considerations and Ac measurements. However, the superstructures observed along c (6 R for a second-stage phase, for example) are explained more easily with the classical model (some disorder along c for the occupied domains is certainly favored in the model shown in Fig. 4). The A + ions repel each other and at sufficiently low temperatures this should lead to an ordering between A + and vacancies in the van der Waals gap. Such an ordering is well known in the case of graphite compounds. However, in dichalcogenides no superstructure related to such ordering is observed by x-ray diffraction. Statistically distributed microdomains may exist and this hypothesis is justified since penetration of the ions through the edges of crystals is reflected in concentration gradients depending on time and distance from the crystal edges. By x-ray diffusion techniques the existence of ordered structures has recently been proved in a TiS, single crystal electrochemically intercalated with NaL3.Various superstructures are observed for specific compositions such as x = 6, 1 1 3 9 , 6 , 4 , etc.

+

-C --graphite

-

lisp

-'F;:--Fi!

---? b-

NbS2

Ta2S2C

-

Bi Te

Bi2Te3

Figure 3. Various slabs according to the number of atomic layers.

He8 r t

16.4. The Formation of Sheet Structures 16.4 3. Dichalcogenides 16.4.3.2. Alkali-Metal Intercalates.

293

(b)

Figure 4. Third-stage derivative (a), in the classical model, and (b) in a pleated-layer model (after ref. 12). The formation of an intercalation compound is by an ionic diffusion process between the slabs of the host structure. NMR studies on Li14,15and Na intercalate^^^^'^ establish a substantial ionization of the alkali metal, as expected considering the chemical behavior of the compounds and the Ac values, and bring information concerning the mobility of the A’ ions. Quadrupole coupling constants e2qQ/h, spin-lattice relaxation rate 1/T, and Kisohave been measured. In Li,TiS, the very small Kisovalues for 7Li (3-12 ppm compared to 240 ppm in Li metal) imply electron donation to the layers upon intercalation. Lithium is ionized, but the ionization is less complete for the highest amounts of intercalation as reflected by the increase in Kisoand the leveling off of e2qQ/h (the latter term expresses also the elongation of the A.T.P. sites). Measurements of l/Tl vs. temperature by a pulse technique lead to very low activation energies: 0.11 eV for Li,,,,TiS, and 0.22 eV for Li,,,,ZrS, Is. The valuesare only slightly dependant on x. The Li’ ions have a higher mobility in the less ionic TiS, host structure. Two classes of spectra are obtained for Na,TiS, compounds according to the two structural models (Ia and Ib). A typical second-order quadrupolar coupling is observed, but Na,,,,TiS,, with the Ib structure, shows a strongly broadened line with a stronger quadrupolar coupling compared with the line obtained for the Ia type NaTiS, (the line-width was about 10 G instead of 6 G at 40 MHz). The donation of the Na 3s electron to the TiS,t2g band is probably almost complete according to small positive values of K,,,. A comparison with the Na-W bronzes can be made. However, the shifts are negative in the bronzes and are attributed to a core polarization hyperfine field induced by W 5d electrons at the Na sites. The Na,TiS, data suggest that the constant energy surface has some residual s character in this case. The electrostatic properties of the Na sites, as reflected in e2qQ/h (1.50MHz at 300K for NaTiS,) are comparable in all the dichal~ogenides’~ and similar to those of p-alumina ( ~ 2 . MHz)” 0 and the Na,WO, system where e2qQ/h = 1.5 MHz at low temperature. “Deintercalation” chemistry is a very interesting field leading to a new way of chemical synthesis. To introduce this point it is important to emphasize that intercalation can be regarded as a competition between different redox systems. When intercalation is performed the host structure is reduced: x Li+ + x e-

+ host

-

Li: hostx-

(b)

294

16.4. The Formation of Sheet Structures 16.4.3. Dichalcogenides 16.4.3.2. Alkali-Metal Intercalates. Reagents

Electrode materials

I-

>I

cn

Li NH3-

I

NiPS3

[MOO2

I LiAl

R

-Li

Figure 5. Electropotential scale for reagents and guests relative to Li/Li+. DDQ is 2,3-dichloro4,5-dicyanobenzoquinone; bzph, benzophenone; naph, naphthalene (after ref. 21). but oxidizing agents stronger than the host can oxidize the lithium intercalation compound and remove the lithium ions from the structure. For example, lithium can be intercalated in TiS, by using n-BuLi solutions in hexane but lithium can be removed from Li,TiS, by using iodine solutions in acetonitrile: Li:(TiS,)”-

+ X-2 I,

-

xLiI

+ TiS,

Many reagents have been tested for their ability to allow intercalation or deintercalation. It all depends on their position relative to the host structures on an electropotential scale such as the one in Fig. 5”. A very important consequence of these reversible processes is the ability to prepare new chalcogenides. Despite the fact that an M,S, chalcogenide is unknown, solid-state techniques allow the preparation of A,M,S, derivatives sometimes isotypic with A,MS, intercalation compounds. Deintercalating by means of an appropriate oxidizing reagent results in the M,S, phase. This can also be done electrochemically (see 516.4.3.3). For example, VS, was unknown. Only V, +xS2forms were known; but LiVS, was prepared at 700°C by solid-state techniques (CS, iLiVO, or Li,C03 + V,O, under H,S), and then lithium was removed by iodine or electrochemically”. Likewise, CrSe, has been obtained”. It is possible to prepare Li,Fe,S, phases at 850°C by direct combination of Li,S and FeS or FeS,. In Li,FeS, resulting from the reaction between Li,S and FeS, it is then possible to remove one lithium to getz3 Li,FeS, compounds with 0.2 5 x 5 2. (J. ROUXEL)

16.4. The Formation of Sheet Structures 16.4.3. Dichalcogenides 16.4.3.3. Electrochemical Formation

295

1. W. Rudorff, Chimia, 19, 489 (1965). 2. J. Cousseau, L. Trichet, J. Rouxel, Bull. Soc. Chim. Fr., 3, 872 (1973). 3. J. Rouxel, in Physics and Chemistry of Layered Materials, Vol VI, F. Levy, ed., D. Reidel, Dordrecht, 1979, p. 201-250. 4. M. B. Dines, Muter. Res. Bull., 10, 287 (1975). 5. W. Omboo, F. Jellinek, J. Less-Common Met., 20, 121 (1970). 6. G. A. Wiegers, R. Van der Meer, H. Van Heiningen, H. J. Kloosterboer, A. J. A. Albernik, Mater. Res. Bull., 9, 1261 (1974). 7. R. Schollhorn, A. Lerf, J. Less-Common Met., 42, 89 (1975). 8. M. Danot, J. Bichon, J. Rouxel,.Comp. Rend., 276, 1283 (1973). 9. A. Leblanc, M. Danot, L. Trichet, J. Rouxel, Mater. Rex Bull., 9, 191 (1974). 10. J. Rouxel, J . Solid State Chem., 17, 223 (1976). 11. R. Brec, J. Ritsma, G. Ouvrard, J. Rouxel, Inorg. Chem., 16, 660 (1977). 12. A. Herold in Physics and Chemistry of Layered Materials, Vol. 6, F. Levy, ed., D. Reidel, Dordrecht, 1979. p. 401. 13. T. Hibma, J. Solid. State Chem., 34, 97 (1980). 14. B. G. Silbernagel,M. S. Whittingham, J. Chem. Phys., 64, 3670 (1976). 15. C. Berthier, L. Trichet, Soc. Fr. Phys., Poitiers meeting, 1977. 16. B. G. Silbernagel,M. S. Whittingham, Muter. Res. Bull., 11, 29 (1976). 17. L. Trichet, D. Jerome, J. Rouxel, C. R. Hebd. Seances Acad. Sci., 280C, 1025 (1975). 18. J. P. Boilot, L. Zuppiroli, G. Delplanque, D. Jerome, Phil. Mag., 32, 343 (1975). 19. G. Bouera, F. Borsa, M. L. Grippa, A. Rigamonti, Phys. Rev., 134, 52 (1971). 20. D. W. Murphy, P. A. Christian, Science, 205, 651 (1979). 21. D. W. Murphy, J. N. Carides, F. J. Di Salvo, C. Cros, J. V. Waszczak, Mater. Res. Bull., 12, 825 (1977). 22. G. Wiegers, Physica, B99, 166 (1980). 23. L. Blandeau, G. Ouvrard, Y. Calage, R. Brec, J. Rouxel, J. Phys. Chem., Solid State Phys., 20, 4271 (1987).

16.4.3.3. Electrochemical Formation. Electrochemical processes are very convenient preparation methods. The intercalates are formed during the discharge of batteries in which the TX, chalcogenide serves as the cathode and the alkali metal, or a source of alkali metal, as the anode. Such cells form the basis of new battery systems and, therefore, deserve special attention. The electrolytic method has been intensively used in the case of sodium and lithium derivatives. To intercalate Ti:-,S, single crystals (x = 0.002; 0.010; 0.020), for example', the crystals are located in a recess machined in a four-bore alumina sheath and held in place by the pressure of a spring contacting wire. The reference electrode is pure lithium or pure sodium, although in this latter case a sodium-vanadium bronze (Na,,,,V,O,) is more convenient, sodium activity having been measured beforehand. Figure 1 shows how this reference material is held in position in the cell. The counterelectrode was either sodium amalgam or lithium metal. The liquid electrolyte phase consists of a saturated solution of either sodium iodide or lithium perchlorate in propylene carbonate. A system currently used to intercalate powders of host structures is the button electrochemical cell (Figure 2). The cathode, made of dichalcogenide mixed with graphite and a binding latex, is pressed in the cover of the cell. The anode is made of a lithium sheet in contact with a nickel grid. The electrolyte is a glass fiber wool soaked with a solution of LiClO, (2 mol L - l ) in a polar organic solvent (propylene carbonate or 1,3-dioxolane for example). In these systems the anode provides the intercalating ions. It is an efficient and mild method in which the intercalation reaction can be controlled through the current passed. The reaction occurs spontaneously in a chemical concentration gradient.

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc. 16.4. The Formation of Sheet Structures 16.4.3. Dichalcogenides 16.4.3.3. Electrochemical Formation

295

1. W. Rudorff, Chimia, 19, 489 (1965). 2. J. Cousseau, L. Trichet, J. Rouxel, Bull. Soc. Chim. Fr., 3, 872 (1973). 3. J. Rouxel, in Physics and Chemistry of Layered Materials, Vol VI, F. Levy, ed., D. Reidel, Dordrecht, 1979, p. 201-250. 4. M. B. Dines, Muter. Res. Bull., 10, 287 (1975). 5. W. Omboo, F. Jellinek, J. Less-Common Met., 20, 121 (1970). 6. G. A. Wiegers, R. Van der Meer, H. Van Heiningen, H. J. Kloosterboer, A. J. A. Albernik, Mater. Res. Bull., 9, 1261 (1974). 7. R. Schollhorn, A. Lerf, J. Less-Common Met., 42, 89 (1975). 8. M. Danot, J. Bichon, J. Rouxel,.Comp. Rend., 276, 1283 (1973). 9. A. Leblanc, M. Danot, L. Trichet, J. Rouxel, Mater. Rex Bull., 9, 191 (1974). 10. J. Rouxel, J . Solid State Chem., 17, 223 (1976). 11. R. Brec, J. Ritsma, G. Ouvrard, J. Rouxel, Inorg. Chem., 16, 660 (1977). 12. A. Herold in Physics and Chemistry of Layered Materials, Vol. 6, F. Levy, ed., D. Reidel, Dordrecht, 1979. p. 401. 13. T. Hibma, J. Solid. State Chem., 34, 97 (1980). 14. B. G. Silbernagel,M. S. Whittingham, J. Chem. Phys., 64, 3670 (1976). 15. C. Berthier, L. Trichet, Soc. Fr. Phys., Poitiers meeting, 1977. 16. B. G. Silbernagel,M. S. Whittingham, Muter. Res. Bull., 11, 29 (1976). 17. L. Trichet, D. Jerome, J. Rouxel, C. R. Hebd. Seances Acad. Sci., 280C, 1025 (1975). 18. J. P. Boilot, L. Zuppiroli, G. Delplanque, D. Jerome, Phil. Mag., 32, 343 (1975). 19. G. Bouera, F. Borsa, M. L. Grippa, A. Rigamonti, Phys. Rev., 134, 52 (1971). 20. D. W. Murphy, P. A. Christian, Science, 205, 651 (1979). 21. D. W. Murphy, J. N. Carides, F. J. Di Salvo, C. Cros, J. V. Waszczak, Mater. Res. Bull., 12, 825 (1977). 22. G. Wiegers, Physica, B99, 166 (1980). 23. L. Blandeau, G. Ouvrard, Y. Calage, R. Brec, J. Rouxel, J. Phys. Chem., Solid State Phys., 20, 4271 (1987).

16.4.3.3. Electrochemical Formation. Electrochemical processes are very convenient preparation methods. The intercalates are formed during the discharge of batteries in which the TX, chalcogenide serves as the cathode and the alkali metal, or a source of alkali metal, as the anode. Such cells form the basis of new battery systems and, therefore, deserve special attention. The electrolytic method has been intensively used in the case of sodium and lithium derivatives. To intercalate Ti:-,S, single crystals (x = 0.002; 0.010; 0.020), for example', the crystals are located in a recess machined in a four-bore alumina sheath and held in place by the pressure of a spring contacting wire. The reference electrode is pure lithium or pure sodium, although in this latter case a sodium-vanadium bronze (Na,,,,V,O,) is more convenient, sodium activity having been measured beforehand. Figure 1 shows how this reference material is held in position in the cell. The counterelectrode was either sodium amalgam or lithium metal. The liquid electrolyte phase consists of a saturated solution of either sodium iodide or lithium perchlorate in propylene carbonate. A system currently used to intercalate powders of host structures is the button electrochemical cell (Figure 2). The cathode, made of dichalcogenide mixed with graphite and a binding latex, is pressed in the cover of the cell. The anode is made of a lithium sheet in contact with a nickel grid. The electrolyte is a glass fiber wool soaked with a solution of LiClO, (2 mol L - l ) in a polar organic solvent (propylene carbonate or 1,3-dioxolane for example). In these systems the anode provides the intercalating ions. It is an efficient and mild method in which the intercalation reaction can be controlled through the current passed. The reaction occurs spontaneously in a chemical concentration gradient.

16.4. The Formation of Sheet Structures 16.4.3. Dichalcogenides 16.4.3.3. Electrochemical Formation

296

Argon atmosphere

Sodium-Vanadium Bronze Liquid electrolyte \

Araldite insulator Sodium Amalgam counterelectrode

Figure 1. Electrochemical cell (after ref. 1). A second type of electrolytic method consists in electrolyzing an electrolyte salt233. The most commonly used are aqueous or ammonia solutions, which present the drawback of giving solvent cointercalated species. Defined intercalation stages from cathodic reduction of MoS, in D N S O and D M E electrolytes containing alkali cations, A', are shown in Figure 3. For all ions except Li', a sharp step in potential is found close to 75 mA mg-', corresponding to a composition A~,,,,MoS,. The calculation of stoichiometries from the amount of charge consumed until a step in potential is reached is only valid on the assumptions that (i) no current loss arises by side reactions and that (ii) the intercalation proceeds at a rate that allows a quasi-equilibrium distribution of A t ions in the host at any time3. Electrolytic processes allow the preparation of phases with a definite stoichiometry. However, in cases of two-phase systems it can be difficult to recognize the phase limits (an EMF plateau in the discharge curve means a two-phase region). This can arise from the disruption of the host. Besides the fact that the solvent can play a role, the slabs have to shift during the intercalation process and behave more and more independently of each other. After a great number of cycles a battery involving an A element normally leading to a two-phase region, may show no plateau at all, owing to the disruption of the host. These are the so-called pseudo-monophased systems. Only the first discharge curves under conditions very close to the equilibrium can specify the phase limits. Electrointercalation allows thermodynamic quantities to be derived concerning the intercalation reaction. At the anode: A-At

+e-

Figure 2. Button generator.

(a>

297

16.4. The Formation of Sheet Structures 16 4.3. Dichalcogenides 16.4.3.3. Electrochemical Formation

-1 .o

cs+

-1.5

-2.0 I

n

W

0

I

I

I

I

I

-1.0 -1.5

-2.0 -1.0 -1.5 -2.0 -1 -0

N a*

-1.5

-2.0 -1 .o -1.5

I

Li+

-2.0 -2.5

A0.125Mos 2

0

50

'

100

150

Figure 3. Defined intercalation stages from cathodic reduction of MoS, in dimethylsulfoxide electrolyte solutions containing A + ions (after ref. 3). at the cathode:

MS,

+ e-

and the overall reaction is:

xA

+ MS,

-

-

MS;

A:MSi-

By setting an opposite external voltage, it is possible to control the reaction. If this voltage increases, then the rate of reaction decreases. Finally, intercalation stops when an

16.4. The Formation of Sheet Structures 16.4.3. Dichalcogenides 16.4.3.3. Electrochemical Formation

298

equilibrium voltage has been reached. This is simply related to the free energy of the reaction:

AG = -E,,.F

(4

where F is 96,489 A . s . Figure 4 shows the variation of AG vs. non-stoichiometry in the Li-TiS, system4. In this particular case a continuous decrease illustrates a continuous process of filling the empty sites of the host. When x increases, AG decreases because of the increasing repulsive interactions between the positively charged Li' ions and also because the electrons must be accommodated on levels of increasing energy. Theoretically these two aspects are not easy to take into account. A basic models can be derived as

2.5

\

-

24-

1.9

F5!

1.*\

w

-

@\

- 1 Q)

0

30

0

76th 60 cycle 90

\

-

1.8 0

I

I

0.2

t

I

I

0.4 0.6 X IN Li,Ti S2

I

I

0.8

Figure 4. Discharge curve for Li-TiS, (after ref. 4).

I

1.0

299

16.4. The Formation of Sheet Structures 16.4.3. Dichalcogenides 16.4.3.3. Electrochemical Formation

The variation of the chemical potential of the ions as a function of the filling of the available sites is: p: = p i +

where x,,

+ RTln

number of occupied sites number of empty sites

(el

represents the upper limit for intercalation. This may be rewritten: pi =pi+

+ RTln-

Yi

1 - Yi

where y, = x/x,,, represents the degree of occupancy of the sites by the A+ ions. A similar expression is derived for the electrons: pe = p:

+ RTln-

Ye

1 - Ye

where ye expresses the filling of the E, level. This is a first approximation. In fact the electrons are fermions and are distributed in the lattice in a band of energy width, L. The total band occupancy, ye, should be expressed as:

where P(E) is the Fermi function and D(E) the density of states. This expression can be simplified to the form written above only if the electron band is narrow (L 0),a logical assumption in the case of largely ionic crystals. Finally the electrode potential is: --f

1 RT E = - - ( p A ++ pe-) = E, - -In F F

Y, l-yi

RT F

__ - -In

~

Y, l-ye

This simple expression neglects problems among which interactions between charged particles are of importance. Additional terms, and particularly interaction terms, have to be added in more sophisticated models. The nergy of intercalation of lithium in TiS, is -206 kJ rnol-ls6.A theoretical energy density 3f 480 W . h kg-' can be deduced for a Li-TiS, battery. Figure 4 shows the first dischargc curve of such a cell and the 76th cycle of the battery, both at lOmA The chemical diffusion coefficient, D,of A + ions in A,MS, intercalates can also be determined electrochemically by using a potentiostatic technique in which the decay of current after a voltage step is measured. This coefficient is related to the self-diffusion coefficient, D*, bys:

where a is the activity and c the concentration of the lithium ions. Chemical diffusion coefficients are found to be very high: lo-' cm2 s - l for LiTiS, (x = 1) at RT '.

300

16.4. The Formation of Sheet Structures 16.4.3. Dichalcogenides 16.4.3.4. Molecular Intercalates.

l

I

eUtilization c i r c u i t

Figure 5. Scheme of a battery using an insertion cathode. The requirements for using a dichalcogenide as a cathode in batteries are: 1. A high free energy of formation of the intercalation compounds, which, associated with a light weight, leads to very high energy densities. 2. Little structural change and good reversibility over a wide range of x values. This is the case of Li,TiS, with a continuous Cd1,-NiAs transition between 0 < x 5 1. 3. The dichalcogenide should be chemically stable toward the electrolyte. 4. A good electronic conductivity (which also favors the ionic diffusion).

Figure 5 gives the general scheme of a battery using an insertion cathode. (J. ROUXEL)

1. 2. 3. 4. 5. 6. 7. 8. 9.

D. A. Winn, J. M. Schemilt, B. C. H. Steele, Muter. Res. Bull., 11, 559 (1976). M. S . Whittingham, J . Chem. Soc., Chem. Commun., 328 (1974). J. 0. Besenhard, H. Meyer, R. Schollhorn, Z. Nuturforsch., Teil B, 31, 907 (1976). M. S. Whittingham, J. Electrochem. Soc., 123, 315 (1976). M. Armand, Thesis Grenoble, France, 1978. M. S. Whittingham, Science, 192, 1126 (1976). M. S. Whittingham, Prog. Solid State Chem., 12, 41 (1978). L. S. Darken, Trans. AIME, 175, (1948). M. S. Whittingham, in Electrode Materials and Processes for Energy Conversion and Storage, J. D. E. McIntyre, ed., The Electrochemical Society, Princeton, New Jersey, 1977.

16.4.3.4. Molecular Intercalates.

A considerable number of molecules have been intercalated in group IVA and VA transition-metal disulfides and diselenides'. No stable compounds have been isolated with the ditellurides. Only electron-donor base-type molecules form intercalation compounds, which is a major difference from graphite.

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc. 300

16.4. The Formation of Sheet Structures 16.4.3. Dichalcogenides 16.4.3.4. Molecular Intercalates.

l

I

eUtilization c i r c u i t

Figure 5. Scheme of a battery using an insertion cathode. The requirements for using a dichalcogenide as a cathode in batteries are: 1. A high free energy of formation of the intercalation compounds, which, associated with a light weight, leads to very high energy densities. 2. Little structural change and good reversibility over a wide range of x values. This is the case of Li,TiS, with a continuous Cd1,-NiAs transition between 0 < x 5 1. 3. The dichalcogenide should be chemically stable toward the electrolyte. 4. A good electronic conductivity (which also favors the ionic diffusion).

Figure 5 gives the general scheme of a battery using an insertion cathode. (J. ROUXEL)

1. 2. 3. 4. 5. 6. 7. 8. 9.

D. A. Winn, J. M. Schemilt, B. C. H. Steele, Muter. Res. Bull., 11, 559 (1976). M. S . Whittingham, J . Chem. Soc., Chem. Commun., 328 (1974). J. 0. Besenhard, H. Meyer, R. Schollhorn, Z. Nuturforsch., Teil B, 31, 907 (1976). M. S. Whittingham, J. Electrochem. Soc., 123, 315 (1976). M. Armand, Thesis Grenoble, France, 1978. M. S. Whittingham, Science, 192, 1126 (1976). M. S. Whittingham, Prog. Solid State Chem., 12, 41 (1978). L. S. Darken, Trans. AIME, 175, (1948). M. S. Whittingham, in Electrode Materials and Processes for Energy Conversion and Storage, J. D. E. McIntyre, ed., The Electrochemical Society, Princeton, New Jersey, 1977.

16.4.3.4. Molecular Intercalates.

A considerable number of molecules have been intercalated in group IVA and VA transition-metal disulfides and diselenides'. No stable compounds have been isolated with the ditellurides. Only electron-donor base-type molecules form intercalation compounds, which is a major difference from graphite.

16.4. The Formation of Sheet Structures 16.4.3. Dichalcogenides 16.4.34. Molecular Intercalates.

301

~~~

Molecular intercalates are easy to prepare. Current methods are: 1. Direct reaction of the dichalcogenide and the gaseous molecule under normal pressure at RT or higher. 2. Sealed tube conditions: for example liquid ammonia at RT in thick-walled borosilicate glass tubes. 3. Electrolysis methods using salts, amino for example. 4. Indirect intercalation achieved by first intercalation with NH, or N,H,, and subsequent treatment with the intercalate. 5. Using solutions such as benzene, hydrazine hydrate, aq or n-butanol. The ability to intercalate and the stability of the resulting products depend on such factors as the size of the molecule, its basicity and the electronic properties of the host. Ammonia, for example, is easily intercalated in 2 H TaS, (25"C, 1 d), whereas larger molecules such as pyrimidine require 25 d at 200°C The basicity of the molecules can be expressed by the dissociation constant, pKa. It appears that in TaS,, for example, only molecules with pKa values greater than 4 usually form stable complexes2. Metallic dichalcogenides such as 2 H NbS, or 2 H TaS, lead to numerous stable complexes, which is not true of dichalcogenides with a marked semiconducting character. Only a few complexes of ZrS, and HE, are known with simple molecules such as NH, or N,H,. Nitrogen-donor molecules such as ammonia, amines and pyridines lead also to the most stable species. Detailed structural studies have been performed only in the case of the smallest molecules such as NH,, Ammonia is found to occupy trigonal prismatic sites between the slabs of the host3, which means that the slabs have been shifted from their initial position, but the threefold axis is not perpendicular to the MX, layers as expected. Proton NMR studies on TaS,, NH, and ND,, and structural investigation by means of x-ray and neutron diffraction on TiS,, NH, and TaS, . ND, 5 , 6 show the NH, symmetry axis to be parallel to TaS, or TiS, layers (Fig. 1). An ordering of the intercalated molecules in the van der Waals gap (516.4.3.1)can occur at sufficiently low temperatures. This is the case of TaS,(ND,) at 4 K, but not at 77 K. Intercalate ordering has been noted in TaS,-aniline and NbSe,-cyclopropylamine complexes at RT by electron diffraction7.In TaS,-pyridine the py molecules form a 2 4 a x 13a superlattice in the van der Waals gap8. The py molecules are perpendicular to the MX, slabs, but their C-N axis is parallel to these layers (Fig. 1). Second-stage phases have been prepared for NH, or simple organic molecules. Detailed structural knowledge is lacking. It seems that stacking faults, twinning and

'.

Figure 1. Orientation of NH, and pyridine between the slabs of the host (after ref. 19).

16.4. The Formation of Sheet Structures 16.4.3. Dichalcogenides 16.4.3.4. Molecular Intercalates.

302

TABLE1. a AND c VALUESFOR A FEW 2 H TaS, MOLECULAR INTERCALATES

Intercalate; Stoichiometry

a(pm)

2 H-TaS, Ammonia; 1 Methylamine; 0.50 Ethylene diamine; 0.25 Pyridine, py; 0.50 2-amino-py; 0.50 2-ethyl-py; 0.29 2-phenyl-py, 0.25

331.6 331.9 332.4 332.3 332.6 332.5 332.8 332.8

c(pm) 2 2 2 2 2 2 2 2

x 603.5 x 910 x 923.6

x 952 x 1202 x 1209

x 964 x 1192

screw and edge dislocations, which are very common in the dichalcogenides, play an important role in the formation of the first-stage and second-stage molecular intercalates. The formation of a molecular intercalate is determined by a drastic change in the c parameter of the host according to the size of the intercalated molecule. The a parameter is slightly affected. Table 1 gives a and c values for a few 2 H TaS, intercalates. A characterisiic Ac increase can be associated with each molecule whatever the MX, structure is (-310pm for NH,, for example). The slight increase of the a axis is also characteristic of the intercalated molecule. Concerning the stacking of the slabs, 1 T forms change to 3 R on intercalation, 2 H and 4 H polymorphs are not altered despite lateral shifting of the slabs. Long-chain molecules can lead to a stacking disorder. In the case of the smallest molecules (NH,) for which sufficient space is available in the van der Waals gap 1:l complexes are obtained. With increasing size of the molecule, (molecule),MX, complexes are formed in which x seems to be expressed by l/n, n being an integer'. Considering a series of homologous derivatives, n-alkyl amines for example, interesting results can be deduced from the values of Ac. Two situations can occur.

1. For short molecules (1-5 carbon atoms) Ac is constant, corresponding to the dimensions of molecules oriented with their C-C bonds parallel to the slabs of the host. 2. For longer molecules Ac increases with the number of carbon atoms. Calculations made in this case show the molecules to have their -C-Cskeleton tilted at an angle toward the MX, slabs. This angle tends toward 90" when the number of carbon atoms increasesg310Figure 2 shows the situation in the case of NbS,. Bilayer arrangements of alkylamine chains are present in the van der Waals gap (Fig. 1, 516.4.3.1) with the N ending directed toward the slabs as shown in Fig. 3'. The kinetic studies performed for NH,, ND, or N,H, intercalates11312show that the intercalation process involves a nucleation step. Neutron diffraction studies indicate that several high-stage complexes are formed at the beginning of the intercalation process before the first-stage intercalation is achieved13.A mechanism of reaction consistent with the experimental observations takes into account the intermediate formation of a Langmuir adsorption adduct, producing a basal coverage 8, 12,14,1s : (MX,),, 1

+ I (gaseous intercalant)

hi

b-

+

(MX2)N,.INnBex

,#O

(MX,),,

(a)

16.4. The Formation of Sheet Structures 16.4.3. Dichalcogenides 16.4.3.4. Molecular Intercalates.

I

50

I

I

I

I

I

303

I

I

-

4030 20.:

-

-a

40 -

50

30 20

-

I

4

1

6

I

8

1

10

I

12

I

14

I

16

Nr of C atoms aliphatic chain (n,)

1

18

Figure 2. Interlayer spacing of the dichalcogenides-n-alkylamines complexes, 0 observed, and A calculated according t o different orientations of the alkyl chain in respect to the slabs (after refs. 9 and 10).

16.4 The Formation of Sheet Structures 16.4.3. Dichalcogenides 16.4.3.4. Molecular Intercalates.

304

-1 d

I

1

I

A-

_(a I

(b) Figure 3. Bilayers of n-alkylamines (after ref. 9).

where b = b+/b- is the Langmuir adsorption coefficient. The layered crystal is considered a stacking of N planar macromolecules at a depth i from the surface (No, surface; Ni inside). The energy transferred by gas collisions onto the surface activates coverage and diffusion into the bulk crystal via cracks. Each of these cracks becomes a growth nucleus for a neighboring van der Waals gap. Then, the molecules diffuse convectively into the layers already opened up. The question of the real guest molecule-host molecule interaction is interesting. The orientation of the C, axis of NH, molecules is obviously unfavorable to direct host-guest interaction. An indirect interaction involving weak covalent mixing between the nitrogen lone-pair orbital and electronic layer states has been proposed16. Extensive calculations have been made based on charge-transfer c~mplexes'~. However a detailed investigation of the intercalation reaction products is in favor of a redox process involving the oxidation of a part of the guest molecule's~'g.The resulting electrons are transferred to the host. Considering TaS,NH, :

2 NH,-N,

+ TaS, H + + NH,

x e-

-

-

+ 6 H f + 6 e-

(b)

TaSi-

(c)

NHf

(4

The result is a compound expressed as [NHf],NH,,,_,,[TaS,]'-. In this case x 0.1. The ammonia intercalate appears as an ionic compound in which NH: ions are solvated by a large amount of NH, (Fig. 4). This scheme explains the orientation of the threefold axis of the NH, molecules as a consequence of dipole interactions; NHf ions can be exchanged, and they play the role of A + ions in the A+-ammonia solvated compounds. Both NH, and NHf are highly mobile at RT. Deintercalation at higher temperature is associated with the decomposition of NHf cations.

305

16.4. The Formation of Sheet Structures 16.4.3. Dichalcogenides 16.4.3.4. Molecular Intercalates

C

Figure 4. Metallocene In TaS, (after ref. 23).

A similar model applies to pyridine intercalates with the formation of dipyridine and protonated pyridine moleculesz0: X

xPY--

2

x py

+x H+

x e-

-

(py - py)

+ TaS,

+ x H + + xex (pyH+) TaSi-

-

The resulting product is ( p ~ )- ,,[pyH+],(py ~ , ~ - py)x,2[TaS,]"-, with x 0.2. Guest-guest interactions via hydrogen bonding are present in the van der Waals gap. The ionic model with solvated molecules explains the dependence of the molecular reactivity on the corresponding pKa values by the formation of protonated species. A lot of work in the field of molecular intercalates has been motivated by interest in superconductivity properties. Experimental verification of the excitonic mechanism of superconductivity" can be found in these series. On the other hand, intercalated molecules produce a separation of the slabs of the host, and it is of interest to observe the effect on the superconducting properties of the host ( 2 H NbS, T, = 6.3 K; 3 R NbS, T, = 5.5 K; NbSe, T, = 7.1 K). The slabs may become quasi-independent superconducting circuits and a tunneling effect could be observed. No clear understanding exists concerning the mechanism which the T, is influenced by molecular intercalation. However, a dependence of T, on the orientation of the intercalated n-alkylamines C,H,, + ,NH, has been found". In TaS, complexes, T, decreases monotonically from 4.2 to 1.7 K for n = 1-8, then jumps from 1.7 to 3 K for n = 9 and remains essentially

306

16 4 The Formation of Sheet Structures 16.4.3. Dichalcogenides 16.4.3.5. Hydrated and Solvated Alkali-Metal Phases.

constant for n = 10-18. The jump in T, corresponds to the range (n = 9-11), where the orientation of the alkyl chain bilayer changes in that case from 56 to 75°C. A special case concerns the intercalation of organometallic molecules, and particularly metallocenes. Cobaltocene, for example, intercalates in TaS, with the arrangement shown in Fig. 4 and the composition [CoCp,], 25TaS223. Diffusion is observed at RT. It stops at low temperatures, but the cyclopentadienyl rings still rotate, even at 4 K. Metallocinium ions are formed with an electronic transfer to the host. Three criteria for direct intercalation have been identifiedz4: the organometallic compounds that have been successfully intercalated all have ionization potentials below 6.2 eV, all have sandwich structures with parallel rings and all the corresponding sandwich cations are stable to isolation. (J. ROUXEL)

1. Subba Rao, M. W. Shafer, in Physics and Chemistry of Materials with Layered Structure, Vol. 6, F. A. Levy, ed., D. Reidel, Dordrecht, 1979. An excellent review. 2. F. R. Gamble, J. H. Osiecki, F. J. Di Salvo, J. Chem. Phys., 55, 3525 (1971). 3. J. Cousseau, L. Trichel, J. Rouxel, Bull. Soc. Chim. Fr., 872 (1973). 4. B. G. Silbernagel, F. R. Gamble, Phys. Rev. Lett., 32, 1436 (1974); J. Chem. Phys., 63, 2544 (1975). 5. R. R. Chianelli, J. C. Scanlon, M. S. Whittingham, F. R. Gamble, Inorg. Chem., 14, 1691 (1975). 6. C. Rieckel, D. Hohlwein, R. Schollhorn, J. Chem. Soc., Chem. Commun., 863 (1976). 7. A. R. Beal, W. Y. Liang, Phil. Mug., 27, 1397 (1973). 8. G. S. Parry, C. B. Scruby, P. M. Williams, Phil. Mug., 29, 601 (1974). 9. A. Weiss, R. Ruthardt, Z. Nuturforsch, Teil B, 28, 249 (1973). 10. R. Schollhorn, E. Sick, A. Weiss, Z. Nuturforsch., Teil B, 28, 168 (1973). 11. M. Dines, R. Levy, J. Phys. Chem. 79, 1979 (1972). 12. J. V. Acrivos, C. Delios, N. Y. Topsoe, J. R. Salem, J. Phys. Chem., 79, 3003, (1975). 13. C. Riekel, R. Schollhorn, Muter. Res. Bull., 11, 369 (1976). 14. A. R. Beal, J. V. Acrivos, Phil. Mug., 37, 409 (1978). 15. J. V. Acrivos In Physics and Chemistry of Materials with Layered Structures, Vol. 6, F. A. Levy, ed., D. Reidel, Dordrecht, 1979. 16. F. R. Gamble, B. G. Silbernagel, J. Chem. Physics, 63, 2544 (1975). 17. J. V. Acrivas, J. R. Salem, Phil. Mug., 39-3, 603 (1974). 18. R. Schollhorn, H. D. Zagefka, Angew. Chem., Int. Ed. Engl., 16, 199 (1977). 19. R. Schollhorn, Physicu B, 99, 89 (1980). 20. R. Schollhorn, H. D. Zagefka, T. Butz, A. Lerf, Muter. Res. Bull., 14, 369 (1979). 21. W. A. Little, Phys. Rev. A , 134, 1416 (1964). 22. F. R. Gamble, J. H. Osiecki, M. Cais, R. Pisharody, F. J. Di Salvo, T. H. Geballe, Science, 174, 493 (1971). 23. M. B. Dines, Science, 188, 1210 (1975). 24. W. B. Davies, M. L. H. Green, A. J. Jacobson, J. Chem. Soc., Chem. Commun, 781 (1976).

16.4.3.5. Hydrated and Solvated Alkali-Metal Phases.

When exposed to moisture alkali-metal intercalates lead to hydrated products A,(H20),MS,. Structural distinguish between water monolayer (x y N 1) and water bilayers (y N 2) compounds (Fig. 1). Each type is characterized by a typical expansion of the c parameter (Table 1). The a axis of the host remains unchanged. A bilayer structure is found only for the smallest cations: this can be associated with the higher hydration energies, being sufficient in the case of lithium and sodium to offset the loss in electrostatic lattice energy on separation of the cations from the anionic sulfide layers’. The hydrated products are also formed by reacting the host chalcogenide with

+

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc. 306

16 4 The Formation of Sheet Structures 16.4.3. Dichalcogenides 16.4.3.5. Hydrated and Solvated Alkali-Metal Phases.

constant for n = 10-18. The jump in T, corresponds to the range (n = 9-11), where the orientation of the alkyl chain bilayer changes in that case from 56 to 75°C. A special case concerns the intercalation of organometallic molecules, and particularly metallocenes. Cobaltocene, for example, intercalates in TaS, with the arrangement shown in Fig. 4 and the composition [CoCp,], 25TaS223. Diffusion is observed at RT. It stops at low temperatures, but the cyclopentadienyl rings still rotate, even at 4 K. Metallocinium ions are formed with an electronic transfer to the host. Three criteria for direct intercalation have been identifiedz4: the organometallic compounds that have been successfully intercalated all have ionization potentials below 6.2 eV, all have sandwich structures with parallel rings and all the corresponding sandwich cations are stable to isolation. (J. ROUXEL)

1. Subba Rao, M. W. Shafer, in Physics and Chemistry of Materials with Layered Structure, Vol. 6, F. A. Levy, ed., D. Reidel, Dordrecht, 1979. An excellent review. 2. F. R. Gamble, J. H. Osiecki, F. J. Di Salvo, J. Chem. Phys., 55, 3525 (1971). 3. J. Cousseau, L. Trichel, J. Rouxel, Bull. Soc. Chim. Fr., 872 (1973). 4. B. G. Silbernagel, F. R. Gamble, Phys. Rev. Lett., 32, 1436 (1974); J. Chem. Phys., 63, 2544 (1975). 5. R. R. Chianelli, J. C. Scanlon, M. S. Whittingham, F. R. Gamble, Inorg. Chem., 14, 1691 (1975). 6. C. Rieckel, D. Hohlwein, R. Schollhorn, J. Chem. Soc., Chem. Commun., 863 (1976). 7. A. R. Beal, W. Y. Liang, Phil. Mug., 27, 1397 (1973). 8. G. S. Parry, C. B. Scruby, P. M. Williams, Phil. Mug., 29, 601 (1974). 9. A. Weiss, R. Ruthardt, Z. Nuturforsch, Teil B, 28, 249 (1973). 10. R. Schollhorn, E. Sick, A. Weiss, Z. Nuturforsch., Teil B, 28, 168 (1973). 11. M. Dines, R. Levy, J. Phys. Chem. 79, 1979 (1972). 12. J. V. Acrivos, C. Delios, N. Y. Topsoe, J. R. Salem, J. Phys. Chem., 79, 3003, (1975). 13. C. Riekel, R. Schollhorn, Muter. Res. Bull., 11, 369 (1976). 14. A. R. Beal, J. V. Acrivos, Phil. Mug., 37, 409 (1978). 15. J. V. Acrivos In Physics and Chemistry of Materials with Layered Structures, Vol. 6, F. A. Levy, ed., D. Reidel, Dordrecht, 1979. 16. F. R. Gamble, B. G. Silbernagel, J. Chem. Physics, 63, 2544 (1975). 17. J. V. Acrivas, J. R. Salem, Phil. Mug., 39-3, 603 (1974). 18. R. Schollhorn, H. D. Zagefka, Angew. Chem., Int. Ed. Engl., 16, 199 (1977). 19. R. Schollhorn, Physicu B, 99, 89 (1980). 20. R. Schollhorn, H. D. Zagefka, T. Butz, A. Lerf, Muter. Res. Bull., 14, 369 (1979). 21. W. A. Little, Phys. Rev. A , 134, 1416 (1964). 22. F. R. Gamble, J. H. Osiecki, M. Cais, R. Pisharody, F. J. Di Salvo, T. H. Geballe, Science, 174, 493 (1971). 23. M. B. Dines, Science, 188, 1210 (1975). 24. W. B. Davies, M. L. H. Green, A. J. Jacobson, J. Chem. Soc., Chem. Commun, 781 (1976).

16.4.3.5. Hydrated and Solvated Alkali-Metal Phases.

When exposed to moisture alkali-metal intercalates lead to hydrated products A,(H20),MS,. Structural distinguish between water monolayer (x y N 1) and water bilayers (y N 2) compounds (Fig. 1). Each type is characterized by a typical expansion of the c parameter (Table 1). The a axis of the host remains unchanged. A bilayer structure is found only for the smallest cations: this can be associated with the higher hydration energies, being sufficient in the case of lithium and sodium to offset the loss in electrostatic lattice energy on separation of the cations from the anionic sulfide layers’. The hydrated products are also formed by reacting the host chalcogenide with

+

16.4. The Formation of Sheet Structures 16.4.3. Dichalcogenides 16.4.3.5.Hydrated and Solvated Alkali-Metal Phases.

307

Figure 1. Monolayer and bilayer hydrated phases (after [5]).

-

aqueous solutions of the alkali-metal hydroxides or electrochemically by the electrolysis of aqueous salt solutions at chalcogenide cathodes4: x K+

+ y H,O + x e- + TaS,

(K+),(H,O),[TaS,]"-

(a)

In the product KO50(H,0)o,,o[TaS,]0 5 - the TaS, slabs have shifted in order to form trigonal prismatic sites which are shared by K + and H,O. In the same first system multistage compounds were observed: K~,2,(H,0)o,,,[TaS,]o~z1 )o second stage, stage, K~,,,(H,O), 39[TaS,]o,11- or K o . ~ 2 ( H ~ O,,(TaS,), K~o,(H,O)o26[TaS,]0.07- or K o , ~ ~ ( H , O78(TaS2)3 )o third stage, all having a similar van der Waals gap population. NMR shows a two-dimensional diffusion of the water molecules without any activation of the rotation around the C, axis'. The anisotropic mobility is due to the interaction of the positively polarized hydrogen atoms with the negatively charged MX, layers. The negatively polarized ends of the H,O molecules are engaged in cation dipole interactions with neighboring cations. Neutron diffraction studies on Nao,6(D,0)2VS, show that the oxygen atoms are in hexagonal close packing with the sulfur, sodium being octahedrally surrounded by TABLE 1. INTERLAYER SPACINGS IN COMPOUNDS

THE

CASE OF A,(H,O),TiS,

C spacings (pm) Alkali intercalate

Monolayer

TiS,

569

Bilayer

a axis (pm)

340

308

16.4. The Formation of Sheet Structures 16.4.3. Dichalcogenides 16.4.3.6. Intercalation-Substitution Compounds and Ionic Mobility

oxygen'. Both K t and OH- ions can also be intercalated from KOH solutions and the limit form, K&OH,,[TaS,]1/3-, of a cation-anion intercalation compound has been reached4. Anions other than OH- can be intercalated. Monolayers of NH, molecules are also probably present in all the alkali-ammonia products (Ac 2: 3.08 A). Their formation is directly related to the polarization power of the alkali metal: the biggest alkali metals do not give ammonia-containing compounds, even if those are nonstoichiometric with available free space. In order to remove the ammonia molecules, the products have to be heated to higher temperatures for lithium compounds than for sodium or potassium derivatives. (J, ROUXEL)

1. R. Schollhorn, A. Weiss, Z. Nuturforsch., Teil B, 28, 711 (1973). 2. M. S. Whitlingham, Mater. Res. Bull., 9, 1681 (1974). 3. G. A. Wiegers, R. Van der Meer, H. Van Heininzen, H. J. Kloosterboer, A. J. A. Albernik, Mater. Res. Bull., 9, 1261 (1974). 4. R. Schollhorn, Physica B, 99, 89 (1980). 5. U. Roder, W. Muller-Warmuth, R. Schollhorn, J. Chem. Phys., 70, 2864 (1979). 6. A. J. A. Bos Alberink, R. J. Haange, G. A. Wiegers, J. Less-Common Met., 63, 69 (1979).

16.4.3.6. IntercalationSubstitution Compounds and Ionic Mobility When intercalated in semiconducting dichalcogenides such as ZrS, or Hf S, alkali metals can induce a metallic behavior by transferring one electron to the t2g band. It is possible to suppress this induced electronic conductivity by means of substitutions in the slabs performed along with intercalation between thein','. The compounds NaXML3 'M;?,S, are formed by heating pressed pellets of NaM'S, and ZrS, at 600-800°C:

x NaMS,

+ (1 - x) ZrS,

-

NaxM,Zr, -xS,

(a)

where M = Y, In, for example. In this series three structural models exist for the first-stage compounds. For decreasing values of x starting from NaM'S, (x = l), one finds at first a 3 R Ia phase with an octahedral coordination (more exactly, antitrigonal prismatic, ATP) of the alkali metal, then a 2 H phase and finally a 3 R Ib phase, both presenting a trigonal prismatic (TP) environment of the alkali metal (Fig. 1). The width of these phases depends on the nature of the M'3+ substituted cation: the greater the fractional ionicity of the M'-S bond, the more stable the 3 R octahedral form These ionic conductors could probably not be used as solid electrolytes in operating batteries because when brought into contact with lithium or sodium they would lead to true intercalation compounds A,Na,M:M, -$, (y N 1 - x) with an electronic contribution. However, they provide models for studying ionic mobility in a truly twodimensional space. The absence of electronic conductivity allows complex impedance measurements and not only NMR studies to be performed. The activation energy depends on three factors: (i) the amount of intercalation with an optimum [Na+]/[O] ratio in the van der Waals gap, (ii) the effect of the covalency of the host structure (the greater the fractional ionicities of the M-S and MI-S bonds, the higher the activation energy; this corresponds to what is normally expected if the depths of the potential wells is considered) and (iii) the symmetry of the alkali-metal site.

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc. 308

16.4. The Formation of Sheet Structures 16.4.3. Dichalcogenides 16.4.3.6. Intercalation-Substitution Compounds and Ionic Mobility

oxygen'. Both K t and OH- ions can also be intercalated from KOH solutions and the limit form, K&OH,,[TaS,]1/3-, of a cation-anion intercalation compound has been reached4. Anions other than OH- can be intercalated. Monolayers of NH, molecules are also probably present in all the alkali-ammonia products (Ac 2: 3.08 A). Their formation is directly related to the polarization power of the alkali metal: the biggest alkali metals do not give ammonia-containing compounds, even if those are nonstoichiometric with available free space. In order to remove the ammonia molecules, the products have to be heated to higher temperatures for lithium compounds than for sodium or potassium derivatives. (J, ROUXEL)

1. R. Schollhorn, A. Weiss, Z. Nuturforsch., Teil B, 28, 711 (1973). 2. M. S. Whitlingham, Mater. Res. Bull., 9, 1681 (1974). 3. G. A. Wiegers, R. Van der Meer, H. Van Heininzen, H. J. Kloosterboer, A. J. A. Albernik, Mater. Res. Bull., 9, 1261 (1974). 4. R. Schollhorn, Physica B, 99, 89 (1980). 5. U. Roder, W. Muller-Warmuth, R. Schollhorn, J. Chem. Phys., 70, 2864 (1979). 6. A. J. A. Bos Alberink, R. J. Haange, G. A. Wiegers, J. Less-Common Met., 63, 69 (1979).

16.4.3.6. IntercalationSubstitution Compounds and Ionic Mobility When intercalated in semiconducting dichalcogenides such as ZrS, or Hf S, alkali metals can induce a metallic behavior by transferring one electron to the t2g band. It is possible to suppress this induced electronic conductivity by means of substitutions in the slabs performed along with intercalation between thein','. The compounds NaXML3 'M;?,S, are formed by heating pressed pellets of NaM'S, and ZrS, at 600-800°C:

x NaMS,

+ (1 - x) ZrS,

-

NaxM,Zr, -xS,

(a)

where M = Y, In, for example. In this series three structural models exist for the first-stage compounds. For decreasing values of x starting from NaM'S, (x = l), one finds at first a 3 R Ia phase with an octahedral coordination (more exactly, antitrigonal prismatic, ATP) of the alkali metal, then a 2 H phase and finally a 3 R Ib phase, both presenting a trigonal prismatic (TP) environment of the alkali metal (Fig. 1). The width of these phases depends on the nature of the M'3+ substituted cation: the greater the fractional ionicity of the M'-S bond, the more stable the 3 R octahedral form These ionic conductors could probably not be used as solid electrolytes in operating batteries because when brought into contact with lithium or sodium they would lead to true intercalation compounds A,Na,M:M, -$, (y N 1 - x) with an electronic contribution. However, they provide models for studying ionic mobility in a truly twodimensional space. The absence of electronic conductivity allows complex impedance measurements and not only NMR studies to be performed. The activation energy depends on three factors: (i) the amount of intercalation with an optimum [Na+]/[O] ratio in the van der Waals gap, (ii) the effect of the covalency of the host structure (the greater the fractional ionicities of the M-S and MI-S bonds, the higher the activation energy; this corresponds to what is normally expected if the depths of the potential wells is considered) and (iii) the symmetry of the alkali-metal site.

309

16.4. The Formation of Sheet Structures 16.4.4. Trichalcogenohypophosphates.

0

s

O

M

@

Alkali metal

0

0 b

0 D

0 1TMS2

3RtIaI

2H [Ib]

Figure 1. Structural types of Na,M~M,-,S, compounds (after refs. 1 and 2). Activation energies are found to be lower in T P sites than in 0, sites [0.22eV and 0.28 eV, respectively in the Na(1n - Zr) series]. Sodium ions have to move along distorted tetrahedral voids in the case of the octahedral structure, whereas large windows exist between adjacent T P sites in the second case. Furthermore the c/a ratio is higher in the case of T P phases, corresponding to more distant sulfur layers and consequently more favorable windows from site to site. Similar conclusions are drawn in the case of lamellar oxides (see 516.4.6). ( J . ROUXEL)

1. L. Trichet, J. Rouxel, Muter. Res. Bull., 12, 345 (1977). 2. M. Danot, P. Colombet, Physzcu B, 99, 117 (1980).

16.4.4. Trichalcogenohypophosphates. Trichalcoenohypophosphates1,2represent a new class of layer materials with general formula MPY, (M = V2', Mn", Fez', Co2+,Ni", Mg2+,Zn2', Cd2+,Sn2' and Y = S, Se). The structure is built up with slabs similar to those in TiS,, except that titanium has been replaced in the octahedral voids by the M atoms and P-P pairs with a ratio of 2/3 and 1/3 (Fig. 1). The formula may be expressed as Mz/3(Pz)l/3Yz. The P-P bond length is 219-233pm. A chalcogen van der Waals gap (Fig. 1, $16.4.3.1) exists between the slabs. In the sulfide derivatives, the stacking of the chalcogen layers is cubic face centered (ABC) and not hexagonal closest packed (AB) as in TiS,. Another difference concerns the positions of the P-P pairs and M atoms from one slab to another in relation to the

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc.

309

16.4. The Formation of Sheet Structures 16.4.4. Trichalcogenohypophosphates.

0

s

O

M

@

Alkali metal

0

0 b

0 D

0 1TMS2

3RtIaI

2H [Ib]

Figure 1. Structural types of Na,M~M,-,S, compounds (after refs. 1 and 2). Activation energies are found to be lower in T P sites than in 0, sites [0.22eV and 0.28 eV, respectively in the Na(1n - Zr) series]. Sodium ions have to move along distorted tetrahedral voids in the case of the octahedral structure, whereas large windows exist between adjacent T P sites in the second case. Furthermore the c/a ratio is higher in the case of T P phases, corresponding to more distant sulfur layers and consequently more favorable windows from site to site. Similar conclusions are drawn in the case of lamellar oxides (see 516.4.6). ( J . ROUXEL)

1. L. Trichet, J. Rouxel, Muter. Res. Bull., 12, 345 (1977). 2. M. Danot, P. Colombet, Physzcu B, 99, 117 (1980).

16.4.4. Trichalcogenohypophosphates. Trichalcoenohypophosphates1,2represent a new class of layer materials with general formula MPY, (M = V2', Mn", Fez', Co2+,Ni", Mg2+,Zn2', Cd2+,Sn2' and Y = S, Se). The structure is built up with slabs similar to those in TiS,, except that titanium has been replaced in the octahedral voids by the M atoms and P-P pairs with a ratio of 2/3 and 1/3 (Fig. 1). The formula may be expressed as Mz/3(Pz)l/3Yz. The P-P bond length is 219-233pm. A chalcogen van der Waals gap (Fig. 1, $16.4.3.1) exists between the slabs. In the sulfide derivatives, the stacking of the chalcogen layers is cubic face centered (ABC) and not hexagonal closest packed (AB) as in TiS,. Another difference concerns the positions of the P-P pairs and M atoms from one slab to another in relation to the

310

16.4. The Formation of Sheet Structures 16 4.4. Trichalcogenohypophosphates.

Figure 1. The MPY, slab.

rhombohedral symmetry. In addition a slight distortion leads to a monoclinic unit cell (space group C2/m). All the selenides present a rhombohedral symmetry and an hexagonal stacking (AB) of the anions with the exception of NiPSe,, which is monoclinic with the ABC stacking like the sulfides. The structural model MPS, can also be compared to the CrC1, one where a third of the octahedral sites remains empty (0)within the slabs Cr2,,!J1,,C12. Mixed cations forms have been prepared in the MPY, series. This is for the case (CuiAg+),,,,Cr~,~,PY3 phases,. The M2' cation can also be replaced by M 3 + ions and vacancies: 2 M3'

+ 0= 3 M2+,

(4

is a scheme that is realized in In,,,O,,,PY, derivatives4. Some of the MPY, compounds, in particular NiPS,, FePS, and FePSe,, are active cathode materials in lithium cells's6. Intercalation can also be achieved through the butyllithium technique (see 516.4.3.2). The reaction: MPX,

+ Li

Li' (through electrolyte) >LixMPS, e- (through external circuit)

(b)

takes place in a battery: Li-LiC10, in propylene carbonate-pressed MPY,. n-Butyllithium solutions in hexane at 25°C lead to: MPY,

+ x LiC4H,

-

Li,MPY3

+ -X2 C,H,,

(c>

The real intercalation process is limited to 1.5 Li per NiPS,. When the product contains a higher ratio of Li (or other alkali metal) to MPS,, this indicates a reaction between the host and alkali metal, not an intercalation process.

16.4 T h e Formation of Sheet Structures 16.4 4. Trichalcogenohypophosphates.

31 1

k L i,N i P S,

2-

1

0

1

I

2

X

*

Figure 2. Discharge curve for a Li,NiPS3 battery. The best cathode material is NiPS,. No parameter expansion is observed during intercalation. The space group is C2/m. Lithium can occupy 2d or 4h octahedral sites or 4i and 8j tetrahedral sites. A localization in the octahedral voids is consistent with the following points: (i) the Li-S distances (254 pm) are very similar to those found in LiTiS, (256 pm), where lithium is also in octahedral sites; (ii) the number of available sites leads to the limit formula Li,,,NiPS3, which is experimentally observed; (iii) the small size of the tetrahedral voids will lead to a parameter expansion. On the discharge curve (Fig. 2 ) there is a small but reproducible discontinuity at x = 0.5. It is related to the filling of the first type of octahedral site (2d sites). On contact with air or moisture, the NiPS, intercalates undergo spontaneous hydration. The x-ray spectra show an expansion of 58 pm in the direction perpendicular to the layers, which is consistent with a bilayer of water around lithium (see $16.4.3.5). Intercalation is not possible in all MPY, phases. The ability to intercalate seems to be related to the band gap of the host structures’. Optical absorption measurements performed on single crystals of NiPS,, FePS,, FePSe,, MnPS,, MnPSe,, CdPS,, show all members of the series to be broad-band semiconductors (Table 1). The three materials NiPS,, FePS, and FePSe,, that have proved to be the best cathodes appear to have energy gaps substantially lower than the materials that do not intercalate or intercalate poorly. Molecular intercalates of MPX, phases have been prepared with amines as well as metallocenes. By heating MnPSe, at 30°C for 4 d under the saturated vapor pressure of pyridine, a complex MnPSe,(py),,, is obtained. This compound can partially lose its pyridine amount to give MnPSe,(py),,,. These two compounds present two different orientations of the intercalated molecules. According to one-dimensional electron density

312

16 4. The Formation of Sheet Structures 16.4 4. Trichalcogenohypophosphates. TABLE1. ABSORPTIONEDGEOF SOMEMPY, PHASES

MnPS, MnPSe, FePS, FePSe, NiPS, ZnPS, CdPS, 1n2/3PS3

1n2/3PSe3

3.0 eV 2.5 eV 1.5 eV 1.3 eV 1.6 eV 3.4 eV 3.5 eV 3.1 eV 1.9 eV

on the c axis and considering the expansion of the parameter in the direction perpendicular to the MnPSe, slabs, the pyridine molecules should be perpendicular to the slabs in ( p ~ ) ~(Ac , , = 590pm) and parallel to the slabs in ( p ~ )(Ac ~ ,=~ 338 pm)'. Intercalation in FePSe, leads to the ( p ~ ) ~derivative /, (Ac = 590 pm). By reaction with sandwich organometallic compounds such as cobaltocene or bisbenzene chromium, MPS, derivatives corresponding to Mn, Zn, Fe, Ni, can undergo intercalationg.Intercalation depends on the host structure (semiconducting band gap) and how the metallocenes are allowed to react. When treated with metallocene solutions in toluene, MnPS, and ZnPS, accept only cobaltocene. The increase (by pm's E 530 pm) of the basal spacings compares with the value observed when cobaltocene is intercalated in dichalcogenides (535pm in ZrS,). FePS, and NiPS, with narrower band gaps intercalate cobaltocene but also bis-benzenechromium [(q6-C6H6),Cr]. The high stoichiometries, e.g., Z~PS,[CO(~~-C,H,),],,,~, MnPS,[Co(y5-C,H,),],,,, and F~PS,[CO(~~-C,H,),],,,~, are consistent with the guest molecules being closely packed within the layers and in an arrangement similar to dichalcogenide, with the rings perpendicular to the host slabs. Using benzene as a solvent, (qb-C6H6),Cr can be intercalated in MnPS, (also in ZnPS,). A small amount of benzene is cointercalated. This is also the case of cobaltocene, which leads to MnPS,[(q5-C,H5),Co], 24[C6H6]0,12with a Ac increased as compared with the benzene free derivative. New possibilities of intercalation are provided by using organometallic salts; e.g., cobaltocene, but also chromocene is intercalated in MnPS, using solutions of iodide salts in dry ethanol or water. As in the case of dichalcogenides or oxyhalides, metallocenes with low first ionization potentials are found to intercalate easily into MPS, phases. This suggests an ionization scheme [(q5-Cp),M] '[MPS,] , which is also consistent with the intercalation of lithium in NiPS, and FePS,. However, intercalation in ZnPS, cannot be explained in the same way. Interactions between the cyclopentadienyl ligands and the chalcogen layers may play an important role. The mechanism of the intercalation reaction may also involve a dismutation of the [P2S,] groups (which is also possible in the case of lithium intercalates). In fact, IR and UV-visible absorption spectroscopies clearly confirm that the intercalated guest metallocene is fully cationic, but the donated electrons to the host appear to be localized and paired according the diamagnetic properties". (J ROUXEL)

16.4. The Formation of Sheet Structures 16.4.5. Transition-Metal Oxyhalides. 1. 2. 3. 4. 5. 6. 7. 8. 9. 10.

313

W. Klingen, G. Eulenberger, H. Hahn, Naturwzssenschaften, 57, 88 (1970). C. D. Carpentier, R. Nitsche, Mat. Res. Bull., 9, 1097 (1974). A. Le Blanc, J. Rouxel, C. R . Hebd. Seances Acad. Scz., Ser. C, 274, 786 (1972). S. Soled, A. Wold, Mater. Res. Bul., 11, 657 (1976). A. Le Mehaute, G. Ouvrard, R. Brec, J . Rouxel, Mater. Res. Bull., 12, 1191 (1977). A. H. Thompson, M. S. Whittingham, Mater. Res. Bull., 12, 741 (1977). R. Brec, D. M. Schleich, G. Ouvrard, A. Louisy, J. Rouxel, Inorg. Chem., 18, 1814 (1979). S. Otani, M. Shimada, F. Kanamaru, M. Koizumi, Inorg. Chem., 19, 1249 (1980). R. Clement, M. L. H. Green, J. Chem. Soc., Dalton Trans., 10, 1566 (1979). J. P. Audiere, R. Clement, Y. Mathey, C. Mazieres, Physica, 99B, 133 (1980).

16.4.5. Transition-Metal Oxyhalides. Iron oxychloride FeOCl presents a layered structure encountered in the various MOX derivatives of 3d elements that have been prepared, i.e., TiOCl, VOCI, CrOCl, FeOC1, TiOBr, VOBr, CrOBr. Table 1 shows the parameters of the orthorhombic unit cell. The structure is built up with sheets made of two corrugated metal-oxygen layers surrounded by two halogen layers (Fig. 1). The M 3 +ions are octahedrally surrounded by four oxygen and two chlorine ions as shown in Fig. 2 for CrOCI. These octahedra share edges in order to form the idealized model of Fig. 3'. The FeOCl structure is met with in cations that prefer octahedral coordination. It can be approximated to some distortion of the PbFCl type. In the tetragonal PbFC1-type compounds, the corrugated M-0 layers are replaced by planar M-F layers. The PbFCl structure is also found for oxyhalides (for example, in the LnOCl series from manthanum to erbium). The FeOC1-type oxyhalides can be intercalated by lithium and various molecules. Molecular intercalates of FeOCl and AlOCl (which has a different layer structure) were However, the chemistry of lamellar in fact obtained before those of dichal~ogenides~.~. oxyhalides is distinguished from that of lamellar dichalcogenides by a substitution reaction in which the outer layers of the slabs (ie., the chlorine layers) are easily replaced by other ion layers such as OH-, [NHJ-. The structure thus lends itself not only to reversible topochemical reactions by intercalation, but also to irreversible topochemical

TABLE1. UNITCELLPARAMETERS OF SOMEMOX OXYHALIDES WITH FeOCl STRUCTURAL TYPE b (pm) FeOCl TiOCl VOCl CrOCl TiOBr VOVr CrOBr

378 379 377 386 378 377 386

792 803 79 1 769 853 842 836

330 338 330 318 348 338 323

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc. 16.4. The Formation of Sheet Structures 16.4.5. Transition-Metal Oxyhalides. 1. 2. 3. 4. 5. 6. 7. 8. 9. 10.

313

W. Klingen, G. Eulenberger, H. Hahn, Naturwzssenschaften, 57, 88 (1970). C. D. Carpentier, R. Nitsche, Mat. Res. Bull., 9, 1097 (1974). A. Le Blanc, J. Rouxel, C. R . Hebd. Seances Acad. Scz., Ser. C, 274, 786 (1972). S. Soled, A. Wold, Mater. Res. Bul., 11, 657 (1976). A. Le Mehaute, G. Ouvrard, R. Brec, J . Rouxel, Mater. Res. Bull., 12, 1191 (1977). A. H. Thompson, M. S. Whittingham, Mater. Res. Bull., 12, 741 (1977). R. Brec, D. M. Schleich, G. Ouvrard, A. Louisy, J. Rouxel, Inorg. Chem., 18, 1814 (1979). S. Otani, M. Shimada, F. Kanamaru, M. Koizumi, Inorg. Chem., 19, 1249 (1980). R. Clement, M. L. H. Green, J. Chem. Soc., Dalton Trans., 10, 1566 (1979). J. P. Audiere, R. Clement, Y. Mathey, C. Mazieres, Physica, 99B, 133 (1980).

16.4.5. Transition-Metal Oxyhalides. Iron oxychloride FeOCl presents a layered structure encountered in the various MOX derivatives of 3d elements that have been prepared, i.e., TiOCl, VOCI, CrOCl, FeOC1, TiOBr, VOBr, CrOBr. Table 1 shows the parameters of the orthorhombic unit cell. The structure is built up with sheets made of two corrugated metal-oxygen layers surrounded by two halogen layers (Fig. 1). The M 3 +ions are octahedrally surrounded by four oxygen and two chlorine ions as shown in Fig. 2 for CrOCI. These octahedra share edges in order to form the idealized model of Fig. 3'. The FeOCl structure is met with in cations that prefer octahedral coordination. It can be approximated to some distortion of the PbFCl type. In the tetragonal PbFC1-type compounds, the corrugated M-0 layers are replaced by planar M-F layers. The PbFCl structure is also found for oxyhalides (for example, in the LnOCl series from manthanum to erbium). The FeOC1-type oxyhalides can be intercalated by lithium and various molecules. Molecular intercalates of FeOCl and AlOCl (which has a different layer structure) were However, the chemistry of lamellar in fact obtained before those of dichal~ogenides~.~. oxyhalides is distinguished from that of lamellar dichalcogenides by a substitution reaction in which the outer layers of the slabs (ie., the chlorine layers) are easily replaced by other ion layers such as OH-, [NHJ-. The structure thus lends itself not only to reversible topochemical reactions by intercalation, but also to irreversible topochemical

TABLE1. UNITCELLPARAMETERS OF SOMEMOX OXYHALIDES WITH FeOCl STRUCTURAL TYPE b (pm) FeOCl TiOCl VOCl CrOCl TiOBr VOVr CrOBr

378 379 377 386 378 377 386

792 803 79 1 769 853 842 836

330 338 330 318 348 338 323

314

16.4. The Formation of Sheet Structures 16.4.5. Transition-Metal Oxyhalides.

0

CI

0 0 1i z=o 0 M

CI

M

Figure 1. The MOCl (FeOC1) structural

-

reactions involving the complete replacement of the external layers of the sheets, leaving only the inner layers unchanged: FeOCl

+ NH,

+ NH,Cl

FeONH,

-

(a)

The reaction with water, which leads to y-FeO(0H) with a similar structure, also belongs to this type: FeOCl

+ H,O

FeO(0H)

+ HCl

(b)

Lithium intercalation in oxyhalides can be realized either by the butyllithium technique or electrochemically. The upper limit of the reaction: x C,H,Li

+ MOCl

+ LixMOCl

!!C,H,, 2

-

(c)

corresponds to x N 1. This reaction is performed at RT by contact between FeOCl and n-butyllithium solutions in hexane. Electrochemically the reaction is:

x Lit

+ x e- + MOCl

LixMOCl

(4

Figure 4 shows various voltametric sweeps4.In the case of FeOCl two well-defined waves corresponding to the intercalation and deintercalation processes are apparent. The value of the potential at the cathodic current peak gives E,(FeOCl) = -270mVvs. Ag/Agt and indicates the case of reduction of the Fe3+ species. The reoxidation wave at - 80 mV corresponds to a one-cycle charge recuperation > 80 %. This is in agreement with the

16.4. The Formation of Sheet Structures 16.4.5. Transition-Metal Oxyhalides.

't.

315

03

C

"

Figure 2. The CrO,C1, octahedron in CrOCI.

Figure 3. Idealized MOX sheets (after ref. 1).

02

316

16.4. The Formation of Sheet Structures 16.4.5.Transition-Metal Oxyhalides

FeOCl

CH,CN

-270

I

V

Figure 4. Various voltametric sweeps (after ref. 3).

good reversibility observed chemically. Since the boundary conditions are not the same, only the reduction peak obtained with uncycled FeOCl is to be taken as E,. However, the small difference ( < 200 mV) between the anodic and cathodic peak potentials gives a qualitative clue of a fast electrochemical process illustrating a fast diffusion of the Li' ions in the solid. The two chromium oxyhalides are characterized by the stability of the Cr3' ion. The chloride gives E,(CrOCl) = - 1340mV and the bromide, E,(CrOBr) = -940mV. As can be predicted, the C1- ions stabilize the higher valency of chromium than the Br- ions more effectively, as reflected by the more negative potential of reduction. No reoxidation peak appears, indicating that the electrochemical process is irreversible in the solvent used. Similar conclusions can be drawn concerning the stability of V3' ions compared to V2' ions and the nonreversibility of the process for VOCl. The structure of Li,MOCl derivatives is unknown. It is, however, highly probable that the Li' ions occupy octahedral positions between M,O,Cl, sheets, where they are surrounded by five C1- and one 0'- ion. In FeOCl four Li-Cl distances of 252 pm, one Li-Cl of 245 pm, and one Li-0 distance of 247 pm can be compared with Li-Cl = 256 pm in LiCl and Li-0 = 230 pm in Li,O. A galvanostatic reduction of FeOCl in DMSO electrolytes solutions5 results in nonstoichiometric intercalation compounds M,(DMSO),(Fe: ?,Fel +OCl). An expanded lattice is observed owing to the presence of the solvated M + cations between the slabs of the host. From the length of the first reduction plateau upper limits of x = 0.8 - 0.1 for alkali metals, x = 0.5 for M + = H + and x N 0.03 for M + = tetraalkylammonium are found.

16.4. The Formation of Sheet Structures 16.4.5. Transition-Metal Oxyhalides.

317

TABLE2. BASALSPACINGSIN FeOCl (M)(l/n) INTERCALATES (AFTER REF 6). __ PK, of the guest b (pm) ~~~~

FeOCl FeOCl (py) FeOCl (DMP) i FeOCl (AP) FeOCl (TMP) FeOCl (PA)

4 4

5.2 6.8 9.2 9.6 10.5

792 1327 1498 1357 1179 1189

As expected the formation of molecular intercalates in MOX structures results in a significant expansion of the b parameter in the direction perpendicular to the M,0,X2 slabs. Table 2 gives some basal spacings in FeOC1-py derivatives complexes6. As in the case of dichalcogenides intercalation depends on the pKa of the guest molecules. y-Ray recoilless resonances fluorescence spectroscopy shows a significant electronic transfer from guest to FeOCl layers. The resistivity of FeOCl crystals strongly decreases from lo6 C2 cm to values of about 102-103,owing to a rapid electronic exchange between Fez+ and Fe3+ ions. The planes of the pyridine rings are perpendicular to the host layers so that the nitrogen atoms face the layers. However in the case of substituted molecules or some alkylamines the molecules can be tilted slightly from a plane perpendicular to the layers6. Both NMR and neutron diffusion studies on picolines intercalated in VOCl show the guest molecules to be tilted. Above 250 K molecular reorientation processes are observed and have the effect of switching between a number of equivalent orientations that are related to each other by the symmetry operations of the cage7. Organometallic derivatives, and particularly ferrocene and cobaltocene, can be intercalated in FeOC18,9. The van der Waals gap (Fig. 1,516.4.3.1)expands by about 500 pm. In addition there is a translation of the FeOCl slabs resulting in a 2b superstructure. Table 3 shows the structural parameters of intercalates in FeOCl and also TiOCl and VOC19. The Ab expansions are consistent with a structural model where the cyclopentadiene rings of the guest are perpendicular to the host layers. TABLE3. METALLOCENE INTERCALATES IN FeOC1, TiOCl Host FeOCl FeOCl FeOCl FeOCl VOCl VOCl TiOCl TiOCl

Guest C~(Y~-CP), Fe(vS-Cp), Fe(EtMe,Cp-$), C~(Y’-~P), CoW-Cp),

AND

VOCl (AFTER REF. 10).

a (Pm)

b (Pm)

c (Pm)

377.0 380.3 379.2

790.7 2 x 1284.9 2 x 1303.2 2 x 1546 791.1 2 x 1277 803 2 x 1316

330.2 333.5 332.4

376.5 384.6 379 382

329.7 330.8 336 334

Layer expansion (P4 -

494 513 755 486 -

512

318

16.4. The Formation of Sheet Structures 16.4.6. Lamellar Oxides.

According to the available space a maximum stoichiometry of FeOClFe[(Cp-y'),] 1,6 can be inferred, which is exactly observed. ?-Ray recoilless resonance fluorescence spectra of FeOCl [Fe(Cp-q5),] 1,6 show an electronic transfer from guest to FeOCl hostg~". Ferricinium ions are formed between the slabs, which can be expressed by the ionization scheme [Fe(Cp-y5),];,,6 [FeOC1]0,'6-. Electrical measurements on intercalated FeOCl single crystals in the direction parallel to the slabs show a large decrease on resistivity as compared to the host structure value. (J. ROUXEL)

1. F. Hulliger, Structural Chemistry oflayer-type Phases, F. Levy, ed., D. Reidel, Dordrecht, 1976, p. 18, An excellent review of layered structures. 2. P. Hagenmuller, J. Rouxel, C. R. Hebd. Seances Acad. Sci., 250, 1859 (1960). 3. P. Hagenmuller, J. Rouxel, J. Portier, C.R. Hebd. Seances Acad. Sci., 254, 2000 (1962). J. Rouxel, J. Portier, C.R. Hebd. Seances Acad. Sci., 254, 2000 (1962). 4. M. Armand, L. Coi'c, P. Palvadeau, J. Rouxel, J. Power Sources, 3, 137 (1978). 5. H. Meyer, A. Weiss, Mater. Res. Bull., 13, 913 (1978). 6. S . Kikkawa, F. Kanamaru, M. Koizumi, Bull. Chem. SOC.Jpn., 52, 4, 963 (1979). 7. S. Clough, J. P. Venien, P. Palvadeau, unpublished work. 8. Thomas R. Halbert, J. Scanlon, Mater. Res. Bull., 14, 415 (1979). 9. L. Coi'c, Dissertation, Nantes, France, 1978. 10. Thomas R. Halbert, D. C. Johnston, L. E. McCandlish,A. H. Thompson, J. C. Scanlon, Physica, 99B, 128 (1980). 11. Schafer-Stahl,R. Abele, Z. Anorg. Allg. Chem., 465, 147 (1980).

16.4.6. Lamellar Oxides. Two-dimensional materials are built up with slabs in which there is strong ionic, covalent or metallic bonding, whereas they are separated by rather large distances in agreement with weak van der Waals intelayer bonding (see $16.4.3.1). The van der Waals gap (between the slabs) is bounded on each side by atomic arrangements of the same nature. Its width depends on the electronegativity of these atoms. For a given structural type to be stable the bonding through the van der Waals gap (Fig. 1, $16.4.3.1) must stabilize the structure in front of the repulsion between the same atomic layers situation on each side. This explains why the layered structures are to be found in the lower ionicities region in a classical ionicity-structure diagram (Fig. 1). Such a diagram, here for MX, derivatives, consists of plotting an average principal quantum number n (from the individual principal quantum numbers of the components) vs. the M-X electronegativity difference'. In the case of oxides the repulsion between adjacent layers would be quite strong and very destabilizing. Instead of the layered model observed for numerous transitionmetal dichalcogenides, most of the MO, oxides take the rutile arrangement. However, A,MO, derivatives are often isostructural with A,MS, intercalation compounds'. The sodium ions located in the van der Waals gap stabilize the two-dimensional arrangement. In the A,MO, layer oxides M is a transition metal with two oxidation states or a mixture of tetravalent and trivalent (or divalent) elements. A wide range of composition, as observed in the case of chalcogenides (0 < x 5 1) has not been found. The higher ionic character of the bonds requires x > 0.5.

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc. 318

16.4. The Formation of Sheet Structures 16.4.6. Lamellar Oxides.

According to the available space a maximum stoichiometry of FeOClFe[(Cp-y'),] 1,6 can be inferred, which is exactly observed. ?-Ray recoilless resonance fluorescence spectra of FeOCl [Fe(Cp-q5),] 1,6 show an electronic transfer from guest to FeOCl hostg~". Ferricinium ions are formed between the slabs, which can be expressed by the ionization scheme [Fe(Cp-y5),];,,6 [FeOC1]0,'6-. Electrical measurements on intercalated FeOCl single crystals in the direction parallel to the slabs show a large decrease on resistivity as compared to the host structure value. (J. ROUXEL)

1. F. Hulliger, Structural Chemistry oflayer-type Phases, F. Levy, ed., D. Reidel, Dordrecht, 1976, p. 18, An excellent review of layered structures. 2. P. Hagenmuller, J. Rouxel, C. R. Hebd. Seances Acad. Sci., 250, 1859 (1960). 3. P. Hagenmuller, J. Rouxel, J. Portier, C.R. Hebd. Seances Acad. Sci., 254, 2000 (1962). J. Rouxel, J. Portier, C.R. Hebd. Seances Acad. Sci., 254, 2000 (1962). 4. M. Armand, L. Coi'c, P. Palvadeau, J. Rouxel, J. Power Sources, 3, 137 (1978). 5. H. Meyer, A. Weiss, Mater. Res. Bull., 13, 913 (1978). 6. S . Kikkawa, F. Kanamaru, M. Koizumi, Bull. Chem. SOC.Jpn., 52, 4, 963 (1979). 7. S. Clough, J. P. Venien, P. Palvadeau, unpublished work. 8. Thomas R. Halbert, J. Scanlon, Mater. Res. Bull., 14, 415 (1979). 9. L. Coi'c, Dissertation, Nantes, France, 1978. 10. Thomas R. Halbert, D. C. Johnston, L. E. McCandlish,A. H. Thompson, J. C. Scanlon, Physica, 99B, 128 (1980). 11. Schafer-Stahl,R. Abele, Z. Anorg. Allg. Chem., 465, 147 (1980).

16.4.6. Lamellar Oxides. Two-dimensional materials are built up with slabs in which there is strong ionic, covalent or metallic bonding, whereas they are separated by rather large distances in agreement with weak van der Waals intelayer bonding (see $16.4.3.1). The van der Waals gap (between the slabs) is bounded on each side by atomic arrangements of the same nature. Its width depends on the electronegativity of these atoms. For a given structural type to be stable the bonding through the van der Waals gap (Fig. 1, $16.4.3.1) must stabilize the structure in front of the repulsion between the same atomic layers situation on each side. This explains why the layered structures are to be found in the lower ionicities region in a classical ionicity-structure diagram (Fig. 1). Such a diagram, here for MX, derivatives, consists of plotting an average principal quantum number n (from the individual principal quantum numbers of the components) vs. the M-X electronegativity difference'. In the case of oxides the repulsion between adjacent layers would be quite strong and very destabilizing. Instead of the layered model observed for numerous transitionmetal dichalcogenides, most of the MO, oxides take the rutile arrangement. However, A,MO, derivatives are often isostructural with A,MS, intercalation compounds'. The sodium ions located in the van der Waals gap stabilize the two-dimensional arrangement. In the A,MO, layer oxides M is a transition metal with two oxidation states or a mixture of tetravalent and trivalent (or divalent) elements. A wide range of composition, as observed in the case of chalcogenides (0 < x 5 1) has not been found. The higher ionic character of the bonds requires x > 0.5.

319

16.4. The Formation of Sheet Structures 16.4.6. Lamellar Oxides.

0

2

1 electronegativity

difference

3

AX

Figure 1. Ionicity structure diagram for MX, compounds (after ref. 1). The structural types are similar to those described in $16.4.3.2and 16.4.3.6for alkalimetal intercalates of dichalcogenides. There are: (i) the 3 R octahedral exactly antitrigonal prismatic (ATP) structure with the ABCABC stacking of the oxygen layers [this model, well illustrated by a-NaFeO,, is obtained with a small vacancy content (x < l)]; (ii) the 3 R trigonal prismatic (TP) structure with the sequence AABBCC; (iii) the 2 H T P structure corresponding to an ABBA stacking. In all cases the M element remains octahedrally surrounded between A and B, C and A and B and C layers in the 3 R ATP model; between A and B, B and C and C and A in the 3 R T P model; and between A and B and B and A in the 2 H structure. These edge-sharing (MO,) octahedra build (MO,), sheets. The alkali-metal ions separate these sheets due to their location between B and C, A and B and C and A in the 3 R ATP model; between A and A, B and B and C and C, in the 3 R T P model and between B and B and A and A in the 2 H T P structure (Fig. 2). The AMO, stoichiometric forms are prepared by heating mixtures of A,O and M,O, oxides at 450"C, e.g.: Na,O

+ Cr,O,

-

2 NaCrO,

Either Na,O, or Na,CO,, easier to handle, can replace Na,O.

(a)

320

16.4. The Formation of Sheet Structures 16.4.6. Lamellar Oxides.

3 Rla (ATP)

2 H (TP)

3 Rlb (TP)

Figure 2. Stacking modes in layer oxides (after ref. 7).

321

16.4. The Formation of Sheet Structures 16.4.6. Lamellar Oxides.

Nonstoichiometric forms result from preparations involving oxides corresponding to different oxidation states of the same M metal or of two or more different metals3 at 450°C:

x Na,O

+ 2(1-

x) MnO,

at 1000°C

x K,O

+ x In,O, + 2(1 - x) SnO,

at 450°C

3x Na,O,

5

+ 2 Co,O,

-

+ x Mn,O,

-

6 Na,CoO,

for x > at 450°C under oxygen pressure: 3x Na,O,

+ 2 Co30, + (2 - 3x) 0,

for x 5 3 at 450°C: (2 - X)KO,

2 Na,MnO,

(b)

2 K,(In,Sn, -x)Oz

(c)

+ (3x - 2) 0,

(4

6 Na,CoO,

(el

-

+ ( 2 -~ 1) KZO + 3 COO

3 K,CoO,

(f)

Gold tubes sealed under vacuum are generally used. Other preparation methods can be used:

-

1. Non-stoichiometric K,CrO, is formed from KCrO, by heating at 700°C according to: KCrO,

K,CrO,

+ (1 - x) K

-

(9)

2. Reaction of alkali hydroxides with M oxides:

2~ KOH

+ x M i 0 3 + 2(1 - X)MO,

2 K,(M:M, -,)O,

+ x H2O

(h)

This reaction needs two stages: heating 5 h at 500°C under an oxygen flow, then heating 15 h at 1000°C in sealed gold tubes. The AMO, phases generally decompose when heated at high T. At T > lOOO"C, NaCrO, gives Na,O Cr,03. Heating under oxygen leads to an increase of the M 4 + ions rate. The charge compensation is realized through vacancy formation in both cationic sublattices: AXMO,

+ Y 0,

-

+

(1 + Y) A,,, fyMl,l+ y o 2

0)

In the case of A,MnO, the process can be complete with oxidation at all the Mn3+ ions. Not all the A,MO, phases present layered structures. For the lowest values of x tunnel structures may be observed, as in the case of manganese. Also, if the radius ratio rA+/rM3+ > > 1, a structural model related to cristobalite and well represented by KFeO, (with potassium in the vacant 12-coordination site of Si0,P) is observed. Lattice energy calculations show the upper limit for the a-NaFeO, structure to be stable with regard to the K[FeO,] arrangement corresponding to a cation/anion (a/c) ratio of 5.g4. Using an ionicity-structure diagram (Fig. 3) similar to the one used for lamellar chalcogenides (§16.4.3.2), the layered A,MO, phases can be classified according to the coordination (octahedral-ATP or TP) of the A' ions.

322

16.4. The Formation of Sheet Structures 16 4.6. Lamellar Oxides.

,~i+

0

0

0.70-

Figure 3. Classification of the A,MO, phases in an ionicit-structure diagram (after ref. 7).

The ionic mobility of the alkali ions inserted between the sheets strongly depends on their environment. Conductivity is lo3 to lo4 greater when the alkali ion is in trigonal prismatic environment than when it is octahedrally surrounded. This is to be related to larger windows between trigonal prismatic sites and to less deep potential wells in that less ionic situation. A 2 M 0 3oxides can also present a layered structure (Fig. 4) for small differences in the cation sizes. An NaC1-related type is observed that derives from a-NaFeO, according to the formulation A(A1,3M2,3)02 4,5. With a large difference between the radii other structural types are encountered, but the lamellar structures may be formed under high pressures (except K,Ti03, for which the difference between R A L and RM4+is too large). Another class of layered oxides is the Li,M4'06, Li,MSf0, and L i 6 M i + 0 6 ~ e r i e s ~e.g., . ~ ,Li,SnO,, Li,NbO, and Li,In,O,. Their structure involves layers built up with edge-sharing octahedra. These octahedra are occupied both by the M element and a part of the lithium ions. Excess lithium occupies tetrahedral holes between the layers. The general formula is A,MO-,, which corresponds to Liytra,(Li2,3M1,3)OCf02 in the case of Li,MO,, Li~""(Li,,3Ml,3U1~3)""'0,(where 0 is an empty space) in the case of Li7M0, and Li~t'a(In,i30 1,3)OCt02 in the case of Li,M,O, phases. The charge difference between Li' ions and M3+,M4' or M5' ions induces an ordering in the sheets. (J. ROUXEL)

323

16.4. The Formation of Sheet Structures 16 4.7. Lamellar Silicates and Related Materials

NaCI-related phases 0 K2Zr03 0 Rb2Ti03 0 0 o

0

0.9o.j Ce -Pb Tb ,Zr

8

0

8

Ir

0

Mn

0

/'i

0

0

0

/

y00

0

K

cs

0

0.50 Li

1

05

I

09

Na

Rb

I

13

1

1

17

0 -

RA+(A)

Figure 4. Structural types of various A,MO, phases (after ref. 7). 1. 2. 3. 4. 5. 6. 7.

E. Mooser, W. B. Pearson, Acta Crystallogr., 12, 1015 (1959). C. Fouassier, C. Delmas, P. Hagenmuller, Mater. Sci. Eng., 31, 297 (1977). C. Fouassier, C. Delmas, P. Hagenmuller, Muter. Res. Bull., 10, 443 (1975). R. Hoppe, Bull. SOC.Chim.Fr., 1115 (1965). P. Hagenmuller, M. Devalette, J. Claverie, Bull. SOC.Chim. Fr., 1581 (1966) R. Hoppe, R. M. Braun, Z . Anorg. Allg. Chem., 433, 181 (1977). C . Delmas, C. Fouassier, P. Hagenmuller, Physica, 9B, 81 (1980).

16.4.7. Lamellar Silicates and Related Materials This type of structure is present in naturally occurring silicates and aluminosilicate minerals. Ionic and molecular exchanges in these materials have been known for a long time. They are essential to controlling plants' biological processes. These same processes of exchange, sorption and intercalation are responsible for the ion exchange or catalyst properties of some groups of silicates and notably of zeolites. Zeolites are discussed in $16.3.3. Zeolites present a porous but quasi-rigid framework, the geometry of which is only slightly modified upon intercalation. The layered silicates differ; these are typically two-dimensional materials presenting parallel lamellae bonded to each other by nonbonding interaction forces. Very often interlayer cations are present and provide an electrostatic bonding as found in the case of A,MO, oxides (see 516.4.6). (J. ROUXEL)

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc.

323

16.4. The Formation of Sheet Structures 16 4.7. Lamellar Silicates and Related Materials

NaCI-related phases 0 K2Zr03 0 Rb2Ti03 0 0 o

0

0.9o.j Ce -Pb Tb ,Zr

8

0

8

Ir

0

Mn

0

/'i

0

0

0

/

y00

0

K

cs

0

0.50 Li

1

05

I

09

Na

Rb

I

13

1

1

17

0 -

RA+(A)

Figure 4. Structural types of various A,MO, phases (after ref. 7). 1. 2. 3. 4. 5. 6. 7.

E. Mooser, W. B. Pearson, Acta Crystallogr., 12, 1015 (1959). C. Fouassier, C. Delmas, P. Hagenmuller, Mater. Sci. Eng., 31, 297 (1977). C. Fouassier, C. Delmas, P. Hagenmuller, Muter. Res. Bull., 10, 443 (1975). R. Hoppe, Bull. SOC.Chim.Fr., 1115 (1965). P. Hagenmuller, M. Devalette, J. Claverie, Bull. SOC.Chim. Fr., 1581 (1966) R. Hoppe, R. M. Braun, Z . Anorg. Allg. Chem., 433, 181 (1977). C . Delmas, C. Fouassier, P. Hagenmuller, Physica, 9B, 81 (1980).

16.4.7. Lamellar Silicates and Related Materials This type of structure is present in naturally occurring silicates and aluminosilicate minerals. Ionic and molecular exchanges in these materials have been known for a long time. They are essential to controlling plants' biological processes. These same processes of exchange, sorption and intercalation are responsible for the ion exchange or catalyst properties of some groups of silicates and notably of zeolites. Zeolites are discussed in $16.3.3. Zeolites present a porous but quasi-rigid framework, the geometry of which is only slightly modified upon intercalation. The layered silicates differ; these are typically two-dimensional materials presenting parallel lamellae bonded to each other by nonbonding interaction forces. Very often interlayer cations are present and provide an electrostatic bonding as found in the case of A,MO, oxides (see 516.4.6). (J. ROUXEL)

16.4. The Formation of Sheet Structures 16.4.7. Lamellar Silicates and Related Materials 16.4.7 1. Lamellar Silicates.

324

16.4.7.1. Lamellar Silicates.

Examples of layered silicates are of course micas and chlorites, but also talc, pyrophillite, vermiculites, smectites, kandites, serpentines and polygorskite. The basic structural feature of these materials is the presence of the classical (SiO,) tetrahedra of the silicates arranged in order to form layers with an hexagonal symmetry. Each (SiO,) tetrahedron shares three corners with other tetrahedra to give systems of composition [(Si,O,),]zn"- or [{(Si, Al)205}n]x-.Two kinds of sheets may be formed: a six-ring sheet characteristic of the micas and the clay minerals and a sheet with alternate 4- and 8membered rings which is found in apophyllite KF, Ca,Si,O,, (Fig. 1). These simple models are very rare. What is generally found is a more complicated sheet that forms when tetrahedral sheets are condensed together or with various octahedral sheets. A tetrahedral sheet may be considered the result of the condensation-polymerization of %(OH), molecules. A common corner is introduced by eliminating H,O between two Si(OH),: OH OH HOSi-O~~-SiOH OH OH

-

Si,O(OH),

+ H,O

(a)

and finally the formulation of the sheet will be SiO,,,(OH), i.e., Si,O,(OH),, with the three shared oxygen atoms and one hydroxyl group on the free corners. If the free vertices of the tetrahedra point to the same side of the sheet further condensation may occur, with another sheet either tetrahedral or octahedral. In the latter case the process can continue with another tetrahedral sheet. Two-layer or three-layer sheet silicates (aluminosilicates) are obtained. The simplest two-layer sheet is formed when two layers of tetrahedra, with appropriate OH orientations, condense. The formulation corresponds to: O,,,SiO/H

(a

1

+ HOISiO,,,

-

Si208,,

+ H,O

(b)

Figure 1. Six (a) and four-eight (b) tetrahedral sheets.

(b)

16.4. The Formation of Sheet Structures 16.4.7. Lamellar Silicates and Related Materials 16.4.7.1. Lamellar Silicates.

325

Slab 0

Si(AI)

0

0

@

ca2+

Slab

Figure 2. Two-layer tetrahedral sheets in CaAl,Si,O,

In fact, such a slab would be neutral. Substitution of aluminum by silicon allows the presence of cations between the slabs, which they bind together, e.g., CaAl,Si,O, and BaAl,Si,O,. Figure 2 shows the structure of CaAl,Si,O, with the Ca2+ ions between two-layer tetrahedral sheets. The most important family of layer silicates is characterized by composite two- or three-layer sheets resulting from the condensation between sheets of tetrahedra and octahedra. The octahedra sheet may be provided by the layers of Mg(OH), or Al(OH), octahedra in the brucite Mg(OH), or bayerite (hydrargillite) Al(OH), structures. Figure 3 shows the formation of a bilayer sheet (a) and a three-layer sheet (b). For aluminum the two-layer sheet has the composition Al,Si,O 5(OH)4. This sheet is present in kandites (kaolinite, dickite, nacrite and halloysite). The silicaous side presents a surface of oxygen atoms; whereas the aluminous side provides a surface of hydroxyl groups. The threelayer sheet is expressed as Al,Si,O,,(OH), and is found with various substitutions in pyrophyllite, micas, vermicullites and smectites. Talcs correspond to the Mg(OH), group with a three-layer sheet Mg,Si,O,o(OH),. Al(OH), and Mg(OH), differ in the number of octahedral holes occupied by metal atoms. As a consequence the composite sheets they make reproduce this situation. Other di- or trivalent ions of appropriate size can replace Mg2 or A13 ions. Additional complications may arise from different condensation modes or from interleaving different sheets in the same structure. The first case is illustrated by the polygorskites group of fibrous clay minerals, which includes attapulgite and sepiolite. Figure 4 shows the structural arrangement in attapulgite, with slit-shaped channels running parallel to the fiber axis. The ideal formulation is Mg5Si80,,(OH),(H,0), . 4 H,O, with a distinction between zeolitic and crystallization (or bound) water. The second situation is observed in chlorites. The minerals are made +

+

326

16.4. The Formation of Sheet Structures 16.4 7. Lamellar Silicates and Related Materials 16.4.7.1. Lamellar Silicates.

0 oxygen @ OH Si 0 Al


or Mg

intercalated cation

a water

Figure 3. Formation of a bilayer (a) or a three-layer (b) tetrahedral-octahedral sheet: (a) is exemplified by kaolinite A1,Si2O,(OH), or chrysotile Mg,Si,O,(OH),, (b) is exemplified by talc Mg,Si,OIo(OH), and substituted forms of A1,Si40,0(OH), in micas.

up of three-layer sheets with compositions ranging from [Mg,(AlSi,O,,)(OH),]to [Mg2Al(Al,Si,0,,](OH)2] - , alternating with an octahed-ral positive layer resulting from the replacement of one-third of the Mg in a brucite Mg(OH), layer by Al, i.e., Mg,Al(OH); (Fig. 5). Table 1 shows various layered silicates according to structural type. (J. ROUXEL)

16.4. The Formation of Sheet Structures 16.4.7. Lamellar Silicates and Related Materials 16 4.7.1. Lamellar Silicates.

0

Si

0

Al,Mg

0

0

@

OH

0

7 1 327

13.4

H20

A

Figure 4. Complex arrangement of the three layers (tetrahedral-octahedral-tetrahedral) sheets in attapulgite.

'

@

MgAl in positive sheets

MgAl lin negative sheets

0 SiAl

Figure 5. Structure of chlorite: Mg,Al(OH); octahedral positive layers alternate with (MgAl)(SiAI) octahedral-tetrahedral sheets. In phlogopite K t ions replace the positive layers.

328

16.4. The Formation of Sheet Structures 16.4 7. Lamellar Silicates and Related Materials 16.4.7.2. Intercalation Products.

TABLE1. SOMELAYERED SILICATES

Mineral

I. Two-layer-sheets, tetrahedra-octahedra: Kandites Kaolinite, dickite, nacrite (differing in the sequence of layer^)^ Amesite (Mg-M'" and Si-A1 substitutions) Halloysite (hydrated) Serpentines Chrysotile, antigorite 11. Three-layer-sheets, actahedra-tetrahedra-octahedra: Pyrophyllite Talc Some Micas Muscovite Paragonite Phlogopite Biotite Margarite Vermiculites Some Smectites Montmorillonite Saponite Hectorite * Fluorheclorite

Ideal formula Al,(OH),Si,O,

Mg,AI(SiAIO,)(OH), Al,(Si,O,)(OH),~2 H,O

A12(Si4010)(0H)2 Mg3(Si4010)(0H)2

1

no interlayercation

16.4.7.2. Intercalation Products. The presence of cations between the slabs of numerous layered silicates allows various exchange reactions, including those involving organic cations. These types of layered silicates can be considered as intercalation compounds by themselves. However, intercalation chemistry of lamellar silicates reveals a particular richness and originality regarding molecular intercalates. Despite the numerous studies recently centered on dichalcogenides and graphite, the largest number and variety of molecular intercalates have been prepared as silicates, and particularly as layered silicate host structures. Several situations can be distinguished depending on whether the host structure contains cations between the slabs or not, and on the species to be intercalated: neutral molecules, salts, polarizable molecules, inorganic complexes, water, etc.' . Various organic molecules, including hydrazine, formamide, acetamide and urea, have been intercalated between the neutral sheets of kandite minerals2. Attapulgite can also accommodate various molecules in its channels. However, the process seems to be limited to sites situated at the channel entrances and is strongly dependant on the dehydration processes applied to the material. More important is the fact that attapulgite can present a selectivity concerning the fixation of n-paraffins: when outgassed at 70°C attapulgite shows the sequence n- > 1- > t-pentane,. Structural studies in the case of formamide intercalates of kaolinite show that the > CO is directed toward the hydroxyl side of the two-layer host sheet (external octahedra layer surface). Salts such as acetates

16.4. The Formation of Sheet Structures 16.4.7. Lamellar Silicates and Related Materials 16.4.7.2. Intercalation Products. ~

~

329

~~~~

or propionates of K, Rb, Cs and NH, may also intercalate in kandites. Intercalation depends on the pH of the solution used and is facilitated when another molecule able to intercalate independently is used simultaneously. Hydrazine can play this role. The difficulty of nucleating intercalation is reflected in the kinetic curves, which are sigmoid with induction periods. When cations are present between the sheets of the host, many polar molecules can be intercalated, as in vermiculites and smectites. Micas have their three-layer sheets very close together and do not swell easily unless a preliminary ionic exchange with an organic cation takes place (for example alkylammonium ions replacing potassium). Mono- and bilayer complexes may be observed in which organic molecules such as alcohols, ketones and nitriles form layers one or two molecules thick. For example, smectites with double layers of glycerol (d,,, N 1770pm) have been prepared, but if large monovalent cations are present only monolayer derivatives (d,,, ?: 1400 pm) are formed. This is a common feature of dichalcogenides (see 516.4.3.5).Intercalation is not limited to light molecules: for example monolayer and bilayer complexes have been prepared between Na-, Ca- and Mg-montmorillonites and long-chain esters. Pyridine has been intercalated and structural studies performed in the case of Na-rn~ntmorillonite~ show that once more (see 516.4.3.4) the plane of the rings is perpendicular to the silicate sheets with the cation-cation long axis parallel to these sheets. Another analogy with dichalcogenides is observed in the case of alkylammonium salts. With up to eight carbon atoms in the alkyl chains, 1-n-alkylpyridinium bromides lead to cationic-exchange processes with Na- or Ca-montmorillonites. With longer alkyl chains and also with alkylammonium salts, intercalation occurs in addition to the exchange process. When the number of C-atoms increases, the organic chain changes

Figure 1. Cetyltrimethylammonium bromide or chloride in montmorillonite (after ref. 5; cited in ref.1).

16.4. The Formation of Sheet Structures 16.4.7 Lamellar Silicates and Related Materials 16.4.7.2.Intercalation Products.

330

from an orientation parallel to the silicate sheets to a position where they are steeply orientated toward these sheets5. Figure 1 represents the case of cetyltrimethylammoni.um bromide or chloride in montmorillonite (see Fig. 3 in 516.4.3.4). The behavior toward water is also very similar to what has been observed in dichalcogenides and described in 516.4.3.5. (Note that the observations on layered silicates were made before those on dichalcogenides.) From the many studies performed on smectites and vermiculites, it is observed that the d(,,,, spacings are step functions of H,O vapor pressure with H,O monolayer and bilayer situations. Figure 2 shows the basal spacings observed in the case of natural Na-montmorillonite6. Penetration of water between the host slabs is directly related to the decrease in free energy when the guest molecules solvate the cations. There is a competition between this term and the increase in free energy associated with the separation of the sheets'. If the cationic population and the charges borne by these cations in the interslab region are high, the slabs are tightly binded together. A high energy is needed for the separation, and hydration may be impossible, as in the cation-rich silicates such as margarite. On the other hand, the absence of interlayer cations causes severe problems of nucleation that appear particularly in the case of kandites. Apart from this limitation, the ability to intercalate is directly related to the interlayer surface per unit charge (Table 1 and ref. 7). A specific intercalation process in layered silicates, which has not yet been reproduced in dichalcogenides, concerns transition-metal complexes. Most of the exam-

:[ oa

Q

u)

a Q

,O

17t

15 14 -

-

16

2 layer hydrate - 0 i ' ,#O

< ,

12 -

o,e

-

/'

I

13

11

,*'

'.

o'+

,'

0 '

o?. #'

-

1 layer hydrate

,( I I'

Figure 2. Basal spacings observed for monolayer and bilayer hydrates in the case of Na-montmorillonite.

16 4. The Formation of Sheet Structures 16.4.8. The Chemical Reactivity of Low-Dimensional Solids

331

TABLE1. CHARGE DENSITY AND SWELLING BY HYDRATION IN VARIOUSSILICATES (AFTERREF. 7) ~

~~~

Mineral

Interlayer surface per unit charge (pm)2

Extent of swelling in A for the Na’ form

Margarite Muscovite Biotite Vermiculite Saponite Montmorillonite Talc

12 x 104 24 x 104 24 x 104 36 x 104 42 x 104 75 x 104 co

0 190 190 510 490 u3

0

ples refer to Cu2+ complexes but Co3+,N i z f , Cd2+,Z n Z +and Ag’ complexes have also been found. The starting point is an exchanged layered silicate, particularly montmorillonite, which presents present the cation conveniently and allows the complex to form in situ. For example, Cuz montmorillonite forms complexes with ethylenediamine, pyridine, rubeanic acid, benzene, toluene, xylene and many other reagents’. Their intercalation from solution provides a way of collecting and concentrating small amounts of metals; e.g., silver can form complexes with urea. These complexes are intercalated with a high selectivity and with a displacement of interlayer cations such as N a + , Ca”, A13+ The field of intercalation chemistry in layered silicates is huge. It includes many types of intercalates, and a large variety of host structures that can be modified before intercalation by various exchange reactions. Organoclays, for example, represent a family of compounds resulting from the introduction of organic cations by ion exchange and which also allow intercalation. +

’.

(J. ROUXEL)

1. R. M. Barrers, Zeolites and Clay Minerals as Sorbents and Molecular Sieves,Academic Press, r\ W York, 1978. An excellent review. Recommended reading. 2. A. Weiss, W. Thielpape, G. Goring, W. Ritter, H. Schafer, International Clay Conference,Vol. 4, Pergamon, London; 1963, p. 287. 3. R. M. Barrer, N. MacKenzie, D. M. MacLeod, J. Phys. Chem., 58, 568 (1954). 4. J. M. Adams, J. M. Thomas, M. J. Walters, J. Chem. Soc., Dalton Trans., 1459 (1975). 5 . P. Franzen, Clay Minerals Bull., 2, 223 (1954). 6. F. H. Gillery, Am. Mineral., 44, 806 (1959). 7. A. Weiss, Chem. Ber., 91, 487 (1958). 8. J. Pleysier, A. Cremer, J. Chem. SOC.,Faraday Trans. I, 71, 256 (1975).

16.4.8. The Chemical Reactivity of Low-Dimensional Solids Low-dimensional solids have a rich topotactic chemistry at rather low temperature (often at RT). This “soft chemistry” (chimie douce) can be an intercalation-disintercalation chemistry associated to redox processes as described in the above sections, an acidobasic chemistry involving a condensation-reconstruction of the structure or a graft chemistry using the van der Waals (916.4.3.1.)gaps separating slabs or fibers as internal surfaces of such solids. All these reactions are largely governed by the electronic structure of the low-dimensional material. (J ROUXEL)

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc. 16 4. The Formation of Sheet Structures 16.4.8. The Chemical Reactivity of Low-Dimensional Solids

331

TABLE1. CHARGE DENSITY AND SWELLING BY HYDRATION IN VARIOUSSILICATES (AFTERREF. 7) ~

~~~

Mineral

Interlayer surface per unit charge (pm)2

Extent of swelling in A for the Na’ form

Margarite Muscovite Biotite Vermiculite Saponite Montmorillonite Talc

12 x 104 24 x 104 24 x 104 36 x 104 42 x 104 75 x 104 co

0 190 190 510 490 u3

0

ples refer to Cu2+ complexes but Co3+,N i z f , Cd2+,Z n Z +and Ag’ complexes have also been found. The starting point is an exchanged layered silicate, particularly montmorillonite, which presents present the cation conveniently and allows the complex to form in situ. For example, Cuz montmorillonite forms complexes with ethylenediamine, pyridine, rubeanic acid, benzene, toluene, xylene and many other reagents’. Their intercalation from solution provides a way of collecting and concentrating small amounts of metals; e.g., silver can form complexes with urea. These complexes are intercalated with a high selectivity and with a displacement of interlayer cations such as N a + , Ca”, A13+ The field of intercalation chemistry in layered silicates is huge. It includes many types of intercalates, and a large variety of host structures that can be modified before intercalation by various exchange reactions. Organoclays, for example, represent a family of compounds resulting from the introduction of organic cations by ion exchange and which also allow intercalation. +

’.

(J. ROUXEL)

1. R. M. Barrers, Zeolites and Clay Minerals as Sorbents and Molecular Sieves,Academic Press, r\ W York, 1978. An excellent review. Recommended reading. 2. A. Weiss, W. Thielpape, G. Goring, W. Ritter, H. Schafer, International Clay Conference,Vol. 4, Pergamon, London; 1963, p. 287. 3. R. M. Barrer, N. MacKenzie, D. M. MacLeod, J. Phys. Chem., 58, 568 (1954). 4. J. M. Adams, J. M. Thomas, M. J. Walters, J. Chem. Soc., Dalton Trans., 1459 (1975). 5 . P. Franzen, Clay Minerals Bull., 2, 223 (1954). 6. F. H. Gillery, Am. Mineral., 44, 806 (1959). 7. A. Weiss, Chem. Ber., 91, 487 (1958). 8. J. Pleysier, A. Cremer, J. Chem. SOC.,Faraday Trans. I, 71, 256 (1975).

16.4.8. The Chemical Reactivity of Low-Dimensional Solids Low-dimensional solids have a rich topotactic chemistry at rather low temperature (often at RT). This “soft chemistry” (chimie douce) can be an intercalation-disintercalation chemistry associated to redox processes as described in the above sections, an acidobasic chemistry involving a condensation-reconstruction of the structure or a graft chemistry using the van der Waals (916.4.3.1.)gaps separating slabs or fibers as internal surfaces of such solids. All these reactions are largely governed by the electronic structure of the low-dimensional material. (J ROUXEL)

332

16.4. The Formation of Sheet Structures 16.4.8.The Chemical Reactivity of Low-Dimensional Solids 16 4 8.1 Transfer and Formation of Cationic Intercalation Compounds.

16.4.8.1. The Electronic Transfer and the Formation of Cationic Intercalation Compounds.

The intercalation of cations in a given host structure is the result of a reversible ion-electron transfer reaction:

x A'

+ x e- + host=

A,

+ hostx-

(a)

where A is an alkali metal. Various structural arrangements are associated with the A t intercalated cations (616.4.3.2.).Electronic transfer, often not considered significant, plays an essential role in the formation of intercalation compounds and the occurrence of phase transitions in a given system. Even the so-called Ac parameter expansion, perpendicular to the slabs of a layered host structure, has an electronic component. A classification of intercalation reactions into three groups is based on electronic transfer. These three groups correspond to three different steps in the delocalization of the transferred electrons. The level of aceptance can be (i) a discrete atomic one; (ii) molecular, as a discrete polyatomic entity existing in the structure; and (iii) part of a conduction band. Case (i) corresponds to a reduction of a given ion in the host network, e.g., Fe3+ to Fez+ in Li,FeOCl intercalates'. Such systems, which are associated with the most ionic host structures (oxides, fluorides, chlorides), are largely governed by redox competition between A/At and M"+/M("-l)Ccouples. Lithium can be intercalated in FeOCl but not in CrOCl. If the number of available sites is sufficient, the chemical formulation of the interclate can be deduced from among the common oxidation states. However if a fast electronic exchange between M"+ and M'"")' is possible (particularly if the geometry allows it), it may stabilize a particular situation. This is well illustrated by the intercalation compounds that can be made from aRuCl,, which has a vacant layered dichalcogenide structural type with a slab built up from partially occupied edge-sharing octahedra according to the formulation [ClO 1,3R~2,,C1].There is one empty octahedral site per slab in the van der Waals gap (Fig. 1, 616.4.3.1.),i.e., 1.5 per site per ruthenium atom, so the geometry allows a chemical formulation A,,,RuCl,. Of course the limited reduction of Ru3+ to Ru" stops the process when the A,RuCl, composition is reached2. A yellow phase, CuRuCI,, e.g., is obtained, but it is unstable and easily oxidized in air to black Cu,,,,RuCl, and Cu,O. Obviously the electronic exchange between Ru3+ and RuZ+stabilizes the mixed-valence compound that forms which is favored by a geometry of edge-sharing octahedra in RuCl,. The second situation is well illustrated by phases that are built up from clusters of Mo, octahedra enclosed in pseudo-cubes of eight chalcogen atoms3. For small ions the structure formally provides at least 6, or even 12, possible lattice sites, but the molecular levels of the clusters have 20 electrons within a possible maximum population of 24 electrons. This leads to an electronic limit with a maximum of four intercalated monovalent cations4. Changing the chemistry of the cluster by substituting, e.g., Ru for MOchanges the intercalation chemistry according to the new electronic population of the cluster. Full electronic delocalization is illustrated by transition-metal dichalcogenides. In TiS, or ZrS,, with the metal in a 4 + oxidation state, octahedral symmetry and consequently a broad and empty t,, band, there is no electronic limitation to the intercalation process. Chemistry is governed only by the geometry, i.e., the number of

16.4. The Formation of Sheet Structures 16.4.8. The Chemical Reactivity of Low-Dimensional Solids 16.4.8.2. Structural Charges Induced by Electronic Transfer.

333

empty sites in the van der Waals gap. With NbS,, trigonal prismatic slabs and d1 configuration lead to a half-filled narrow dz2 band, which is more difficult to intercalate. In MoS, the dzZ band is completely filled and the transferred electrons must be accommodated on a higher level. The MoS, intercalates are difficult to obtain and very unstable. Using band structure nomenclature for the host solids allows a much more detailed discussion of the intercalation reaction. The electronic transfer is influenced not only by the number of empty levels but also by the structure of the band itself, i.e, the number of electrons in orbitals, or density of states. A pseudo-plateau in the potential variation of a discharge curve may be related to the filling of a zone of high density of states. This pseudo-plateau does not mean a two-phase region but occurs because the difference in energy of electrons in the alkali metal, i.e., before intercalation, and in the host, i.e, after intercalation, remains quite constant. (J. ROUXEL)

1. 2. 3. 4.

J. Rouxel, P. Palvadeau, Rev. Chim. Mineral., 19, 317 (1982) R. Steffen, R. Schollhorn, Solid State Ionics, 22, 31 (1986). R. Chevrel, M. Sergent, J. Solid State Chem., 3, 8807 (1971). R. Schollhorn, Angew. Chem., Int. Ed. Engl., 19, 983 (1980).

16.4.8.2. Structural Charges Induced by Electronic Transfer. The electronic transfer can induce phase transitions concerning guest positions as well as the host skeleton itself. As noted in 516.4.3.2, there are two first-stage Na-TiS, phases, one with Na in trigonal prismatic (TP) sites and the other with Na in octahedral sites between TiS,Slabs. Qualitative reasons have been proposed to explain the transition and have led to an ionicity-structure diagram. Studies of phase transition in Na-TiS, using x-rays and Na NMR’ and band structure calculation’ show that going from phase Ib to Ia cell volume remains nearly constant, with a 3.9% decrease of the c parameter (perpendicular to the slabs) and 2.2% increase of the a parameter (parallel to the slabs). Such a transition costs elastic energy because the a plane of the slabs has more stiffness than does the c direction. To compensate this cost in elastic energy the transformation must lead to a lowering of the electronic energy. It appears that the Na Kisoundergoes a 48-ppm increase when passing from the Ib to the Ia phase. In addition Kax,a,is negligible in the TP phase but not in the octahedral one, which implies an increase of the number of electrons in orbitals during the transition and an anisotropic distribution of charge around Na in the octahedral phase. Band structure calculations performed as a function of a parameter 8, which characterizes the TiS, slab deformation during intercalation (Fig. la), confirm these implications. For the TP phase the energy is the lowest for special point M(1/2,0a* as in TiS,, but for the octahedral phase this is observed for point (O,O)a*. The eM- and e,- levels vary as a function of 0 (Fig. lb) and e, becomes more stable for 6 > 56’4’. This critical value corresponds to the structural discontinuity associated with the transition. Band structure calculations also confirm the increase in the number of electrons in orbitals suggested by K,,, measurements. Furthermore, analysis of the contribution of d-orbitals to the eM- and e,- levels shows that eM- level has contributions from all d levels, whereas e, has no dzZ contribution, which explains the variation of Kaxialat the transition. The most spectacular effect concerning phase transition induced in the host skeleton is seen in MoS, Layered MS, chalcogenides are not formed by the elements on the

’.

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc. 16.4. The Formation of Sheet Structures 16.4.8. The Chemical Reactivity of Low-Dimensional Solids 16.4.8.2. Structural Charges Induced by Electronic Transfer.

333

empty sites in the van der Waals gap. With NbS,, trigonal prismatic slabs and d1 configuration lead to a half-filled narrow dz2 band, which is more difficult to intercalate. In MoS, the dzZ band is completely filled and the transferred electrons must be accommodated on a higher level. The MoS, intercalates are difficult to obtain and very unstable. Using band structure nomenclature for the host solids allows a much more detailed discussion of the intercalation reaction. The electronic transfer is influenced not only by the number of empty levels but also by the structure of the band itself, i.e, the number of electrons in orbitals, or density of states. A pseudo-plateau in the potential variation of a discharge curve may be related to the filling of a zone of high density of states. This pseudo-plateau does not mean a two-phase region but occurs because the difference in energy of electrons in the alkali metal, i.e., before intercalation, and in the host, i.e, after intercalation, remains quite constant. (J. ROUXEL)

1. 2. 3. 4.

J. Rouxel, P. Palvadeau, Rev. Chim. Mineral., 19, 317 (1982) R. Steffen, R. Schollhorn, Solid State Ionics, 22, 31 (1986). R. Chevrel, M. Sergent, J. Solid State Chem., 3, 8807 (1971). R. Schollhorn, Angew. Chem., Int. Ed. Engl., 19, 983 (1980).

16.4.8.2. Structural Charges Induced by Electronic Transfer. The electronic transfer can induce phase transitions concerning guest positions as well as the host skeleton itself. As noted in 516.4.3.2, there are two first-stage Na-TiS, phases, one with Na in trigonal prismatic (TP) sites and the other with Na in octahedral sites between TiS,Slabs. Qualitative reasons have been proposed to explain the transition and have led to an ionicity-structure diagram. Studies of phase transition in Na-TiS, using x-rays and Na NMR’ and band structure calculation’ show that going from phase Ib to Ia cell volume remains nearly constant, with a 3.9% decrease of the c parameter (perpendicular to the slabs) and 2.2% increase of the a parameter (parallel to the slabs). Such a transition costs elastic energy because the a plane of the slabs has more stiffness than does the c direction. To compensate this cost in elastic energy the transformation must lead to a lowering of the electronic energy. It appears that the Na Kisoundergoes a 48-ppm increase when passing from the Ib to the Ia phase. In addition Kax,a,is negligible in the TP phase but not in the octahedral one, which implies an increase of the number of electrons in orbitals during the transition and an anisotropic distribution of charge around Na in the octahedral phase. Band structure calculations performed as a function of a parameter 8, which characterizes the TiS, slab deformation during intercalation (Fig. la), confirm these implications. For the TP phase the energy is the lowest for special point M(1/2,0a* as in TiS,, but for the octahedral phase this is observed for point (O,O)a*. The eM- and e,- levels vary as a function of 0 (Fig. lb) and e, becomes more stable for 6 > 56’4’. This critical value corresponds to the structural discontinuity associated with the transition. Band structure calculations also confirm the increase in the number of electrons in orbitals suggested by K,,, measurements. Furthermore, analysis of the contribution of d-orbitals to the eM- and e,- levels shows that eM- level has contributions from all d levels, whereas e, has no dzZ contribution, which explains the variation of Kaxialat the transition. The most spectacular effect concerning phase transition induced in the host skeleton is seen in MoS, Layered MS, chalcogenides are not formed by the elements on the

’.

334

16.4. The Formation of Sheet Structures 16.4.8. The Chemical Reactivity of Low-Dimensional Solids 16.4 8.2. Structural Charges Induced by Electronic Transfer.

Figure 1. (a) The angle B is a measure of a TiS, slab deformation during intercalation. (b) eMand e, levels as a function of 6'.

right side of the periodic table, starting with manganese, as a consequence of a redox competition between the top of the sp anionic band and d levels that progressively drop and enter the sp band [reduction of the cation and formation of anionic pairs occur, driving the structure from a layered type to a pyrite M 2 + , [S,]'arrangement]. With Mn a d3 electronic configuration would have led to an octahedral coordination of the metal in the layered MS, phase, which cannot be made directly. The stabilization through a TP distortion associated with d1 and d 2 configurations of Nb4+ and Mg4+, respectively, no longer exists. Indeed such a situation has been achieved indirectly by adding one electron to MoS, by intercalating Li. A phase transition is observed, leading to octahedral MoS,. The MoS, transition can be expressed through simple translations of sulfide layers relative to each other. What is frequently observed is that the transition metal changes its site in the host structure. This has been well characterized in a few systems involving iron or an other cation able to serve as a local probe for x-ray recoilless resonance fluorescence or NMR studies. (J. ROUXEL)

1. P. Molinie, L. Trichet, J. Rouxel, C. Berthier, Y. Chabre, P. Segransan, J. Phys. Chem. Solids, 451, 105 (1984). 2. M. H. Whangbo, L. Trichet, J. Rouxel, Inorg. Chem., 24, 1824 (1985). 3. M. A. Py, R. R. Hawing, Can. J. Phys., 61, 76 (1983).

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc.

16.4. The Formation of Sheet Structures 16.4.8. The Chemical Reactivity of Low-Dimensional Solids 16.4.8.3. Strong Ion-Electron Coupling.

335

16.4.8.3. Strong Ion-Electron Coupling.

Two questions arise immediately. (i) Is it possible to continue to use the band structure of the host, i.e., a rigid band model for strong electron-ion couples? (ii) Then, if it is possible, does the intercalation process result simply in a progressive rise in the constant energy surface level? Self-consistent band calculations have been made' on LiTiS, *. A comparison of the numbers of electrons in orbitals for TiS, and LiTiS, shows that the shapes of the bands are very similar. However, if the d bands are set aligned to serve as a reference, the s-p band shifts to lower energies (- 1 eV) in the intercalated material because of the rather strong polarization of the electronic density of S2- anions by Li+ ions. The shape of the d bands remains unchanged during intercalation. The fundamental question is the evolution of the constant energy surface level, which is explicated' in terms of the screened-impurity model3. This model proposes that the d bands move down in energy as electrons are added in order to keep constant energy surface constant in position. This assertion is indeed supported by experiment particularly by intercalation in modified phases that are built up from clusters of Mo, octahedra enclosed in pseudo-cubes of eight chalcogen atoms. An example is the semiconductor Ru,Mo,Se,. Each Ru adds two more electrons, producing filled bands (24 electrons instead of 20). Intercalating into samples of Ru,Mo,-,Se, initially causes the progressive filling of the same band. A drop in potential when four electrons have been added, by either Ru or Li, indicates that the band is full. However, the first plateau is observed at the same potential in all Li,Ru,Mo,~,Se, samples (x Il), and this is possible only if the band drops in energy when electrons are added2z4. The screening-impurity model implies a strong-electron coupling. The strong polarization of the sp electronic density of the chalcogen by Li+ ions induces a negative charge defect behind the transition-metal site and reinforces the local positive potential around it. This favors a localization of the transferred electron to this site. The more polarizable the anion, e.g., the Li-ZrSe, system, the more important is the effect. With the stoichiometry ZrSe,,,,, zirconium diselenide is a semiconductor with an empty t,, band (516.4.3.1). The Li,ZrSe,,,, intercalation compounds show a phase transition when x reaches 0.40. A continuous filling of the octahedral sites of the van der Waals gap (see Fig. 1, 516.4.3.1.) is observed over the whole composition domain (0 < x 5 1) according to a classical Cd1,-NiAs transition. Electrical, electrochemical, magnetic, NMR and EPR measurements5 affirm that it is a purely electronic transition. Below x = 0.40 the electrons are localized on zirconium centers, reducing Zr4+ ions Z r 3 + ;above x = 0.40 a metallic behavior is observed. In the metallic region at room temperature, the 77Se NMR lineshape versus temperature shows a narrow single peak which resolves into two peaks below 250 K. This is due to the influence of Li+ ions in the neighborhood of selenium. Computer simulation shows a fair agreement with a hypothesis of alternative rows of empty and occupied sites of lithium. The lineshape is then a picture of the immobilization of Li+ ions at low T in a classical A,,,,MS, ordering. At a higher temperature the screening effect increases and the Se nucleus experiences an average coupling. In the semiconducting domain (x = 0.29) the line has three peaks. The simulation is in agreement with a statistical distribution of Li+ ions, leading to Se atoms having respectively zero, one or two Li+ ions as nearest neighbors. A computer simulation, including the NMR results, illustrates the distribution of the lithium ions on their

336

16.4. The Formation of Sheet Structures 16.4.8. The Chemical Reactivity of Low-Dimensional Solids 16.4.8 3. Strong Ion-Electron Coupling. I

1

I

I

6.6 6.5

/

's 6.46.3 6.2 w

/

h

B,'

y

0

-

+

-

/

w

A-

;

I

1

1

I

1

I

I

I

,

I

.

1

B

3.74

-

3.72

-

0

3.70

-,

'

-

-

1

-

--;\.,*

78:3q+-3 3.76 -

9 (

1

/

I'

1

1

1

/

t

I

-

&'-A

.

,4'

-

3.80

1

B ',

.

I

I

.

I

I

-

x in Li,ZrSey Figure 1. a and c parameters versus x in Li,ZrSe,. (A) for y

=

1.94, (B) for y = 1.85.

triangular lattice, i.e., the absence of three Li+ nearest neighbors of a selenium atom. Isolated groups of lithiums are formed as x increases up to the value of 0.42, which represents a critical percolation rate for the lithium sublattice. The transition seems related to this percolation value, illustrating a strong ion-electron coupling in that system. With a lower stoichiometry ZrSe, (y I1.90) is metallic. A continuous behaviour is observed, as illustrated in Fig. 1, which gives the c parameter change as a function of intercalation for ZrSe,,,, (curve A) and ZrSe,,,, (curve B)6-8. A semiconducting structure therefore may accommodate a certain quantity of lithium without important

16.4. The Formation of Sheet Structures 16.4.8.The Chemical Reactivity of Low-Dimensional Solids 16.4.8.4.Microdomains in MPS, Phases.

337

parameter expansion. Electronic delocalization triggers expansion of the c parameter. In metallic samples (ail the single crystals prepared so far are metallic) the expansion of the c parameters occurs at the initiation of intercalation. (J. ROUXEL)

C. Umngar, D. E. Ellis, D. Wang, H. Krakaner, M. Postenakk, Phys. Rel;. Ser. B, 26,4935 (1982). W. R. McKinnon, Chemical Physics oflntercalation, NATO AS1 Series B 172, 181-194 (1987). J. Friedel, Adv. Phys.; 3, 446 (1954). W. R. McKinnon, L. S. Selwyn, Phys. Rev., Ser. B, 35, 1275 (1987). C. Berthier, Y. Chabre, P. Segransan, P Chevalier, L. Trichet, A. Le Mehaute, Solid State Ionics, 5, 379 (1981). 6. For a critical discussion see P. Denlard, L. Trichet, Y. Chabre, NATO ASI, Vol. 172,387 (1987). 7. J. R. Dahn, W. R. McKinnon, C. Levy-Clement, Solid State Commun.,54, 245 (1985). 8. Y. Onuki, T. Hirai, K. Shibutani, T. Komatsubara, J. Inclusion Phenomena, 2, 279 (1984).

1. 2. 3. 4. 5.

16.4.8.4. Microdomains in MPS, Phases.

An NMR study of the resonance line of 31Pin Li-intercalated NiPS, (516.4.4) does not show any modification from its characteristics in pure NiPS,. Its shape, which is a typical powder pattern with dipolar splitting, position and relaxation time (tl = 16 ms) remain constant. However, a new resonance line appears and increases at the expense of the first one; this line is no longer a doublet, its shift is the same as that of 31Pin CdPS, and its relaxation time is increased by about two orders of magnitude. In addition the new line does not disappear at 155 K as the first one does when magnetic ordering takes place. All these observations are consistent with the coexistence in the intercalates of paramagnetic domains (cf. NiZf in NiPS,) and diamagnetic domains parallel to a reduction of Ni2' to Nio (cf. Cd2+ in CdPS,). Magnetic measurements confirm this model. EXAFS studies show Nio to be in tetrahedral holes in the corresponding microdomains. Figure 1

I/

NI Tetra

0

0.6

1 x In L I x NIPS3

1.6

Figure 1. Nickel-0 contents in tetrahedral holes as a function of x.

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc.

16.4. The Formation of Sheet Structures 16.4.8.The Chemical Reactivity of Low-Dimensional Solids 16.4.8.4.Microdomains in MPS, Phases.

337

parameter expansion. Electronic delocalization triggers expansion of the c parameter. In metallic samples (ail the single crystals prepared so far are metallic) the expansion of the c parameters occurs at the initiation of intercalation. (J. ROUXEL)

C. Umngar, D. E. Ellis, D. Wang, H. Krakaner, M. Postenakk, Phys. Rel;. Ser. B, 26,4935 (1982). W. R. McKinnon, Chemical Physics oflntercalation, NATO AS1 Series B 172, 181-194 (1987). J. Friedel, Adv. Phys.; 3, 446 (1954). W. R. McKinnon, L. S. Selwyn, Phys. Rev., Ser. B, 35, 1275 (1987). C. Berthier, Y. Chabre, P. Segransan, P Chevalier, L. Trichet, A. Le Mehaute, Solid State Ionics, 5, 379 (1981). 6. For a critical discussion see P. Denlard, L. Trichet, Y. Chabre, NATO ASI, Vol. 172,387 (1987). 7. J. R. Dahn, W. R. McKinnon, C. Levy-Clement, Solid State Commun.,54, 245 (1985). 8. Y. Onuki, T. Hirai, K. Shibutani, T. Komatsubara, J. Inclusion Phenomena, 2, 279 (1984).

1. 2. 3. 4. 5.

16.4.8.4. Microdomains in MPS, Phases.

An NMR study of the resonance line of 31Pin Li-intercalated NiPS, (516.4.4) does not show any modification from its characteristics in pure NiPS,. Its shape, which is a typical powder pattern with dipolar splitting, position and relaxation time (tl = 16 ms) remain constant. However, a new resonance line appears and increases at the expense of the first one; this line is no longer a doublet, its shift is the same as that of 31Pin CdPS, and its relaxation time is increased by about two orders of magnitude. In addition the new line does not disappear at 155 K as the first one does when magnetic ordering takes place. All these observations are consistent with the coexistence in the intercalates of paramagnetic domains (cf. NiZf in NiPS,) and diamagnetic domains parallel to a reduction of Ni2' to Nio (cf. Cd2+ in CdPS,). Magnetic measurements confirm this model. EXAFS studies show Nio to be in tetrahedral holes in the corresponding microdomains. Figure 1

I/

NI Tetra

0

0.6

1 x In L I x NIPS3

1.6

Figure 1. Nickel-0 contents in tetrahedral holes as a function of x.

338

16.4. The Formation of Sheet Structures 16.4.8. The Chemical Reactivity of Low-Dimensional Solids 16.4.8.5. Electronic Transfer and Ac Parameter Expansion.

shows that the number of Ni atoms in tetrahedral holes increases as x/2, the amount of Li This is coherent with the Ni2" + Nio reduction. The tetrahedral environment of Nio is usual in coordination chemistry. An x-ray recoilless resonance fluorescence spectroscopy study of Li,FePS, intercalates confirms the above results'. Microdomains of reduced iron in tetrahedral holes are found involving 20-200 Fe atoms. The unreduced octahedral Fe microdomains behave as if the first ones were not present. The magnetic ordering temperature and the hyperfine field do not differ from the pure FePS,. A third type of Fe is found in low concentration, considered to be located at the boundaries between reduced and unreduced domains. Local strain effects account for a distortion of the geometry of the sites that is reflected in the spectroscopic parameters. This rules out the existence of two macroscopic phases, which is also excluded by x-ray diffraction studies. (J. ROUXEL)

1. P. Colombet, G. Ouvrard, 0. Antson, R. Brec, J . Mugn. Mugn. Muter. 71, 100 (1987).

16.4.8.5. Electronic Transfer and Ac Parameter Expansion.

The electronic transfer from guest to host contributes to Ac by two opposite effects: an increasing repulsion between slabs that are now negatively charged and an attraction between these slabs and the A" positive layers. If the A' ion is small, it does not separate the slabs sufficiently to minimize the repulsion and the slab to slab repulsion may be the most important factor. In the Li-TiS, system, e.g., the lattice expansion on intercalation is associated with a charge donation to the host rather than with propping of the layers by Li' ions'. Such a situation is also favorable to obtaining a dispersion of the slabs, as in a dispersion of polymers, in a solvent. In Li,Mo,Se,, e.g. Lit ions are located between Mo,Se, infinite chains built up from a condensation of Mo,Se, clusters (516.4.8.1). Both Li,Mo,Se, and Na,Mo,Se, can be dissolved' in highly polar solvents such as DMSO or N-methylformamide. In contrast to Li,Mo,Se, the absence of swelling in the M,Mo,X, compounds, with M = K, Rb, Cs and X = Se, Te, as well as the presence of macroscopic rod particles with Na,Mo,Se, are clearly the result of competition between solvation energy and a lattice energy, which increases with increasing size of the cationic species. Indeed, if the alkali metal is bigger, the balance between slab to slab repulsion and coulombic attraction between A' layers and negatively charged slabs is obviously in favor of the latter term: True layer MO, oxides do not exist because of the strong repulsion between 0' - layers, but Na,MO, derivatives exist that are isostructural with A,MS, intercalates ($16.4.6). In the exchange reactions between layered silicates or other two-dimensional structures a smaller cation tends to be replaced by a bigger one. Finally, layered oxides can exist in only three cases: (i) if extra cations are present between the slabs, as described just above; (ii) if hydrogen bonds are present to stabilize the structure (such a situation, as is the first one, is often encountered in silicates) or (iii) if the oxidation state of the metal ion is high enough to polarize the oxygen atoms so that the repulsion between adjoining layers is minimized (e.g., MOO,). (J. ROUXEL)

1. A. H. Thompson, C. R. Symon, Solid State lonics, 3-4, 175 (1981). 2. J. M. Tarascon, F. J. Di Salvo, C. H. Chen, P. J. Caroll, M. Walsh, L. Rupp, J. Solid State Chem., 58, 290 (1985).

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc. 338

16.4. The Formation of Sheet Structures 16.4.8. The Chemical Reactivity of Low-Dimensional Solids 16.4.8.5. Electronic Transfer and Ac Parameter Expansion.

shows that the number of Ni atoms in tetrahedral holes increases as x/2, the amount of Li This is coherent with the Ni2" + Nio reduction. The tetrahedral environment of Nio is usual in coordination chemistry. An x-ray recoilless resonance fluorescence spectroscopy study of Li,FePS, intercalates confirms the above results'. Microdomains of reduced iron in tetrahedral holes are found involving 20-200 Fe atoms. The unreduced octahedral Fe microdomains behave as if the first ones were not present. The magnetic ordering temperature and the hyperfine field do not differ from the pure FePS,. A third type of Fe is found in low concentration, considered to be located at the boundaries between reduced and unreduced domains. Local strain effects account for a distortion of the geometry of the sites that is reflected in the spectroscopic parameters. This rules out the existence of two macroscopic phases, which is also excluded by x-ray diffraction studies. (J. ROUXEL)

1. P. Colombet, G. Ouvrard, 0. Antson, R. Brec, J . Mugn. Mugn. Muter. 71, 100 (1987).

16.4.8.5. Electronic Transfer and Ac Parameter Expansion.

The electronic transfer from guest to host contributes to Ac by two opposite effects: an increasing repulsion between slabs that are now negatively charged and an attraction between these slabs and the A" positive layers. If the A' ion is small, it does not separate the slabs sufficiently to minimize the repulsion and the slab to slab repulsion may be the most important factor. In the Li-TiS, system, e.g., the lattice expansion on intercalation is associated with a charge donation to the host rather than with propping of the layers by Li' ions'. Such a situation is also favorable to obtaining a dispersion of the slabs, as in a dispersion of polymers, in a solvent. In Li,Mo,Se,, e.g. Lit ions are located between Mo,Se, infinite chains built up from a condensation of Mo,Se, clusters (516.4.8.1). Both Li,Mo,Se, and Na,Mo,Se, can be dissolved' in highly polar solvents such as DMSO or N-methylformamide. In contrast to Li,Mo,Se, the absence of swelling in the M,Mo,X, compounds, with M = K, Rb, Cs and X = Se, Te, as well as the presence of macroscopic rod particles with Na,Mo,Se, are clearly the result of competition between solvation energy and a lattice energy, which increases with increasing size of the cationic species. Indeed, if the alkali metal is bigger, the balance between slab to slab repulsion and coulombic attraction between A' layers and negatively charged slabs is obviously in favor of the latter term: True layer MO, oxides do not exist because of the strong repulsion between 0' - layers, but Na,MO, derivatives exist that are isostructural with A,MS, intercalates ($16.4.6). In the exchange reactions between layered silicates or other two-dimensional structures a smaller cation tends to be replaced by a bigger one. Finally, layered oxides can exist in only three cases: (i) if extra cations are present between the slabs, as described just above; (ii) if hydrogen bonds are present to stabilize the structure (such a situation, as is the first one, is often encountered in silicates) or (iii) if the oxidation state of the metal ion is high enough to polarize the oxygen atoms so that the repulsion between adjoining layers is minimized (e.g., MOO,). (J. ROUXEL)

1. A. H. Thompson, C. R. Symon, Solid State lonics, 3-4, 175 (1981). 2. J. M. Tarascon, F. J. Di Salvo, C. H. Chen, P. J. Caroll, M. Walsh, L. Rupp, J. Solid State Chem., 58, 290 (1985).

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc. 16.4. The Formation of Sheet Structures 16.4.8. The Chemical Reactivity of Low-Dimensional Solids 16.4.8.6. Acidobasic Topochemical Reactions.

339

16.4.8.6. Acidobasic Topochemical Reactions.

The acidobasic behavior of layered materials has led to elegant new syntheses illustrated by TiO2(B)I7,a new form of W 0 , 2 , or a mixed Nb-Ti oxide3. The starting material for TiO,(B) is potassium tetratitanate, K,Ti,09, with KL' ions between slabs made up of zigzag ribbons of four TiO, octahedra (Fig. 1). Potassium ions can be exchanged in an acidic medium: e.g., K,~,(H,O)nTi,O,-x(OH), where (0 < x 2 1). This causes protonation of the angular oxygen atoms of the octahedra blocks. The electronic density of such oxygen atoms is not as readily available for bonding by the cations inside the slabs. They represent more basic sites, which are protonated. Then, when heated, H,O condenses from two O H - groups, forming TiO,(B), which has the ionocovalent framework of Na,TiO,bronzes. The principle of the mechanism is a selective protonation of sites according to their basicity. This basicity depends on the geometry of the 0'- position and on the nature of the associated cation. This trend is shown by the intercalation of an interesting illustration of that point of view can be found in bases such as amines in the starting layered oxides. This intercalation is favored by a more acidic framework. Substitution in the slabs also changes the chemical and thermal stabilites of the materials. (J. ROUXEL)

Figure 1. The layered structure of K,Ti,,O,.

340

16.4. The Formation of Sheet Structures 16.4.8. The Chemical Reactivity of Low-Dimensional Solids 16.4.8.7. “Internal Surfaces” Grafting Chemistry.

1. R. Marchand, L. Brohan, M. Tournoux, Muter. Res. Bull., 15, 1129 (1980). 2. B. Gerand, G. Nowogrocki, J. Guenot, M. Figlarz, J. Solid State Chem., 29, 429 (1979). 3. H. Rebbah, G. Desgardins, B. Raveau, Muter. Res. Bull., 14, 1125 (2979).

16.4.8.7. “Internal Surfaces” Grafting Chemistry.

The third type of reaction, arising in low-dimensional solids, depends on surface chemistry. The van der Waals gaps are considered to provide reactive internal surfaces. The chemistry of layered oxihalides is distinguished from that of layered chalcogenides by a substitution reaction that allows easy replacement of the outer layer of the slabs, i,e., the chlorine layer, by other groups. The structure fends itself not only to reversible topochemical reactions by intercalation (§16.4.5.), but also to irreversible topochemical reactions involving the complete replacement of the external layers of the sheets, leaving only the inner layers unchanged. The Fe-CI bond can be broken, leading to grafting, pillaring and other types of reactions (Fig. 1). Simple grafting reactions have been discussed for FeOCl (816.4.5). The can be carried out as follows:

-

60°C + FEOCl acetone KCL + FeO-R 2 RNH, + FeOCl [R[NH,]]Cl+ FeONHR ROH + FeOCl HCI + FeO-OR

KR

(b) (c)

Aliphatic amines, aliphatic hydrocarbon potassium salts and other reagents have been used, e.g., K[O,CC,H,,], to obtain pendant groups on each side of the slabs. Double grafting on two different slabs of the van der Waals gap leads to stacks built up with inorganic floors and pillars that are usually organic’ : K-R-K

+ 2 FeOCl

-

2 KCl

+ FeOROFe

(4

Whereas this reaction occurs, it does not systematically lead to the formation of bridges between the sheets. A bending and folding of a molecule doubly fixed to the same sheet can be observed. To avoid this phenomenon it is necessary to chose more rigid organic molecules with either multiple bonds or rings. Such features are present in, e.g., p-KO-C,H,-C0,K. In this case the reaction is rapid and occurs in two stages: (i) intercalation with increase of the b parameter from 7.9 to more than 13 A, followed by

Figure 1. Simple (a) and double (b) grafting reactions in FeOCL.

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc. 340

16.4. The Formation of Sheet Structures 16.4.8. The Chemical Reactivity of Low-Dimensional Solids 16.4.8.7. “Internal Surfaces” Grafting Chemistry.

1. R. Marchand, L. Brohan, M. Tournoux, Muter. Res. Bull., 15, 1129 (1980). 2. B. Gerand, G. Nowogrocki, J. Guenot, M. Figlarz, J. Solid State Chem., 29, 429 (1979). 3. H. Rebbah, G. Desgardins, B. Raveau, Muter. Res. Bull., 14, 1125 (2979).

16.4.8.7. “Internal Surfaces” Grafting Chemistry.

The third type of reaction, arising in low-dimensional solids, depends on surface chemistry. The van der Waals gaps are considered to provide reactive internal surfaces. The chemistry of layered oxihalides is distinguished from that of layered chalcogenides by a substitution reaction that allows easy replacement of the outer layer of the slabs, i,e., the chlorine layer, by other groups. The structure fends itself not only to reversible topochemical reactions by intercalation (§16.4.5.), but also to irreversible topochemical reactions involving the complete replacement of the external layers of the sheets, leaving only the inner layers unchanged. The Fe-CI bond can be broken, leading to grafting, pillaring and other types of reactions (Fig. 1). Simple grafting reactions have been discussed for FeOCl (816.4.5). The can be carried out as follows:

-

60°C + FEOCl acetone KCL + FeO-R 2 RNH, + FeOCl [R[NH,]]Cl+ FeONHR ROH + FeOCl HCI + FeO-OR

KR

(b) (c)

Aliphatic amines, aliphatic hydrocarbon potassium salts and other reagents have been used, e.g., K[O,CC,H,,], to obtain pendant groups on each side of the slabs. Double grafting on two different slabs of the van der Waals gap leads to stacks built up with inorganic floors and pillars that are usually organic’ : K-R-K

+ 2 FeOCl

-

2 KCl

+ FeOROFe

(4

Whereas this reaction occurs, it does not systematically lead to the formation of bridges between the sheets. A bending and folding of a molecule doubly fixed to the same sheet can be observed. To avoid this phenomenon it is necessary to chose more rigid organic molecules with either multiple bonds or rings. Such features are present in, e.g., p-KO-C,H,-C0,K. In this case the reaction is rapid and occurs in two stages: (i) intercalation with increase of the b parameter from 7.9 to more than 13 A, followed by

Figure 1. Simple (a) and double (b) grafting reactions in FeOCL.

16.4. The Formation of Sheet Structures 16.4.8. The Chemical Reactivity of Low-Dimensional Solids 16.4.8.7. “Internal Surfaces” Grafting Chemistry.

341

(ii) substitution of chlorine atoms. This causes a decrease in the b parameter, which stabilizes at 11.33 A, in good agreement with the structure of the bulk molecule. These pillaring reactions are reminiscent of the pillaring in layered silicates using the [A1,304(0H)]247+cation. However, pillaring is not always an easy way to prepare new catalytic materials. The pillars may occupy most of the space between the slabs and impede the diffusion of molecules. It is necessary to separate the pillars sufficiently to allow this diffusion and, eventually, to add a size-selectivity property to the catalytic properties by the choice of the active sites introduced when the pillars are chosen. Simple grafting is often more interesting. The flexibility of the two-dimensional organization is maintained. Grafting depends on the electronic structure of the slab. It is easy to work with FeOCl, but not with CrOCI. The particular stability of Cr3+ ions in octahedral surrounding may impede the formation of an intermediate complex. (J. ROUXEL)

1. J. Rouxel, P. Palvadeau, J. P. Venien, J. Villieras, P. Janvier, B. Bujoli, Muter. Res. Bull., 22, 1217 (1987).

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc.

Abbreviations abs a.c. Ac acac acacH AcO Ad ads AIBN Alk am amt Am amu anhyd aq Ar asym at atm av BBN bcc BD biPY bipyH bp Bu

Bz

C-

ca. catal CDT cf. Ch. CHD Chx ChxD CI Cob COD COE conc const. COT COTe CP CPE

absolute alternating current acetyl, CH,CO acetylacetonate anion acetylacetone, CH,C(0)CH,C(O)CH3 acetate anion, CH,C(O)O adamant yl adsorbed 2,2-azobis(isobutyronitrile), 2,2-[(CH,),CCN],N, alkyl amine amount a w l , C,H,, atomic mass unit anhydrous aqueous aryl asymmetrical, asymmetric atom (not atomic, except in atomic weight) atmosphere (not atmospheric) average 9-Borabicyclo[3.3.1]nonane body-centered cubic butadiene 2,2’-bipyridyl protonated 2,Y-bipyridyl boiling point butyl, C,H, benzyl, C,H,CH, cyclo (used in formulas) circa, about, approximately catalyst (not catalyzing, catalysis, catalyzed, etc.) cyclododecatriene compare chapter 1,3-cycloheptadiene cyclohexyl 1,3-cyclohexadiene configuration interaction cobalamine cyclooctadiene cyclooctene concentrated (not concentration) constant cyclooctatriene cyclooctatetraene cyclopentadienyl, C,H, controlled-potential electrolysis

343

344

Abbreviations

CPm CT

counts per minute charge-tr ansfer cyclic voltammetry chemical vapor deposition continuous wave day, days

cv

CVD CW d DABIP DBA d.c. DCM DCME DCP DDT dec DED depe DIAD diars dien diglyme dil diop

dipda diphos Div. DMA dme DME DMF DMG DMP dmpe DMSO dpam dpic DPP dPPb

N,N’-diisopropyl-1,4-diazabutadiene

dibenzylideneacetone direct current dicyclopentadienylmethane CI,CHC(O)CH, 1,3-dicyclopentadienylpropane dichlorodiphenyltrichloroethane, l,l,l,’-trichloro-2,2-bis-(4chloropheny1)ethane decomposed

1,l-bis(ethoxycarbonyl)ethene-2,2-dilhiolate, “~Hsc,oc(o>l,c=c~,l

1,2-bis(diphenylphosphino)ethene,

(C,H,),PCH=CHP(C,H,), diindenylanthracenyl 1,2-bis(dimethylarsino)benzene,o-phenylenebis(dimethy1arsine), ~ , ~ - ( C H , ) , A S C ~ H , A ~ ( C H ~ ) ~ diethylenetriamine, [H,N(CH,),],NH diethyleneglycol dimethylether, CH30(CH,CH,0)CH, dilute 2,3-O-isopropylidene-2,3-dihydroxy1,4-bis(diphenylphosphino)butane, (C,H,),PCH,CH[OCH(CH,)=CH,ICH [OCH(CH,)=CH,]CH,P(C,HS)z p-i-PrC,H,CH=CHC,H,-c-p

1,2-bis(diphenylphosphino)benzene, 1,2-(C6H,)2PC6H,P(C6H5)2

division dimethylacetamide dropping mercury electrode 1,2-dimethoxyethane,glyme, CH,O(CH,),OCH, N,N-dimethylformamide, HC(O)N(CH,), dimethylglyoxime, CH,C(=NOH)C(=NOH)CH, 1,2-dimethoxybenzene,1,2-(CH,0),C6H, 1,2-bis(dimethylphosphino)ethane,(CH3)2P(CH,),P(CH3), dimethylsulfoxide, (CH,),SO bis(diphenylarsino)methane, [(C,H5),As],CH, dipicolinate ion differential pulse polarography 1,4-bis(diphenylphosphino)butane,

~,~-(C,HS)ZP(CHZ)~P(C,H~)~

1,2-bis(diphenylphosphino)ethane, 1,2~(C6H5)2P(CHZ)2P(c6H5~2

bis(diphenylphosphino)methane, [(C6H5)zP],CH,

bis(diphenylphosphory1)ethane 1,3-bis(diphenylphosphino)propane, 1~3-(C6H5)2P(CH2),P(c6H5)2

Abbreviations

DTA DTBQ DTH DTS ed. eds. EDTA e.g. EHMO emf en enH EPR equimol equiv EPR Eq. ERF ES ESR esu Et etc. Et,0 EtOH et seq. eu fac Fc fcc ff. Fig F1 FP fP g g-at GLC glyme graph GS h H-Cob HD

hept Hex HdMDB hmde MHI

1,2-bis(di-p-tolylphosphino)ethane, 1,2-(4-CH,CeHJ,P(CH,),P (C,H,CH3-4), differential thermal analysis 3,5-di-t-butyl-o-benzoquinone l,&dithiahexane, butane-1,4-dithiol, 1,4-HS(CH2),SH dithiosquarate edition, editor editors ethylenediaminetetraacetic acid, CHoC(o)lzN(CH,),"C(o)oHl, exempli gratia, for example extended Huckel molecular orbital electromotive force ethylenediamine, H,N(CH,),NH, protonated ethylenediamine electron paramagnetic resonance equimolar equivalent electron paramagnetic resonance equation effective reduction factor excited state electron-spin resonance electrostatic unit ethyl, CH,CH, et cetera, and so forth diethyl ether, (C,H,),O ethanol, C,H,OH et sequentes, and the following entropy unit facial ferrocenyl face-centered cubic following figure fluoren yl ~5-C,H,Fe(CO), freezing point gas gram-atom gas-liquid chromatography 1,2-dimethoxyethane,CH,O(CH,),OCH, graphite ground state hour, hours cobalamine 1,5-hexadiene heptyl hexyl hexamethyl(Dewar benzene) hanging mercury drop electrode heptamethylindenyl

345

346 HMPA HOMO HPLC i.e. Im inter alia IPC IR irrev ISC isn 1 L LC LF LFER liq LMCT Ln LSV LUMO m max M MC Me Men mes MeOH mer mhP min MLCT MO mol mP MV n.a. naPY NBD neg rihe NMR No. nP NP Nuc NPP NQR NTA 0

obs

Abbreviations hexamethylphosphoramide, [(CH,),N] ,PO highest occupied molecular orbital high-pressure liquid chromatography id est, that is imidazole among other things isopinocamphylborane infrared irreversible intersystem crossing isonicotinamide liquid ligand ligand centered ligand field linear free-energy relationship liquid ligand-to-metal charge transfer lanthanides, rare earths linear-scan voltammetry lowest unoccupied molecular orbital meta maximum metal metal centered methyl, CH, menthyl mesitylene, 1,3,5-trimethylbenzene derivative methanol, CH,OH meridional; the repeating unit of an oligomer or polymer 2-hydroxy-6-methylpyridine, 2-HO, 6-CH3C,H,N minimum, minute, minutes metal-to-ligand charge transfer molecular orbital molar melting point methyl viologen, l,l’-dimethyL4,4’-bipyridiniurn dichloride not available naphthyridine norbornadiene, [2.2.l]bicyclohepta-2,5-diene negative normal hydrogen electrode nuclear magnetic resonance number tris-[2-(diphenylphosphino)ethyl]amine, ~CCH,CH,P(C,H,),I, naphthyl nucleophile normal pulse polarography nuclear quadrupole resonance nitrilotriacetate ortho observed

Abbreviations

347

Oct OCP 0, 0, oq ox. P P. P Pat. pet. Ph phen Ph,PPy PiP PMDT

octyl octaethylporphyrin oxidation factor octahedral oxyquinolate oxidation para Page pressure patent petroleum phenyl, C6H5 1,lO-phenanthroline 2-(diphenylphosphino)pyridine, 2-(C6H5)2PC,H,N piperidine, C,HI,N pentamethyldiethylenetriamine,

PMR Pn POS Po-tol, PP. PPb PPm PPn PPt Pr PSS PVC PY PYr PZ PZE rac R RDE RE red. Redox ref. rev rf RF RF rh rms rPm RT

proton magnetic resonance propylene-1,3-diamine, 1,3-H,NCH,CH,CH2NH, positive tri-o-tolylphosphine pages parts per billion parts per million bis(diphenylphosphino)amine, [(C,H,),P],NH precipitate propyl, C3H2 photostationary state poly(viny1 chloride) pyridine, C5H,N pyrazine pyrazolyl potential of zero charge racemic mixture, racemate organic group; universal gas constant rotated disk electrode rare earths, lanthanides reduction reduction-oxidation reactions reference reversible radio frequency reduction factor R group with substituted F rhombohedra1 root mean square revolutions per minute room temperature second, seconds; solid saturated calomel electrode standard calomel electrode secondary sepulcrate, 1,3,6,8,10,13,16,19-octaazabicyclo[6.6.6]eicosane

S

sce SCE sec SeP

(CH~)2N(CH2)~N(CH3)(cH2~2N(cH3~2

348 Sia SMAD soh soh SP STP sub1 Suppl. sYm L

T Td TCNE TEA terPY tetraphos TGA TGL THF THP THT Thx TLC TMED tmen TMP TMPH To1 Tos TPA TPPO triars triphos trien

uv

V

Vi viz. vol., Vol. VPE vs. wk. wt X xs Y Yr. § Y

Abbreviations Diisamyl solvated metal-atom dispersed solution solvated specific standard temperature and pressure sublimes supplement symmetrical, symmetric time; tertiary temperature tetrahedral tetracyanoethylene tetraethylammonium ion, [(C2H,),N] 2,22"-terpyrid yl Ph,PCH2CH2PPhCH,CH2PPhCH2CH,PPh, thermogravimetric analysis triethyleneglycol dimethylether tetrahydrofuran tetrahydropyran tetrahydrothiophene thexyl thin-layer chromatography N,N,N’,N’-tetramethylethylenediamine, (CH3)2N(CH2)2N(CH3)2

’

N,N,N,N-tetramethylethylenediamine

2,2,6,6-tetramethylpiperidyl 2,2,6,6-tetramethylpiperidine, 2,2,6,6-(CH,),C,H,N tolyl, C,H,CHs, p-tolyl tosyl, tolylsulfonyl, 4-CH3C,H,S0, tetraphenylarsonium ion, [(C,H,),As] triphenylphosphineoxide bis-[-(dimeth y1arsino)phenyllmethylarsine, [~-(CH,)~ASC~HJ~ASCH~ +

1,1,l-tr~s(diphenylphosphinomethyl)ethane,

I(C6H5)2PCH213CCH3 triethylenetetraamine, H,N(CH2)2NH(CH,),NH(CH2)2NH, ultraviolet vicinal

(E)-[2-(CH3),NCH,C6H,]C=C(CH3)C,H,CH3-4

videlical, that is to say, namely volume vapor-phase epitaxy versus week weight halogen or pseudohalogen excess often used for S, Se year section hapto designator

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc.

349

Author Index

Author Index

The entries of this index were derived directly by computer program from the lists of references. The accuracy of the references was the sole responsibility of the authors. No editorial check, except for format and journal-title abbreviation,was applied. Consequently, errors occurring in authors’ names in the references will recur in this index. Each entry in the index refers to the appropriats section number. A Abele, R. 16.4.5 Abou El Ela, A. H. 15.2.2.2.13 Abraham, K. M. 15.2.10.4 Achenbach, H. 15.2.3.2 Achtsnit, H. 15.2.13.2.2 Acrivas, J. V. 16.4.3.4 Adams, J. M. 16.4.7.2 Ademu-John, C. 15.2.14.5.2 Adigezalora, N. 15.2.13.4 Aftergut, S. 15.2.14.1 Agaskar, P. A. 15.2.8.2 Agranat, I. 16.4.2.1.1 Ahlrichs, R 15.2.12.5 Ahrens, L H. 16.2.3 Aitken, C. T 15 2.4.5. Akhtar, M. 15.2 12.4

Alange, G. G. 15.2.12.5 Albernik, A. J. A. 16.4.3.2 16.4.3.5 Albright, T. A. 15.2.3.1 Alexander, R. P. 15.2.7.2 15.2.7.3 Allcock, H. R. 15.1.1.1 15.1.1.2 15.1.1.3.1 15.1.1.3.2 15.1.2.2.4 15.2.4.5. 15.2.7.3 15.2.11.1 15.2.11.2 15.2.11.3 15.2.11.4 15.2.13.1 15.2.13.2.3 15.2.13.3.2 15.2.14.5.2 15.2.14.1 Allen, R. W. 15.2.11.3 15.2.11.4 15.2.14.5.2 Allred, A. L. 15.2.4.2.2 Aloi, M. J. 15.2.13.2.1

15.2.13.5 Altman, M. B. 15.2.1 Amiell, J. 16.4.2.3 16.4.2.8.2 Anderson, R. 15.2.13.2.2 Andrianov, K. A. 15.2.8.2 15.2.9.2 15.2.9.3 15.2.9.5 15.2.13.3.2 15.2.13.3.4 Angoso Catalina, A. 16.4.2.5.3 Anhaus, J. 15.2.12.2 Ansul, G. R. 15.2.8.2 Antimonov, A. F. 16.4.2.6.3 Antson, 0. 16.4.8.4 Appelman, E. H. 16.4.2.1.2 Aragon de la Cruz, F. 16.4.2.5.3 Archer, R. D. 15.2.13.1 15.2.13.2.3 ATCUS,R. A. 15.2.11.1

350 Arif, A. M. 15.2.3.1 Arkles, B 15.2.13.2.2 15.2.4.1.2 Armand, M. 16.4.2.8.3 16.4.3.3 16.4.5 Armitage, D. A. 15.1.1.1 Armitage, F. 15.2.6.3 Arnal, H. 15.2.12.2 15.2.12.3 Aronson, S. 16.4.2.2.1 16.4.2.8.1 Artemova, V. S. 15.2.14.1 Ashby, E. C. 15.2.5.1.1 Ashe, 111, A. J. 15.2.3.1 Asher, R. C. 16.4.2.2.1 Asnovich, E. Z. 15.2.13.3.2 Aspey, S. 15.2.14.5.5 Aspey, S. A. 15.2.14.1 Assouik, J. 16.4.2.2.2 Aten, A. H. W. 15.2.1 15.2.2.2.3 Ates, M. 15.2.3.4 Atkinson, I. B. 15.2.5.1.1 15.2.5.5.1 Attwood, P. A. 16.4.2.5.3 16.4.2.7.3 Atwal, R. 15.2.13.3.1 Atyaksheva, L. F. 16.4.2.7.3 Aubke, F. 16.4.2.1.2 Audiere, J. P. 16.4.4 Ault, B. S. 15.2.6.2 Avdeev, V. V. 16.4.2.2.1 Averbuch-Pouchot, M. T. 15.2.10.2

Author Index Avogadro, A. 16.4.2.1.4

B

Bacon, R. F. 15.2.1 15.2.2.2.1 Badachape, R . B. 16.4.2.5.1 16.4.2.5.3 16.4.2.6.1 16.4.2.6.3 Baddour, R. F. 16.4.2.6.1 Baes Jr., C. F. 15.2.6.1.1 Bagieu-Beucher, M. 15.2.10.2 Bagnall, K. W. 15.2.1 Bailar, J. C. 15.2.1 15.2.14.1 15.2.14.4 Bailey, D. L. 15.2.8.4 Bailey, M. W. 15.2.2.2.1 Bailey, W. I. 15.2.12.3 Bajah, S. T. 15.2.13.5 Balesdent, D. 16.4.2.2.2 Ballard, D. 15.2.4.2.2 Bally, T. 15.2.4.1.1 Ban, H. 15.2.4.5. Baney, R. H. 15.2.4.5. Banister, A. J. 15.2.12.2 15.2.12.3 15.2.12.4 15.2.12.5 Banks, E. 16.3.2.2.1 16.3.2.2.2 16.3.2.3.2 16.3.2.3.3 Banthia, A. K. 15.2.8.4 Bard, A. J. 16.4.2.7.3 Barduhn, A. J. 16.2.2.2 Barendrecht, E. 16.4.2.7.3

Barker, K. A. 15.2.7.2 Barrer, R. M. 16.2.1 16.2.2.2 16.3.1.1 16.3.1.6 16.4.7.2 Barry, A. J. 15.2.8.1 Bart, J. C. J. 15.2.9.4 Bartholmei, S. 15.2.9.1 Bartlett, B. K. 15.2.5.3 Bartlett, M. 16.4.2.7.2 Bartlett, N. 16.4.1. 16.4.2.1.1 16.4.2.1.2 16.4.2.6 16.4.2.8.3 Bartnitskaya, T. S. 16.4.2.1.2 Barton, L. 15.2.6.2 Barton, S.S. 16.4.2.7.3 Basenhard, J. 0. 16.4.2.5.2 16.4.2.5.3 16.4.2.7.2 Basu S. 16.4.2.2.1 Batallan, F. 16.4.2.1.4 Baudler, M. 15.2.3.2 Bauer, M. E. 15.2.2.2.10 Bauer, S. H. 15.2.5.1.2 15.2.6.2 Baughman, R. H. 15.2.12.4 Bautista, R. G. 16.4.2.6.1 Bazuin, C. G. 15.2.14.3 Beachley, 0. T. 15.2.5.5.2 Beal, A. R . 16.4.3.4 Beard, C. D. 15.2.7.2 Beatty, H. A. 15.1.3.3 15.1.3.3.1

Author Index Beck, C. A. 15.2.5.1.1 15.2.5.5.1 Beck, F. 16.4.2.8.3 Becke-Goehrmg, M. 15.1.1.3.3 15.2.12.2 15.2.12.8 Becker, G. 15.2.3.4 Bkguin, F. 16.4.2.3 16.4.2.8.2 Beine, H. 15.2.1 Bekasova, N. I. 15.2.5.5.2 15.2.7 15.2.7.3 15 2.14.1 Belash, I. T. 16.4.2.2.1 Belhaes, P. 16.4.2.8.2 Belin, M. 16.4.1. Bell, B. 15.2.12.5 Bell, M. G . 16.4.3.1 Bellama, J. M. 15.2.3.3 Bellavance, D. 16.3.2 16.3.2.3.3 Bendriss-Rerhrhaye, A. 16.4.2.2.2 16.4.2.4.2 16.4.2.4.3. Bennion, D. N. 16.4.2.8.3 Bergbreiter, D. E. 16.4.2.8.2 Berger, A. 15.2.13.2.1 Berger, D. 16.4.2.2.1 Berlin, A. A. 15.2.14.1 Bernal, I. 15.2.3.5 Bernard, G. 16.4.2.8.2 Bernhard, W. J. 15.2.9.4 Bernheim, M. Y. 15.2.11.4 Bertaut, F. 16.4.3.1

Berthelot, J. 16.4.2.8.3 Berthet, M. P. 15.2.12.2 Berthier, C. 16.4.3.2 16.4.8.2 16.4.8.3 Berzins, T. 16.3.1.6 Besenhard, J. 0. 16.4.2.1.1 16.4.2.2.1 16.4.2.3 16 4.2.5.2 16.4.2.5.3 16.4.2.7.1 16.4.2.7.2 16.4.2.7.3 16.4.2.8.1 16.4.2.8.2 16.4.2.8.3 16.4.3.3 Best, L. R. 15.2.9.4 Best, R. J. 15.2.11.1 15.2.1 1.3 Bestul, A. B. 15.2.2.2.1 1 Bett, J. A. S. 16.4.2.7.3 Beurskene, G. 16.2.2.3 Beyer, R. T. 15.2.2.2.2 Biagoni, R . N. 16.4.2.1.1 16.4.2.1.2 Bianconi, P. A. 15.2.4.5. Bibcrachen, W. 16.4.2.7.2 Bibcracker, W. 16.4.2.7.2 Biberacher, W. 16.4.2.8.1 Bichon, J. 16.4.3.2 Bielefeldt, D. 15.2.12.5 Bierbaum, M. 15.2.4.3 Bigley, N. 15.2.14.5.2 Billaud, D. 16.4.2.1.1 16.4.2.2.1 16.4.2.2.2 16.4.2.4.2

351 16.4.2.8.3 Bingham, R. L. 15.2.12.4 Bird, J. D. 15.2.13.2.2 Bird, R. J. 16.4.2.5.3 16.4.2.7.3 Bissell, E. C . 15.1.1.3.2 Bissot, T. C . 15.2.5.1.2 Blackman, L. C. F. 16.4.2.8.3 Blandeau, L. 16.4.3.2 Blanke, W. 16.4.2.7.3 Blaxall, H. 15.2.13.3.5 Blicke, F. F. 15.2.3.3 Blinka, T. A. 15.2.4.1.1 15.2.4.1.2 15.2.4.3 Blix, M. 15.2.9.4 15.2.9.6 Block, B. P. 15.2.13.3.5 15.2.14.5.1 Blount, J. 15.2.4.1.1 Blurton, K. F. 16.4.2.7.3 Blyumenfel’d, A. B. 15.2.7.2 Boberski, W. G. 15.2.4.2.2 Bochkarev, V. N. 15.2.9.3 Boddeker, K. W. 15.2.5.4 15.2.5.1.2 15.2.5.5.1 Boeck, A. 16.4.2.1.2 Boehm, H. P. 16.4.2.1.1 16.4.2.1.4 16.4.2.5.1 16.4.2.5.2 16.4.2.5.3 16.4.2.5.4 16.4.2.6 16.4.2.7.1 16.4.2.7.3 16.4.2.8.1 16.4.2.8.3

352 Boese, R. 15.2.5.4 15.2.5.1.1 Boeyens, J. C. A 16.3.1.5 Bogusch, E. 15.2.9.2 15.2.9.3 Bohan, J. J. 15.2.7.2 Bohler, J. 15.2.13.2.2 Boilot, J. P. 16.4.3.2 Bonamico, M. 16.2.2.3 Bonnetain, L. 16.4.2.1.1 16.4.2.1.3. 16.4.2.3 16.4.2.8.2 16.4.2.8.3 Borsa, F. 16.4.3.2 Bos Alberink, A. J. A. 16.4.3.5 Bottomley, M. J. 16.4.2.1.4 16.4.2.5.2 16.4.2.8.1 16.4.2.8.3 Boudeulle, M. 15.2.12.3 Bouera, G. 16.4.3.2 Bourrie, D. B. 15.2.13.2.2 Bouyad, B. 16.4.2.8.1 Bradford, J. L. 15.2.5.5.2 Bradley, D. C. 15.2.13.3.2 Brauer, G. 15.2.12.7 Brauman, J. I. 16.4.2.1.2 16.4.2.1.3. 16.4.2.6.1 Braun, R. M. 16.4.6 Brec, R. 16.4.3.2 16.4.4 16.4.8.4 Breed, L. W. 15.2.9.2 15.2.13.3.3 Brennan, G. L. 15.2.5.4

Author Index Brenner, A. 16.4.2.8.3 Brenner, D. 15.2.14.5.2 Bresadola, S. 15.2.7.3 15.2.13.2.3 Bretschneider, 0 16.4.2.6.1 Breunig, H. J. 15.2.3.1 15.2.3.4 Brewer, S . D. 15.2.9.1 Briegleb, G. 15.2.2.2.3 Brinker, C. J. 15.2.6.1.1 25.2.13.3.2 Brodie, B. C. 16.4.2.5.1 16.4.2.5.2 Brohan, L. 16.4.8.6 Bronnikov, A. D. 16.4.2.2.1 Brooke, P. 15.2.13.1 15.2.13.2.3 Brotherton, R. J. 15.2.5.1.1 Brough, L. F. i5.2.4.1.2 15.2.4.3 Brown, B. K. 16.4.2.5.2 Brown, E. D. 15.2.8.4 Brown, G. P. 15.2.14.1 Brown Jr., J. F. 15.2.8 15.2.8.1 15.2.8.2 15.2.8.3 15.2.13.3.4 Bruchhaus, R. 15.2.12.3 Bruck, T. 15.2.3.5 Bruenig, H. J. 15 2.3.1 15.2.3.4 Brunner, H. 15.2.3.5 Buchanan, A. S. 15.2.6.2 Buchanan, 111, A. C. 15.2.4.3 Budde. W. L. 15.2.9.2

Bujoli, B. 16.4.8.7 Bulgakova, I. A. 15 2.14.1 Bulychova, E. G. 15.2.7.3 Bunker, B. C. 15.2.6.1.1 Bunting, R. K. 15.2.5.1.2 15.2.5.3 15.2.5.4 15.2.5.5.1 Burch, G. M. 15.1.3.3.1 Burch, J. E. 15.2.5.5.1 Burdett, J. K. 15.2.12.3 Burg, A. B. 15.2.3.2 15.2.3.4 Burger, H. 15.2.9.2 Burger, M. 15.2.12.2 Burgmeister, W. 15.2.9.4 Burkhard, C. 15.2.4.1.2 Burrish, G. 15.2.13.2.1 Burt, F. P. 15.2.12.2 Burwell, J. T. 15.2.1 Buscarlet, E. 16.4.2.1.1 16.4.2.1.3. 16.4.2.8.3 Buschfeld, A. 15.2.9.6 Bush, R. P. 15.2.9.2 Butler, C. 15.2.13.5 Buty, T. 16.4.2.7.2 16.4.2.8.1 Butz, T. 16.4.3.4 Buxton, R. L. 15.2.10.2 15.2.10.3

C

Cadenbach, G 16.4.2.2.1 Cadman, P. 16.4.2.6.1

353

Author Index 16.4.2.7.3 Caglioti, V. 15.2.11.1 Cais, M. 16.4.3.4 Calabrese, J. C. 15.2.4.3 Calage, Y. 16.4.3.2 Calderazzo, F. 15.2.3.1 Calingaert, G. H. 15.1.3 3 15.1.3.3.1 Callis, C. F. 15.2.10.1 Camilii, A. 15.2.8.2 Cano Ruiz, J. 16.4.2.5.3 Carberry, E. 15.1.1.3.1 15.2.4.1.2 15.2.4.3 Carides, J. N 16.4.3.2 Carlsohn, B. 15.2.3.2 Carlson. C. W 15.2.4.1.2 15.2.4.1.3 15.2.4.2.2 15.2.4.2.3 15.2.4.3 Carlson, L. 15.2.12.7 Carlsson, J. 15.2.13.2.3 Carmichael, J. B. 15.1.3 15.2.8.3 Carol], P. J. 16.4.8.5 Carpentier, C. D. 16.4.4 Carr, K. E. 16.4.2.5.1 Carraher, C. 15.2.13.1 15.2.13.2.1 15.2.13.2.2 15.2.132.3 15.2.13.3.1 15.2.13.3.5 15.2.13.4 15.2.13.5 15.2.14.5.2 15.2.14.5.3 15.2.13.6 Carraher, H. 15.2.13.1

15.2.13.2.3 Carraher, S. 15.2.13.5 Carton, B. 16.4.2.2.1 Cerutis, D. R. 15.2.14.5.2 Cerutis, D. S. 15.2.14.5.3 Cervents, R. 15.2.13.5 Chaabouni, M. 16.4.2.2.1 Chabre, Y. 16.4.8.2 16.4.8.3 Champetier, G. 15.2.9.4 Chang. S. S. 15.2.2.2.11 Chang, T.-H. 15.2.12.6 Chantrell, P. G. 15.2.13.2.2 Chapman, D. R. 15.2.7.2 Charles, R. 15.2.14.5.4 Charles, R. G. 15.2.14.5.4 15.2.14.1 Chen, C. H. 16.4.8.5 Chen, H. Y. 16.3.1.6 Chen, S. M. 15.2.4.1.1 15.2.4.1.2 15.2.4.1.3 15.2.4.1.4 15.2.4.3 Chen, W. 15.2.14.5.2 Chen, Y-L. 15.2.4.5. Chen, Y.-S. 15.2.4.1.3 Chenite, A. 16.4.2.8.3 Cherin, P. 15.2.1 Chevalier, P. 16.4.8.3 Chevrel, R. 16.4.8.1 Chevreton, M. 16.4.3.1 Chianelli, R. R. 16.4.3.4 Chiang, C. K. 15.2.12.4

Chien, J. C. W. 15.2.12.10 Chingas, G. C. 16.4.1. Chivers, T. 15.2.12.2 15.2.12.3 15.2.12.9 Choudhury, P. 15.2.3.3 15.2.3.4 Christ, C. L. 15.2.6.1.1 Christensen, M. J. 15.2.13.2.1 Christian, P. A. 16.4.3.2 Chu, C. T.-W. 15.2.11.4 Chung, D. D. L. 16.4.2.1.5 16.4.2.7.1 Churchill, M. R. 15.2.3.1 15.2.3.5 Clapp, D. B. 15.2.5.1.1 Clark, D. 15.2.13.3.2 Clark, H. A. 15.2.13.2.2 Clark. J. R. 15.2.6.1.1 Clarke, R. 16.4.2.2.1 Clausen, C. A. 16.4.1. Clauss, A. 16.4.2.5.3 16.4.2.5.4 Clauss, H. 15.2.9.4 Claussen, W. F. 16.2.1 16.2.2 16.4.1 Clavene, J. 16.4.6 Clegg, W. 15.2.9.1 15.2.9.2 Clement, R. 16.4.4 Clements, K. E. 16.4.2.1.1 Clough, S. 16.4.5 Clymo, R. 15.2.14.5.3 Coblenz, W. S. 15.2.7.2

354 Codding, P. W. 15.2.12.3 Coetzee, C. J. 16.3.1.5 Cohen, A. D. 16.4.2.1.1 Cohen, M. J. 15.2.12.3 Cohen, M. S. 15.2.7.2 Coic, L. 16.4.5 Colclough, R. 0. 15.2.11.1 Colin, G. 16.4.2.1.5 Colin, M. 16.4.2.4.1. Collins, D. H. 16.4.2.6.1 Collins, S. 15.2.4.2.3 15.2.4.4. Collins, W. T. 15.2.8.2 15.2.8.3 Colombet, P. 16.4.3.6 16.4.8.4 Conard, J. 16.4.2.4.2 Connolly, M. S. 15.2.11.4 Cook, C. B. 15.2.13.3.1 Cook, C. F. 16.4.2.1.5 Cook, S. H. 15.2.10.4 Cook, W. J. 15.2.11.1 15.2.11.2 15.2.11.4 15.2.14.5.2 Cooke, D. J. 15.2.8.2 Cooper, G. D. 15.2.8 Cooper, W. C. 15.2.1 Corbridge, D. E. C. 15.2.10 15.2.10.2 15.2.10.3 15.2.10.4 Cordes. A. W. 15.2.12.2 Cordischi, D. 15 2.11.1 Costello, A. J. R. 15.2.10.1

Author Index 15.2.10.2 Costello, J. R. 15.2.10.3 Cotter, R. J. 15.2.13.3.4 Cousins, W. 15.2.2.2.3 Cousseau, J. 16.4.3.2 16.4.3.4 Cowley, A. H. 15.2.3.1 15.2.3.2 15.2.13.1 15.2.13.2.3 Cox, D. M. 16.4.1. Craighead, P. W. 15.2.9.4 15.2.9.6 Craigie, J. 15.2.14.5.3 Cremer, A. 16.4.7.2 Crist, C. L. 15.2.6.1.1 Croft, R. C. 16.4.2.1.2 16.4.2.1.3. Crooke, W. 15.2.14.5.3 Cros, C. 16.2.1 16.2.3 16.2.3.1 16.4.3.2 Crowley, J. I. 15.2.12.4 Csalan, E. 16.4.2.5.3 Culbertson, B. 15.2.13.1 15.2.13.2.3 Curl, R. F. 16.4.1. Currell, B. R. 15.2.5.1.1 15.2.5.4 15.2.5.5.1 15.2.13.3.1 Curry, J. E. 15.2.13.2.2 Cutler, M. 15.2.2.2.4 15.2 2.2.5

D

Dahl, G. H. 15.2.5.1.1

15.2.5.2 15.2.5.4 Dahn, J. R. 16.4.8.3 Dalgaard, G. A. P. 15.2.12.6 Dalle-Molle, E. 16.4.2.7.3 Daly, J. J. 15.2.3.3 15.2.9.4 Damjanovic, A. 16.3.2.1 16.3.2.2.2 16.3.2.3.2 16.3.2.3.3 Dammeier, R. L. 15.2.13.4 Danilenko, A. M. 16.4.2.6.1 Danot, M. 16.4.3.2 16.4.3.6 Darken, L. S. 16.4.3.3 Datta, A. 15.2.2.2.11 Daudt, W. H. 15.2.8.1 Daumas, N. 16.4.2.3 David, L. D. 15.2.4.1.4 15.2.4.3 15.2.9.3 Davidov, D. 16.4.2.1.1 Davies, W. B. 16.4.3.4 Davis, P. B. 15.2.7.2 Davison, J. B. 15.2.13.3.3 Davy, H. 16.2.1 Dawson, J. W. 15.2.5.1.1 15.2.5.1.2 Day, V. W. 15.2.8.2 De Jouge, R. 16.4.3.1 De Neef, T. 15.2.2.2.6 de Pape, R. 16.3.1.2 De Young, D. J. 15.2.4.2.3 Dehnicke, K. 15.2.12.4

Author Index Dele-Dubois, M. L. 16.4.2.1.4 Delhaes, P 16.4.2.3 16.4.2.7.1 16.4.2.7.2 Delios, C. 16.4.3.4 Delmas, C. 16.4.6 Delplanque, G. 16.4.3.2 Delpy, K. 15.2.5.1.1 15.2.5.4 Demant, U. 15.2.12.4 Denton, D. L. 15.2.5.3 Derbyshire, W. 15.2.1 15.2.2.2.4 15.2.2.2.5 Derksen, J. C. 16.4.2.5.3 Desgardins, G. 16.3.1.3 16.4.8.6 DeSorcie, J. L. 15.2.1 1.4 Devalette, M. 16.4.6 Di Salvo, F. J. 16.4.3.2 16.4.3.4 16.4.8.5 Dierdorf, D. S. 15.2.3.2 Dietrich, H.J. 15.2.7.2 Dilanjan Soysa, H. S. 15.2.9.4 Dima, M. 15.2.14.1 DiMaio, A.-J. 15.2.3.5 Dines, M. B. 16.4.3.2 16.4.3.4 DiNunzio, J. 15.2.14.5.3 Dion, M. 16.3.1.4 DiPetro, H. R. 15.2.14.1 Dippel, K. 15.2.9.2 Dislich, H. 15.2.13.3.2

Ditter, J. F. 15.2.7 15.2.7.2 Dobinski, S. 15.2.2.2.1 Dolgoplaosk, S. B. 15.2.13.3.3 Domicone, J. J. 15.2.8.1 Donahue, J. 15.2.4.4. Donnet, J. B. 16.4.2.7.3 Donnet, J. P. 16.4.2.7.3 Dorofeenko, L. P. 15.2.7.2 Douglas, C. M. 15.1J.3.1 Downing, J. W. 15.2.4.5. Drager, M. 15.2.3.4 Dresselhaus, G. 16.4.2.2.1 Dresselhaus, M. S. 16.4.1. 16.4.2.1.5 16.4.2.2.1 Drone, F. J. 15.2.3.1 Dryoff, D. R. 15.2.10.2 DuBois, D. A. 15.2.5.5.1 Dudley, E. A. 15.2.14.1 15.2.14.5.5 Dumont, E. 15.2.13.2.2 Dungan, C. H. 15.1.3.3.2 15.2.9.5 15.2.10.1 Dunham, M. L. 15.2.8.4 Dunks, G. B. 15.2.7.2 Dunnarant, W. R. 15.2.13.3.3 Dunnavant, W. R. 15.2.13.2.2 Dunning, J. S. 16.4.2.8.3 Dupree, R. 15.2.2.2.9 Durasov, V. B. 16.4.2.6.3 Durif, A,. 15.2.10.2

355 Dvornic, P. R. 15.2.13.3.3 Dyroff, D. R. 15.2.10.3 Dzurus, M. L. 16.4.2.7.2

E

Eaborn, C. 15.2.8 15.2.8.1 Earnest, T.R. 15.2.14.3 Eberlein, J. 15.2.13.2.3 Ebert, F. 16.4.2.6.1 Ebert, L. B. 16.4.1. 16.4.2.1.1 16.4.2.1.2 16.4.2.1.3. 16.4.2.1.4 16.4.2.1.5 16.4.2.6 16.4.2.6.1 16.4.2.7.1 16.4.2.7.2 16.4.2.8.1 Eckel, M. 16.4.2.5.2 16.4.2.8.1 Economy, J. 15.2.1 3.4 Edeling, M. 15.2.2.2.14 Edwards, J. W. 15.2.10.3 Eeles, W. T. 16.4.2.1.5 Egert, E. 15.2.9.2 Egsgaard, H. 15.2.12.7 Ehrburger, P. 16.4.2.7.3 Ehrhardt, C. 15.2.12.5 16.4.2.1.1 Eisenberg, A. 15.2.2.1 15.2.2.2.1 1 15.2.14.3 Eklund, P. C. 16.4.2.1.5 El Makrini, M. 16.4.2.2.2 Elalem, N. E. 16.4.2.4.2

356 16.4.2.4.3. Elansari, L 16.4.2.4.2 16.4.2.4.3. Elbel, S. 15.2.12.7 Eliel, E. L. 15.2.4.3 Elliott, J. R. 15.2.8 Elliott, R. L. 15.2.9.2 15.2.13.3.3 Ellis, D. E. 16.4.8.3 Ellis, J. A. 15.2.9.4 Elmes. P. S. 15.2.3.3 15.2.3.5 El-Shady, M. F. 15.2.3.4 Elter., G. 15.2.5.4 Elutin, V. P. 16.4.2.1.3. Elving, P. J. 16.4.2.7.3 Emeleus, H. J. 15.2.1 15.2.12.5 Emel’yanova, G. I. 16.4.2.7.3 Emerson, G. F. 15.2.3.5 Emmett. P. H 16.4.2.7.3 Endo, M. 16.4.2.6.1 Enterling, D. 15.2.9.2 Epstein, B. D. 16.4.2.7.3 Ercolani, C. 15.2.8.2 Erhardt, C. 16.4.2.8.3 Ermachkova, I. F. 15.2.7.2 Essaddek, A. 16.4.2.2.2 Estrade-Szwarckopf, H. 16.4.2.4.2 Etienne, Y. 15.2.9.4 Eulenberger, G. 16.4.4 Evans, T. L. 15.2.11.1 15.2.11.4

Author Index Ewing, A. G. 15.2.11.4

F

Facchini, L. 16.4.2.3 Failli, A. 15.2.12.6 Falardeau, E. R. 16.4.2.1.1 Falloner, J. D. 16.3.1.1 16.3.1.6 Fanelli, R. 15.2.1 15.2.2.2.1 Faraday, M. 16.2.1 Farmer, B. L. 15.2.4.5. Farmer, J. B. 15.2.6.1.1 15.2.6.4 Farr, C. 15.2.1 Fedorov, G. G. 16.4.2.7.3 Fedorova, N. L. 15.2.7.2 Feher, F. 15.2.9.4 Feil, D. 16.2.2.3 Fein, M. M. 15.2.7.2 Feld, W. 15.2.14.5.3 Fergusson, R. C. 15.2.3.3 Ferretti, A. 16.3.2.3.1 Fettinger, J. C. 15.2.3.1 Fewell, L. L. 15.2.7.3 Fialkov, A. S. 16.4.2.7.2 Ficalora, P. J. 16.4.2.6.1 16.4.2.6.3 Fickes, G. N. 15.2.4.5. Fielder, A. J. 15.2.12.2 Figlarz, M. 16.4.8.6 Findlay, R. H. 15.2.12.3 Fink, M. J. 15.2.8

15.2.4.2.3 Fink, W. 15.2.9.1 15.2.9.2 Fischer, J. E. 16.4.1. 16.4.2.2.1 16.4.2.8.1 16.4.2.8.3 Fischer, R. 15.2.2.2.12 Fitzer, E. 16.4.2.7.3 Flagg, E. E. 15.2.13.3 5 Flanders, P. 16.4.2.2.1 Flandrois, S. 16.4.2.2.1 16.4.2.8.1 Flath, R. A. 15.2.9.4 Fleischmann, C. W. 16.3.2.2.1 16.3.2.2.2 16.3.2.3.2 16.3.2.3.3 Fleischmann, M. 16.4.2.1.5 Fleitmann, T. 15.2.10 Fleming, W. 15.2.4.5. Flogel, P. 15.1.3.3.2 Flory, P. J. 15.1.3 Fluck, E. 15.1.1.3.3 15.1.3.3.1 15.2.12.8 15.2.12.9 Foley, M. J. 15.2.3.5 Fomina, L. P. 15.2.13.3.3 Forder, R. A. 15.2.12.5 Forestier. J. 15.2:1 Fornes. R. E. 16:4.2.7.3 Forsman, W. C. 16.4.2.8.3 Forstner, J. A. 15.2.9.4 Fortman, J. 15.2.14.5.2 Foster, R. 16.4.2.1.5

357

Author Index Foster, V. 15.2.13.2.1 15.2.13.5 Foster, W. E. 15.2.13.3.2 Fouassier, C. 16.4.6 Foug, W. 16.4.2.7.2 Fouletier, M. 16.4.2.8.3 Fountain, M. E. 15.2 3.5 Fountain, M. L. 15.2.3.5 Fourquet, J. L. 16.3.1.2 Foutasse, C . A. 15.1.1.3.4 Fowler, D. L. 16.2.2.3 Fraenkel, G. K. 15.2.1 15.2.2.2.5 15.2.2.2.6 Frankl, G. 16.4.2.8.1 Franzen, P. 16.4.7.2 Frazier, S. E. 15.1.1.3.1 15.2.11.2 Fredenhagen, K. 16.4.2.2.1 Fredlein, R. A. 16.3.2.1 16.3.2.2.2 16.3.2.3.2 16.3.2.3.3 Frenkel’, M. M. 15.2.9.4 Frenklach, M. 16.4.1. Frenzel, A. 16.4.2.5.3 Freyland, W. 15.2.2.2.5 Fricke, A. L. 15.2.5.5.1 Fridland, D. C. 15.2.9.4 Fridland, D. V. 15.2.9.4 Friedel, J. 16.4.8.3 Friedlich, H. B. 16.4.2.1.5 Friend, A. 15.2.14.5.2 Frishberg, C. 16.4.2.8.1

Fritz, H. P. 15.2.12.3 16.4.2.5.2 16.4.2.5.3 16.4.2.7.3 16.4.2.8 16.4.2.8.1 16.4.2.8.3 Frolova, Z. M. 15.2.7.3 Fruhbuss H. 16.2.2.2 Frunze, T. M. 15.2.7 Frye, C. L. 15.2.8 15.2.8.1 15.2.8.2 15.2.8.3 15.2.8.4 15.2.8.5 15.2.13.3.4 Fuchs, J. 15.2.9.2 Fuerst, C. D. 16.4.2.2.1 Fuerst, C . J. 16.4.2.2.1 Fug, G. 16.4.2.7.1 16.4.2.7.2 Fugellier, H. 16.4.2.8.1 Fuji, R. 16.4.2.8.3 Fuji, Y. 16.4.2.6.1 16.4.2.6.2 Fukatsu, K. 15.2.14.1 Fukuda, M. 16.4.2.6.1 Fukui, K. 15.2.12.2 Fukushima, K. 15.2.4.1.3 Fuller, T. J. 15.2.11.4 Fullgrabe, H. J. 15.2.5.5.2 Furdin, G. 16.4.1. 16.4.2.1.1 16.4.2.1.5 16.4.2.8.3 Fuzellier, H. 16.4.2.1.3. 16.4.2.1.4 16.4.2.8.3

G

Gaines, D. F. 15.2.5.1.2 Galkin, A. F. 15.2.5.5.2 Gallagher, P. K. 16.4.2.1.1 16.4.2.1.2 16.4.2.1.5 16.4.2.6 16.4.2.6.1 Gallmeier, J. 16.2.1 16.2.3.2 Gamble, F. R. 16.4.3.4 Gani, 0. 16.4.2.1.1 16.4.2.1.2 Garcia, A. R. 16.4.2.1.1 Gard, G. L. 15.2.12.5 Gardner, D. M. 15.2.1 15.2.2.2.5 15.2.2.2.6 Gardner, J. A. 15.2.2.2.4 Gardner, J. E. 15.2.11.1 15.2.11.4 Garito, A. F. 15.2.12.2 15.2.12.3 Gartaganis, P. A. 15.2.10.1 Garten, V. A. 16.4.2.7.3 Gaspar, P. P. 15.2.4.1.3 Gatehouse, B. M. 15.2.3.5 Gaul, Jr., J. H. 15.2.4.5. Gaur, U. 15.2.2.2.11 Gauthier, S. 16.4.1. Geanangel, R. A. 15.2.5.1 1 15.2.5.1.2 15.2.5.5.1 Geballe, T. H. 16.4.3.4 Gebelein, C. 15.2.13.3.5 15.2.14.5.2

358 Gebhard, L. A. 16.4.2.1.3. Gee, G . 15.2.2.1 15.2.2.2.8 15.2.11.1 Gehrke, T. 15.2.13.5 Geigl, K.-H. 16.4.2.7.3 Geiser, V. 16.4.2.2.1 16.4.2.8.2 Gerand, B. 16.4.8.6 Gerber, A. H. 15.2.13.2.1 15.2.13.3.5 15.2.14.1 15.2.14.4 15.2.14.5.4 Gerber, G. 15.2.13.2.1 Gerhart, F. J. 15.2.7.2 Gerlach, E. 15.2.1 Gernez, M. D. 15.2.1 Gerrard, W. 15.2.5.2 15.2.5.3 15.2.5.4 15.2.5.5.1 15.2.5.5.2 15.2.6.2 Gertner, D. 15.2.13.4 Gesundheit, N. 15.2.9.2 Geymayer, P. 15.2.9.2 15.2.9.3 Gier, T. E. 16.3.1.6 Gilbert, A. R. 15.2.8 Gilbert, R. D. 16.4.2.7.3 Gilkey, J. W. 15.2.8.1 Gill, D. 15.2.13.5 Gill, W. D. 15.2.12.2 15.2.12.3 Gillery, F. H 16.4.7.2 Gillespie, D. J. 16.4.2.7.3

Author Index Gillham, J. K. 15.2.7 Gilman, H. 15.2.4.1.1 15.2.4.1.2 15.2.4.2.2 Ginderow, D. 16.4.2.3 16.4.2.8.2 Giron, D. 15.2.13.5 15.2.14.5.2 Giron, D. J. 15.2.13.5 Glauss, A 16.4.2.5.1 Glemser, 0. 15.1.1.3.1 15.2.12.5 15.2.12.6 Glonek, T. 15.2.10 15.2.10.1 15.2.10.2 15.2.10.3 15.2.10.4 Cole, J. 16.4.2.8.2 Golubenkova, L. I. 15.2.7.2 Golubtsov, S . A. 15.2.7.2 Gordon, D. J. 15.2.8.3 Goring, G. 16.4.7.2 Gossedge, G. M. 16.4.2.6.1 Gotcher, A. J. 15.2.7 15.2.7.2 Could, R. 0. 15.2.6.1.1 Could, S . E. B. 15.2.6.1.1 Graf zu Stollberg, N. 15.1.1.3.1 Graham, W. A. G. 15.1.1.1 15.2.13.3.6 Grain, R. D. 15.2.13.3.6 Grandits, D. M. 16.3.2.3.3 Granditz, D. M. 16.3.2.3.1 Grankin, V. M. 16.4.2.6.3 Grannec, Y. J. 16.4.2.1.1

Grant, D. 15.1.3.3.2 15.2.10.1 Grant, L. R. 15.2.3.4 Grant, P. M. 15.2.12.3 15.2.12.2 Gray, A. P. 15.2.13.2.2 15.2.7.2 Green, J. 15.2.7.2 Green. M. L. H, 16.4.3.4 16.4.4 Greene, R. L. 15.2.12.2 15.2.12.3 Greigger, P. P. 15.2.11.4 Gridina, V. F. 15.2.7.2 Griffith, E. J. 15.2.10 15.2.10.1 15.2.10.2 15.2.10.3 Grigutch, F. D. 16.4.2.1.1 Grimm, G. A. 15.2.6.2 Grimmer, A. R. 15.2.10.3 Grimpel, M. 15.2.13.2.2 Grinblat, M. P. 15.2.11.2 Grippa, M. L. 16.4.3.2 Groenweghe, L. C. D. 15.1.3 15.1.3.3 15.1.3.2.1 15.1.3.2.2 15.1.3.3.1 15.1.3.3.2 15.2 10.1 Gross, J. 16.4.2.2.1 16.4.2.7.1 16.4.2.7.2 Gross, P. 15.2.1 Gross, R. 15.2.3.5 Grushkin, B. 15 2.12.9 Guenot, J. 16.4.8.6

359

Author Index Guenther, F. 0. 15.2.8.2 Guentherodt, H. J. 16.4.2.1.1 GuBrard, D. 16.4.2.1.1 16.4.2.2.1 16.4.2.2.2 16.4.2.4.2 16.4.2.4.3. 16.4.2.4.4. Guggenheim, H. S. 16.4.2.1.1 Guitel, J. C. 15.2.10.2 Gulec, S. 15.2.3.4 Gdntherodt, H. J. 16.4.2.2.1 16.4.2.8.2 Gupta, S. K. 15.2.6.4 Gupta, V. K. 15.2.3.3 Gurrits, E. D. 15.2.13.3.2 Guselnikov, L. E. 15.2.8.2 Gutmann, V. 15.2.5.1.1 15.2.5.5.2 Guymarc’h, A. 16.3.1.3

H

Haange, R. J. 16.4.3.5 Haas, A. 15.2.9.4 15.2.9.6 15.2.12.5 Haas, C. 16.4.3.1 Haase, M. 15.2.9.2 Habei1e.K 15.2.3.4 Haber, C. P. 15.2.9.1 Haberland, G. 15.2.8.2 Hachim, L. 16.4.2.1.1 Hadd, K. A. 15.2.13.2.2 Haddon, W. F. 15.2.9.5 Hadigui, S. E. 16.4.2.4.2

Haduc, I. 15.2.10.2 Haenni, P. 16.4.2.5.2 Haering, R. R. 16.4.8.2 Hagenmuller, P. 16.2.1 16.2.3 16.2.3.1 16.3.2.3.3 16.4.5 16.4.6 Hagnauer, G. L. 15.2.11.4 Hahn, G. A. 15.2.5.3 Hahn, H. 16.4.4 Haiduc I. 15.1.1.1 15.2.6.2 15.2.6.3 15.2.9.1 15.2.9.2 15.2.9.3 15.2.9.4 15.2.9.5 15.2.9.6 Halbert, Thomas R. 16.4.5 Halett, J. C. 15.2.3.3 Halferlich, F. 16.3.1.1 Hall, E. A. H. 16.4.2.8.2 Hallensleben, M. L. 15.2.13.2.1 Haller, K. J. 15.2.4.1.4 15.2.4.3 15.2.8 Hammerschmidt, E. 16.2.2.2 Hammick, D. L. 15.2.2.2.3 Hampton, J. F. 15.2.8.2 Hamwi, A. 16.4.2.1.1 16.4 2.3 Hanic, F. 15.2.10.2 15.2.10.3 Hanlon, L. R. 16.4.2.1 1 Hanyawa, Y. 15.2.4.1.1 Hark, S. K. 16.4.2.2.2

Harold, G. 15.2.5.1.1 Harper, J. R. 15.2.3.5 Harrell, A. L. 15.2.4.1.1 Harris, J. J. 15.2.5.1.1 Harris, P. J. 15.2.11.2 15.2.11.4 Harris, R. E. 15.2.1 15.2.2.2.3 Harrison, B. H. 16.4.2.7.3 Harrison, D. E. 15.2.2.2.1 Harrod, J. F. 15.2.4.5. Hartley, F. 15.2.13.1 15.2.13.2.3 Hasegama, Y. i5.2.13.2.2 Hashimoto. M 15.2.14.1 Hashimoto, T. 15.2.4.3 Hassler, J. C. 15.2.5.5.1 Hauptman, Z. V. 15.2.12.2 15.2.12.3 Haworth, D. T. 15.2.5.1.1 Hawthorne, M. F. 15 2.5.1.1 Hayase, S. 15.2.4.5. Hayashi, J. 15.2.4.5. Hazell, A. C. 15.2.12.6 Hazell, R. G. 15.2.12.6 Heal, H. G. 15.2.12.1 15.2.12.5 15.2.12.6 15.2.12.7 Heald, S. M. 16.4.2.1.5 Heath, J. R. 16.4.1. Hedaya, E. 13.2.7.2 Heeger, A. J. 15.2.12.2 15.2.12.3

360 15.2.12.4 Heine, M. 16.4.2.7.3 Heitmann, D. 16.4.2.1.1 Helferlich, F. 16.3.1.6 Helle, W. 16.4.2.1.4 16.4.2.8.3 Heller, G. 15.2.6.1.1 Helmer, B. 15.2.4.3 Helmer, B. J. 15.2.4.1.3 15.2.4.1.4 Hench, L. 15.2.13.3.2 Hendra, P. J. 16.4.2.1.5 Hengge, E. 15.2.4.1.1 15.2.4.1.2 15.2.4.1.4 15.2.4.2.1 15.2.4.2.2 15.2.4.5. Henneberg, W. 15.2.10 Hennig, G. R. 16.4.2.1.5 16.4.2.5.3 16.4.2.7.1 16.4.2.7.2 Henrich, G. 16.4.2.7.3 Hensel, F. 15.2.2.2.7 15.2.2.2.12 15.2.2.2.14 Herold, A. 16.4.1. 16.4.2.1.1 16.4.2.1.2 16.4.2.1.3. 16.4.2.1.4 16.4.2.1.5 16.4.2.2.1 16.4.2.2.2 16.4.2.3 16.4.2.4.1 16.4.2.4.2 16.4.2.4.3. 16.4.2.4.4. 16.4.2.1.2 16.4.2.8.2 16.4.2.8.3 16.4.3.2 Herran, J. 16.4.2 8.1

Author index Herring, D. L. 15.1.1.3.1 Herwig, H. 15.2.13.2.2 Herzog, A. H. 15.2.10.3 Hess, G. 15.2.14.5.2 Hesse, M. 15.2.9.2 Hey, R. G. 15.2.12.2 Heying, T. L. 15.2.7 15.2.7.2 15.2.7.3 Heyling, T. L. 15.2.13.2.2 Heymach, G. J. 16.4.2.6.3 Hibbs, J. D. 16.4.2.1.5 Hibma, T. 16.4.3.2 Hickam Jr., C. W. 15.2.5.3 Higuchi, K. 15.2.4.4. 15.2.4.1.2 Hiltv. ,, T. K. 15.2.4.5. Hirai, T. 16.4.8.3 Hluchy, H. 15.2.9.1 15.2.9.2 Hochino, Y. 15.2.4.4. Hoebbel, D. 15.2.8.2 Hoess, E. A. 15.2.13.2.2 Hofer, D. 15.2.4.5. 15.2.13.3.2 Hofer, R. 15.2.12.6 Hoffman, R. 15.2.12.10 15.2.3.5 Hoffmann, R. 15.2.3.5 Hofmann, U. 16.4.2.1.4 16.4.2.5.1 16.4.2.5.3 16.4.2.5.4 16.4.2.7.3 Hofmeister, H. K. 15.1.3.3.1

Hohlwein, D. 16.4.2.1.1 16.4.3.4 Hohmann, E. 16.2.3.1 Hohnstedt, L. F. 15.2.5.1.1 Hojo, M. 15.2.12.7 Hoio. * I N. 15.2.14.1 Holbrook, G. W. 15.2.8.4 Holliday, L. 15.2.13.4 15.2.14.3 Holmes, W. B. 15.2.1 Holt, E. M. 15.2.12.9 Holt, S. L. 15.2.12.9 Holtman, M. S. 15.2.4.2.2 Hong, H. Y. P. 16.3.1.4 Honle, W. 15.2.3.2 15.2.3.5 Hooks, H. 15.2.7.2 Hooley, J. G. 16.4.2.1.1 16.4.2.1.2 16.4.2.1.3. 16.4.2.1.4 16.4.2.1.5 16.4.2.7.1 16.4.2.7.2 16.4.2.7.3 Hoppe, R. 16.4.6 Horiguchi, R. 15.2.4.5. Horn, D. 16.4.2.1.4 16.4.2.8.3 Horowitz, E. 15.2.14.1 Horsma, D. A. 15.2.2.2.10 Houtman, J. P. W. 15.1.3.3 Hsu, C. 15.2.12.2 Hsueh, L. 16.4.2.8.3 Hu, Y. C. 16.2.2.2 Hua-Cheng, S. 15.2.2.2.11

361

Author Index Huang, C. T. 15.2.8 2 Huber, M. 16.4.3.1 Hudson, H. R. 15.2.5.5.2 Hueber, F. 16.4.2.7.3 Huggins, R. A. 16.3.2.3.1 16.3.2.3.3 16.4.2.1.2 16.4.2.1.3. 16.4.2.6.1 Huheey, J. E. 16.4.2.1.2 Huisman, R. 16.4.3.1 Hulliger, F. 16.4.2.12 16.4.5 Hulme, R. 16.4.2.6.1 Hummers, W. S. 16.4.2.5.2 Hunger, H. F. 16.4.2.6.3 Huntsman, B. 15.2.14.5.3 Hurt, C. J. 15.2.4.2.2 15.2.4.3 Huttner, W. 16.4.2.7.3

I

Iijima, T, 16.4.2.6.1 Ikeda, H. 15 2.4.3 Ikeda, K. 15.2.10.4 I1 Nam Jung, 15.2.9.4 Il'in, A. S. 15.2.6.1.1 Imoto, H. 16.4.2.6.1 Inagaki, M. 16.4.2.1.1 16.4.2.7.1 16.4.2.7.2 Indriksons, A. 15.2.4.4. Inkson, R. 15.2.14.5.3 Ino, T. 16.4.2.6.1 Inone, M. 15.2.14.1

hose, J. 15.2.4.1.3 Inoue, S. 15.2.4.2.2 Interrante, L. V 16.4.2.1.1 Iqbal, Z . 15.2.12.4 Isackson, F. J. 15.2.8.3 Isenberg, W. 15.2.12.9 Ishikawa, M. 15.2.4.1.2 15.2.4.2.1 15.2.4.3 Iskander, B. 16.4.2.1.4 Iskenderor, M. 15.2.13.4 Islamor, T. Kh. 15.2.8.2 Ismailov, B. A. 15.2.9.2 15 2.9.3 Iwamore, S . 15.2.13.2.1

J

Jaboric, M. 15.2.13.2.2 Jackson, K. F. 15.2.7.2 Jacktiess, W. 15.2.5.1.1 Jacobson, A. J. 16.4.3.4 Jacobson, H. 15.1.3 Jager, H. 16.4.2.7.3 Jambaya, L. M. 15.2.13.5 Jameson, R. F. 15.2.10.1 Jannakoudakis, A. D. 16.4.2.5.3 16.4.2.7.3 Janvier, P. 16.4.8.7 Jeffrey, G. A. 16.2.2 16.2.2.1 16.2.2.3 Jegoudez, J. 16.4.2.1.5 16.4.2.3 Jellinek, F. 16.4.3.1

16.4.3.2 Jemmis, E. D. 15.2.3.5 Jensen, D. 15.2.1 Jerome, D. 16.4.3.2 Jobert, A. 16.4.2.8.3 Joesten, B. L 15.2.7.2 Johannson, 0. K. 15.2.8 15.2.8.1 15.2.8.4 Johns, I. B. 15.2.14.1 Johnson 11, A. D. 15.2.5.3 Johnson, W. D. 16.4.2.2.1 Johnston, D. C. 16.4.5 Jolly, W. L. 15.2.12.2 15.2.12.5 Jonck, P. 16.4.2.7.2 Jones, M. E. B. 15.2.14.1 Jones, P. G. 15.2.12.9 Jones, R. 15.2.4.4. Jost, F. 16.4.2.1.1 Jost, K. 15.2.10.4 Jotter, R. 15.2.12.10 Judd, M. L. 15.2.14.1 Junge, H. 16.4.2.8.3 Juza, R. 16.4.2.7.2 16.4.2.7.3

K

Kabe, Y. 15.2.4.4. Kageyame, M. 15.2.4.2.2 15.2.4.1.3 Kaim, W. 15.2.3.5 Kain, I. 16.4.2.3 Kaiser, W. 16.4.2.8.3

362 Kajlwara, M. 15.2.14.1 Kaldor, A. 16.4.1, Kalliney, S. Y. 15.2.10.2 Kalnin, I. L. 16.4.2.7.3 Kamarchik P. 16.4.2.6.1 16.4.2.6.3 Kameda, I. 16.4.2.6.1 Kamimura, H. 16.4.1. Kan, P 15.2.13.2.1 Kanamaru, F. 16.4.4 16.4.5 Kanda, S. 15.2.14.1 Kaner, R. B. 16.4.1. Kang, S.-K. 15.2.3.1 Kantor, S. W. 15.2.8 Kao, H. I, 15.2.12.3 Karagounis, G. 15 2.8.2 Karatsu, T. 15.2.4.5. Karl-Kroupa, E. 15.1.3.3 Karunanithy, S. 16.4.2.1.2 Kasper, J. S. 16.2.1 16.2.3 Kato, H. 15.2.12.2 Kato, M. 15.2.4.2.3 Katti, A. 15.2.4.1.1 15.2.4.1.2 15.2.4.1.3 15.2.4.3 Katyshkina, V. V 15.2.3.3 Katz, J. R. 16.4.2.5.3 Kawakami, J. H. 15.2.7.2 Kawamura, T. 16.4.2.6.1 Kazakova, V. V. 15.2.9.2

Author Index Kaz’mina, T. K. 16.4.2.7.2 Kbis Ariguib, N. 15.2.10.2 Keenan, A. G. 16.4.2.7.3 Keezer, R. C . 15.2.2.2.1 Kemp, W. 16.4.2.7.3 Kendrick, A. G. 15.2.12.3 Kennett, F. A. 15.2.12.3 Kertesz, M. 15.2.3.5 Kesting, R. E. 15.2.7.2 Kewcloh, N. 15.2.9.1 Khairetdinov, E. F. 16.4.2.6.1 Khalil, M. H. 15.2.12.6 Khananashvili L. M. 15.2.9.5 Khodabocus, M. 15.2.5.4 Khusidman, M. B. 16.4.2.1.2 Kikkawa, S. 16.4.5 Kilkha, A. 15.2.13.4 Killough, J. M. 16.4.2.8.2 Kim, H. K. 15.2.4.5. Kim, Y. K. 15.2.13.2.2 King, M. 15.2.14.3 Kinoshita, K. 16.4.2.7.3 Kinsinger, J. B. 15.1.3 Kirkpatrick, R. J. 15.2.6.1.1 Kischkel, H. 15.2.3.4 Kiselev, V. F. 16.4.2.7.3 Kishita, M. 15.2.14.1 Kispert, L. D. 16.4.2.1.5 Kita, Y. 16.4.2.6.1 16.4.2.6.2 Kitahara, T. 15.2.4.1.2

Klages, F. 15.2.3.4 Klebanskii, A. L. 15.2.7.2 15.2.11.2 15.2.13.3.3 Klein, H. F. 16.4.2.3 16.4.2.2.1 16.4.2.7.1 16.4.2.7.2 Klein, J. 16.4.1. Klein, R. M. 15.2.14.1 Klemm, W. 16.2.3.1 Klemperei, W. G. 15.2.8.2 Kleppinger, J. 15.2.12.3 15.2.12.4 Kleps, R. A. 15.2.10 15.2.10.2 15.2.10.4 Kliebisch, U. 15.2.9.1 15.2.9.2 Klingebiel, U. 15.2.8 15.2.9.1 15.2.9.2 Klingen, W. 16.4.4 Klingenbiel, U. 15.2.9.1 Kloosterboer, H. J. 16.4.3.2 16.4.3.5 Klosowski, J. M. 15.2.8 15.2.8.4 15.2.8.5 15.2.13.3.4 Kloth, B. 15.2.3.2 Klusmann, E. B. 15.2.7.2 Kluver, H. 15.2.12.6 Knappwost, A. 16.4.2.1.1 Knight, A. 15.2.14.5.3 Knoll, F. 15.2.3.3 Knollmueller, K. 0. 15.2.7.2 Knowles, D. J. 15.2.6.2

363

Author Index Knozinyer, H. 16.4.2.7.3 Kobayashi, M. 15.2.4.1.2 Koch, B. 15.2.3.2 Koch, D. 15.2.3.2 Koenig, P. E. 15.2.13.3.2 Koenig, P. F. 15.2.13.3.6 Kohlschiitter, V. 16.4.2.5.2 Kohne, R. 15.2.8.2 Koike, H. 15.2.4.4. Koizumi, M. 16.4.4 16.4.5 Kojima, N. 15.2.4.4. Kolesnikova, I. P. 16.4.2.5.3 Kollman, G. 15.2.4.2.2 Komarova, L. G. 15.2.5.5.2 Komarova, L. I. 15.2.7 Komatsubara, T. 16.4.8.3 Komm, R. 15.2.5.5.1 Konigsberger, D. C. 15.2.2.2.5 15.2.2.2.6 Kononov, A. M. 15.2.9.2 15.2.9.3 Kopf, P. W. 15.2.7.2 Kopylov, V. M. 15.2.9.3 Korkish, J. 15.2.14.5.3 Korol’ko, V. V. 15.2.7.2 Korshak, V. V. 15.2.5.5.2 15.2.7 15.2.7.3 15.2.13.2.3 15.2.14.1 15.2.14.5.4 Kosydar, K. M. 15.2.11.2 Kotake, H. 15.2.4.3

Kotloby, A. 15.2.7.2 Kotrelev, G. V. 15.2.9.2 15.2.9.3 Kottke, T. 15.2.9.2 Kougo, Y. 15.2.4.1.3 15.2.4.2.3 Kouvetakis, J. 16.4.1. Kovar, D. 15.2.4.1.1 15.2.4.1.2 15.2.4.2.1 Kovar, R. A. 15.2.5.1.1 Kovol, E. N. 15.2.9.4 Koyama, S. 16.4.2.6.1 Kraft, M. Ya. 15.2.3.3 Krakaner, H. 16.4.8.3 Krampe, C. 15.2.9.1 Krannick, L. K. 15.2.3.3 Krantz, K. W. 15.2.13.3.4 Krapchev, T. 16.4.2.2.1 Krasovskaya, T. A. 15.2.7.2 Krauss, C. A. 16.2.2.3 Krebs, H. 15.2.2.2.1 Krebs, K. 15.2.1 Kriegsmann, H. 15.2.9.4 Krishnan, V. 15.2.2.2.11 Krivonishchenkor, V. V. 15.2.13.3.2 Krohn, H. 16.4.2.8.3 Kroto, H. W. 16.4.1. Kruger, C. R. 15.2.9.3 Krupnova, L. E. 15.2.7.2 Kuballa, M. 15.2.2.2.11 Kubo, K. 15.2.14.1

Kugel, R. L. 15.2.11.1 15.2.11.4 Kullmann R. 15.2.9.4 Kumada, K. 15.2.4.2.1 Kumada, M. 15.2.4.1.2 15.2.4.2.1 15.2.4.3 15.2.4.4. Kunnmann, W. 16.3.2.3.1 Kurashera, N. A. 15.2.13.3.2 Kuriakose, A. K. 16.4.2.6.1 Kuteinikova, L. I. 15.2.13.3.2 Kuwabara, M. 15.2.4.2.3 Kuyper, J. 15.2.12.4 15.2.12.9 Kuzmany, H. 15.2.4.5. Kuznetsov, Ye. V. 15.2.6.1.1 15.2.6.4 Kwasnik, H. 15.2.7.2 Kwiatkowski. G. T. 15.2.7.2 Kyoda, J. 15.2.4.3

L

Laali, K. 16.4.2.1.1 Labes, M. M. 15.2.12.2 15.2.12.3 Labib, H. H. A. 15.2.2.2.13 Lagow, R. J. 15.2.12.3 16.4.2.5.1 16.4.2.5.3 16.4.2.6.1 Lagrange, P. 16.4.2.2.1 16.4.2.2.2 16.4.2.4.1. 16.4.2.4.2 16.4.2.4.3. 16.4.2.4.4. Laidlaw, W. G. 15.2.12.3

364 Lam, D. K. 16.4.2.6.1 Lampe, F. W. 15.1.1.1 Lang, R. 15.2.13.3.6 Langer, J. 15.2.11.1 Langford, E. 15.2.2.2.3 Langguth, R. P. 15.2.10.3 Lanzen, E. 15.2.4.1.4 15.2.4.2.1 Larchan, T. B. 15.2.7.2 Lau, W. M. 15.2.12.3 Laubengayer, A. W 15.2.5.5.2 Lauginie, P. 16.4.2.4.2 Lavin, K. D. 15.2.11.4 Lavrukhin, B. D. 15.2.13.3.2 Lawrenson, I. J. 15.2.5.3 Le Blanc, A. 16.4.4 Le Mehautt, A. 16.4.4 16.4.8.3 Lebedev, E. P. 15.2.9.4 Leblanc, A. 16.4.3.2 Lee, C. L. 15.2.8 15.2.8.1 15.2.8.4 Lee, G. H. 15.2.6.2 Lee, L. H. 15.2.13.5 Legendre, J. J. 16.4.3.1 Legrand, A. P. 16.4.2.2.1 16.4.2.3 Leith, R. M. A. 16.4.2.7.2 Lelaurain, M. 16.4.1. 16.4.2.1.1 16.4.2.8.1 Lelaurin, M. 16.4.2.1.5 Lemanski, M. F. 15.2.4.1.1

Author Index 15.2.4.2.2 Lenk, C. 15.2.13.2.1 Lenz, R. W. 15.2.13.3.3 Leonta, D. 15.2.14.1 Leonyuk, N. I. 15.2.6.1.1 Lerf, A. 16.4.3.2 16.4.3.4 16.4.2.7.2 16.4.2.8.1 Letoffe, J. M. 16.4.2.3 Levin, V. Y. 15.2.13.3.4 Levitskii, M. M. 15.2.8.2 Levy, F. 16.4.2.7.2 Levy, R. M. 15.1.3 15.1.3.2.2 15.1.3.3.3 16.4.3.4 Levy-Clement, C. 16.4.8.3 Lewis, J. E. 15.2.3.3 Liang, W. Y. 16.4.3.1 16.4.3.4 Liblong, S . W. 15.2.12.3 Lidy, W. 15.2.12.9 Lienhard, K. 15.2.9.2 Liepins, R. 15.2.5.5.1 Lin, C.-H. 16.4.2.1.1 Lin, T.-P. 15.2.12.6 Lin, Y. 15.2.8.2 Linville, R. 15.2.13.2.1 15.2.13.3.5 15.2.13.5 Lipscomb, W. N. 15.2.7.1 Lisitsa, V. V. 16.4.2.5.3 Litscher, J. 15.2.4.1.1 Litteral, C. J. 15.2.8

Little, W. A. 16.4.3.4 Liu, Y. 16.4.1. Lloyd, D. J. 15.2.3.5 Lloyd, N. C. 15.2.8.2 Loebenstein, W. V. 16.2.2.3 Lopez, I. 15.2.13.2.1 15.2.13.5 15.2.14.5.2 Lorimer, J. W. 15.2.13.3.2 Lorriaux-Rubbens. A. 16.4.2.1.4 Lotmar, W. 15.2.11.1 Louisy, A. 16.4.4 Love, P. 15.2.12.3 Lowde, D. R. 16.4.2.5.3 Lucovsky, G. 16.4.3.1 Ludwig, Jr., E. G. 15.2.3.1 Lueders, K. 16.4.2.1.1 Lund, H. 16.4.2.8.2 Lundberg, B. 16.4.2.1.1 Lundberg, R. D. 15.2.14.3 Lundin, R. E. 15.2.9.5 Lunera, L. K. 15.2.13.2.3 15.2.14.1 Liipschen, R. 15.2.9.4 Luski, S. 16.4.2.1.1

M

Maaroufi, A. 16.4.2.2.1 Maass, W. 15.2.14.5.3 MacDiarmid, A. G. 15.2.12.2 15.2.12.3 15.2.12.4 MacDonald, A. L. 15.2.12.7

365

Author Index MacEwan, D. M C 16.4.2.5.3 Mack, D. P. 15.2.11.4 15.2.14.5.2 MacKenzie, N. 16.4.1.2 MacKnight, W. J. 15.2.14.3 MacLean, G. K. 15.2.12.3 Macleod, D. B. 15.2.1 MacLeod,D M. 16.4.1.2 Maeda, S. 15.2.4.1.2 Maeda, Y. 16.4.2.8.2 Magerl, A. 16.4.2.1.4 Maguire, K. D. 15.2.12.5 Mahler, W. 15.2.3.2 Mai, C. 16.4.2.3 Maire, J. 16.4.2.7.2 Majling, J. 15.2.10.2 Makrini, M El 16.4.2.2.1 Malachesky, P. A. 16.4.2.6.2 Malling, J. 15.2.10.3 Mangold, D. J. 15.2.1 Manini, C. 16.4.2.1.1 Marcehe, J. F. 16.4.2.1.5 March, C. S. 15 2.14.1 Marchand, R. 16.4.8.6 Marcu, M. 15.2.14.1 Marcus, B. 16.4.2.8.2 Marcus, J. 16.3.2.3.1 16.3.2.3.2 16.3.2.3.3 Markhe, J. F. 16.4.1. 16.4.2.1.1 16.4.2.2.1 16.4.2.4.4.

Margrave, J. L. 16.4.2.5.1 16.4.2.5.3 16.4.2.6.1 16.4.2.6.3 Maricich, T. J. 15.2.12.6 Marketz, H. 15.2.4.1.1 Markiewicz, R. S. 16.4.2.1.1 Markle, R. A. 15.2.13.2.2 15.2.9.3 Marov, I. 16.4.2.1.5 Marsh, R. E. 16.2.1 16.2.2 Marsmann, H. C. 15.2.3.3 Martin, K. V. 15.2.14.1 15.2.14.5.5 Martin, M. M. 15.2.14.1 Martiu, C. 16.4.2.6.1 Marvel, C. S . 15.2.14.1 15.2.14.5.5 Masamune, S. 15.2.4.1.1 15.2.4.1.3 15.2.4.2.3 15.2.4.4. Mason, J. H. 15.2.13.4 Massen, C. H. 15.2.1 15.2.2.1 15.2.2.2.4 15.2.2.2.5 Mathais, L. 15.2.13.3.5 Mathews, J. F. 16.4.2.8.3 Mathey, Y. 16.4.4 Matsner, M. 15.2.13.3.4 Matsumoto, H. 15.2.4.4. 15.2.4.1.3 Matsumoto, M. 15.2.4.1.3 Matsumura, K. 15.2.4.1.2 15.2.11.4 Matsuo, K. 16.4.2.8.3

Matsuyaki, S. 16.4.2.3 Matsuzawa, T. 15.2.13.2.2 Mattson, C. E. 15.2.14.1 Mattson, J. S. 16 4.2.7.3 Matty, L. 16.4.2.1.3. 16.4.2.1.5 Matula, D. W. 15.1.3 15.1.3.3 15.1.3.2.1 15.1.3.2.2 15.1.3.3.2 15.1.3.3.4 15.2.10 1 Matyjaszewski, K. 15.2.4.5. Maude, H.-J. 15.2.1 Maxka, J. 15.1.1.3.4 15.2.4.5. May, J. F. 15.2.12.8 Maya, L. 15.2.6.1.1 Mayerle, J. J. 15.2.12.9 Maves, N. 15.2.7.2 Mavfield. D. L ’ 15.2.9.4 Mayikres, C. 16.4.2.3 Mazdiyasni, K. S. 15.2.9.3 Mazieres, C. 16.4.2.3 16.4.2.1.5 16.4.4 McCandlish, L. E. 16.4.5 McCarron, E. M. 16.4.2.1.1 16.4.2.8.3 McCarron, G. 16.4.2.1.2 McCoy, R. E. 15.2 5.1.2 McCullough, J. F. 15.2.10.1 15.2.10.2 McDoual, G. J. 16.3.1.5 McGee, Jr., H. A. 15.2.5.5.1

366 McGilvery, J. D. 15.2.10.3 McGrath, J. E. 15.2.8.4 McInerney, E. F. 15.2.13.2.1 15.2.13.3.5 15.2.14.1 15.2.14.4 15.2.14.5.4 McIntyre, D. B. 15.2.8.4 McKean, D. R. 15.2.4.5. McKee, D. W. 16.4.2.1.1 McKinley, A. J. 15.2.4.5. McKinnon, W. R. 16.4.8.3 McLean, J. A. 15.2.14.1 McManns, S . 15.2.13.2.1 McManus, S. 15.2.13.2.1 McMullan. R. K. 16.2.2.1 16.2.2.3 McNeely, C. 15.2.13.5 McNeil, D. W. 15.2.7.2 McNicol, B. D. 16.4.2.5.3 16.4.2.7.3 McQuillan, B. W. 16.4.2.6 16.4.2.1.1 16.4.2.1.2 16.4.2.8.3 McRae, E. 16.4.1. 16.4.2.1.1 16.4.2.1.4 16.4.2.1.5 16.4.2.2.1 Meadowcroft, T. R. 15.2.10.1 Medda, P. 15.2.3.2 Meier, H.-U. 15.2.5.1.1 15.2.5.4 Meier, W. 15.2.3.5 Meinhold, W. 16.2.2.2 Meister, J. J. 15.2.11.3

Author Index Meites, L. 16.3.2.2.1 16.3.2.2.2 16.3.2.3.2 16.3.2.3.3 Mele, A. 15.2.11.1 Melin, J. 16.4.2.1.1 16.4.2.8.3 Meller, A. 15.2.5.1.1 15.2.5.5.2 15.2.5.4 15.2.9.1 15.2.9.2 15.2.13.2.3 Mellies, R. 15.2.9.2 Mel’nik, 0. A. 15.2.7 Memory, J. D. 16.4.2.7.3 Mercando, P. 15.2.3.5 Mering, J. 16.4.2.7.2 Merker, R. L. 15.2.8.2 15.2.13.3.3 Merle, G. 16.4.2.3 Mertway, H. E. 16.4.2.8.3 Meshri, D. T. 16.4.2.6.1 Mesmer, R. E. 15.2.6.1.1 Meston, A. 15.2.13.2.2 Metoki, N. 16.4.2.2.1 Metrot, A. 16.4.2.1.4 16.4.2.2.1 16.4.2.4.2 16.4.2.4.3. 16.4.2.4.4. 16.4.2.8.1 16.4.2.8.3 Metz, W. 16.4.2.1.3. Mews, R. 15.2.12.5 15.2.12.6 15.2.12.9 Meyer, B. 15.2.1 Meyer, H. 16.4.3.3

16.4.5 15.2.12.5 Meyer, K. H. 15.2.11.1 Meyer-Base, K. 15.2.12.9 Meyers, M. B. 15.2.2.2.3 Meyer-Spasche, H. 16.4.2.1.3. Michalczyk, M. J. 15.2.4.2.3 Michel, C. 16.3.1.2 16.3.1.3 Michl. J. 15.2.4.2.3 15.2.4.5. 15.2.4.8 Middlecamp, C. H. 15.2.4.2.2 Middleton, S. 15.2.3.3 Midgley, H. G. 15.2.13.3.1 Migdal, S . 15.2.13.4 Mikhailov, B. M. 15.2.5.5.2 15.2.6.3 15.2.6.4 Mikulski, C. M. 15.2.12.2 15.2.12.3 Millard, M. M. 15.2.9.5 Miller, R. 15.2.4.5. 15.2.13.3.2 Miller, R. D. 15.2.4.5. Miller, R. E. 15.1.3.3.1 Milliken, J. 15.2.12.4 16.4.1. Mills, D. R. 16.4.2.1.1 16.4.2.1.5 Minakov, V. T. 15.2.1.2 Mirkamilova. M. S . 15.2.14.1 Miroshnikova, I. I. 15.2.7.2 Mitsuhashi, T. 16.4.2.1.1 Mitter, F. K. 15.2.4.5. 15.2.4.2.1

367

Author Index Miyamoto, H. 15.2.4.4. Moedritzer, K. 15.1.3.1 15.1.3.3.1 15.1.3.3.2 15.1.3.3.3 15.1.3.3.4 15.2.9.5 Moeller, T. 15.2.12.6 Moffitt, R. B. 15.2.7.2 Mohler, R. L. 16.3.2.3.2 Mohlwald, H. 16.4.2.1.1 Mohwald, H. 16.4.2.5.2 16.4.2.7.1 16.4.2.7.2 16.4.2.8.1 16.4.2.8.2 16.4.2.8.3 Molher, R. L. 16.3.2.3.3 Molinie, P. 16.4.8.2 Molloy, H. M. 15.2.13.5 Monroe, R. F. 15.2.13.3.5 Monteil, Y. 15.2.12.2 15.2.12.5 Mooney, E. F. 15.2.5 2 15.2.5.5.1 15.2.5.5.2 Moore, A. W. 16.4.1. 16.4.2.6 Moore, G. Y. 15.2.11.2 15.2.11.4 Mooser, E. 16.4.6 Mordkovich, V. Z. 16.4.2.2.1 Morgan, G. T. 15.2.14.4 Morgan, W. E. 15.2.10.1 15.2.10.2 Morgunova, M M. 15.2.9.3 Morley, J. G. 16.4.2.7.3 Moroni, R. 15.2.13.2.2

Morrillo, A. 15.2.3.1 Mourald, H. 16.4.2.3 Mudgett, M. 15.2.10.1 15.2.10.2 Muetterties, E. L. 15.1.1.3.2 15.2.5.1.2 15.2.9.4 16.4.2.1.2 Mukherjee, S . P. 15.2.9.3 Miillen, H. R. 16.2.2.1 Miiller, D. 15.2.3.1 Muller, H. 15.2.9.4 Miiller, H. 15.1.3.3.1 Miiller, H. R. 16.2.2 16.2.2.2 Muller, M. 16.4.2.1.1 Muller, R. 15.2.8.2 Muller-Schiedmayer, G. 15.2.3.2 Miiller-Warmuth, W. 16.4.3.5 Munch, V. 16.4.2.1.1 16.4.2.1.2 Miinchen T. H. 16.4.2.8.3 Mundt, 0. 15.2.3.4 Murakami, S. 15.2.4.1.1 15.2.4.1.3 15.2.4.2.3 Muraoka, T. 15.2.4.1.3 15.2.4.2.2 Murphy, D. W. 16.4.3.2 Murphy, K. E. 15.2.1 Murray, R. W. 16.4.2.7.3 Musgrave, T. R. 15.2.14.1 Myakishev, K. G. 15.2.5.1.1 Myer, G. H. 15.2.12.3

Myers, M. B. 15.2.2.2.8 Myers, T. C. 15.2.10 15.2.10.1 15.2.10.2 15.2.10.3 15.2.10.4 Mykytiuk, P. 15.2.13.5

N

Nadi, N. E. 16.4.2.1.1 Nagai, N. 15.2.4.4. Nagai, Y. 15.2.4.1.2 15.2.4.1.3 15.2.4.2.3 15.2.4.4. Nahlovsky, B. D. 15.2.13.3.3 Nakajima, T. 16.4.2.5.2 16.4.2.6.1 Nakane, K. 16.4.2.6.1 Nametkin, N. S. 15.2.8.2 15.2.13.2.2 Naoi, Y. 15.2.4.4. Naoshima, Y. 15.2.13.2.1 15.2.13.5 Narula, C. K. 15.2.5.5.2 Nazarov, A. S. 16.4.2.5.3 16.4.2.6.1 16.4.2.6.3 Neemann, J. 15.2.9.1 Neilson, R. H. 15.1.1.3.4 15.2.1 1.3 15.2.1 1.4 Nelmec, P. J. 15.2.6.1.1 Neureuther, A. 15.2.13.3.2 Neuse, E. 15.2.13.2.1 Newman, J. M. 15.2.7.2 Newman W. P. 15.1.1.3.1 15.1.1.3.3

368 Nichols, L. F. 15.2.12.3 Nickl, J. J. 16.4.2.1.1 16.4.2.5.2 16.4.2.7.1 16.4.2.7.2 16.4.2.8.1 16.4.2.8.2 16 4.2.8.3 Nickless, G. 15.1.3.3 Niedenzu, K. 15.2 5.1.1 15.2.5.1.2 Nielson, R. H 15 1.1.3.1 15.1.1.3.4 15.2.5.5.1 15.2.11.4 Nikonorov, Y. I. 16.4.2.6.1 Nishina, Y. 16.4.1. Nishitani, N. 16.4.2.2.1 Nishitani, R 16.4.1. Nitsche, R. 16.4.4 Nixon, D. E. 16.4.2.2.1 Nixon, J. F. 15.2.3.2 Noda, K. 15.2.12.2 Nogai, Y 15.2.4.2.2 Noll, W. 15.2.8 15.2.8.1 15.2.8.4 Noltemeyer, M. 15.2.9.1 15.2.9.2 15.2.12.9 Noltes, J. 15.2.13.2.3 Nomine, M. 16.4.2.3 Norman, N. C. 15.2.3.1 Norval, S. 15 1.3.3.1 15.1.3.3.2 15.2.10.1 Noth, H. 15.1.1.3.1 Noth, J. 15.2.5.1.1

Author Index Novikov, Yu. N 16.4.2.7.2 16.4.2.8.2 Nowogrocki, G. 16.4.8.6 Nweke, S . 0. 15.2.10.4 Nyholm, R. 15.2.1

0

Oakley, R. T. 15.2.12.2 15.2.12.3 15.2.12.9 Obenland, C. B. 15.2.7.2 Obenland, C. 0. 15.2.7.3 O’Brian, J. P. 15.2.7.3 O’Brien, E. L. 15.2.13.2.2 15.2.7.2 O’Brien, J. P. 15.2.11.2 15.2.11.4 15.2.14.5.2 O’Brien, S.C. 16.4.1 Oddous, J. 16.4.2.7.3 Offeman, R. E. 16.4.2.5.2 Offermann, W. 15.2.3.4 Ohana, I. 16.4.2.1.1 Ohashi, S. 15.1.3.3 15.2.10.3 15.2.10.4 Ohhashi, K. 16.4.2.1.2 Ohlerich, G. 16.4.2.7.3 Okamura, K. 15.2.13.2.2 Okamuro, K. 15.2.4.5. Okawa, T. 15.2.4.2.3 Oleksyszyn, J. 15.2.3.1 Olsson, K. 15.2.8.2 Omboo, W. 16.4.3.2 Omori, M. 15.2 4.5.

Onishi, Y. 15.2.4.5. Onuki, Y. 16.4.8.3 Oommen, T. V. 15.2.1 Opalovskii, A. A. 16.4.2.6.1 Osiecki, J. H. 16.4.3.4 Osterheld, R. K. 15.2.10.3 Ostertag, W. 16.4.2.3 Otani, S. 16.4.4 Ouchi, A. 15.2.12.6 Oufkir, A. 16.4.2.4.2 16.4.2.4.3. Ouvrard, G. 16.4.3.2 16.4.4 16.4.8.4 Owen, W. J. 15.2.8.2 Owens, D. A 15.2.7.2

P

Paetzold, P. 15.2.5.1.1 15.2.5.4 Paire, R. T. 15.2.5.5.2 Pakulski, M. 15.2.3.1 Palin, D. E. 16.2.1 Pall, D. B. 16.2.2.3 Palmer, M. H. 15.2.12.3 Pal’nichenko, A. V. 16.4.2.2.1 Palvadeau. P. 16.4.5 16.4.8.1 16 4.8 7 Panayotov, I. M. 16.4.2.3 Panckhurst, D. J. 15.2.9.4 15.2.9.6 Pankow, G. W. 15.2.11.1 Panzer, R. E. 16.4.2.7.3

369

Author Index Papetti, S 15.2.7.2 15.2.7.3 15.2.13.2.2 Parks, G. D. 16.4.2.6.1 16.4 2.6.3 Parry, G. S . 16.4.2.1.4 16.4.2.2.1 16.4.2.5.2 16.4.2.8.1 16.4.2.8 3 16.4.3.4 Parry, R. W. 15.2.5.1.2 Parsonage. J. R. 15.2.13.3.1 Parthasarathy, A. 15.2.1 Pascal, P. 15.2.1 Passmore, J. 15.2.12.3 15.2.12.5 Patterson, D. B 15.2.11.1 15.2.11.2 15.2.11.3 15.2.11.4 Patterson, W. 15.2.13.2.1 Patzelt, H. 15.2.13.3.2 Pauer, F. 15.2.9.2 Paul, V. 15.2.9.2 Pauling, L. 16.2.1 16.2.2 Paushkin, Y. 15.2.13.2.1 Payne, J. H. 15.2.10.1 Pazdernik, L. J. 15.2.9.5 Pazdernik; L. J. 15.2.9.5 Pearce, C. A. 15.2.9.2 15.2.13.2.2 Pearson, J. 16.4.2.1.5 Pearson, W. B. 16.4.6 Peka, I. 16.4.2.6.1 Pelizzi, G. 15.2.3.1

Pendrys, L. A. 16.4.1. Pentenrieder, R. 16.4.2.6 Perlstein, J. H. 15.2.12.2 Person, W. B. 16.4.2.1 5 Pesce, B. 15.1.3.3 Peter, W. 15.2.4.2.1 Peters, E. N. 15.2.7 15.2.7.2 Peters, K. 15.2.4.1.3 Peterson, D. J. 15.2.4.1.1 Peterson, G. 15.2.13.2.1 Peterson, L. K. 15.2.3.3 Peterson, W. 15.2.4.1.2 15.2.13.2.2 Petrzila, V. 16.4.2.6.1 Pettit, R. 15.2.3.5 Pfisterer, H. 15.2.3.5 Pfluger, P. 16.4.2.2.1 16.4.2.8.2 Philipp, A. 16.4.2.1.1 Phillip, G. 15.2.13.3.2 Piccoli, W. A. 15.2.8.2 Piekos, R. 15.2.9.4 Pierce, 0. R. 15.2.13.2.2 15.2.8.4 Pieters, R. 15.2.12.9 Pietronero, L. 16.4.2.1.4 Piffard, Y. 16.3.1.4 Pines, A. N. 15.2.8.4 Pisharody, R. 16.4.3.4 Pittman, C . 15.2.13.2.1 15.2.13.2.3 15.2.13.3.5

15.2.13.1 15.2.13.4 Plass, R. 16.4.2.5.1 16.4.2.5.3 Platteeuw, J. C. 16.2.2.2 Platzer, A. 16.4.2.2.1 Plekhanova, K. 15.2.13.4 Plet, F. 16.3.1.2 Pleysier, J. 16.4.7.2 Plotho, C. von 15.2.5.1.1 15.2.5.4 Poggi, G. 15.2.4.5. Pohl, S. 15.2.4.2.3 Poli, R. 15.2 3.1 Pollard, F. H. 15.1.3.3 Porter, E. F. 15.2.6.2 15.2.6.4 Porter, R. F. 15.2.6.2 Portier, J. 16.4.5 Postenakk, M. 16.4.8.3 Pouchard, M. 16.2.1 16.2.3 16.2.3.1 Poulet, R. J. 15.2.12.5 Poulis, J. A. 15.2.1 15.2.2.1 15.2.2.2.4 15.2.2.2.5 Powell, H. M. 16.2.1 Pratt, D. E. 15.2.5.2 Prescott, P. I. 15.2.8.1 Prevedorou-Demas, C. 15.2.13.3.2 Prigozhina, M. P. 15.2.7.3 Pritzlaff, B. 16.4.2.1.1 Pron, A. 16.4.2.1.1

370 16.4.2.8.3 Prons, V. N. 15.2.11.2 Pukhov, A. A. 15.2.5.1.1 Puri, B. R. 16.4.2.7.1 16.4.2.7.3 Pusatcioglu, S. Y. 15.2.5.5.1 Py, M. A. 16.4.8.2

Q

Quimby, 0. T. 15.2.10.3 Quinton, M. F 16.4.2.3

R

Rabet, F. 15.2.9.2 Rabinovitz, M. 16.4.2.1.1 Rabolt, J. F. 15.2.4.5. Raghuveer, K. S. 15.2.3.1 Ramakrishnan, S . 15.2.12.10 Ramamoorthy, B. 15.2.3.5 Ramirez Garcia, A. 16.4.2.5.3 Ramsdell, L. 16.4.3.1 Randin, J. P. 16.3.2.2.2 16.3.2.3.3 Randin, J.-P. 16.4.2.7.3 Rao, B. M. L. 16.4.2.6.2 Rao, M. N. S . 15.2.12.3 R a m , W. 15.2.3.4 Rashkov, I. B. 16.4.2.3 Rattay, W. 15.2.6.4 Rauscher, W. 15.2.13.2.1 Raveau, B. 16.3.1.2 16.3.1.3 16.4.8.6 Ray, P. 15.2.14.1

Author Index Rebbah, H. 16.4.8.6 Reddy, A. C. 15.2.3.5 Redl, G. 15.2.9.3 Redmond, J. P. 16.4.2.7.3 Reichmann, K. C. 16.4.1. Reids, A. F. 16.3.2.3.1 Reids, A. R. 16.3.2.3.3 Reikhsfel’d, V. 0. 15.2.9.4 Reimer, J. T. 15.2.13.4 Reimer, M. 16.4.2.1.3. Reiner, J. R. 15.2.7.3 Reitzer, C. 16.4.2.7.3 Remy, F. G. 15.2.10.4 Resing, H. A. 16.4.1. Reyerson, L. H. 16.4.2.7.2 Reynolds, J. 15.2.13.2.3 15.2.13.1 Rheingold, A. L. 15.1.1.1 15.2.3.1 15.2.3.3 15.2.3.4 15.2.3.5 Rice, M. J. 16.4.2.1.4 Rice, R. G. 15.1.1.3.1 15.2.11.2 Rice, R. W. 15.2.7.2 Richardson, F. D. 15.2.10.1 Richardson, J. F. 15.2.12.9 Richmond, J. R. 15.2.9.4 Richter, A. 15.2.5.1.1 Riding, G. H. 15.2.11.4 Rieckel, C. 16.4.3.4 Riekel, C. 16.4.3.4

Riess, G. 16.4.2.7.3 Rieter, P. C. U. 15.2.2.2.6 Riffle, J. S. 15.2.8.4 Rigamonti, A. 16.4.3.2 Rikhter, L. Ya. 15.2.5.2 Ring, E. M. 15.2.2.2.2 Ritchie, R. J. 15.2.11.2 15.2.11.4 Ritsma, J. 16.4.3.2 Ritter, W. 16.4.7.2 Robert, C. 16.3.1.3 Robertson, A. S. 16.4.2.1.1 16.4.2.1.2 16.4.2.8.3 Robinson, J. 16.4.2.1.5 Rochow, E. G. 15.2.9.2 15.2.9.3 15.2.13.2.3 15.2.13.3.3 Rode, V. V. 15.2.14.1 Roder, U. 16.4.3.5 Roesky, H. W. 15.2.12.2 15.2.12.9 Rogers, D. B. 16.3.2.3.1 Rohwer, E. F. C. H. 16.3.1.5 Roland, A. 15.2.12.6 Roller, M. B. 15.2.7 Roozeboom, H. W. B. 16.2.2.1 Rose, S. H. 15.2.11.4 Rosenberg, H. 15.2.13.2.1 15.2.13.3.3 Rosenman, I. 16.4.2.1.4 Rossetto, F. 15.2.13.2.3 15.2.7.3 Roth, A. 15.2.3.2

37 1

Author Index Roth, C. A. 15.2.8.2 Roth, G. 16.4.2.1.1 Roth, S. 16.4.2.1.1 Rothgery, E. F. 15.2.5.1.1 Rotinjanz, L. Z. 15.2.1 Rottig, G. 15.2.9.4 Rouillon, J. C. 16.4.2.7.1 16.4.2.7.2 Rousseaux, F. 16.4.2.2.2 16.4.2.4.2 16.4.2.4.3. 16.4.2.8.1 Roussel, M. 16.4.2.6.1 Rousset, S. 16.4.1. Rouxel, J. 16.4.3.2 16.4.3.4 16.4.3.6 16.4.4 16.4.5 16.4.8.1 16.4.8.2 16.4.8.7 Rozmyslova, A. 15.2.13.2.1 Ruano Casero, R. J. 16.4.2.5.3 Rubisch, D. 16.4.2.3 Rubisch, 0. 16.4.2.8.2 Rudner, B. 15.2.5.1.1 Rudorff, G. 16.4.2.6.1 Rudorff, W. 16.4.2.1.2 16.4.2.1.4 16.4.2.2.1 16.4.2.3 16.4.2.5.3 16.4.2.6.1 16.4.2.6.3 16.4.2.8.3 16.4.3.2 Ruess, G. 16.4.2.5.1 16.4.2.5.3 Ruff, 0. 16.4.2.6.1

16.4.2.8.3 Ruisinger, B. 16.4.2.1.4 16.4.2.8.3 Rukhadze, E. G. 15.2.14.1 Rumba, G. Ya. 15.2.9.2 15.2.9.3 RUPP, L. 16.4.8.5 Russo, P. J. 15.2.12.2 Ruthardt, R. 16.4.3.4 Ryan, J. W. 15.2.8.2 Rzaigui, M. 15.2.10.2

S

Sacks, W. 16.4.1. Saehr, D. 16.4.2.2.1 16.4.2.4.1. Safran, S . 16.4.2.2.1 Sairo, H. 15.2.14.1 Saito, M. 16.4.2.7.2 Sakamoto, A. 15.2.4.1.3 Sakamoto, K. 15.2.4.3 Sakamoto, S . 15.2.4.1.2 Sakharova, A. A. 15.2.7 Sakharova, L. N. 15.2.7.2 Salem, J. R. 16.4.3.4 Salzano, F. J. 16.4.2.2.1 Samuel, E. 15.2.4.5. Samyannikov, P. P. 16.4.2.6.3 Sanchez, M. G. 15.1.1.3.1 15.2.11.2 Sano, M. 16.4.2.3 Sanz, F. 15.2.3.3 Saraceno, R. A. 15.2.11.4

Saran, M. S. 15.2.12.2 15.2.12.3 Sarishvili. I. G 15.2.7 15.2.7.3 Sartori, G. 15.2.8.2 Sasaki, Y. 16.4.1. Satka, S . 15.2.13.3.2 Sattler, M. L. 16.4.1. Saunders, G. A. 16.4.2.7.2 Sauter, R. F. 16.4.2.5.3 16.4.2.7.3 Savost’yanova, N. A. 16.4.2.7.2 Saylor, J. C. 15.2.8.4 Scanlon, J. 16.4.5 Scanlon, J. C. 16.4.1. 16.4.2.1.1 16.4.2.1.3. 16.4.2.1.4 16.4.2.1.5 16.4.3.4 16.4.5 Schaaf, R. 15.2.13.2.1 Schaeffer, B. B. 15.2.7.2 15.2.13.2.2 Schaeffer, R. 15.2.5.1.1 15.2.5.1.2 15.2.5.2 15.2.5.3 15.2.5.4 15.2.5.5.2 Schafer, A. 15.2.4.1.3 15.2.4.2.3 Schafer, H. 16.2.1 16.2.3.2 Schafer, R. 16.2.3.1 Schafer-Stahl, 16.4.5 Schaffling, 0. G. 15.2.7.2 Scharff, P. 16.4.2.1.1 16.4.2.8.3

372 Schats, J. 15.2.13.1 15.2.13.2.3 Scheibler, C. 16.3.2.3.3 Schemilt, J. M. 16.4.3.3 Schenk, P. W. 15.2.1 15.2.12.8 Scherer, 0. J. 15.2.3.5 15.2.12.10 Schlegel, R. 15.2.5.5.2 Schleich, D. M. 16.4.4 Schlogl, R. 16.4.2.6 Schmack, L. 15.2.13.3.6 Schmeckenbecher, A. 16.4.2.7.2 Schmeisser, M. 15.2.9.4 Schmid, G . 15.2.5.1.1 15.2.5.4 Schmidbam, H. 15.2.13.2.3 Schmidt, D. 15.2.9.2 Schmidt, D. L. 15.2.13.3.5 Schmidt, H. 15.2.13.3.2 Schmidt, M. 15.2.1 15.2.13.3.2 Schmidt, U. 15.2.3.2 Schmitz-DuMont, 0. 15.2.11.1 15.2.11.2 Schmutz, J. L. 15.2.11.2 15.2.11.4 Schmutzler, R. W. 15.2.2.2.12 15.2.2.2.14 Schneider, G. M. 15.2.2.2.11 Schneider. N. S. 15.2.11.4 Schoff, C. K. 15.2.7 Schogl, R. 16.4.2.1.1 Schollhorn, R. 16.4.3.2

Author Index 16.4.3.3 16.4.3.4 16.4.3.5 16.4.8.1 Scholz, W. 16.4.2.5.1 16.4.2.5.2 16.4.2.5.3 16.4.2.8.1 Scholze, H. 15.2.13.3.2 Schragle, W. 15.1.1.3.1 Schramm, E. P. 15.2.4.1.1 15.2.4.2.2 Schreurs, J. 16.4.2.7.3 Schriver, M. J. 15.2.12.5 Schroder, E. 15.2.5.1.1 Schroder, H. 15.2.12.5 Schroeder, B. B. 15.2.7.2 Schroeder, H. A. 15.2.7 15.2.7.2 15.2.7.3 15.2.13.2.2 15.2.13.2.3 Schroeder, H. J. A. 15.2.7.2 Schroeder, J. 15.2.13.5 15.2.14.5.2 15.2.14.5.3 Schroeder, J. A. 15.2.13.5 Schroer, R. 15.2.3.2 Schiilke. U. 15.2.10.2 Schulze, E. 16.4.2.2.1 16.4.2.3 Schulze, I. 15.2.10.3 Schumann, H. 15.2.3.2 Schumb, W. C. 15.2.9.4 Schurmans. H. 15.2.13.1 15.2.13.2.3 Schuster, H. G . 15.2.4.2.1 Schwalb, J. 15.2.3.5

Schwark, J. M. 15.1.1.3.4 Schwarz, M. 15.2.13.5 Schwarz, R. 15.2.13.5 Schwarzmann, E. 15.2.10.1 Schwebke, G. L. 15.2.4.1.1 Scopelianos, A. G. 15.2.7.3 15.2.11.2 15.2.11.4 Scort, M. J. 15.2.13.3.3 Scott, A. E. 15.2.10.3 Scott, J. D. 16.4.2.7.3 Scott, R. N. 15.2.7.2 Scott, w. 15.2.14.5.2 Scruby, C. B. 16.4.3.4 See, R. F. 15.2.3.1 Seeger, K. 16.4.2.1.1 Seel, F. 15.1.1.3.1 15.2.11.1 15.2.12.6 Segransan, P. 16.4.8.2 16.4.8.3 Seiler, B. C. 16.4.2.5.3 Selig, H. 16.4.2.1.1 16.4.2.1.2 16.4.2.1.5 16.4.2.6 16.4.2.6.1 16.4.2.7.2 16.4.2.8.1 Selwyn, L. S. 16.4.8.3 Semenenko, K. N. 16.4.2.2.1 Semerneva, G. A. 15.2.13.3.2 Semlyen, J. A. 15.2.8.2 Sen, D. K. 15.2.4.2.2 Sergent, M. 16.4.8.1 Setaka, N. 16.4.2.6.1

Author index Setton, R. 16.4.1. 16.4.2.3 16.4.2.1.5 16.4.2.2.1 16.4.2.8.2 Severengiz, T. 15.2.3.4 Seyferth, D. 15.1.1.3.4 15.2.9.3 Shanks, H. R. 16.3.2.3.1 16.3.2.3.2 16.3.2.3.3 Shannon, R. D. 16.3.1.6 Shapatin, A. S. 15.2.7.2 Sharaf. K. A. H. 15.2.2.2.13 Sharikov, 0. V. 16.4.2.2.1 Shaw, S. Y. 15.2.5.5.1 Shawl, E. T. 15.1.1.3.2 Sheats, J. 15.2.3.1 15.2.13.2.1 15.2.13.2.3 15.2.13.3.5 15.2.13.4 Sheikh, S. U. 15.2.5.5.1 Sheldrick, G. M. 15.2.9.1 15.2.9.2 15.2.12.5 15.2.12.9 Shen, C. Y. 15.2.10.2 15.2.10.3 Sherle, A. I. 15.2.14.1 Shibutani, K. 16.4.8.3 Shido, H. 15.2.13.2.1 15.2.13.5 Shimada, M. 16.4.4 Shimp, L. A. 16.4.2.6.1 Shinjo, T. 16.4.2.1.2 Shirai, H. 15.2.14.1 Shishkova, V. 16.4.2.3

Shishkova, V. C. 16.4.2.3 Shkambarnaya, N. I. 15.2.7.2 Sholette, W. P. 15.2.6.2 Shore, S. G. 15.2.5.1.1 15.2.5.1.2 15.2.5.3 15.2.5.4 15.2.5.5.1 Short, R. T. 16.4.2.5.3 16.4.2.7.3 Shreeve, J. M. 15.2.12.5 Sick, E. 16.4.3.4 Sick, H. H. 16.4.2.8.2 Siebert, W. 15.2.1 Siecke, W.-F. 16.4.2.8.3 Sieckhaus, J. F. 15.2.7.2 15.2.7.3 Siegman, D. 15.2.14.5.2 Siegmann, D. 15.2.14.5.2 Silbernagel, B. G. 16.4.3.2 16.4.3.4 16.4.3.4 Silbiger, J. 15.2.9.2 Silverman, M. S. 15.2.11.1 15.2.11.2 Simmler, W. 15.2.8.2 Simon, C. 16.4.2.1.4 Simon, G. 15.1.1.3.1 15.2.12.6 Simonet, J. 16.4.2.8.2 16.4.2.8.3 Sinclair, R. G. 15.2.13.2.2 Singler, R. E. 15.2.11.4 Sirotkina, N. 2. 15.2.6.1.1 Siryatskaya, V. N. 15.2.7.2 Sisler, H. H. 15.1.1.3.1

373 15.2.11.2 Sitzmann, H. 15.2.3.5 Skoda, L. 15.2.9.2 Skowronski. J. M. 16.4.2.1.3. Slabaugh, W. H. 16.4.2.5.3 Sladkov, A. 15.2.13.2.3 Sladkov, A. M. 15.2.14.1 Slaten, B. L. 15.2.1 Sliwinski, S. 15.2.8.2 Slusarczuk, G. M. J. 15.2.8 15.2.8.3 Small, R. W. H. 15.2.12.3 Smaller, B. 16.4.2.1.5 Smalley, R. E. 16.4.1. Smazhok, M. P. 15.2.7.2 Smeltz, K. M. 15.2.11.1 15.2.11.4 Smeltz, L. A. 15.2.11.4 Smit, J. Van R. 16.3.1.5 Smith, A. 15.2.1 Smith. F. D. 15.2.3.3 Smith, H. D. 15.2.7.3 Smith, N. J. 15.2.14.4 Smith, N. R. M. 15.2.12.2 15.2.12.4 Snell, K. D. 16.4.2.7.3 Snow, J. T. 15.2.4.1.1 Sobolevskii, M. V. 15.2.7 15.2.7.3 Solch, J. 15.2.14.5.3 Soled, S. 16.4.4 Solimine, G. 15.2.13.5 Solin, S. A. 16.4.2.2.1

374 16.4.2.2.2 Sollradl, H. P. 15.2.4.2.1 Solomatina, A. I. 15.2.7.3 Sommer, J. 16.4.2.1.1 Sommer, L. H. 15.2.8.2 Sommer, R. 15.2.4.2.2 Soneda, Y. 16.4.2.1.1 Song, J. J. 16.4.2.1.5 Sooriyakumaran, R. 15.2.4.5. Soroos, H. 15.1.3.3 Sotnikov, E. E. 15.2.7.2 Soulen, J. R. 15.2.11.1 15.2.11.2 Sowerby, D. B. 15.2.9.4 15.2.9.5 15.2.9.6 Speiling, L. 15.2.13.5 Spielvogel, D. E. 15.2.8.1 15.2.8.3 Spring, C. 15.2.3.4 Sprung, M. M. 15.2.8.2 Stahl, H. 16.4.2.1.1 Stalke, D. 15.2.9.1 15.2.9.2 Stallings, W. 15.2.4.4. Stang, I. 16.4.2.1.1 Stanislawski, D. A. 15.2.4.3 Staudenmaier, L. 16.4.2.5.2 Staveley, L. A. K. 16.2.1 Steele, B. C. H. 16.4.3.3 Steele, K. 15.2.9.5 Steffen, R. 16.4.8.1 Stein, C. 16.4.2.8.2

Author Index Steinberg, H. 15.2.5.1.1 15.2.6.2 Stelzer; 0. 15.2.3.2 Stern, E. A. 16.4.2.1.5 Stern, R. L. 15.2.13.2.3 Steudel, R. 15.2.1 Steuer, H. 15.2.5.4 Stevie, F. A. 16.4.2.1.2 16.4.2.1.5 16.4.2.6.1 Stevison, D. 15.2.13.2.1 15.2.13.5 Stewart, D. D. 15.2.7.2 Stickney, P. B. 15.2.13.2.2 Stock, A. 15.2.7.1 Stockmayer, W. H. 15.1.3 Stokes, H. N 15.2.11.1 Stolz, s. 16.4.2.2.1 16.4.2.8.2 Stone, F. 15.2.13.1 15.2.13.2.3 Stone, F. G. A. 15.1.1.1 15.2.13.3.6 Storch, W. 15.2.5.1.1 Storey, 0. W. 16.4.2.5.2 Strahl, H. 16.4.2.1.1 Strassler, S. 16.4.2.1.4 Street, G. B. 15.2.12.2 15.2.12.3 15.2.12.4 15.2.12.9 Streifinger, L. 16.4.2.6 Stroh, E. G. 15.2.11.1 Strong, S. L. 16.4.1. Strother, R. 15.2.13.5

15.2.14.5.2 Stroyer-Hansen, T. 15.2.1 Stuart, W. I. 16.2.2.2 Stiiger, H. 15.2.4.2.2 15.2.4.5. Stumpf, W. 15.2.7 Stumpp, E. 16.4.2.1.1 16.4.2.8.3 Sturm, W. 15.2.6.4 Subba Rao, Such, J. E. 15.2.10 Suda, K. 16.4.2.2.1 Suematsu, H. 16.4.2.2.1 Sukegawa, K. 15.2.4.5. Sulleu, J. 16.4.2.6.1 Sullivan, P. J. 15.2.3.5 Sunder, W. A. 16.4.2.1.2 16.4.2.1.5 16.4.2.6.1 Sundermeyer, J. 15.2.12.9 Sundermeyer, W. 15.2.12.6 15.2.12.9 Sundqvist, B. 16.4.2.1.1 Surikova, M. A. 15.2.7 Suszko, P. R. 15.2.11.4 Suvorov, A. L. 15.2.13.3.2 Suzuki, A. 15.2.14.1 Sverdlov, L. M. 15.2.5.2 Swarowsky, H. 15.2.3.5 Sweeton, F. H. 15.2.6.1.1 Swepston, P. N. 15.2.12.2 Sykes, K. W. 16.4.2.7.3 Symon, C. R. 16.4.8.5

Author Index Symons, M. C. R. 16.4.2.1.5 Syrtsova, Zh. S. 15.2.9.3

T

Tagawa, S. 15.2.4.5. Tagliavini, G. 15.2.7.3 15.2.13.2.3 Takoudjou, C. 16.4.2.4.2 Tallant, D. R. 15.2.6.1.1 Tanaguchi, M. 16.4.2.3 Tanaka, K. 15.2.12.2 Tanzella, F. 16.4.2.1.2 Tarascon, J. M. 16.4.8.5 Tarasevich, B. P. 15.2.6.1.1 15.2.6.4 Tate, D. P. 15.2.11.4 Tauc, J. 15.2.2.2.7 Taylor, B. E. 16.3.1.6 16.3.2.3.2 16.3.2.3.3 Taylor, M. 15.2.14.5.3 Tebeneva, N. A. 15.2.9.2 Terent’ev, A. P. 15.2.14.1 Theodoridou, E. 16.4.2.5.2 16.4.2.5.3 16.4.2.7.3 16.4.2.8.1 Thewalt, U. 15.2.12.2 Thielpape, W. 16.4.7.2 Thijssen, T. 15.2.5.1.1 Thilo, E. 15.1.3.3.2 15.2.10.1 15.2.10.2 15.2.10.3 15.2.10.4 Thomas, J. M. 16.4.2.7.3 16.4.7.2

Thompson, A. c. 16.4.2.1.1 16.4.2.1.2 Thompson, A. H. 16.4.4 16.4.5 16.4.8.5 16.4.2.1.1 Thompson, T. E. 16.4.1. 16.4.2.1.1 16.4.2.8.3 Thornton, D. A. 15.2.14.1 Thummler, U. 15.2.1 Tiedemann. W. H. 16.4.2.8.3 Tiernan, T. 15.2.14.5.3 Tisinger, L. 15.2.13.2.1 15.2.13.4 15.2.13.5 Tisinger, W. 15.2.13.4 Titus, J. A. 16.4.2.8.3 Tobita, H. 15.2.4.1.1 15.2.4.1.3 15.2.4.2.3 Tobolsky, A. V. 15.2.2.1 Toeniskoetter, R. H. 15.2.5.2 Tollefson, N. M. 15.2.7.3 Tomasi, R. A. 15.2.4.1.1 15.2.4.1.2 Tomic, E. A. 15.2.14.1 Tomita, A. 16.4.2.5.3 Topsoe, N. Y. 16.4.3.4 Torjman, I. 15.2.10.2 Torosyan, Zh. K. 15.2.7.2 Torre, L. P. 15.2.13.6 Tougain, Ph. 16.4.2.5.2 Tournoux, M. 16.3.1.4 16.4.8.6 Touro, F. J. 15.2.2.2.1

375 Touzain, P. 16.4.2.1.1 16.4.2.1.3. 16.4.2.1.4 16.4.2.3 16.4.2.8.2 16.4.2.8.3 Towlson, H. E. 16.2.2.2 Trefonas, P. 15.2.4.5. Trefonas, R. 15.2.4.5. Trembley, M. 15.2.13.6 Tremmel, W. 15.2.3.5 Tretzel, J. 15.2.13.2.2 Trevor, D. J. 16.4.1. Treylazova, T. I. 16.4.2.7.3 Trichet, L. 16.4.3.2 16.4.3.4 16.4.3.6 16.4.8.2 16.4.8.3 Trillat, J. J. 15.2.1 Trombley, M. 15.2.14.5.2 Trotman-Dickenson, A. F. 15.2.1 Trotter, J. 15.2.12.7 Trsic, M. 15.2.12.3 Trulla, F. F. 15.2.7.2 Tsang, T. 16.4.1. Tsuji, S. 15.2.14.5.3 Tsuzuku, T. 16.4.2.7.2 Tulis, R. W. 15.2.7.2 Tunker, G. 15.2.13.3.2 Turnbull, J. A. 16.4.2.1.5 Turner, C. 15.2.14.5.2 Turner, H. S . 15.2.5.1.1 15.2.5.2 15.2.5.3

376

U

Ubbelohde, A. R. 16.4.2.1.4 16.4.2.5.2 16.4.2.7.2 16.4.2.8.1 16.4.2.8.3 Uden, P. C. 15.1.3.3 Uenishi, M. 15.2.13.5 Ueno, K. 16.4.2.6.1 Ulrich, D. 15.2.13.3.2 Uminskii, A. A 16.4.2.6.1 Umrigar, C. 16.4.8.3 Underhill, C. 16.4.2.2.1 Underwood, G . R. 15.2.3.5 Ungar, P. 15.2.1 Uno, Y. 16.4.2.2.1

V

Vaknin, D. 16.4.2.1.1 Valan, K. J. 15.2.11.1 15.2.11.4 Valerga, A. J. 16.4.2.6.1 16.4.2.6.3 Vallet, G. 15.2.12.8 van Attekum, P. M. 16.4.2.1.1 van den Leeden, P. 15.2.2.2.5 van der Kerk, G . J. M. 15.2.13.2.3 van der Leeden, P. 15.2.1 15.2.2.2.5 van der Meer, R. 16.4.3.2 16.4.3.5 van Ghemen, M. 15.2.3.2 van Heiningen, H. 16.4.3.2 van Heininzen, H. 16.4.3.5 van Ooij, W. J. 15.1.3.3

Author Index van Wazer, J. R 15.1.3 15.1.3.1 15.1.3.2.1 15.1.3.2.2 15.1.3.3 15.2.3.3 15.1.3.3.1 15.1.3.3.2 15.1.3.3.3 15.1.3.3.4 15.2.9.5 15.2.10 15.2.10.1 15.2.10.2 15.2.10.3 15.2.10.4 van Wolput, J. H. M. C. 15.2.2.2.6 Vandi, A. 15.2.12.6 Vangelisti, R. 16.4.2.1.1 Varezhkin, Yu. M. 15.2.9.3 Vasile, M. J. 16.4.2.6.1 Vasil'eva, L. M. 15.2.7.2 Vasilie, M. J. 16.4.2.1.2 16.4.2.1.5 Vasse, R. 16.4.2.8.3 Vast, P. 16.4.2.1.4 Vater, N. 15.2.9.1 15.2.9.2 Vdovin. V. M 15.2.8.2 15.2.13.2.2 Venezky, D. L. 15.2.13.3.6 Venien, J. P. 16.4.5 16.4.8.7 Ventatachalam, R. 15.2.13.5 Verbeek, W. 15.2.12.9 Verhoest, J. 15.2.13.1 15.2.13.2.3 Verhovodka, L. 15.2.13.2.1 Vericad Raga, J. B. 16.4.2.5.3 Villa, M. 16.4.2.1.4

Villeno-Blanca, M. 15.2.12.2 Villieras, J. 16.4.8.7 Vincent, H. 15.2.12.2 15.2.12.5 Vinogradora, S. 15.2.14.5.4 Vinogradova, S. V. 15.2.7.3 15.2.14.1 Vishnevskii, F. N. 15.2.7.2 Vogel, F. L. 16.4.1. 16.4.2.1.1 16.4.2.2.2 16.4.2.8.3 Vogt Jr., L. H. 15.2.8.1 15.2.8.2 15.2.13.3.4 Volkov, V. S. 15.2.5.1.1 Voll, M. 16.4.2.7.3 Voloshin, A. G. 16.4.2.5.3 15.2.13.2.1 Vol'pin, M. E. 16.4.2.8.2 von Lempe, F. 15.2.10.4 von Plotho, C. 15.2.5.4 von Schnering, H. G. 15.2.4.1.3 von Schnering, H.-G. 15.2.3.2 15.2.3.5 von Stackelberg, M. 16.2.2 16.2.2.1 16.2.2.2 von Strumpf, W. 15.2.13.2.2 Vongehr, M. 15.2.9.4

W

Wachnik, R. 16.4.1. Wachter, J. 15.2.3.5 Wada, N. 16.4.2.2.1 Wadsworth, C. J. 15.2.4.3 Wadsworth, C. L. 15.2.4.1.4

377

Author index ~~

Wagner, G. H. 15.2.8.4 15.2.12.6 Wagner, L. J. 15.2.11.4 Wagner, R. I. 15.2.5.5.2 Walatka Jr., V. V. 15.2.12.2 Walker Jr., P. L. 16 4.2.7.3 Wallace, W. 15.1.3.3.2 Wallraf, G. M. 15.2.4.5. Walsh, M. 16.4.8.5 Walters, M. J. 16.4.1.2 Wang, D. 16.4.8.3 Wang, K. S. 15.2.13.6 Wang, P. J. 15.2.8.2 Wang, W. 16.4.2.1.1 Wannagat, U. 15.2.9.1 15.2.9.2 15.2.9.3 Ward, A. T. 15.2.2.2.3 15.2.2.2.8 Ward, K. J. 15.2.6.1.1 Warne, R. J. 15.2.5.1.1 15.2.5.2 15.2.5.3 Warren, B. E. 15.2.1 Warren Jr., W. W. 15.2.2.2.7 15.2.2.2.9 15.2.2.2.14 Waszczak, J. V. 16.4.3.2 Watanabe, H. 15.2.4.1.2 15.2.4.1.3 15.2.4.2.2 15.2.4.2.3 Watanabe, N. 16.4.2.6.1 16.4.2.6.2 Watson, K. J. 15.2.12.9 Watts, J. A. 16.3.2.3.1

16.3.2.3.3 Watts, L. 15.2.3.5 Webb, R. F. 15.2.14.1 Weber, W. P. 15.2.9.4 Wechsberg, M. 15.2.5.1.1 Wegerhoff, A. 15.2.13.2.2 Weidenbruch, M. 15.2.4.1.3 15.2.4.2.2 15.2.4.2.3 15.2.9.4 Weidman, T.W. 15.2.4.5. Weintraub, E. 16.4.2.2.1 Weiss, A. 16.2.1 16.2.3.2 16.4.3.4 16.4.3.5 16.4.5 16.4.7.2 Weiss, D. E. 16.4.2.7.3 Weiss, R. 16.4.2.7.3 Weiss, W. 16.2.3.2 Weller, P. F. 16.3.2.3.1 16.3.2.3.2 16.3.2.3.3 Wells, E. J. 15.2.3.3 Welsh, K. M. 15.2.4.2.3 Weltner, W. 16.4.2.7.2 Wertheim, G. K. 16.4.2.1.1 Wertz, J. E. 16.4.2.7.2 Weser, G. 15.2.2.2.7 15.2.2.2.14 Wesolowski, J. 15.2.2.2.1 Wessely, M. J. 15.2.3.4 Wesson, J. P. 15.2.13.2.2 West, B. 0. 15.2.3.3 15.2.3.5 West, R. 15.1.1.3.1

15.1.1.3.4 15.2.4.1.1 15.2.4.1.2 15.2.4.1.3 15.2.4.1.4 15.2.4.2.2 15.2.4.2.3 15.2.4.3 15.2.4.4. 15.2.4.5. 15.2.8 15.2.9.3 Westman, A. E. R. 15.2.10.1 Westwood, N. P. C. 15.2.12.3 Weyts, A. G. L. M. 15.2.2.2.4 15.2.2.2.5 Whangbo, M. H. 15.2.12.10 16.4.8.2 White, P. 16.4.2.7.3 White, P. S. 15.2.12.3 White, R. M. 16.4.3.1 Whitehead, M. E. 15.2.13.3.3 Whitehurst, H. 16.4.2.7.2 Whitmire, K. H. 15.2.3.1 Whittingham, M. S. 16.3.2.3.1 16.3.2.3.3 16.4.2.6.1 16.4.3.3 16.4.3.2 16.4.3.3 16.4.3.4 16.4.3.5 16.4.4 Whittle, R. R. 15.2.7.3 Wiberg, E. 15.2.6.4 15.2.8.2 Wiberg, V. E. 15.2.5.1.1 Wiberley, S.E. 15.2.6.2 Wiegel, K. 15.2.9.2 Wiegers, G . 16.4.3.2 Wiegers, G. A. 16.4.3.2 16.4.3.5

378 Wieker, W. 15.2.8.2 15.2.10.2 15.2.10.3 Wiewiorowski, T. K. 15.2.1 15.2.2.2.1 Wilberg, E. 15.2.3.2 Wilkes, G. L. 15.2.8.4 Wilkins, C. J. 15.2.9.4 15.2.9.6 Wilkins, C . L. 15.2.3.3 Wilkinson, G. 15.2.13.2.3 15.2.13.1 Wilkus, E. V. 15.2.13.2.1 Williams, D. J. 15.2.4.1.1 15.2.4.4. Williams, J. 0. 16.4.2.5.3 Williams, M. 15.2.13.2.1 15.2.13.5 Williams, P. M. 16.4.3.4 Williams, R. E. 15.2.7 15.2.7.2 Williams. T. C. 15.2i3.2.2 Willmann, P. 16.4.2.1.4 16.4.2.8.3 Wills, R. R. 15.2.9.3 Willson, C. 15.2.13.3.2 Willson C. G. 15.2.4.5. Wilson, J. A. 16.4.3.1 Winn, D. A. 16.4.3.3 Winter, G. 15.2.5.1.1 Wirbelauer, W. 15.2.9.4 Wiseman, G. H. 15.1.1.3.4 15.2.7.2 15.2.9.3 Wisian-Nielson, P. 15.1.1.3.1 15.1.1.3.4

Author Index 15.2.11.3 15.2.11.4 Wismar, H.-J. 15.2.9.2 Witt H. 16.4.2.3 Witt, M. W. R. 15.2.12.3 Wloka, K. 16.4.2.8.3 Wofler, D. 15.2.4.2.2 Wohrer, L. C. 15.2.13.4 Wojnowski, M. 15.2.9.4 Wojnowski, W. 15.2.4.1.2 15.2.4.2.2 15.2.4.3 15.2.9.4 Wold, A. 16.3.2 16.3.2.3.3 16.4.4 Wolf, C. J. 15.2.5.2 Wolf, K. 16.4.2.7.3 Wolff, A. R. 15.2.4.5. Wolmershauser, G. 15.2.3.5 15.2.12.10 Wong, C.-M. 15.2.12.3 Woo, K. C. 16.4.2.2.1 Wood, J. L. 16.4.2.5.1 16.4.2.5.3 16.4.2.6.1 16.4.2.6.3 Woodward, R. B. 15.2.12.10 Workman, J. H. 15.2.13.5 Worsfold, Ch. D. J. 15.2.4.5. Worzala, H. 15.2.10.4 Wudy, E. 16.4.2.7.2 Wunderlich, B. 15.2.1 15.2.2.2.11 Wiirtenberg, S. 15.2.5.1.1 Wynne, K. J. 15.1.1.1

15.2.4.5. 15.2.13.1 15.2.13.2.3 15.2.13.3.2 15.2.13.3.3 Wynne-Jones, W. F. K. 16.4.2.8.1

x

Xavier, J. 15.2.14.1

Y

Yacoby, Y. 16.4.2.1.1 Yagami, R. 16.4.2.5.2 Yajima, S. 15.2.4.5. 15.2.13.2.2 Yakovlev, I. I. 16.4.2.5.3 16.4.2.6.1 16.4.2.6.3 Yamabe, T. 15.2.12.2 Yamaguchi, S. 16.4.2.6.1 Yancey, J. A. 15.2.7.2 Yankovlev, I. I. 16.4.2.5.3 Yasaitas, E. L. 16.4.2.1.5 Yilgor, I. 15.2.8.4 Yoffee, A. D. 16.4.3.1 Yoldas, B. 15.2.13.3.2 York, B. R. 16.4.2.2.2 Yoshihiko, S. 15.2.14.1 Yoshizumi, K. 15.2.4.1.3 Young, D. A. 16.4.2.1.4 16.4.2.1.5 16.4.2.5.2 16.4.2.7.2 16.4.2.8.1 16.4.2.8.3 Young, M. A. 15.2.5.3 Yoyami, R. 16.4.2.8.2 Yu, Y.-F. 15.1.1.3.4

379

Author Index Yudanov, N. F. 16.4.2.5.3

Z

Zageka, H. D. 16.4.3.4 Zanini, M. 15.2.2.2.7 Zarif'yants, Ju. A. 16.4.2.7.3 Zav'yalov, V. I. 15.2.13.2.2 Zeigler, J. M. 15.2.4.5. Zeigler, M. L. 15.2.3.5 Zeldin, M. 15.2.13.2.3 15.2.13.3.2

15.1.1.1 15.2.4.5. 15.2.13.1 Zeller, C. 16.4.2.2.1 Zeller, H. R. 16.4.2.1.4 Zeng, Y. J. 15.2.8.2 Zhang, Q.L. 16.4.1. Zhang, X. H. 15.2.4.5. Zhang, Z.-H. 15.2.4.1.3 Zharikov, 0. V. 16.4.2.2.1 Zhdanov, A. A. 15.2.8.2

15.2.13.3.4 Zhigach, A. F. 15.2.7 15.2.7.2 Zhigach, A. V. 15.2.7.3 Zhinkin, D. Ya. 15.2.9.3 Zhuikova, T. N. 16.4.2.7.2 Zilber, R. 15.2.10.2 Zingaro, R. A. 15.2.1 Zuppiroli, L. 16.4.3.2 15.2.14.1 15.2.14.2

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc.

Compound Index

Compound Index This index lists individual, fully specified compositions of matter that are mentioned in the text. It is an index of empirical formulas, ordered according to the following system: the elements within a given formula occur in alphabetical sequence except for C, or C and H if present, which always come first. The formulas are ordered alphanumerically without exception. The index is augmented by successively permuted versions of all empirical formulas. As an example, C,H,A10, will appear as such and, at the appropriate positions in the alphanumeric sequence, as H,AlO,*C,, AlO,*C,H, and O,*C,H,Al. The asterisk identifies a permuted formula and allows the original formula to be reconstructed by shifting to the front the elements that follow the asterisk. Whenever an empirical formula does not show how the elements are combined in groups, it is followed by a linearized structural formula, which reveals the connectivity of the compound(s) underlying the empirical formula and serves to distinguish substances which are identical in composition but differ in the arrangement of elements. The nonpermuted empirical formulas are followed by keywords. They describe the context in which the compounds represented by the empirical formulas are discussed. Section numbers direct the reader to relevant positions in the book.

Formation synthesis: 15.2.5.1.1 Al

Al Preparation of borazines: 15.2.5.1.1 AICI, AICI, Catalyst for polymerization of R,SiX,: 15.2.14.2.2

Intercalation into graphite: 16.4.2.8.3 Reaction with (SN),: 15.2.12.2 AICsF,Mg CsMgAIF, Structure: 16.3.1.2 AIF, AIF, Reaction with B,O,: 15.2.6.2

381

382

Compound Index

~

AIF,MgRb RbMgAIF, Structure: 16.3.1.2 AIH,Na,O, $is Na,[A1Si501,].4 H,O Structure: 16.3.1.1 AIH ,MgO,,Si, Mg[AISi,01,]*6 H,O Structure: 16.3.1.1 *c 1SH 1 5 A~,BaH,o0,,Si6 Ba[AISi,O,],*5 H,O Structure: 16.3.1.1 AI,CaH,O I ,Si4 Cd[AiSi,O,],*2 H,O Structure: 16.3.1.1 AI,CaH,O,,Si, Ca[AI,Si,O,,]*3 H,O Structure: 16.3.1.1 AI,CaH,O,,Si, Ca[AISiO,],.4 H,O Structure: 16.3.1.1 AI,CaH,,O,,Si, Ca[AISi,O,],.5 H,O Structure: 16.3.1.1 AI,CaH, ,01$i4 Ca[A1SiZO,],*6 H,O Structure: 16.3.1.1 AI,CaH,,O,,Si, Ca[AISi,O,],.6 H,O Structure: 16.3.1.1 A&CaH,,O,,Si,o Ca[AISi,O,,],*8 H,O Structure: 16.3.1.1 AI,CaH,,Na,O,,SiS [Na,Ca][A1,Si,01,].9 H,O Structure: 16.3.1.1 AI,CaH,,O,,SiS Ca[AI,Si,01,]*9 H,O Structure: 16.3.1.1 AlzH4Na2O,,Si2 Na,[AISi,O,],*2 H,O Structure: 16.3.1.1 AI,H409Si, AI,Si,O,(OH), Structure: 16.4.7.1 AI,H,,Na,Q,8Si4 Na,[AISi,O,],.6 H,O Structure: 16.3.1.1 AI,H, ,Na,O,,Si, Na,[AISi,O,] -6 H,O Structure: 16.3.1.1

Ar Ar Clathration by water: 16.2.2.1 As*CH, AsCl ,*CH AsCI,*C,H, AsCl, AsC1, Redistribution reactions: 15.1.3.3.1 AsCs*C, AsCs*C, AsF, AsF, Reaction with As,O,: 15.1.3.3.2 AsF, AsF, Reaction with graphite: 16.4.2.1.1 AsF,*C, AsF,*C, AsF,K KAsF, Anodic intercalation into graphite: 16.4.2.8.3 AsF,O, O,CAsF,l Reaction with graphite: 16.4.2.1.1 ASH, ASH, Clathration by water: 16.2.2.1 AsH,04 H,AsO, Anodic oxidation of graphite: 16.4.2.8.3 AsI,*CH, AsN,*C,Hl, AsNa,O,*CH, AsO,*C,H,

,

Reaction with AsF,: 15.1.3.3.2 As,MoO,*C,,H,, As,MoO,*C,,H,, As,Ti*C,,H,, As4F206*C10H12 As4F602

F,AsOAs(F)OAs(F)AsF, Formation: 15.1.3.3.2 As5*C5H15 As,Mo,*C10H10 As5MoZ04*C19H25 As6*C36H30 A~,MOO,*C,~H,~ A~,MOO,P~*C,~H,~

B B

Reaction with H,O: 15.2.6.1 BBr*C,H, BBr, BBr, Reaction with B,O,: 15.2.6.1 Reaction with (HSBS),: 15.2.6.2 Reaction with H,S: 15.2.6.2, 15.2.6.4 B*C,H9 BCI,*C,H, BCI, BCI, Reaction with B,O,: 15.2.6.1 Reaction with RNH,: 15.1.1.3.1 Reaction with (HSBS),: 15.2.6.2 Reaction with H,S 15.1.1.3.1 Synthesis of borazines: 15.2.5.1.1 BCl3N*C4H1 BC1,N*C6H7 BCl3N*C6Hl, BCl3N*C1,H2, BF,H,N H,NBF, Preparation of polymer: 15.2.5.5.1 BF3 BF, Intercalation into graphite: 16.4.2.8.3 Reaction with B,O,: 15.2.6.1 Reaction with C n F 16.4.2.6.1 Reaction with graphite: 16.4.2.1.1 Synthesis of borazines: 15.2.5.1.1 BF,H,N NH,BF, Pyrolysis: 15.2.5.5.1 BF,N*C,H, BF4*C16 BF,H HBF, Anodic oxidation of graphite: 16.4.2.8.3 BF,K KBF, Anodic intercalation into graphite: 16.4.2.8.3 BF,NO, “o2lCBF41 Reaction with graphite: 16.4.2.1.1 BH,N HN=BH Formation: 15.1.1.3.1 BH303 WOW3 Dehydration: 15.2.6.1

BH,N H,NBH, Formation: 15.1.1.3.1 Preparation of linear B-N polymers: 15.2.5.3 Synthesis of polymer: 15.2.5.1.2, 15.2.5.5.1 BH6N NH,BH, Pyrolysis: 15.2.5.5.1 BN BN Reaction with S,O,F,: 16.4.2.1.2 BN*CH, BN*CH8 BN*C,H,, BN*C,H,, BN*Cl OH 14 BN2*C7H, BN2*CI,H21 BN, *C14H25 BN3*C6H18 BO,*C,H, B03*C18H15 BZHZS, (HSBS), Formation: 15.2.6.2, 15.2.6.4 B2H6 B2H6

Preparation of boranamine polymers: 15.2.5.1.2 Pyrolysis: 15.1.1.3.2 Reaction with RSH: 15.2.6.2 Reaction with RNH,: 15.1.1.3.1 Reaction with NH,: 15.1.1.3.1 Reaction with 0,: 15.2.6.1 Synthesis of B-N polymers: 15.2.5.3 B,H,N H2NB2H5

Preparation of linear B-N polymers: 15.2.5.3 B,H*N, (H,NBH,), Formation: 15.2.5.1.2 Produced from pentamer: 15.2.5.4 Transformation to trimer: 15.2.5.4 B,N*CH, B,N*C,Hll B2N2*C4H16 B2N2S2*C4H 1Z B2°3 B2°3

Reaction with BR,: 15.2.6.1

384 Reaction with BX,: 15.2.6.1 Reaction with H 2 0 : 15.2.6.1 B2S2*Cl2HlO B2S3 B2S3

H,S: 15.2.6.4 B2S4*C4H111 B3Br303 (BI.€3 0)3 Formation: 15.2.6.1 B3C12N3*C4H12 B3C12N3*C6H16 B3C12N3*C9H14 B3C13F1 sN3*'1 8 B3CW3N3 (HNBCI), Conversion to BN dimer: 15.2.5.4 B3Cl3N3*C,H9 B3C13N3*C1S H 1 5 B3CI303 (CIBO), Formation: 15.2.6.1, 15.2.6.4 B3 c13s3 (SBCI), Formation: 15.1.1.3.1 B3F3H3N3 (HNJW3 Hydrolytic stability: 15.2.5.2 B3F303 (FW3 Formation: 15.2.6.1 B3F3S3 (FBS), Formation: 15.2.6.2 J33H303 (HBO), Formation: 15.2.6.1 B3H3S6

(HSBS), Formation: 15.2.6.2, 15.2.6.4 Reaction with BX,: 15.2.6.2 B3H6N3

WNBH), Formation: 15.2.5.1.1 Thermal and hydrolytic stability: 15.2.5.2

B3H12N3

(H2NBH2)3 Conversion from dimer: 15.2.5.4 Formation: 15.2.5.1.2 Thermal and hydrolytic stability: 15.2.5.2

Compound Index B3183 (IBS), Formation: 15.2.6.2 B3Li2N3*C4Hl, B3N3*C3H12 B3N3*C3H,,

B3N3*C18H18 B3N3*C30H42 B3N3S3*C6H18 B3N5*C7H22 B3N5*C11H30 B3N5*C19H30 B3N6*C21H2'7 B3P3*C12H36 B3S3*C3H1 5 B3S3*C12H33 B3S3*C18H15 B3S6*C6H

15

B4C14N4*C16H36 B4C1404 (c1w4 Conversion to (CIBO),: 15.2.6.4 B4FN403S CBNl4CSO3Fl Formation: 16.4.2.1.2 B4H16N4

(H2NBH2)4 Formation: 15.2.5.1.2 B4N4*C40H56 B,H,N, N5BA Formation: 15.2.5.5.2 BSH, B5H9

Formation: 15.1.1.3.2 BSH2ONS (H2NBH2)5 Formation: 15.2.5.1.2 Preparation of dimer: 15.2.5.4 B6H10

B6H 10

Formation: 15.1.1.3.2

B6H10N6

N6B6H10

Formation: 15.2.5.5.2

10*C2H 14

B10H14 B10H14

Formation: 15.1.1.3.2 Reaction with R3N: 15.1.1.3.2 nido-BloH14 Reaction with R2S: 15.1.1.3.2

Compound Index B,OLi,*C,HlO B10S*C4H22 BaH Si,*AI, BaNaO,P, NaBa[P,O,] Formation: 15.2.10.2 Ba2030P10Zn3

2 Ba0.3 Z n 0 . 5 P,O, Formation: 15.2.10.2 BiCI, BiCI, Intercalation into graphite: 16.4.2.8.3 BiCs*C, BiCs*C, Br*CH, Br*C,H, Br*C,H,B Br*C,H, BrCl BrCl Clathration by water: 16.2.2.1 BrCI,*C BrCrO CrOBr Electrochemistry: 16.4.5 Structure: 16.4.5 BrF, BrF, Reaction with graphite: 16.4.2.6.1 BrF, BrF, Reaction with graphite: 16.4.2.1.5 BrI IBr Reaction with (SN),: 15.2.12.4 Reaction with graphite: 16.4.2.1.5 BrN*C,,H,, BrNPSi*C,H, BrOTi TiOBr Structure: 16.4.5 BrOV VOBr Structure: 16.4.5 BrSi*C,H, BrSi,*C,,H,, Br2 Br, Clathration by water: 16.2.2.1 Graphite residue compound: 16.4.2.7.1, 16.4.2.7.2 Reaction with (SN),: 15.2.12.4

,

385

Reaction with carbon surface: 16.4.2.7.3 Reaction with graphite: 16.4.2.1.5 Br,Ge*C,H, Br,Si*C,H Br3*B Br,O,*B, Br,S,Si, (Br,SiS), Formation: 14.2.9.4, 14.2.9.6 Br6N3P3

(N=PBr,), Pyrolysis: 15.2.11.3 Br6Si2 Si,Br, Reaction with SF,: 14.2.9.4 Br,Si,*C,,H,, C Cgraph,te Reaction [NO,][BF,]: 16.4.2.1.1 Reaction with O,[AsF,]: 16.4.2.1.1 Reaction with AsF,: 16.4.2.1.1 Reaction with BF,: 16.4.2.1.1 Reaction with CrO,: 16.4.2.1.3 Reaction with CrO,F,: 16.4.2.1.2 Reaction with CrO,CI,: 16.4.2.1.2 Reaction with CrO,: 16.4.2.1.2,16.4.2.1.4 Reaction with FeC1,: 16.4.2.1.1 Reaction with Fe: 16.4.2.1.3 Reaction with GeF,: 16.4.2.1.1 Reaction with KrF,: 16.4.2.1.1, 16.4.2.1.2 Reaction with K[MnO,]: 16.4.2.1.4 Reaction with MoOC1,: 16.4.2.1.2 Reaction with MOO,: 16.4.2.1.3 Reaction with MoF,: 16.4.2.1.1 Reaction with NbF,: 16.4.2.1.1 Reaction with OsF,: 16.4.2.1.1 Reaction with PbO,: 16.4.2.1.4 Reaction with PF,: 16.4.2.1.1 Reaction with RdS: 16.4.2.1.3 Reaction with SbF,: 16.4.2.1.1 Reaction with Sb,S,: 16.4.2.1.3 Reaction with Sb,O,: 16.4.2.1.3 Reaction with H,SeO,: 16.4.2.1.4 Reaction with SeO,: 16.4.2.1.3 Reaction with Cu: 16.4.2.1.3 Reaction with N,O,: 16.4.2.1.3 Reaction with SO,: 16.4.2.1.3 Reaction with H,SO,: 16.4.2.1.4 Reaction with [NH,],[S,O,]: 16.4.2.1.4 Reaction with S,O,F,: 16.4.2.1.2 Reaction with TaF,: 16.4.2.1.1 Reaction with TiF,: 16.4.2.1.1

386

Compound Index

Reaction with TI$: 16.4.2.1.3 Reaction with UO,CI,: 16.4.2.1.2 Reaction with UF,: 16.4.2.1.1 Reaction with VOF,: 16.4.2.1.2 Reaction with WS,: 16.4.2.1.3 Reaction with XeOF,: 16.4.2.1.2 Reaction with XeF,: 16.4.2.1.1 Reaction with XeF,: 16.4.2.1.1 Reaction with XeF,: 16.4.2.1.1, 16.4.2.1.2 Reaction with X,: 16.4.2.1.5 Reaction with HNO,: 16.4.2.1.4 Reaction with IOF,: 16.4.2.1.2 Reaction with HIO,: 16.4.2.1.4 Reaction with HCIO,: 16.4.2.1.4 Reaction with HIO,: 16.4.2.1.4 Reaction with C1,0,: 16.4.2.1.3 Reaction with F,: 16.4.2.1.1 Reaction with F,, CIF,, BrF,: 16.4.2.6.1 Reaction with CIF,: 16.4.2.1.5 Reaction with BrF,: 16.4.2.1.5 Reaction with CIF,: 16.4.2.1.5 Reaction with IF,: 16.4.2.1.5 Reaction with IF,: 16.4.2.1.2, 16.4.2.1.5 Reaction with IBr: 16.4.2.1.5 Reaction with ICI: 16.4.2.1.5 Reaction with CIF: 16.4.2.1.1 Reaction with group-IA metals: 16.4.2.2.1 CBrCI, CC1,Br Clathration by water: 16.2.2.2 CCI, CCI, Clathration by water: 16.2.2.2 CF, CF4 Formation: 16.4.2.6.1 CHCI, CHCI, Clathration by water: 16.2.2.2 CHF,O,S CF,SO,H Anodic oxidation of graphite: 16.4.2.8.3 CH,CI, CH,CI, Clathration by water: 16.2.2.2 CHZNZ CH2N2 Methylation of graphite oxide: 16.4.2.5.3 Reaction with surface oxides on carbon: 16.4.2.7.3 CH,AsCI, CH,AsCI, Reaction with RAsH,: 15.2.3.3

CH,AsI, CH,AsI, Formation: 15.2.3.3 Reaction with RAsH,: 15.2.3.3 CH,AsNa,O, CH,As(O)(ONa), Reduction: 15.2.3.3 CH,Br CH,Br Clathration by water: 16.2.2.1, 16.2.2.2 CH,CI CH,CI Clathration by water: 16.2.2.1 CH,F CH,F Clathration by water: 16.2.2.1 CH,I CH,I Clathration by water: 16.2.2.2 CH,NO, CH,NO, Anodic oxidation of graphite: 16.4.2.8.3 CH, CH4 Clathration by water: 16.2.2.1 CH,O CH,OH Intercalation into graphite oxide: 16.4.2.5.3 CH,S CH,SH Clathration by water: 16.2.2.1 CH,As CH,AsH, Reaction with RAsX,: 15.2.3.3 Reaction with R,Hg: 15.2.3.3 CH,BN CH,NHBH, Formation: 15.2.5.1.2 CH,CIN [CH,NH,]CI Reaction with LiBH,: 15.2.5.1.1 Reaction with BC1,: 15.2.5.1.1 CH,BN CH,NH,BH3 Pyrolysis: 15.2.5.1.2 CH,B,N CH,NHB,H5 Preparation of linear B-N polymers: 15.2.5.3

cos cos

Reaction with NF,: 15.2.12.1

Compound Index

ocs

Clathration by water: 16.2.2.2

CO,

co2

Clathration by water: 16.2.2.1

CS,

cs2

Clathration by water: 16.2.2.2 Formation of carbon surface sulfides: 16.4.2.7.3

‘ZF6

C2F6 Formation: 16.4.2.6.1 C2HF302

CF,COOH Anodic oxidation of graphite: 16.4.2.8.3 C,H* C2H2 Clathration by water: 16.2.2.1 C A N CH,CN Anodic oxidation of graphite: 16.4.2.8.3

C2H4

C2H4 Clathration by water: 16.2.2.1 C2H4C12

CH2C1-CH2CI Clathration by water: 16.2.2.2

C2H402

CH,C02H Intercalation into graphite oxide: 16.4.2.5.3 Reaction with (HNBCI),: 15.2.5.4 C,H,AsCI, C,H,AsCI, Reaction with RASH,: 15.2.3.3 C,H,Br C2H,Br Clathration by water: 16.2.2.2 C,H,CI C2H,CI Clathration by water: 16.2.2.2 C,H,F C2HJ Clathration by water: 16.2.2.1 C,H,NaO NaCOC2H51 Neutralization of acidic surface oxides on carbon: 16.4.2.7.3 C2H6

C2H6 Clathration by water: 16.2.2.1

387

C,H6BBr (CH3)2BBr Reaction with R,PH: 15.1.1.3.1 C,H6Br,Ge (CH3)2GeBr2 Reaction with R,GeS: 15.1.3.3.2 C,H,CI,Si (CH,),SiCI, Reaction with R2SiC12-Li: 15.2.4.1.4 Reaction with Li: 15.2.4.1.3 Reaction with Li, Na, K: 15.2.4.1.2 Reaction with Na-K: 15.1.1.3.1 Reaction with carboranes: 15.2.7.3 Reaction with cyclosilazanes: 14.2.9.3 C,H6CI,Si, C12(CH,)SiSi(CH,)C12 Reaction with H,O: 15.2.8.2 ‘ZH6’

(CH3)20 Clathration by water: 16.2.2.2 C2H,0H Intercalation into graphite oxide: 16.4.2.5.3 C,H60S (CHd2SO Cathodic reduction of graphite: 16.4.2.8.1, 16.4.2.8.2 C2H603S

(CH30)2S0 Anodix oxidation of graphite: 16.4.2.8.3 C2H604S

(CH30)2S02 Reaction with surface oxides on carbon: 16.4.2.7.3 C2H6S

(CH3)2S Clathration by water: 16.2.2.2 C,H,CISi (CH,),SiHCI Reaction with LI: 15.2.4.1.2 Reaction with H,O: 15.2.8.1 C,H,P (CH3)2PH Reaction with R2BX: 15.1.1.3.1 HP(CH3)2 Formation: 15.2.11.2 CZHIOBN (CH,),NHBH3 Pyrolysis: 15.2.5.1.2 C,H,oB,oLi, Li2CC,B,oH,ol Reaction with R,SiX 15.2.7.3

388

Compound Index

Li2C1,7-C2B,oHioI Reaction with RSiX: 15.2.7.2 C,HllB,N (CH3)2NB2H5 Preparation of linear B-N polymers: 15.2.5.3 C2H14B10 B10C2H14

Formation: 15.1.1.3.2 C,Li LiC, Formation: 16.4.2.2.1 C,Na NaC, Formation: 16.4.2.2.1

C3H10P2

(CH,),PPHCH, Formation: 15.2.11.2

C3H11NP2

(CH.3)2PP(CH3)NH2 Pyrolysis: 15.2.11.2 C3H,,B3N3 (CH,NBH), Formation: 15.1.1.3.1, 15.2.5.1.1, 15.2.5.1.2 (HNBCH,), Formation: 15.1.1.3.1 C3H15B3S3

(H2BSCH3)3 Formation: 15.2.6.3

C3H2F9P2

C3H18B3N3

CF,(H)P(CF,)P(H)CF, Formation: 15.2.3.2 C3H,Br n-C,H,Br Clathration by water: 16.2.2.2 C3H8 C,H* Clathration by water: 16.2.2.2 C3H8CI,Si CH,(C,H,)SiCI, Reaction with Li: 15.2.4.1.3 CI(CH,),SiCH,CI Reaction with Mg: 15.1.1.3.2 C3HgAs03 As(OCH,), Redistribution reactions: 15.1.3.3.1 C3H9B B(CH,), Reaction with (HSBS),: 15.2.6.2 Reaction with NH,: 15.1.1.3.1 C3HgBF,N (CH,),NBF, Abstraction of HF: 15.2.5.1.1 C3H9BO3 B(OCH,), Reaction with (HSBS),: 15.2.6.2 C3H9B3CI3N3 (CH,NBCl), Formation: 15.2.5.1.1 C3HgBrSi (CH,),SiBr Reaction with (NSCI),: 15.2.12.4 c3H 903p P(OCH,), Redistribution reactions: 15.1.3.3.1 C3HP P(CH,), Formation: 15.2.11.2

(CH3NHBH2)3 Formation: 15.2.5.1.2 C3Li LiC, Formation: 16.4.2.2.1 C3Na NaC, Formation: 16.4.2.2.1 C4AsCs CsAsC, Formation: 16.4.2.2.2 C,BiCs CsBiC, Formation: 16.4.2.2.2 C4Fi2P2

(CF3)2PP(CF3)2 Formation: 15.2.3.2 C4F12P4

c-(CF,P)4 Formation: 15.2.3.2 Reaction with RPH,: 15.2.3.2 C4FeNa204 N%[Fe(CO)41 Reaction with CIC(O)RC(O)CI: 15.2.14.2.3 C4H603 (CH3C0)20 Acetylation of graphite oxide: 16.4.2.5.3 C,H,CO3 Anodic oxidation of graphite: 16.4.2.8.1, 16.4.2.8.3 C4H80

(CH2)40 Cathodic reduction of graphite: 16.4.2.8.2 C4H802

0(C2H4)20 Intercalation into graphite oxide: 16.4.2.5.3

Compound Index

c4 H 802s

(CH2)4S02 Anodic oxidation of graphite: 16.4.2.8.3 C,H,CISi CH,CH(CH,),SICI Reaction with HSiRC1,: 15.2.8.2 C4H,Li n-C,H,Li Reaction with MPS,, MPSe,: 16.4.4 Reaction with MOCI: 16.4.5 Reaction with transition-metal chalcogenides: 16.4.3.2 C4H10

(CH313CH Clathration by water: 16.2.2.2 C4H10B2S4

(C2HsSBS)2 Formation: 15.2.6.4 C4Hl0Br,Si (C2Hd2SiBr2 Formation: 15.2.4.2.2 C4HloCI,Si CH,(i-C,H,)SiCI, Reaction with Li: 15.2.4.1.3 (C,H5),SiCI2 Formation: 15.2.4.2.2 Reaction with Li: 15.2.4.1.3 Reaction with Na: 15.2.4.2.2 C4Hl0I2Si (C2H5)2Si12 Formation: 15.2.4.2.2 C4H100

(CH&COH 14.2.9.4 Reaction with (C,H5)2O Intercalation into graphite oxide: 16.4.2.5.3 C,H,OO, CH,O(CH,),OCH, Cathodic reduction of graphite: 16.4.2.8.2 C4HIOS CH3(CH2)3SH Elimination from borazine ring: 15.2.5.5.2 C4H,,BCI,N (CH,),CNH,BCI, Preparation of BN tetramers: 15.2.5.1.1 C4HIlN CH3(CH2)3NH2 Reaction with RBCI,: 15.2.5.3 (CH,),CHCH,NH, Reaction with RBCI,: 15.2.5.3

(CH3)3CNH2 Reaction with RBCI,: 15.2.5.3 C4H12B2N2S2

C(CHJ,NBS12 Ring expansion: 15.2.6.4 C4Hz 2B3C4N3 1 CH3N[B(CI)N(CH3)]2BCH, Condensation with [(CH,),Si],NH: 152.5.5.2 Copolymerization: 15.2.5.5.2 C4H12B3Li2N3 1 CH,N[B(CH,)N(Li)],BCH, Copolymerization: 15.2.5.5.2 C4HIZCIN C(CHMW Cathodic reduction of graphite: 16.4.2.8.1 C4H1

Ge(OCH,), Redistribution reactions: 151.3.3.1 C4HI ,Li2N2Si Li[CH,NSi(CH,),NCH,]Li Reaction with R,SiCI,: 15.1.1.3.1 C4Hl,02Si (CHJ2Si(OCH3)2 Formation: 15.2.4.1.2 C,H,,O,Si Si(OCH,), Redistribution with Six,: 15.1.3.3.1 C4H12P2

(CH3)2PP(CH3)2 Formation: 15.2.11.2 C,H12S2~i, C(CH,)2SiS1, Equilibration: 14.2.9.5 Formation: 14.2.9.4 Thermal rearrangement: 15.1.1.3.1 2s6si4

(CH,Si),S, Formation: 14.2.9.4 C*HI,Sb* (CH3)2SbSb(CH3)2 Thermal decomposition: 15.2.3.4 C4Hl,C12NSi2 [CISi(CH3),l2 NH Formation: 14.2.9.3 HN[Si(CH,),CI], Reaction with silylamine: 15.2.9.2 C4H16B2N2

[(CH3)2NBH212 Formation: 15.2.5.1.2 C4H22B10S

B,OH12CS(C,H5)21 Formation: 15.1.1.3.2

389

Compound index

390 Reaction with RC-CR: C4HgK KHgC4 Formation: 16.4.2.2.2 C,HgRb RbHgC, Formation: 16.4.2.2.2 C4K KC4 Formation: 16.4.2.2.1

15.1.1.3.2

C,H,BCI, C,H,BC1, Reaction with Reaction with C,H5CI,Sb C,H,SbCI, Reaction with C,H,F,Si C,H,SiF, Reaction with

(R,Si),S: 15.2.6.4 butylamines: 15.2.5.3 RSbH,: 15.2.3.4 cyclodisilazane: 15.2.9.2

C6H6

C5F15P3

(CF3)2PP(CF3)P(CF3)2 Thermal decomposition: 15.2.3.2 CsFe05 Fe(CO), Reaction with c-(RE),: 15.2.3.5 C5H100~S

CH3(C4H7)S02 Anodic oxidation of graphite: 16.4.2.8.3 C,H, ,CI,Si CH,(t-C,H,)SiCI, Reaction with Li: 15.2.4.1.3 C,H, ,CIOSi,

,

I

(CH,)CISiOSi(CH,),CH,CH, Formation: 15.2.8.2 C,H,5'4ss c-(CH3As), Formation: 15.2.3.3 Reaction with [CpMo(CO),],: 15.2.3.5 Reaction with Fe(CO),: 15.2.3.5 Reaction with R,TiCI,: 15.2.3.5 Reaction with X,: 15.2.3.3 C,H,,BrNPSi (CH,),SiN=PBr(CH,), Pyrolysis: 15.1.1.3.4 C,H,,PSi (CH,),PSi(CH,), Reaction with R,BCI: 15.1.1.3.1 C5H15P5

c-(CH,P), 'Formation: 15.2.3.2 Reaction with Fe(CO),: 15.2.3.5 C,Eu EuC, Formation: 16.4.2.2.1 C,F,Sb C,SbF, Formation: 16.4.2.1.1 C6H2FSN

CtP5NH2 Preparation of perfluoroborazines: 15.2.5.1.1

C6H6

Clathration by water: 16.2.2.2 C,H,BCI,N H2(C6H5)NBC13 Formation: 15.1.1.3.1 C6H7N C6H5NH2

Reaction with BX,: 15.1.1.3.1

C6H7P C6H5PH2

Formation: 15.2.3.2 C,H7Sb C,H,SbH, Reaction with RSbX,: 15.2.3.4 C,H,Si C,H,SiH, Oligomerization: 15.2.4.5 Reaction with S,: 14.2.9.4 C,H,,CI,Si (CH,),CCH,(CH,)SiCI, Reaction with Li: 15.2.4.1.3 (i-C,H,),SiCI, Reaction with Li: 15.2.4.1.3 (n-C,H,),SiCI, Reaction with Li: 15.2.4.1.3 C6H140S

(C3H7)2S0 Intercalation into graphite oxide: 16.4.2.5.3 C,H,,BCI,N (C2H5)3NBC13 Preparation of borazines: 15.2.5.1.1 C6H15B3S6

(C2H5SBS)3 Conversion to dimer: 15.2.6.4 C,H,,FLiNSi [(CH,),HC],SiF(NHLi) Thermal cycloization: 15.2.9.1 C6H1503P

P(OC,Hd3 Redistribution reactions: 15.1.3.3.1

COmDOUnd Index C6H16B3C12N3

I

CH3NCB(Ci)N(CH,)l,BCH(CH,), Condensation with [(CH,),Si],NH: 15.2.5.5.2 C6H16CI,Si, CI(CH3),SiCH,CH,Si(CH3),C1 Reaction with H,O: 15.2.8.2 C6Hl,OSi, (CH,),SiOSi(CH,),CH,CH, Formation: 15.2.8.2

-

C6H16S2Si2

C(CH,),S1SCH,I, Decomposition with hv: 14.2.9.4 C6H18AsN3

As"(CH3)2] 3 Redistribution reactions: 15.1.3.3.1 C6H18BN3

B"(CH3)2I 3 Reaction with (HSBS),: 15.2.6.2 C6H18B3N3S3

C(CHd2NBSI3 Formation: 15.2.6.4 C6H18C1202Si3 (CH3),Si[OSi(CH3),C11, Reaction with carboranes: 15.2.7.2 C6H1 8Ge3S3

C(CH,),GeSJ, Reaction with R,GeX,: 15.1.3.3.2

Reaction with NaOR: 15.1.1.3.2 C6H18SSi2 [(CH3)$1$ Reaction with RBX,: 15.2.6.4 C6H18S2Si2

C(CH3)8iS], Reaction with R,SiCl,: 14.2.9.4 C6H18S3Si3

[(cH3)2sis13 Equilibration: 14.2.9.5 Formation: 14.2.9.4, 15.1.1.3.1 C6H 1 8 2

(CH3)3SiSi(CH3)3 Formation: 15.2.4.2.3 Si2(CH3)6 Reaction with SF,: 14.2.9.4 C6H19Li2N3Si3 [(CH,),SiNH][(CH,),SiNLi], Reaction with R3SiC1: 15.2.9.2

{C(CH,),siNHIC(CH3),siNLi12}

Reaction with amino-fluorosilanes: 15.2.9.2 C6Hl,NSi2 [(CH3)3Sil,NH Condensation with chloroborazines: 15.2.5.5.2 C6H2,LiN3Si,

{C(CH,),SiNHlzl(CH,),SiNLi1}

Reaction with amino-fluorosilanes: 15.2.9.2

C6H1816Si6

(CH,SiI), Reaction with MR: 15.2.4.2.1

C6H18N2Si2

CH3NSi(CH3)2N(CH3)Si(CH3)2 1

Formation: 15.1.1.3.1

C6H18N3P

[(CH 3 )2N13p Solvated alkalimetal graphites: 16.4.2.8.2 P"(CH 31213 Redistribution reactions: 15.1.3.3.1 C6H18N3P3

"=P(CH 31213 Formation: 15.1.1.3.4 Pyrolysis: 15.2.11.2, 15.2.11.3 "P(CH 31213 Formation: 15.1.1.3.1 C6H1802Si2

C(CH3),SiOCH31, Reaction with KOR: 15.2.4.1.2 C6H,803Si3

C(CHJ2SIOI3 Formation: 14.2.9.3 Pyrolysis: 15.I. 1.3.4

39 1

C6H2CIN4Si2

,

[CH,NSi(CH,),NH] Formation: 15.2.9.2

C6H21N3Si3

[(CH3),SiNHI, Pyrolysis: 15.1.1.3.4 Reaction with LiR, R,SiX: 15.2.9.2 Reaction with R,SiCl,: 14.2.9.3 C6K

KC, Formation: 16.4.2.2.1 C6Li LiC, Formation: 16.4.2.2.1 C6Mo06 Mo(C0)6 Reaction with +(RE),: 15.2.3.5 C6Yb YbC, Formation: 16.4.2.2.1 C,H,Fe NaO, Na[.rl-5C,H,Fe(CO),] Reaction with (N=PF,),: 15.2.11.4

392

Compound Index

C7H8 CH3C6H5

Carbon purification: 16.4.2.7.3 C,H,CI,Si C6H5(CH,)SiC1, Reaction with R,SiCI,-Li: 15.2.4.1.4 Reaction with H,O: 15.2.8.1 Reaction with Li: 15.2.4.1.4 C7H9BN2

p-CH3C,H4NBNH, Synthesis of polymer: 15.2.5.5.1 C,H, ,F,NOPSi (CH,),SiN=P(OCH,CF,)(CH,), Pyrolysis: 15.1.1.3.4 C,H,,F,NSi, C(CH,),[Si(CH,),]NSiF, Reaction with cyclotrisilazane: 15.2.9.2 C,H, F,NSi, [(CH,),Si],NSiF,CH, Reaction with cyclotrisilazane: 15.2.9.2 C,H, ,LiZNSi, CH,N[(CH,),SiNCH,Li], Reaction with disilazane: 15.2.9.2 C7H22B3N5

1

CH,N[B(NH2)N(CH,)IzBc4Hg Polymerization: 15.2.5.5.2 C7H22N2Si3

[(CH,),SiNSi(CH,),][NHSi(CH,),I Ring expansion: 15.2.9.2 C,AsCs CsAsC, Formation: 16.4.2.2.2 C,AsF, C,AsF5 Formation: 16.4.2.1.1 C,AsF, C,[AsF6I Formation: .16.4.2.1.1 C,BiCs CsBiC, Formation: 16.4.2.2.2 c,cs csc, Formation: 16.4.2.2.1 C,F,Os C,COsF,jI Formation: 16.4.2.1.1 'SH90

(CH,),C,H,OH Reaction with (SiSJX: 14.2.9.4 C,H,,CI,Si (i-C4Hg),SiCl2 Reaction with Li: 15.2.4.1.3

~,H,,OS (C4H9)2S0 Intercalation into graphite oxide: 16.4.2.5.3 C,H,,FLiNSi [(CH,),C],SiF(NHLi) Thermal cycloization: 15.2.9.1 C,H,,O,Si Si(OC,H5)4 Redistribution with (RO),Si: 15.1.3.3.1 C8H20P2

(C2H5)2PP(C2H5)2 Formation: 15.2.3.2 C,H,,CI,Si, [(CH3),SiCH,],SiC1, Reaction with Li: 15.2.4.1.3 C,H,,CoKP, KC?-2C,H4CoEP(CH,),131 Reaction with graphite: 16.4.2.2.1 C,H,4Cl,N,Si4 [C1(CH3),SiNSi(CH,),1, Reaction with water: 14.2.9.3 [ClSi(CH,),NSi(CH,),], Formation: 14.2.9.3 C8H24GeN4

Ge"(CH,)zI4 Redistribution reactions: 15.1.3.3.1 C8H24N4P4

"=P(CH,)zI4 Pyrolysis: 15.2.11.2 C,Hz4N4Si Si"(CH&.I4 Redistribution with Six,: 15.1.3.3.1 C8H24Si4

[(CH&sila Formation: 15.2.4.1.2 C,H,,N0,Si4 [(CH,),SiO] ,[(CH,),SiN(CH,),SiOH] Formation: 14.2.9.3 C8H28N4si4

C(CHJzSiNHI4 Reaction with RNH,: 15.2.9.2 Reaction with R,SiCI,: 14.2.9.3, 14.2.9.4 C*K KC, Formation: 16.4.2.2.1 Reaction with H,: 16.4.2.4.1 C,Rb RbC, Formation: 16.4.2.2.1 Reaction with H,: 16.4.2.4.1 C9H12BN

(CH ,),,CHNBC,H Synthesis of polymers: 15.2.5.5.1

Compound Index

393

C10H12Fe~06P4

CFe(CO)31zCcatena-q2,~-(CH,P)41 Formation: 15.2.3.5 C10H14BN

CH3CH(CH3)CH2NBC,H, Synthesis of polymers: 15.2.5.5.1 -{[(CH,),FSiNSi(CH,),][(CH,),SiNLi]* (CH,),CHCH,NBC,H, [(CH,),SiNSiF,CH,]} Synthesis of polymers: 15.2.5.5.1 LiF-elimination: 15.2.9.2 (CH,),CNBC,H, C,H,,Li,N3Si4 Synthesis of polymers: 15.2.5.5.1 {C(CH3)3siNsi(CH3)21C(CH,),SiNLil}2 CI OH1SF@, Reaction with cyclotrisilazane: 15.2.9.2 ~ ~ ~ ~ ~ , ~ , ~ ~ ~ ~ ~ ~ ~ ~ , ~ , 1 ~ ~ C 15-C,(CH,),Fe(P,-~5) H , ) , S i N L i 1 , } Formation: 15.2.3.5 Reaction with cyclotrisilazane: 15.2.9.2 C, oH2,CI,Si C9H,,LiN,Si4 C(CH,),CCH,I,SiCI, [(CH,),SiNH][(CH,),SiNLi](CH,),SiN* Reaction with Li: 15.2.4.1.3 WH,), CIOH2SP3 Reaction with RSiF,: 15.2.9.2 (CzHs),PP(C,Hs)P(C,H,), C9H29N3Si4 Thermal decomposition: 15.2.3.2 {C(CH3),siN~i(CH3)21CNHSi(CH3)212} C10H25P5 Formation: 15.2.9.2 c-(CzH,P), I(CH,),~iNSi(CH,),I~NHSi(CH,),I, Formation: 15.2.3.2 Formation: 15.2.9.2 C,oH30F,LiN3Si, (CH,),Si[NSi(CH3)2]2Si(CH3)2NH, { [(CH,),SiFNSi(CH,),],(CH,),SiNLi]} Ringexpansion: 15.2.9.2 Reaction with cyclotrisilazane: 15.2.9.2 CIOCS CSC,, Formation: 16.4.2.2.1 C,oH,oAs,Mo, (q-5C5H5Mo)z(q4,p-~-A~5)

Formation: 15.2.3.5 C, ,H, ,CI,Ti (q5-C,H5),TiC1, Reaction with (q-SHOCH,C5H4),Fe: 15.2.13.2.1 (q-sC5H5)2TiC12 Reaction with c-(RAs),: 15.2.3.5 CI OH1 oco

(q5-CsH5),Co Intercalation with MOCI: 16.4.5 Reaction with TaS,: 16.4.3.4 C,oH,oFe (q5-CsH5)2Fe Intercalation with MOCI: 16.4.5 Reaction with RC(0)H: 15.2.13.2.1 CloHloMo,Ps (4-5C5HsMo),(q4,p-c-ps) Formation: 15.2.3.5 C10H1oRu (q5-C5H5),Ru Reaction with H,C(OR),: 15.2.13.2.1 C10H12As4F206

[Fe(CO),],[~atena-q~,p-(CH~As)~] Formation: 15.2.3.5

C10H30FSN3Si6

{ [(CH,),SiFNSi(CH,),]~[(CH,),SiNSi* F,]} Reaction with cyclotrisilazane: 15.2.9.2

C10H300Si4

[(CH,),Si],SiOCH, Thermal decomposition: 15.2.4.1.3

C10H30Si4

(CH,),SiC(CH,),~il,~~(CH,), Isomerization: 15.2.4.3 [(CH,),Si],SiCH, Formation: 15.2.4.3 I OH30SiS

[(CH,)$il5 Formation: 15.2.4.1.2, 15.2.4.2.3 Redistribution: 15.2.4.3 CIoH3,LiN4Si4

{ HN[Si(CH,),NCH,Si(CH,),],NLi} Reaction with CH,I: 15.2.9.2

C10H32N4Si4

[(CH,)zsiNH(CH,)Nsi(cH3)z1z Ring expansion: 15.2.9.2

10H33N4Si4

{HN[Si(CH3)2NCH3Si(CH3)2],NCH,) Formation: 15.2.9.2

CIOK

KC,, Formation: 16.4.2.2.1

394

Compound Index

CloM~zOlo Mn,(CO),o Reaction with c-(RE),: 15.2.3.5 C10O10Re2 Re2(CO),, Reaction with c-(RE),: 15.2.3.5 CIORb RbC 10 Formation: 16.4.2.2.1 1‘

1F6Mo

C11MoF6

C12HlZF18N306P3

Formation: 16.4.2.1.1

‘1

IH2Z0ZSi3

C,H,(CH,)Si[OSi(CH,)zH]2 Formation: 15.2.8.1

C11H30B3N5

(C6H5)2SiC12 Electrolytic reduction: 15.2.4.1.1 Polymerization: 15.2.14.2.2 Reaction with Li, Na, Mg: 15.2.4.1.1 C,,Hl0Li,Si (C6H,),SiLi, Reaction with (RO),PO: 15.2.4.2.2 C1zH12Cr (q6-C6H6)2Cr Reaction with MPS,: 16.4.4

I

CH3N[B(NHC2H5)N(CH3)1ZBC4H9 Polymerization: 15.2.5.5.2 Cl,H3lN4~i.l

“=P(OCHZCF,)~I, Formation: 15.2.11.2 Pyrolysis: 15.2.11.3 CI zH1 2 p z C6H5(H)PP(H)C6H5 Thermal decomposition: 15.2.3.2 C12H14Fe02

($-HOCH2C5H4),Fe Reaction with CP2TiC1,: 15.2.13.2.1 { [CH,NHSi(CH,)2NSi(CH3)2][NCH3Si* C1zH1,FeNz (CHd212 (q5-H2NCH2C,H4),Fe Formation by ring contraction: 15.2.9.2 Reaction with ClC(O)RC(O)Cl C, ,H33FLiN3Si5 15.2.13.2.1 { [(CH,),SiFNSi(CH,),][LiNSi(CH,),]* CI2HZlBNZ C(CH,),SiNSi(CH,),I~ C(CH~)~CHNHIZBC~H, Reaction with H-acid compounds: Condensation polymerization: 15.2.5.5.1 15.2.9.2 C12H27BC13N { [(CH,),SiNSi(CH,),][(CH,),SiNLiJ[(C* [CH3(CH2)313NBC13 H,),SiNSiF(CH,),]} Pyrolysis: 15.2.5.5.1 LiF elimination: 15.2.9.2 CIZH2,FS { [(CH,),SiNSi(CH,),][Si(CH,),NL(n-C,H,),SF iJ[Si(CH,), NSiF(CH,),]} Clathration by water: 16.2.2.3 Reaction with Six,: 15.2.9.2 C12H28% ClIH33F4N3Si6 (i-C,H,),Si=Si(C,H,-i), { [CH,),SiNSi(CH,),][CH,),SiNFormation: 15.2.4.1.3 SiF,(CH,)I,1 C1ZH30N2si2 Reaction with cyclotrisilazane: 15.2.9.2 (CH,),Si[NSi(CH,),],Si(CH,),NSiF[(CH,)2HC]2SiNHSi[CH(CH3)2]2NH $iF* Formation: 15.2.9.1 (CH3)2 C12H30NZSn2 Formation by ring contraction: 15.2.9.2 [(CHd3CNSn(CH3)212 C, ,H8CIzFeO, Formation of BN dimers: 15.2.5.1.1 [q5-ClC(0)C,H4] ,Fe C12H31N3Si2 Reaction with HOROH: 15.2.13.2.1 C(CH,)3NH“Si(CH3)2I,C(CH,), Reaction with RSiF,: 15.2.9.2 CI ZHIOB2S2 (C6H5BS)2 C12H33B3S3 Formation: 15.2.6.4 (HztBSC,H,-t), Formation: 15.2.6.3 Cl2HloC~zNzPz (C1PNC6H5)2 C12H36B3P3 Formation: 15.1.1.3.1 [(CH~)~BP(CH~)Z], C,,Hl0CI,Si Formation: 15.1.1.3.1 Cl,H,6ClSi6 C12Si(C6H5)2 1 Reaction with [(CH,),SiS],: 14.2.9.4 (CH,)2~i[(CH,)2Si]3(CH3)2SiSi(CH3)2Cl Formation: 15.2.4.2.1

Compound Index C12H36Si5

(CH,),~iC(CH,)2~il,~~(CH,), Isomerization: 15.2.4.3 Pyrolysis: 15.2.4.3 C(CH3),Si14Si Formation: 15.2.4.3 C12H36Si6

(CH,)2Si[(CH,)2Si]3(CH3)SiSi(C~3)3 Formation: 15.2.4.3 [(cH3)2si16 Formation: 15.1.1.3.1, 15.2.4.1.2, 15.2.4.1.3, 15.2.4.2.1 Photolysis: 15.2.4.2.3 Pyrolysis: 15.1L3.4 Reaction with HSiC1,-PtCI,: 15.2.4.2.1 Reaction with R,SiCI,-AICI,: 15.2.4.2.1 Redistribution: 15.2.4.3 [(CH3),Si],S~=Si[Si(CH3),], Formation: 15.2.4.1.3 C12H37N30Si5

{ [CH,OSi(CH,),NSi(CH,),][HNSi(C* H3)2lC(CH,),SiNSi(CH~)zl} Formation: 15.2.9.2

395

{ [(CH,),SiNSi(CH,),][(CH,),SiNSiF(C*

H3)212} Reaction with cyclotrisilazane: 15.2.9.2 CI,HJ,N,OSi, { [C2H,0Si(CH3)2NSi(CH3)2][HNSi(CH,)21C(CH,)3SiNSi (CH,),I} Formation: 15.2.9.2 C13H41 FN4Si6

{ [CH,),SiNH],[(CH,),SiNSiFCH,NCSi(CH,)3121~ Formation: 15.2.9.2 C14CrO3 C14CrO3 Formation: 16.4.2.1.3 C14H14Hg (C6H5CH2)2Hg Reaction with RASH,: 15.2.3.3 C14H16S2Si2

[CH3(C6H5)SiSl2 Equilibration: 14.2.9.5 6si

(C6H5)2Si(CH3)2 Formation: 15.2.4.2.2 C14H21As3M002

Formation by ring contraction: 15.2.9.2 C12H37N3Si5

[(CH,)3SiNSi(CH,)2]2NHSi(CH,), Formation: 15.2.9.2 C(CH,),SiN~i(CH,)212C"Si(CH,),I Formation: 15.2.9.2

(CH,),Si[NSi(CH,),],Si(CH,),NHSi(C*

HJ3 Formation: 15.2.9.2 Formation by ring contraction: 15.2.9.2 Ringexpansion: 15.2.9.2

C13F6U C13UF6

Formation: 16.4.2.1.1 C, ,HI ,As,Ti (v-~C,H,)~T~(A~CH,), Formation: 15.2.3.5

V- ~ C,(CH ,)~MO (CO )~(~-~CH ,A ~A ~A ~C* H3) Formation: 15.2.3.5

C14H25BN2

[CH3CH(CH3)CH2NH],BC6H5 Condensation polymerization: 15.2.5.5.1 C(CH,)zCHCH,NHI,BC,H, Condensation polymerization: 15.2.5.5.1 C(CH3),CNHI2BC6H5 Condensation polymerization: 15.2.5.5.1 C,,H29F,LiN,Si, { [(CH,),FSiNSi(CH,),][(CH,),SiNLi]* [(CH,),SiNSiF,C,H,]} LiF-elimination: 15.2.9.2

Formation: 15.2.4.3 [(CH&zsil, ~C(CH,)3~iN~i(CH,)21CNH~i(CH,)21C(C* Formation: 15.2.4.1.2 H,),SiNSiF,C(CH,),]} Redistribution: 15.2.4.3 Formation: 15.2.9.2 C14H42Si8 (CH,),Si[NSi(CH,)21,Si(CH,)2NH(CH,),,Si, SiF2C* Formation as a cage structure: 15.2.4.4 (CH,), CI 4H44N4Si6 Formation by ring contraction: 15.2.9.2 [(CH,)3SiNHSi(CH3)2NSi(CH3)2]2 Formation: 15.2.9.2 C13H38F2N4Si5 { C(CH,),SiNH],[(CH3),SiNSiF2NC(C* [(CH3),SiNSi(CH,)2NHSi(CH,),]2 Formation: 15.2.9.2 Hd,Si(CH,)31} Formation: 15.2.9.2 C13H37F2N3Si5

396

Compound Index

C15H28N2Si

C16H36FP

(n-C4H9)4PF Clathration by water: 16.2.2.3

(C,HS)CH,S~[N(C~H,)~]~ Reaction with R,SiCI,: 14.2.9.4 C,,H,,CISi, (CH,),C6H2Si[Si(CH,),1,C1 Photolysis: 15.2.4.2.3

C16H3604S2Si2

C15H30Si3

C16H36SbZ

(CH3)3C6H2Si[Si(CH3)31ZH

Photolysis: 15.2.4.2.3 C,5H32FZLiN3Si,

{ [(CH,),SiFNSi(CH,),][(CH,),SiN-

Li][(CH,),SiNSiF (CH3)C6HS1} LiF-elimination: 15.2.9.2 C1SH40Si5

CCH,(C,H,)Sil, Formation: 15.2.4.1.3 C15H45N3Si6

(CH3),Si[NSi(CH3),],Si(CH,),N[Si(C*

Hd312 Formation by ring contraction: 15.2.9.2 [(CH3)3SiNSi(CH3)213 Formation: 15.2.9.2

C16BF4

C16[BF41

Formation: 16.4.2.1.1

C16H10M0206

[v~C~H,MO(CO)JZ Reaction with c-(RAs),: 15.2.3.5 C,6H320Si3

(CH~),C,HZS~[S~(CH,)~~~~CH, Photolysis: 15.2.4.2.3

cl 6H32Si3

(CH~),C,HZS~[S~(CH~),I,CH, Photolysis: 15.2.4.2.3

'1

6H35N02

r("-C,H,),NHl[n-C,H,CO,I Clathration by water: 16.2.2.3 C16H36B4C14N4

[(CH3),CNBCII, Formation synthesis: 15.2.5.1.1 Thermal and hydrolytic stability: 15.2.5.2 C16H36BrN

[("-C4Hd4NIBr Clathration by water: 16.2.2.3 C16H36C1N

[(n-C4H9),NIC1 Clathration by water: 16.2.2.3 C16H36FN

[(n-C4H9)4NIF Clathration by water: 16.2.2.3

[c(cH3)30]2~is}2 Formation: 14.2.9.4 (n-C,H,),SbSb(C,H,-n), Thermal decomposition: 15.2.3.4

C16H36Sb4

C-(t-C,HgSb), Formation: 15.2.3.4

C16H38N2Si2

[(CH,)3C]2SiNHSi[C(CH,),I,NH 1

Formation: 15.2.9.1 C, 6H4,CIzOSi4

C1(C,Hs),SiO[(C,H5),Si]3Cl Formation: 15.2.4.2.2

c1 6H,00Si4

1

(C2H,),Si[(C2Hs)zSiI z(CzHs)2SiO Formation: 15.2.4.2.2 C16H40Si4

Reaction with RC=CR-PdCI,: 15.2.4.2.2 Reaction with HOAC 15.2.4.2.2 Reaction with MCPBA: 15.2.4.2.2 Reaction with ROH: 15.2.4.2.2 Reaction with LiAlH,: 15.2.4.2.2 Reaction with X,: 15.2.4.2.2 Reaction with CCI,-O,: 15.2.4.2.2 Reaction with HX: 15.2.4.2.2 C, ,H4,BrSi4 H[(C~HS)~S~]& Formation: 15.2.4.2.2 C, ,H4,CISi4 H[(CZ.HS)~S~]~CI Formation: 15.2.4.2.2 c1 6H42Si4

H[(CzHs)zSi],H Formation: 15.2.4.2.2 C16H48si7

[(CH3),Si]3SiSi[Si(CH,),I,CH, Formation: 15.2.4.3

C16H48Si8

[(CHJ~~IB Formation: 15.2.4.1.2 Redistribution: 15.2.4.3 16H48Si,

(CH3)i6Si, Formation as a cage structure: 15.2.4.4 C16H48Si10

(CH3)16Si10 Formation as a cage structure: 15.2.4.4

397

Compound Index C18H35F2N3Si3

C17H37N03

[C(CH,)3NSi(CH,)2]2NSiF,C,H,

[(n-C4H9)4NIHCO3 Clathration by water: 16.2.2.3

Formation: 15.2.9.2

C17H39N30Si5

C18H36Si3

H3)2°C6H5 Formation by ring contraction: 15.2.9.2 C17H43N3Sis

C18H40FN

(CH,),Si[NSi(CH,),],Si(CH3),NHSi(%* (CH,),Si[NSi(CH3),],Si(CH,),NHSi(C*

H3),CCC4H9 Formation by ring contraction: 15.2.9.2 { [C4H9CCSi(CH,),NSi(CH,),1 [HNSi(C* H,),IC(CH,),~~N~i(CH,~,I~ Formation: 15.2.9.2

C17H45F4N3Si6

{ [(CH,),SiNSi(CH,),] [(CH,),SiNSiF,C* (CH,),I,} Formation: 15.2.9.2

(CH,),Si[NSi(CH,),],Si(CHJ2N[SiF,C* (CH31312 Formation of ring contraction: 15.2.9.2

'1

(C6F5NBC1)3 Formation synthesis: 15.2.5.1.1 I SH1

A1(0C6H5)3 Product in borazine: 15.2.5.1.1 C18HlSBO3 B(°C6H5)3 Preparation of borazines: 15.2.5.1.1

C18HS4F4N6Si10

{[(CH,),siF"si(CH3)212si(CH,),CNSiFCH,]}, Formation: 15.2.9.2 C18H54Si9

[(CH3)2silg Formation: 15.2.4.1.2 Redistribution: 15.2.4.3 C18H54Si10

C19F6Xe C19XeF6

Formation: 16.4.2.1.1 C19H2sAssMo2O4

[q-5C5H5Mo(CO)2],[catena-q2,p-

C18H15B3S3

(C6H5BS)3 Formation: 15.2.6.3 C18HlSP (C6H5)3P Formation: 15.2.3.2

(CH,As)sl Formation: 15.2.3.5 C19H30B3NS

1

CH3N[B(NHC6H5)N(CH3)12Bc4H9

Polymerization: 15.2.5.5.2

C18H15P3

C19H38Si3

(CH3)3C,H,Si[Si(CH3),l2c4H9-t

c-(C6H5P)3 Formation: 15.2.3.2 CI8Hd33N3 (C6H5NBH)3 Copolymerization: 15.2.5.5.2 (HNBC,H,), Copolymerization: 15.2.5.5.2

Photolysis: 15.2.4.2.3

C19H39N3Si5

(CH3)3Si[NSi(CH,),],Si(CH3)2 NHSi(G*

-

H3)2CCC6H5 Formation by ring contraction: 15.2.9.2

C20H40Si5

Ph(CH3)SiOSi(CH,),0(Ph)(CH,)SiOSi*

[(n-C4H9)3(i-C6H5)NIF Clathration by water: 16.2.2.3

CH,C(O)O[(C,H,)2Sil,H Formation: 15.2.4.2.2

C15CrO,CO,CCF,I, Formation: 16.4.2.1.3

(C6H5NBC1)3 Formation: 15.1.1.3.1

C18H32FN

C18H4402Si4

C19CrF606

C18H1SB3C13N3

(CH3)2 Formation: 15.2.8.1

C(i-C5Hi I ) ~ ( ~ - C ~ H ~ ) ~ N I F Clathration by water: 16.2.2.3

(cH3)18si10 Formation as a cage structure: 15.2.4.4 CI8HS4Sill (cH3)18si11 Formation as a cage structure: 15.2.4.4

8B3C13F1 g N 3

C18H2803Si4

(CH,),C6H,Si[Si(CH3)J2C3H,-i Photolysis: 15.2.4.2.3

I

[CH2(CHMil5 Formation: 15.2.4.1.3 C20H43N02

C(n-C4H,),NICn-C3H,C0,1 Clathration by water: 16.2.2.3

398

Compound Index

C20H44CIN

(i-C,H, 1)4NC1 Clathration by water: 16.2.2.3

C20H60Si10

[(cH3)2silio Redistribution: 15.2.4.3 C20H61F2NSsi9

C20H44FN

[(CH,),Si],NSiFCH,[NSi(CH,),I,(C* H,),SiNHCH,FSiN [Si(CH3),], Formation,by substitution and ring contraction: 15.2.9.2

(ImCSH11)4NF Clathration by water: 16.2.2.3 C20H48Si4

[t-C4HdCH,)Sil, Formation: 15.2.4.1.3

C21H24S3Si3

[CH3(C6H5)SiS13 Equilibration: 14.2.9.5 [(C6H5)CH3SiS13 Formation: 14.2.9.4

C20H50F2N6Si10

{(CH,),SiN[Si(CH,),],"iF2CH3NSiC* H,}Si(CH,),N Si(CHA~NSi(CH3)212 Si(CH,), Formation by substitution and ring contraction: 15.2.9.2 C2oH5oSis [(C~HS)~S~]S Formation: 15.2.4.1.3

C21H24Si2

(C6H5)3SiSi(CH3)3 Reaction with Li: 15.2.4.1.2 C21H27B3N6

(p-CH3C6H4NBNH2)3 Pyrolysis: 15.2.5.5.1

C20H54F3N5Si7

{ [(CH,),Si(CH,),NSiF,NSi(CH,),][Si(C* H,),NSi(CH,),] [CH,SiNSiCH,FNC(C* H3)31} Formation by silylgroup and methanidion migration and ring expansion: 15.2.9.2 C20H54F4LiN5Si,

c2 1H34Si3

(CH3)3C6H2Si[Si(CH3)312C6H5

Photolysis: 15.2.4.2.3

C21H45N02

~

C20H6OF5N5Si9

{ [(CH,),Si],NSiFCH,NSi(CH,),ICSi(C* H,),NSi(CH,),] [CH,SiNSi(CH,),NSi* (CH,),] Formation by silylgroup and methanidion migration and ring expansion: 15.2.9.2

C20H60F6N6Sil

I

{ [(CH,),SiFNSi(CH,),],[(CH,),SiN]}Si* F,N[SiF(CH,),] Si(CH,),[NSi(CH,),],SiF(CH,), Substitution and ring contraction: 15.2.9.2

-

~

4

~

9

~

4

~

1

~

~

1

{(CH,),Si[NSi(CH,),],Si(CH,),NSiF(C*

H3)2}2SiF2 Formation: 15.2.9.2 C22H66N6Si10

{C(CH,),Si"Si(CH,),I,~i(CH,),"si(cH&1}2 Formation: 15.2.9.2

C20H55F4N5Si7

C(CH,),Si(CH,),NSiF,[NSi(CH,),I,Si* (CH,)2NHSiF,NC(CH,),Si(CH,), Formation by substitution and ring contraction: 15.2.9.2 C20H60F2LiN5Si, [(CH,),Si],NSiFCH,[NSi(CH,),I,(C* H,),SiNLiCH,FSiN[Si(CH,),], Reaction with amino-fluorosilanes: 15.2.9.2

~

C22H66F4N6Si1

(CH,),CSi(CH,),NSiF,[NSi(CH3)2]2Si* (CH,),NLiSiF2NC(CH,),Si(CH,), Reaction with amino-fluorosilanes: 15.2.9.2

~

Clathration by water: 16.2.2.3

{(CH3),Si[NSi(CH,),],Si(CH,),[NSi(C*

H,)21}2 Formation by ring contraction and substitution: 15.2.9.2 C22H66Si13

(CH3)22Si13 Formation as a cage structure: 15.2.4.4 2'4' csc24

Formation: 16.4.2.2.1 CuHFz -C24[HF,I Formation: 16.4.2.6.3 C24H20P2

(C6H5)2PP(C6H5)2 Formation: 15.2.3.2 C24H20P4

c-(C,H sp)4 Formation: 15.2.3.2 C24H20S2Si2

[(C6H5)2SiS12 Formation: 14.2.9.4

-

~

,

~

9

~

Compound Index C24H20S6Si4

C27H81F2N6si12

[(CH,),Si],NSiFCH3[NSi(CH3),],(CH-

(C6H5Si)4S6 Formation: 14.2.9.4

,),SiN[SiFCH,N[Si(CH,),] Formation: 15.2.9.2

C24H20Sb2

(C6H5)2SbSb(C6H5)2 Thermal stability: 15.2.3.4 C24H25As3M002

C24H30Fe204

[I-'C,(CH,),Fe(CO)2]2 Reaction with P4: 15.2.3.5 C24H33N3Si3

[(CHd2SiNC6H,I3 Formation: 15.2.9.2 C24H48Si6 I

[CH~(CHZ)~S& Formation: 15.2.4.1.3

C24HS4Si3

[(t-C4H9)2Sil3 Formation: 15.2.4.1.3 C24HS6Si4

[(i-C3H,),Sil4 Formation: 15.2.4.1.3,15.2.4.2.2 HX, I,: 15.2.4.2.2 C24HS7N3Si3

{[(CHd,CI,SiNH)3 Formation: 15.2.9.1 C24H60N4Si4

2]2

C28F4Xe

~-5C5(CH3)5M~(CO)2(p3C,H,AsAsAs*

C6H5) Formation: 15.2.3.5

399

C,,XeF, Formation: 16.4.2.1.1

Formation: 15.2.4.1.3 C28HSSF4N6si10

{(CH,),SiF[NSi(CH,),],Si(CH,),[NSiF*

C6H51}2 Formation: 15.2.9.2 C28H68Si4

C(CH,),CCH,(CH,),Sil4 Formation: 15.2.4.1.3 C28H70si7

[(C#5)2Si] 7 Photolysis: 15.2.4.2.3 C30H2ilP5

c-(C6H5P)5 Formation: 15.2.3.2 Reaction with Mo(CO),: 15.2.3.5 C30H30Si5

(C6H5SiH)5 Formation: 15.2.4.2.1 C30H42B3N3

CCH~(CH~)BNBC~H~IB Produced from tetramer: 15.2.5.4

[(CH3)2HCI2SiNHl4 Formation: 15.2.9.1 C24H60Si6

[(c2H5)2si16 Formation: 15.2.4.2.3 C24K

KC,, Formation: 16.4.2.2.1 Reaction with H,: 16.4.2.4.1,16.4.2.4.4 C24Na

NaC24 Formation: 16.4.2.2.1 C24Rb RbC24

Formation: 16.4.2.2.1

C25HJ0Si5

I

CCH,(CH2)4Sil, Formation: 15.2.4.1.3 C27H72F6N6Si9

(CH,),CSi(CH,),NSiF,[NSi(CH,),I,Si*

(CH3)2NCSiF2NC(CH3)3Si(CH,)31 Formation: 15.2.9.2

Formation: 15.2.4.1.3 C30H64F2N6Si

I0

{ C,H5(CH3)FSi[NSi(CH3),]2Si(CH3)2~

"Si(CH,),ll, Formation: 15.2.9.2 C30H66Si3

{ [(CH3)3CCH,I 2si) 3

Formation: 15.2.4.1.3 Photolysis: 15.2.4.2.3

C30H70SiS

C(n-C,H7)2Sil, Formation: 15.2.4.1.3 C32H3204S2Si2

{[(CH3)2C6H2OI2SiS}2 Formation: 14.2.9.4 C32H48Si2

[2,6-(CH3),C,HJ2Si=Si[C6H3(CH,),-

-

2361 2 Formation: 15.2.4.2.3 C32H64Si8

[CHz(CH2)3Sil8 Formation: 15.2.4.1.3

Compound index

400 C32H72CrN204

[(n-C4H9)4NI,CrO4 Clathration by water: 16.2.2.3 C32H72N204W

[(n-C4H9)4N12W04 Clathration by water: 16.2.2.3

C36H30Si2

(C6H5)3SiSi(C6H5)3 Reaction with Li: 15.2.4.1.2 C36H44Si2

[(CH,),C6H2I2Si=Si[C6H2(cH3)3l Formation: 15.2.4.2.3

C32H72Si4

[(i-c4H9)2sll4 Formation: 15.2.4.1.3 C32H76N4Si4

{[(CH3)3C12SiNH}4 Formation: 15.2.9.1 C34H72NZ04

[(n-C4H9)4NI,C,O4 Clathration by water: 16.2.2.3 C34H7204P2

[(n-C4H9)4P12C204 Clathration by water: 16.2.2.3 C3SH4005Si5

(C,H,SiOCH,), Formation: 15.2.4.2.1

C35H40SiJ

[C6H5(CH3)Si15 Formation: 15.2.4.1.4, 15.2.4.2.1 C35H8SSi5 [CH~(~-C~H,)~S~]S Formation: 15.2.4.1.3 C36H30As6

c-(C6H5As)6 Reaction with Mo(CO),: 15.2.3.5 C36H30Br6Si6

(C6H5SiBr), Formation: 15.2.4.2.1

c36H3016si6

(c6 SSi1)6 Formation: 15.2.4.2.1 Reaction with MR: 15.2.4.2.1 Reaction with ROH: 15.2.4.2.1 Reaction with LiAIH,: 15.2.4.2.1 C36H30N3P3

[N=P(C~HS)~I, Formation: 15.2.11.2 Pyrolysis: 15.2.11.3 C36H30P4

[(C6H5)2P13P Thermal decomposition: 15.2.3.2 c3 6

6 0 ’

c-(C,H 5 Formation: 15.2.3.2 C36H30Sb6 c-(C6H5Sb)6

Formation: 15.2.3.4

Formation: 15.2.4.1.3 C40H30As6M004

(Co)4Mo[?-2(AsC6H5)61 Formation: 15.2.3.5 C40H30As6M004P6

(C0)4Mo[?Z(PAsC6H5)61 Formation: 15.2.3.5

-

C40H56B4N4

[CH3(CH2)3NBC6H514 Conversion to trimer: 15.2.5.4 C40H80Si10

[CH~(CH~)BS~IIO Formation: 15.2.4.1.3 C40H88CrN204

[(i-c5HlJ4Nl2c~o4 Clathration by water: 16.2.2.3 C4ClH.88N204W

C(i-C,H, d4NI2W04 Clathration by water: 16.2.2.3 C40H88Si4

{[(CH3)3CCH2l,Si}4 Formation: 15.2.4.1.3 C42H48Si6

[C6H5(CH3)Si16 Formation: 15.2.4.1.4 (C$5SiCH& Reaction with HX: 15.2.4.2.1 C42H50Sb4

c-[(CH3)3C6H2Sb14‘C6H6 Formation: 15.2.3.4 C48H40r2Si4

1[(c6H5)2si141 Formation: 15.2.4.2.2 Reaction with H 2 0 : 15.2.4.2.2 C48H40Li2Si4 Li[(C6H5)2Si]4Li Formation: 15.2.4.2.2 C4SH400Si4

(C~H~)~$~[(C~HS)~S~],(C~H~)ZS~O Formation: 15.2.4.2.2

‘4SH40Si4

[(C6H5)2Si14

Formation: 15.2.4.1.1,15.2.4.2.2

Reaction with X,, H X 15.2,4.2.2

Compound Index C,,H,,CISi, H[(C6H5)2Si14C1 Formation: 15.2.4.2.2 C48H4202Si4

Ho[(C6H5)ZSi140H Formation: 15.2.4.2.2 C54H45P9 c-(C,H 5 PI9 Formation: 15.2.3.2 C6OH30PlO c-(C6H5 0 Formation: 15.2.3.2 C60HSOSi5 C(C,H,),Sil, Formation: 15.2.4.1.1 C72H60Si6

C(C6H,),Silti Formation: 15.2.4.1.1 Reaction with HX: 15.2.4.2.1

Hs)3]2 : 15.2.4.2.2 C99MoO3 C99MoO3 Formation: 16.4.2.1.3 C, 50Cr03 c1 5 OCQ Formation: 16.4.2.1.3 CaH,Ol,Si,*Al, CaH60,,Si,*A1, CaH,O 12Si2*A1, CaH ,O, lSi,*AI, CaH, ,OI8Si,*A1, CaHl,0,,Si6*AI, CaH, ti032Silo*A1, CaH ,Na,O2,Si5*Alz CaH,,O,,Si,*AI, Ca,FKO,,Si, KF, Ca,Si,O,, Structure: 16.4.7.1 CdCI, CdC1, Structure: 16.4.3.1 CdI, CdI, Structure: 16.4.3.1 CdPS, CdPS, Formation and structure: 16.4.4 CdPSe, CdPSe, Formation and structure: 16.4.4

,

40 1

CI*Br CI*CH3 CI*C2H5 ClCrO CrOCl Electrochemistry: 16.4.5 Structure: 16.4.5 CIF C1F Reaction with graphite: 16.4.2.1.1 CIF0,S S0,CIF Anodic oxidation of graphite: 16.4.2.8.1 CIF, CIF, Preparation of (C,Fx)": 16.4.2.6.3 Reaction with CnF: 16.4.2.6.1 Reaction with graphite: 16.4.2.1.5, 16.4.2.6.1 CIF, ClF, Reaction with graphite: 16.4.2.1.5 ClFeO FeOCl Reaction with NH,: 16.4.5 Structure: 16.4.5 CIH0,S CIS0,H Anodic oxidation of graphite: 16.4.2.8.3 CIHO, HC10, Anodic oxidation of graphite: 16.4.2.8.3 Reaction with graphite: 16.4.2.1.4

c1h4n

"H4lC1 Synthesis of borazines: 15.2.5.1.1 CII ICI Graphite residue compound: 16.4.2.7.2 Reaction with (SN),: 15.2.12.4 Reaction with graphite: 16.4.2.1.5 ClKO KOCl Carbon surface oxidation: 16.4.2.7.3 CIKO, KCIO, Oxidation of graphite: 16.4.2.5.2 CIKO, KC10, Anodic intercalation into graphite: 16.4.2.8.3 CILiO, LiCIO, Electrochemistry: 16.4.5

Compound Index

402

~

CIN*CH, C1N*C,H12 CIN*C,,H,, CIN*CZoH,, ClNS SNCl Formation: 15.2.12.2, 15.2.12.5 CIN,S, [S,N,ICI Formation: 15.2.12.2 S,N,CI Thermal decomposition: 15.2.12.2 CIOSi2*C,HI3 ClOTi TiOCl Structure: 16.4.5 ClOV VOCl Electrochemistry: 16.4.5 Structure: 16.4.5 CIO,

c10,

Clathration by water: 16.2.2.1 CISi*C2H, ClSi*C,H, ClSi3*C,,H,, CISi,*C,6H41 CISi4*C4,H,, CISi,*C,,H,, CI, c 1 2

Clathration by water: 16.2.2.1 Graphite residue compound: 16.4.2.7.2 Reaction with [S,N,CI]Cl: 15.2.12.5 Reaction with (SN),: 15.2.12.5 Reaction with carbon surface: 16.4.2.7.3 Reaction with graphite: 16.4.2.1.5 CI2*CH, CI,*CH,As C12*C,H4 CI,*C,H,As CI,*C,H,B CI2*Cd

CI,CrO, CrO,C1, Graphite residue compound: 16.4.2.7.2 Reaction with graphite: 16.4.2.1.2 Cl,FeO,*C,,H, Cl2NSi2*C4H, C12N,P,*C,,H,, CI,N,S, [S,N,CI]CI Formation: 15.2.12.2

,

Reaction with SO,CI,: 15.2.12.5 Reaction with CI,: 15.2.12.2,15.2.12.5 CI,N,Si4*C,H,, C12N3*C4H12B3 C12N3*C6H16B3 C12N3*C9H14B3 CI,OS SOCI, Reaction with surface oxides on carbon: 16.4.2.7.3 C120Si4*C16H4, CI,O,S SO,CI, Reaction with [S,N,CI]Cl: 15.2.12.5 C1,O,Si,*C,Hl, CI,O,U UO,CI, Reaction with graphite: 16.4.2.1.2 CW, c120, Reaction with graphite: 16.4.2.1.3 CI,S SCI, Reaction with N, to form SN: 15.2.12.1 CIZS, S2CI2 Reaction with NH,: 15.2.12.2 Cl,Sb*C,H, CI,Si*C,H, CI,Si*C,H, Cl,Si*C,H,, CI,Si*C,H,, Cl,Si*C,H,, CI,Si*C,H, CI,Si*C,H,, CI,Si*C,,H,, Cl,Si*C,,Hl, C12Si2*C,H,, CI,Si,*C,H,, CI,Ti*C,,H,, C13*AI CI,*As CI,*B

Cl,*Bi CI,*CBr CI,*CH C13F15N3*C18B3 CI,Fe FeCI, Graphite residue compound: 16.4.2.7.2 Reaction with graphite: 16.4.2.1.1 CI,HNP CI,P=NH Formation: 15.1.1.3.3

Compound Index Reaction with [PCl,][PCl,]: C13H3N,*B, CI,N*C,H, ,B CI,N*C,H,B CI,N*C,H 5B CI,N*C,,H,,B CI,N,*C,H,B,

15.1.1.3.3

C13N3*C18H15B3

CI3N3S3 (NW, Formation: 15.2.12.2, 15.2.12.5 Reaction with Ph,Sb 15.2.12.2 Reaction with R,SiN,: 15.2.12.3 Reaction with R3SiBr: 15.2.12.4 CI30P OPCI, Reaction with H,O: 15.2.10.1 CI,O,*B, CI,P PCI, Reaction with RNH,: 15.1.1.3.1 Redistribution reactions: 15.1.3.3.1 CI,S,*B, CI,*C CI,Ge GeCI, Redistribution reactions: 15.1.3.3.1 CI,MoO MoOCI, Reaction with graphite: 16.4.2.1.2 C14N4*C16H36B4 CI,N,S, (NSCI), Formation: 15.2.12.5 CI,04*B, CI,S,Si, (CI,SiS), Formation: 14.2.9.4, 14.2.9.6 CI,S,Si3 (CI,SiS,),Si Formation: 14.2.9.6 CI,Si SiCI, Redistribution with (RO),Si: 15.1.3.3.1 Redistribution with (R,N),SI: 15.1.3.3.1 CI,Si,*C,H6 CISHN,P, CI,P=NPCl,=NH Formation: 15.1.1.3.3 CI,P PCI, Inhibitor for (NPCI,), polymerization: 15.2.1 1.1

403

Reaction with NH,: 15.1.1.3.3 CI,H,Pt H,PtCI, Hydrosilylation catalyst: 15.2.13.2.1 C16N3P3

[CI,P=N], Reaction with PCI,: 15.1.1.3.2 [N=PCI,], Pyrolysis: 15.1.1.3.4 (N=PCI,), Pyrolysis: 15.2.11.1, 15.2.11.3 C~rIN,P, (N=PCI,), Pyrolysis: 15.2.11.1 CI,S,Si, (CI,SiS), Formation by ring expansion: 14.2.9.4 C~IOP2

[pc141[pc161

Reaction with NH,: 15.1.1.3.3

C111N3P4

C1,P=NPCI,=NPCI,=NPCI4 Formation: 15.1.1.3.2

C112NP3

[CI,P=NPCl,] [PCI,] Formation: 15.1.1.3.3 Reaction with NH,: 15.1.1.3.3

C112N6P6

(N=PCI,), Formation: 15.2.11.1 Co*C10H10 CoCrCsF, CsCoCrF, Structure: 16.3.1.2 CoCsF,O,Ta CsCoTaO,F, Structure: 16.3.1.2 CoCsF,Fe CsCoFeF, Structure: 16.3.1.2 CoKP,*C,H,, COPS, COPS, Formation and structure: 16.4.4 CoPSe, CoPSe, Formation and structure: 16.4.4 c0304

COP4 Reaction with Na,O,: 16.4.6 Cr*C1,H1, CrCsF,*Co

404

Compound Index

CrCsF,Fe CsFeCrF, Structure: 16.3.1.2 CrCuF,K KCuCrF, Structure: 16.3.1.2 CrF,O, CrO,F, Reaction with graphite: 16.4.2.1.2 CrF,KNi KNiCrF, Structure: 16.3.1.2 CrF,O,*C,, CrK,O, K,CrO, Carbon surface oxidation: 16.4.2.7.3 CrLiS, LiCrS, Formation: 16.4.3.2 CrN,O4*C3,H,, C~N,O,*C,,HBB CrNaO, NaCrO, Formation: 16.4.6 CrO*Br CrO*C1 CrO,*CI, CrO, CrO, Reaction with graphite: 16.4.2.1.2, 16.4.2.1.3, 16.4.2.1.4 CrO,*C,, CrO3* c 1 5 0 (3203 Cr203

Reaction with Na,O: 16.4.6 CrP, (320,

Formation: 16.4.2.1.3

Cr2S3 Cr2S3

Reaction with CS,: 16.4.3.2 Cr30, CrAl Formation: 16.4.2.1.3 Cs*C,As Cs*C,Bi cs*c, Cs*C,As Cs*C,Bi CS'C,,

cs*c,,

CsCuF,Fe CsCuFeF, Structure: 16.3.1.2 CsF,MnMo03 CsMnMoO,F, Structure: 16.3.1.2 CsF,MoNiO, CsNiMoO,F, Structure: 16.3.1.2 CsF,Mo03Zn CsZnMoO,F, Structure: 16.3.1.2 CsF,NbNiO, CsNiNbO,F, Structure: 16.3.1.2 CsF,NbO,Zn CsZnNbO,F, Structure: 16.3.1.2 CsF,NiO,Ta CsNiTaO,F, Structure: 16.3.1.2 CsF,O,Ta*Co CsF,*CoCr CsF,Fe*Co CsF,Fe*Cr CsF,FeMn CsFeMnF, Structure: 16.3.1.2 CsMnFeF, Structure: 16.3.1.2 CsF,FeNi CsNiFeF, Structure: 16.3.1.2 CsF,Fe, CsFeFeF, Structure: 16.3.1.2 CsF,Mg*AI CsGe CsGe Thermal decomposition: 16.2.3 CsH CsH Reaction with graphite: 16.4.2.4.2 CsSi CsSi Thermal decomposition: 16.2.3 CS,Si,36 cs7sil

36

Formation: 16.2.3

cu cu Reaction with graphite: 16.4.2.1.3

Compound Index CuF6Fe*Cs CuF6K*Cr cu,o,,p, cUz[p4oi21 Reaction with Na,S: 15.2.10.2 Eu*C6 F*CH3 F*C2H5 F*CI FH HF Anodic oxidation of graphite: 16.4.2.8.3 Preparation of (C4FX)":16.4.2.6.3 FH0,S FS0,H Anodic oxidation of graphite: 16.4.2.8.3 Fluorination of graphite oxide: 16.4.2.5.3 FKOzOSi8*Ca4 FLiNSi*C,H,, FLiNSi*C8H19 FLiN,Si *C H FN*C16H36 FN*Cl,H,, FN*C18H40 FN*C20H44 FN403S*B4 FN4Si6*C1,H4, FNbO ,Rb RbNb0,F Structure: 16.3.1.2 FO,S*CI FO,RbTa, RbTa,05F Structure: 16.3.1.2 FP*C16H36 FS*C12H27 F, F, Fluorination of graphite oxide: 16.4.2.5.3 Preparation of (C4FX)":16.4.2.6.3 Reaction with graphite: 16.4.2.1.1,

,

,,

16.4.2.1.5, 16.4.2.6.1

F2*C24H

F2H,N*B F,I,Si F,SiI, Reaction with HgS: 14.2.9.4 F,Kr KrF, Reaction with graphite: 16.4.2.1.1, 16.4.2.1.2

F,LiN3Si5*CloH,o

F2LiN,Si5*Cl,H3, F,LiN,Si,*C,oH60 F,NSi,*C7H,, F2N3Si?.*C18H35 F2N3Si5*C13H37 F2N3Si6*C13H39 F2N4Si5*C13H38 F2N5Si9*C20H61 F2N6Si10*C20H50 F2N6Si10*C30H64 F2N6Si12*C27H81 F,O,*Cr F2°6*C10H12As4 F,O,S, S2°6F2

Reaction with BN: 16.4.2.1.2 Reaction with graphite: 16.4.2.1.2

F,Si SiF, Formation: 15.1.1.3.2 F,Xe XeF, Reaction with graphite: 16.4.2.1.1 F,*A1 F3*As F,*B F,*Br F,*CI F,H,N*B F,H,N,*B, F,LiN,Si,*CgH,, F,LiN,Si,*C,,H2, F,MnMoO,*Cs F,MoNiO,*Cs F,MoO,Zn*Cs F,N NF, Reaction with COS: 15.2.12.1 F,N*C,H,B F,NOPSi*C,H,, F,NS NESF, Trimerization: 15.1.1.3.1 F,NSi,*C,H,, F,N,O,S, "=S(O)FI, Formation: 15.1.1.3.1 F,N,S, (N=SF), Formation: 15.1.1.3.1 F3N5Si7*C20H54 F,OV VOF, Reaction with graphite: 16.4.2.1.2

405

406

Compound Index

F,02*C2H F3O,*B3 F303S*CH F3S,*B, F,Si*C,H, F,*C F4*C16B F,Ge GeF, Reaction with graphite: 16.4.2.1.1 F4H*B F,K*B F,LiN 5Si,*C20H F4N02*B F4N3Si6*C1lH,3 F4N3Si6*C17H45 F4N5Si7*C20H55 F4N6Si10*C18H54 F4N6Si10*C28H58 F4N6Si1 1*C2.2H66 F,NbNiO,*Cs F,N bO Zn*Cs F,NiO,Ta*Cs F,OS OSF, Reaction with NH,: 15.1.1.3.1 F,OXe XeOF, Reaction with graphite: 16.4.2.1.2 F,O,Ta*CoCs F4S SF, Reaction with Si,Br,: 14.2.9.4 F,S,Si, (F2SW2 Formation: 14.2.9.4 F,Si SiF, Reaction with Si: 15.1.1.3.2 F4Ti TiF, Reaction with graphite: 16.4.2.1.1 F ,Xe XeF, Reaction with graphite: 16.4.2.1.1 F,Xe*CZ8 F,*As F,*Br F5*CBAs F,*Cl FSI

,,

IF5

Reaction with graphite: 16.4.2.1.5

F,IO IOF, Reaction with graphite: 16.4.2.1.2 F5N*C6H2 F5N3Si6*C10H30 F5N5Si9*C20H60 F,Nb NbF, Reaction with graphite: 16.4.2.1.1 FSP PF, Reaction with C n F 16.4.2.6.1 Reaction with graphite: 16.4.2.1.1 F,Sb SbF, Reaction with C n F 16.4.2.6.1 Reaction with graphite: 16.4.2.1.1 F5Sb*C6 F,Ta TaF, Reaction with graphite: 16.4.2.1.1 F6*C2 F,*C8AS F,*CoCrCs F,Fe*CoCs F,Fe*CrCs F,Fe*CsCu F,FeMn*Cs F,FeNi*Cs F Fe * Cs F,K*As F,K*CrCu F,KNi*Cr F6KP KPF, Anodic intercalation into graphite: 16.4.2.8.3

F,KSb KSbF, Anodic intercalation into graphite: 16.4.2.8.3

F6K,Mn K,MnF, Fluorination of graphite oxide: 16.4.2.5.3 F,Mg*AlCs F,MgRb* A1 F6Mo MoF, Reaction with graphite: 16.4.2.1.1 F,Mo*C, 1 F6N3P3

(N=PFz), Pyrolysis: 15.2.11.3

Compound Index Reaction with Na[Fe(CO),Cp-q5]: 15.2.11.4 F6N6Si9*C27H72 F6N6Sii l*C20H60 F60,*As F,O 2* As, F606*ClgCr F60s OsF, Reaction with graphite: 16.4.2.1.1 F,Os*C, F6S

SF, Reaction with Si2R6:14.2.9.4 Reaction with N, to form SN: 15.2.12.1 F6U UF6

Reaction with graphite: 16.4.2.1.1 F6U*C13 F6Xe XeF, Reaction with graphite: 16.4.2.1.1, 16.4.2.1.2 F6Xe*C F,I IF, Reaction with graphite: 16.4.2.1.2, 16.4.2.1.5 F9P2*C3H2 F12P2*C4 F12P4*C4 F l sN3*'1 8B3C13 F15P.3*C5 Fl,N,O,P,*Cl,H12 Fe Fe Reaction with graphite: 16.4.2.1.3 Fe*C10H10 Fe*C13 Fe*CoCsF, Fe*CrCsF6 Fe*CsCuF, FeHO, FeO(0H) Formation: 16.4.5 FeH,NO FeONH, Formation: 16.4.5 FeMn*CsF, FeN2*C12H16 Fe NaO,*C,H, FeNa204*C,

407

FeNa,0,,Si4 Na,[FeSi401 ,] Formation: 16.3.1.6 FeNi*CsF, FeO*C1 Fe02*Cl2H8C12 Fe02*C12H14 FeO,*C, FePS, FePS, Formation and structure: 16.4.4 FePSe, FePSe, Formation and structure: 16.4.4 FeP5*ClOHl5 Fe2*CsF, Fe,04*C24H30 Fe20,P4*C1~H12 Ge Ge Reaction with elemental K: 16.2.3 Ge*C,H,Br, Ge*CI, Ge*Cs Ge*F, GeK KGe Thermal decomposition: 16.2.3 GeN4*C,H2, GeNa NaGe Thermal decomposition: 16.2.3 GeO,*C,H GeRb RbGe Thermal decomposition: 16.2.3 Ge3S3*C6H18

,

Ge46K8 K8Ge46

Formation: 16.2.3 Ge46Rbx RbxGe4, Formation: 16.2.3 NaxGe136

Formation: 16.2.3 H*BF, HCl,*C H*Cs H*F HF2*C24 HF302*C2

Compound index

408

HF,O,S*C HI HI Reaction with B,O,: 15.2.6.2 HIO, HIO, Reaction with graphite: 16.4.2.1.4 HIO, HIO, Anodic oxidation of graphite: 16.4.2.8.3 Reaction with graphite: 16.4.2.1.4 HK KH Reaction with graphite: 16.4.2.4.2 HLI LiH Reaction with graphite: 16.4.2.4.2 HNO, HNO, Anodic oxidation of graphite: 16.4.2.8.3 Carbon surface oxidation: 16.4.2.7.3 Graphite residue compound: 16.4.2.7.2 Intercalation into graphite oxide: 16.4.2.5.3 Oxidation of graphite: 16.4.2.5.2 Reaction with graphite: 16.4.2.1.4 HNP*C13 HN,P,*CI, HNa NaH Reaction with graphite: 16.4.2.4.2 HNa,O,P, Na,H[P,OgI Formation: 15.2.10.2 HO,*Fe HO,S*CI HO,S*F H0,Se HSeO, Reaction with graphite: 16.4.2.1.4 HO,*CI HRb RbH

Reaction with graphite: 16.4.2.4.2

H,

H2 Reaction with MC,: 16.4.2.4.1 Reaction with MC,,: 16.4.2.4.4 Reaction with carbon surface: 16.4.2.7.3 H,*C, H,CI,*C H,F,N*C,

H2F9P2*C3 H,LiN LiNH, Synthesis of B-N polymers: 15.2.5.3 H,N*B H,N*BF, H,NNa NaNH, Preparation of boranamine polymers: 15.2.5.1.2 H,NO*Fe H,N,*C H,NaO,P Na[H,PO,I Dehydration: 15.2.10.1, L 5.2.10.2 H,O H2O Intercalation into graphite oxide: 16.4.2.5.3 HzO, H202 Catalytic decomposition by graphite oxide: 16.4.2.5.3 H,OJ H2S04

Anodic oxidation of graphite: 16.4.2.8.1, 16.4.2.8.3 Graphite residue compound: 16.4.2.7.2 Reaction with graphite: 16.4.2.1.4 H,O,Se H,SeO, Anodic oxidation of graphite: 16.4.2.8.3 H,Pt*Cl, H,S H,S Clathration by water: 16.2.2.1 Formation of carbon surface sulfides: 16.4.2.7.3 Reaction with B,S,: 15.2.6.4 Reaction with RBX,: 15.2.6.2 Reaction with R,SiX,: 15.1.1.3.1 Reaction with BX,: 15.1.1.3.1, 15.2.6.2 Reaction with BBr,: 15.2.6.4 Reaction with V,O,: 16.4.3.2 Reaction with N, to form SN: 15.2.12.1 Reaction with graphite oxide: 16.4.2.5.3 H2S4*B2

H,Se H,Se Clathration by water: 16.2.2.1 H,*As H,AsCl,*C

Compound Index H,AsI,*C H,AsNa,O,*C H,Br*C H,CI*C H,F*C H31*C H3N H3N Reaction with B,H,: 15.1.1.3.1 NH, Cathodic reduction of graphite: 16.4.2.8.1, 16.4.2.8.2 Formation of boranamine polymers: 15.2.5.1.2 Intercalation into graphite oxide: 16.4.2.5.3 Reaction with FeOC1: 16.4.5 Reaction with PCI,: 15.1.1.3.3 Reaction with OSF,: 15.1.1.3.1 Reaction with S,CI,: 15.2.12.2 Reaction with R,SiX,: 15.1.1.3.1 Reaction with (C4FX)":16.4.2.6.3 Reaction with carbon surface: 16.4.2.7.3 H,N*BF, H3N*C2 H,N02*C H,NS,Ta TaS,-ND, Formation: 16.4.3.4 H,N3*B,CI, H,N,*B,F, H303*B H,O,*B, H,O,*As H3P PH, Clathration by water: 16.2.2.1 H,S,*B, H,*C H,*C, H,CI,*C, H,&*B H,N*Cl H,NO,TaW

[NE141

CTaW06i

Formation: 16.3.1.2 H,Na,Ol,Si,*A1, H,O*C H4°2*C2 H,O,SiZ*A1, H,O,,Si,*AI,Ca H,S*C

409

H,As*C H,AsCI,*C, H,BCl,*C, H,Br*C, H,CI*C, H,CI,Sb*C, HSF*C2 H,F,Si*C, H,Fe NaO,*C, H,N,O,TaW [N2HSI[TaWo61

Formation: 16.3.1.2 H,NaO*C2 H,BBr*C, H,BN*C H6*B2 H,Br,Ge*C, H6*C2 H6*C6 H,CIN*C H,Cl,Si*C, H,CI,Si,*C, H,N*B H6N3*B3 H60*C, H,OS*C, H6°3*C4 H6O3S*C, H,O,S*C, H,O,Si, Si,O(OH), Structure: 16.4.7.1 H,O,,Si,*AI,Ca H,S*C, H,S3Si, CH2SiSI3 Formation from [H,SiS],: H,BC13N*C6 H,Br*C, H,CISi*C2 H,N*B2 H,N*C, H7P*C, H,P*C6 H,Sb*C6 H,BN*C H,*C3 H**C, H,CI,FeO,*C, H,CI2Si*C, H,CI2Si*C, HsN,*B,

,

14.2.9.4

410

Compound index

HSN*O,S (NH4)2S04 Catalyst for ring expansion: 15.2.9.2 H,N@SS, “H412CS2081 Reaction with graphite: 16.4.2.1.4 H,N,*B, H8Na2016Si,*AI H80*C4 H8°2*C4 H,02S*C4 H,Ol2Si2*AI2Ca H,Si*C6 H,AsO,*C, H,B*C, H,BF3N*C, HyBN2*C, H,BO,*C, H9B2N*C H,B,CI,N,*C, HY*B5 H,BrSi*C, HyCISi*C4 HyLi*C4 H,O*C, H,O,P*C, H9P*C3 H10As5M02*C10 H,,BN*C, HlOB2S2*C12 H10B2S4*C4 H10*B6 HlOBlOLi2*C2 H ,,Br2Si*C4 H10*C4 HlOC~2N2PZ*ClZ HlOCl2Si*C4 HloCI2Si*Cl2 H, oC12Ti*Cl H 1oco*c 10 H10Fe*C10 Hlo12Si*C4 HloLi2Si*C12 H10M0206*C16 H10M02P5*C10 H10N6*B6 H10°*C4 H10°2*C4 H10°2S*C5 H l o 0 2,Si2*AI2Ca Hl0O2,Si,*Al2Ba H10P2*C3

H,oRu*C,o H10S*C4 HllBC1,N*C4 HllBZN*C2 H l lN*‘4 H l lNP2*‘3 H12As4F206*C10 H12BN*C9 H12B2N2S2*C4 H12B3C12N3*C4 H 2B3Li2N3*C4 12B3N3*C3

H12CIN*C4 H 2C12Si*C5 H12Cr*C12 H12F18N306P3*C12 H12Fe206P4*C10 H12Ge04*C4 H12Li2N2Si*C4 H12Mg0,8Si5*AI H12N3*B3 H 12Na201,Si,*AI, H12Na20,2Si6*A12 HIZN~SO3OPS

N%[P,O241*6 HzO Formation: 15.2.10.2 H1202Si*C4 H1204Si*C4 H, ,O, ,Si4*AI2Ca H12022Si,*A12Ca H12P2*C4 H12P,*C12 H12S2Si2*C4 H12S6Si4*C4 H12Sb2*C4 Hl,C10Si2*C, H13C12NSi2*C4 14BN*C10 H14B3C12N3*C9 H14*B10 H14B10*C2 H ,C1,Si*C6 H14Fe02*C1 2 H14Hg*C14 H140S*C6 H15A103*C1S H15As5*C5 H15BC13N*C6 Hl5BO,*Cl, H15B3C13N3*C18 H15B3S3*C3 H15B3S3*C18

41 1 -

punoduo3

ElP

xepul punodluo=)

Reaction with carbon surface: 16.4.2.7.3 Reaction with graphite: 16.4.2.1.5

--

Graphite residue compound 16.4.2.7.2 Reaction with NBS,, NbSe,, TaS,, TaSe,: 16.4.3.2 Reaction with elemental Si, Ge, Sn:

Compound Index

414

KNi*CrF6 KO*CI KO,*CI KO,*CI K06SbW KSbWO, Structure: 16.3.1.2 K06TaTe KTaTeO, Structure: 16.3.1.2 K0,TaW KTaWO, Structure: 16.3.1.2 K0,,Si,*Ca4F KP*F6 KP,*C,H,,Co KSb*F, KSi KSi Thermal decomposition: 16.2.3 K,Li0,P3 LiK,[P3091 Formation: 15.2.10.2 K,Mn*F6 K,O,*Cr K,OI?% K,S,O* Carbon surface oxidation: 16.4.2.7.3 K2°1

ISb4

K2Sb4011 Structure: 16.3.1.4 K3Na9036P12

NagK,CP30914 Formation: 15.2.10.2 K3°14Sb5 K3SbS014

Structure: 16.3.1.4

K7Si46 K7Si46

Formation: 16.2.3 K8*Ge46 K 8 4 6 K8Si46

Formation: 16.2.3

K8Sn46

KsSn4.5 Formation: 16.2.3 Kr Kr Clathration by water: 16.2.2.1 K'r*F, LaNi, LaNi, Reaction with H,: 16.4.2.4.4

Li

Li

Reaction with NbS,, NbSe,, TaS,, TaSe,: 16.4.3.2 Li*C, Li*C, Li*C,H9 Li*C6 Li*H LiN*H2 LiNSi*C6Hl5F LiNSi*C8H19F LiN3Si,*C6H2, LiN,Si,*C,H,, LiN,SiS*C,H27F, LiN,Si,*C,,H3,F, LiN,SiS*CllH,,F LiN,Si,*C,,H,,F3 LiN,Si5*C15H,,F2 LiN,Si,*C,,H, LiN,Si,*CzoH5,F4 LiN,Si,*C,,H,oF2 Li04*C1 Li0,P3*K, LiS,*Cr LiS,V LiVS, Formation: 16.4.3.2 Li,*C,HlOB10 Li,NSi,*C7H2, Li,N,Si*C,Hlz Li,N,*C,H 2B3 Li,N,Si,*C6Hlg Li,N3Si,*C9H2, Li,Si*C,,H,, Li,Si,*C,,H,, Mg*AICsF, MgO ,sSi5*A1Hl mgp53 MOPS, Formation and structure: 16.4.4 MgP% MgPSe, Formation and structure: 16.4.4 MgRb*AIF6 Mn*CsF,Fe Mn*F6K2 MnMoO,*CsF, MnO, MnO, Reaction with Na,O: 16.4.6 MnO,*K

,

Compound index MnPS, MnPS, Formation and structure: 16.4.4 MnPSe, MnPSe, Formation and structure: 16.4.4 Mn203

Mn203

Reaction with Na,O: 16.4.6 Mn2Ol0*CI0 MO*CllF,j Mo*F6 MoNiO,*CsF, MoO*Cl, MoO,*C,,H,,As, Mo0,*Cz4H,~As, MOO, MOO, Reaction with graphite: 16.4.2.1.3 MoO,*C,, MoO,*CsF,Mn MoO,Zn*CsF, MoO,*C,~H,,A~, MOO~P~*C~OH,OA~~ Mo06*C6 MoS, MoS, Intercalation: 16.4.3.3 Structure: 16.4.3.1 MoTe, MoTe, Structure: 16.4.3.1 Mo,*C1oH10'4s5 Mo,04*Ci9H25As5 Mo206*C16H10 Mo2P5*C10H10 N*B N*BF2H2 N*BF,H3 N*BH, N*BH4 N*BH, N*B,H7 N*CH6B N*CH6C1 N*CH,B N*CH,B, N*C,H, N*C,HloB N*CzHi iBz N*C,HgBF, N*C4H11

N*C,HllBC1, N*C4Hl,C1 N*C,H2F, N*C6H, N*C,H,BCl, N*C,H,,BCl, N*C9HlzB N*C10H14B N*C12H,,BC13 N*C,,H,,Br N*C16H,,C1 N*C16H36F N*C18H32F N*C1 BH40F N*C,oH44CI N*C20H44F N*C1H4 N*F3 N*H,Li N*H, NNa*H2 NNaO, NaNO, Oxidation of graphite: 16.4.2.5.2 NO*FeH, NOPSi*C,H17F3

NO,

n02

Intercalation into graphite oxide: 16.4.2.5.3 NO,*BF, N0,*CH3 No,*' 16H35 N02*C20H43 N02*C21H45 N03*C17H37 NO,*H N0,Si4*C,H2, NO,TaW*H, NP*Cl,H NPSi*C,H ,Br NP2*C3H11 NP,*Cl,,

NS

SN Formation: 15.2.12.1 Polymerization: 15.2.12.1 NS*Cl NS*F, NS,Ta*H3 NS1*C,Hl,FLi NSi*C,H19FLi

415

E Zt O E ff H Eg61~813*E~ b E f f51H813*EN S1dE13Eff813*EN z~3 ~ EP H ~ ~ * ~ N 881H93*EN sV8'H93*' N 213Eff91H93*EN Z E Z I t !'I8 H 3*'N Z~3E~ZT~t3*E~ E 81 E €IH E 21 E ff H 3*'N E~3Eff6~E3*E~ Z 1 ~ E ~ * E ~ 9HE8*EN

E ~ E d E ~ * E ~ E ~ E ~ 3 E ~ * E ~

OE z 1 H 3*'USZN z~3t2~83*v!~Z~

xepui punoduo3

91P

LlP

xapul punodwo3

Compound Index

418

Na,O,*CH,As Na,O,*C,Fe Na,09P3*H Na,O,,Si,*AI,H, Na,O, ,Si,*AIH8 Na,O ,Si,*AI2H Na,0,2Si,*AIZH,2 Na,02 ,Si, *Al,CaH Na309P3 Na,[P,O9] Formation: 15.2.10.2 Na,O,,SeSi, Na,[SeS140121 Formation: 16.3.1.6 Na,O, ,Si,*Fe Na,Ol ,Si,*In Na,O,,Si,Y Na,[YS140i,] Formation: 16.3.1.6

,

Na8030P8*H12

,

Na,Si,, Na,Si,, Formation: 16.2.3 Na9036P1Z*K3 Nb*F, NbNiO,*CsF, NbO,Zn*CsF, Nb05Rb*F NbO,Te*K NbO,W*K NbS, NbS, Reaction with group-IA metals: 16.4.3.2 Structure: 16.4.3.1 NbTe, NbTe, Structure: 16.4.3.1 Ni*CrF,K Ni*CsF,Fe NiO,*CsF,Nb NiO,Ta*CsF, NiO,*CsF,Mo NiPS, NiPS, Cathode material: 16.4.4 Formation and structure: 16.4.4 NiPSe, NiPSe, Formation and structure: 16.4.4 Ni,*La O*BrCr O*CH,

O*C,H,Na O*C,H, O*C,H8 0*C4H10 0*C8H9 O*CICr O*ClFe O*ClK O*Cl,Mo O*F,I O*FeH,N O*H2 O*N, O*Na, OP*CI, OPSi*C,H,,F,N

os*c

OS*C2H6 OS*C6H14 OS*C,H,, OS*CI, OS*F, OSi2*C,HI3CI OSi2*C,H, OSi,*C,,H,, OSi,*C,,H,, OSi,*C, 6H40 0Si,*C1,H4,C1, OSi,*C,8H40 OSi5*C,,H,,N, OSi,*C,,H,,N, OSi,*C,,H,9N, OTi*Br OTi*Cl OV*Br OV*CI OV*F, OXe*F, O,*AsF, O,*As,F, 0,*BF4N o,*c 02*CH3N O,*C,HF, 02*C2H4 02*C4H8 02*C4H1 0 O,*C,H,Fe Na O,*C,,H,Cl,Fe 02*C12H14Fe O,*C,,H,,As,Mo 02*C16H35N

,

Compound Index 02*C20H43N 02*C21H45N 0,*C,,H,5As,Mo 02*Cl O,*Cl,Cr O,*CrF, O,*CrNa O,*CsF,NbNi O,*FeH 02*H2

O,*Mn O,*N O,*Na, 0,Pb PbO, Reaction with graphite: 16.4.2.1.4

0,s so2

Clathration by water: 16.2.2.1 Formation of carbon surface sulfides: 16.4.2.1.3 0,S*C,H8 02S*C5H10

O,S*CIF O,S*CI, O,Si*C,H,, 02SI,*C,HI8 O,Si,*C6HI8C1,

03*C18H15B O,*C,,Mo 03*C150Cr

03*C1K O,*Cr O,*Cr, O,*CsF,MnMo O,*CsF,MoNi O,*HI O,*HN O,*Mn, O,*Mo O,*NNa O,P*C,H, 03P*C6H15 O,P*Na 03s

503

Reaction with graphite: 16.4.2.1.3 O,S*B,FN, O,S*CHF, 03S*C,H6 03S*C1H O,S*FH O,S,*F,N3 0,Se

02Si3*C11H22

SeO, Reaction with graphite: 16.4.2.1.3 O,Se*H

02Si4*C1'dH44

03Si3*C6H18

02Si4*C48H42

O,Ta*CoCsF, O,Ta*CsF,Ni 02U*CI, O,Zn*CsF,Nb 0 3

0,

Carbon surface oxidation: 16.4.2.7.3 O,*As, O,*BH, O,*Bz O,*B,Br, O,*B,CI, 03*B,F, O,*B,H, O,*CH,AsNa, O,*C,H,As O,*C,H,B 03*C,H, 0,*C14Cr 03*C17H37N 03*C18H15A1

0,Si4*C8H,,N

03Si4*C18H28

O,V, v2°3

Reaction with H,S: 16.4.3.2 O,Zn*CsF,Mo 04*AsH, O,*B,Cl, O,*C,FeNa, 04*C,Hl,Ge 04*C19H25As5M02 04*C24H30Fe2

04*C32H72CrNZ 04*C34H72N2 0,*C,,H,oAs6Mo 04*C40H88CrN2 O4*C1H 04*C1K O,*CILi

o,*co,

O,*CrK, 04*HI

419

420

Compound Index

O,*KMn O,P*H,Na 04P2*C34H72 O,P,*BaNa O,P,*C,~H,~A~,M~ O,S*C,H, 04S*H, 0,S*H8N2 04S2Si2*C16H36 04S2Si2*C32H32 04% Sb204

Reaction with graphite: 16.4.2.1.3 O,Se*H, 04Si*C4H 0,Si*C8H20 04W*C32H72N2 04W*C40H88N2 O,*C,Fe O,*Cr, 05*N2

0,Rb'FNb 0,RbTa, *F O~Si5*~3,~40 O,*C,Mo 06*C10Hl

06*C16H10M02 06*C1,CrF, 06P3*C12H12F18N3 06P4*C10H12Fe2 06S2*F2

O,SbW*K O,TaTe*K O,TaW*H,N O,TaW*H,N, O,TaW*K O,Te*KNb 0, W*KNb O,*Cl, 07Si2*H, O,*Cr, 08S,*H,N2 O,S,*K, O,P,*HNa, O,P,*K,Li O,P,*Na, O,Si,*AI,H, O,o*C1oMn, %P4 4'

10

Reaction with R,O: 15.2.10.1 Reaction with H,O: 15.2.10.2

OlOR%*ClO 011Sb4*K2 01,?p4*cu2 0 ,,SeSi,*Na, 0 ,Si,*AI,CaH, 0 ,Si,*FeNa, O,,Si,*InNa, 0 ,Si,Y*Na, 0 ,Si,*AI,CaH, 014Sb5*K3 Ol,Si2*AI,H,Na, 0 ,4Si,*A1,CaH, 0 ,Si,*AlH,Na, O18Si,*AI,CaH, 0 ,Si,*AI,H, ,Na, 0 ,Si,*AIH ,Mg O,,Si,*Ca,FK 0, lSi,*AI,CaH,o O,lSi,*AI,BaHlo O,,Si,*AI,CaH,, O,,Si,*A1,H1,Na2 02,Si,*A12CaHl 0,,Si,*AI,CaHl,Na2 030P8*H12Na8 030P10Zn3*Ba2 O,,Si1~*A1,CaHl, 036P12*K3Na9 Os*C,F, Os*F, P*C2H7 P*C,H, P*C,H,O, P*C6H7 P*C6H1 3 ' 5 P*C6H18N3 P*C16H36F P*C18H15 P*CI, P*Cl,HN P*CI,O P*CI, P*F, P*F6K P*H,NaO, P*H, P*NaO, PS,*Cd PS,*CO PS,*Fe PS,*Mg PS,*Mn PS,*Ni

,

421

Compound Index PS,Sn SnPS, Formation and structure: 16.4.4 PS,V VPS, Formation and structure: 16.4.4 PS3Zn ZnPS, Formation and structure: 16.4.4 PSe,*Cd PSe,*Co PSe,*Fe PSe,*Mg PSe,*Mn PSe,*Ni PSe,Sn SnPSe, Formation and structure: 16.4.4 PSe,V VPSe, Formation and structure: 16.4.4 PSe,Zn ZnPSe, Formation and structure: 16.4.4 PSi*C5H15 PSi*C,H, ,BrN PSi*C,H ,F,NO P2*C3H2F9

P2*C3H10 P2*C3Hl l N P2*C4F 1 2 P2*C4H12 P2*CSH20 PZ*C12H, oC12N2 P2*C12H,2 P2*C24H20 P2*C34H7204

P2*CISHN2 p2*c110 P2PtSi4*Cs4H,o P,*BaNa04 P3*Br6N3 P3*C5F15

P3*C6H18N3 P,*C,HZ2CoK p,*c1 OH2 5 P3*C12H1Z F 1 SN3'6 P3*C12H36B3 3' *1' SH 15 P3*C36H30N3 P3*C16N, P,*CI 2N

P,*F,N, P,*HNa20, P3*K2Li09 P3*Na309 P4*C4F 1 2 P4*C8H24N4 P4*C10H12Fe206 P4*C24H20 P,*C,,H,Ll P4*C1,N4 P4*C111N3 p4*cu2012 '4*OIO

P5*C5H15 1 OH 1OMo2

p5*c

P5*C10H1

5Fe

P5*C10H25 P5*C30H25

P6*C36H30 P6*C40H,oAS6Mo04 P6*C112N6 Pa*Hi 2NaaO3o P9*C54H45 P10*C60H50 P10Zn3*Ba2030 P12*K3Na9036 Pb*O, PdS PdS Reaction with graphite: 16.4.2.1.3 Pt*C16H2 PtSi4*C84H,oP2 Rb*AlF,Mg Rb*C4Hg Rb*Cs Rb*Cio Rb*C24 Rb*FNb05 Rb*Ge Rb*H RbSi RbSi Thermal decomposition: 16.2.3 RbTa2*F05 Rb,*Ge4, Rb,Si,, Rb5Si4, Formation: 16.2.3 Re2*CloOlo Ru*CioHio 5

5

Viscosity: 15.2.2.2.1

Compound Index

422

S2*Hf

SP

Formation: 15.2.1 see also specific allotropes: 15.2.1

s,

Formation: 15.2.1 S*B,FN,03 S*CHF303 S*CH, S*CO S*C,H, S*C,H,O S*C,H,O, S*C2H,0, S*C,H,O, S*C4H10 S*C4H22B10 S*C,HlOOZ S*C6H,,0 S*C,H,,O S*Cl,H,,F S*CIF02 S*C1H03 S*CIN S*CI, S*Cl20 S*CI,O, S*FHO, S*F,N S*F, S*F,O S*F, S*H2 S*H20, S*H,N,O, S*N

s*o,

S*O, S*Pd SSi,*C6H18 STI,

5,s

s*

Reaction with graphite: 16.4.2.1.3

s2

Formation: 15.2.1 S,*C S2*C4H12B2N2 S2*Cl,HlOB2 S,*CI, S2*CrLi s2*F206 S,*H,N,O,

S,*K208

S,*Mo S,*N2 S2*Nb S,Si,*Br, S2Si2*C4Hl, S,Si,*C6H S,Si,*C6H S2Si2*C14H16 S2Si2*C16H3604 S2Si2*C24H20 S2Si2*C32H3204 S,Siz*C14 S,Si,*F, S,Ta TaS, Reaction with NH,: 16.4.3.4 Reaction with metallocenes: 16.4.3.4 S2Ta*H3N S,Ti TiS, Formation: 16.4.3.2 Reaction with group-IA metals: 16.4.3.2 Structure: 16.4.3.1 S,V*Li S2V*Na

s2w

WS, Reaction with graphite: 16.4.2.1.3 S,Zr ZrS, Reaction with group-IA metals: 16.4.3.2 Reaction with NH,: 16.4.3.4 Structure: 16.4.3.1 s3 s3

Formation: 15.2.1 S3*B?. S3*B3C13 s3 *€3 3 F3 s3 *B3 I3 S3*C3H15B3 S3*C6H1 BBJN3 S3*C6H18Ge3 S3*C12H33B3 S3*C18H15B3 S,*CdP S,*CI,N, S3*Cl3N, S,*COP S,*Cr,

Compound Index S,*F,N, S,*F,N,O, S,*FeP S,*MgP S,*MnP S,*NiP S3Si,*C6H18 S3Si3*C21H24 S,Si,*H6 S,Sn*P S,V*P S,Zn*P s4 s4

Formation: 15.2.1 S4*B2H2 S4*C4H10B2 S4*CIN, S4*C14N4 S4*N2 S4*N4 S4*N5 S4Si,*C14 S4S14*CI, SS s5

Formation: 15.2.1 SsSbz %S, Reaction with graphite: 16.4.2.1.3 S6 6 '

Formation: 15.2.1 S6*B3H3 S6*C6H15B3 S6Si4*C4H12 S6S14*C24H2Ll

s, s,

S? Formation, 15.2.1 S8

Allotropic rearrangement: 15.2.1 Formation: 15.2.1 Pyrolysis: 15.1.1.3.4 Solubility: 15.2.1 Thermal rearrangement: 15.1.1.3.1

SIZ

s12

s,,

Formation: 15.2.1

S18 Formation: 15.2.1 Sb*C6F, Sb*C6H,CI2 Sb*C6H, Sb*F, Sb*F6K SbW*K06 Sb2*C4H12 Sb2*C16H36 Sb2*C24H20 Sb2*04 Sb2*S5 Sb4*C16H36 Sb4*C42H50 Sb4*K201 1 Sb5*K3014 Sb6*C36H30

Se

Se Structure: 15.2.1 Viscosity: 15.2.2.2.1 Se*HO, Se*H, Se*H204 Se*O, SeSi4*Na,012 Se,Ta TaSe, Structure: 16.4.3.1 Se,*CdP Se,*CoP Se,*FeP Se,*MgP Se,*MnP Se,*NiP Se,Sn*P Se,V*P Se,Zn*P Se,

Set3 Formation: 15.2.1 Si

Si

s9 s9

Formation: 15.2.1

SIO SlLl Formation: 15.2.1

Reaction with SiF,: 15.1.1.3.2 Reaction with elemental K: 16.2.3 Si*C2H6C12 Si*C,H,Cl

423

424

Compound

Si*C,H,Cl, Si*C,H,Br Si*C4H9CI Si*C4HloBr, Si*C4Hl,Cl2 Si*C4Hlo12 Si*C4H ,Li,N, Si*C4H ,O, Si*C4H1204 Si*C,H12C1, Si*C,H,,BrNP Si*CSHl,P Si*C6HsF3 Si*C6H, Si*C6H14C1z Si*C6H ,FLiN Si*C,H,CIZ Si*C,H1,F3NOP Si*C8H1,Cl2 Si*C8H1,FLiN Si*C,H,,04 Si*C,HZ4N4 Si*C,,H,,C~, Si*C,,Hl ,C12 Si*Cl,H,,Li2 Si*Cl4HI6 Si*ClSH,,N2 Si*CI4 Si*Cs Si*F, Si*F,I, Si*F4 Si*K Si*Na Si*Rb Si,*AI,CaH8012 Si,*AI,CaH ,O, Si,*A1,H4Na,014 Si,*AI2H4Og Si,*Br,S, Si,*Br, SiZ*C2H,C& Si,*C4H ,S2 SiZ*C4H,CI,N SiZ*C,Hl,CIO Si,*C,H S~Z*C~H~,~O Si,*C6H16S2 Si,*C,Hl, Si,*C6H18N, Si,*C,H180, Si2*C6H1 8 s

,

,

Compound Index Si4*C8H2,N03 Si4*C8H2,N4 Si4*C,H2,Li2N, Si4*C,H2,LiN3 Si4*C,HZ9N3 si4*c 10H30

Si4*CloH300 Si4*CloH3 LiN, Si4*C10H32N4 Si4*C10H33N4 si4*c1 1H3 1N4 Si4*C16H40

Si4*C16H&120 Si4*C1,jH4oO Si4*C16H4,Br si4*c 6H4,Cl Si4*C16H42

Si4*C18H2803 Si4*C18H4402 Si4*C20H48

Si4*C24H20S6 Si4*C24HS6 Si4*C24H60N4 Si4*C28H68

Si4*C32H72 Si4*C32H76N4 Si4*C40H88 Si4*C48H40 Si4*C48H4012

Si4*C48H40Li2 Si4*C4,H4,0 Si4*C4,H4,CI Si4*C48H4202

Si4*C8,H7,P2Pt Si4*CI,S4 Si4*FeNa501 Si4*InNa501 Si4*Na5012Se Si4Y*Na50 Si5*A1H8Na,0 6 Si;*A1Hl,MgO18 Si5*A12CaH18Na20,3 Si5*AI,CaH1802, SiS*C9H2,F3LiN3

,

,

Si5*C10H30

Si5*CloH30F2LiN3 Si,*Cl ,H3,FLiN3 SiS*C12H36 SiS*C12H37N3

Si5*Cl,H,,N,O SiS*Cl3H3,F2N3 SiS*C13H38F2N4

425

426

Compound Index

S1g*C27H72F6N6 S19*C36H72

Si ,*AI,CaH 16032 SiI0*Cl6H48 SilO*Ci8H54 S110*C18H54F4N6 Si10*C20H50F2N, Si10*C20H60 Si10*C22H66N6 S~IO*C~EH~EF,N~ Si10*C30~64~2~6 Si10*C40H8~

si1

1*I'

EH 5 4

Sil 1*C20H60F6N6

sil1*C22H66F4N6

Sil,*C27HEF2N6 Si13*C22H66

Si,6*K7 Si,,*K, Si,,*Na8 Si,,*Rb, Sil,,*Cs7 Si, 36*Na, Sn

Sn Reaction with elemental Ik.16.2.3 Sn*PS, Sn*PSe, Sn2*C12H3@N2 Sn,,*K, Ta*CoCsF,O, Ta*CsF,NiO, Ta*F5 Ta*H,NS2 Ta*S, Ta*Se, TaTe*KO, TaW*H,NO, TaW*H,N,O, TaW*KO, Ta,*FO,Rb Te*KNbO, Te*KO,Ta Te,*Mo Te,*Nb

Te,W WTe, Structure: 16.4.3.1 Ti*BrO Ti*Cl ,H 0C12 Ti*C, ,H,,As, Ti*CIO

t1*f4 t1*52 t12*5

U*C13F6 U*CI2O2 U*F, V*BrO V*ClO V*F30 V*LiS, V*NaS, V*PS, V*PSe, vZ*03

W*C32H72N204 W*C40H88N204 W*H4N06Ta W*H,N,06Ta W*KNb06 W*KO,Sb W*K06Ta

w*52

W*Te2 Xe Xe Clathration by water: 16.2.2.1 Xe*ClgF6 Xe*C2,F4 Xe*F, Xe*F4 Xe*F40 Xe*F, Y*Na50,,Si4 Yb*C6 Zn*CsF,MoO, Zn*CsF,NbO, Zn*PS3 Zn*PSe, Zn3*BaZ030P10 Zr*S,

lnorganic Reactions and Methods, Volume I 7 Edited by J.J. Zuckerman, A.P. Hagen Copyright 0 1990 by VCH Publishers, Inc.

Subject Index

Subject Index

This index supplements the compound index and the table of contents by providing access to the text by way of methods, techniques, reaction conditions, properties, effects and other phenomena. Reactions of specific bonds and compound classes are noted when they are not accessed by the heading of the section in which they appear. For multiple entries, additional keywords indicate contexts and thereby avoid the retrieval of information that is irrelevant to the user’s need. Section numbers are used to direct the reader to those positions in the volume where substantial information is to be found.

A

Addition polymerization definition 15.1.1.2 synthesis of borazines 15.1.1.3.1 of cyclic trithiazyls 15.1.1.3.1 Alkali-metal condensed phosphates 15.2.10.1 Alkanes outgassing from attapulgite 16.4.7.2 Antimony catenates 15.2.3.4 Arsenic catenates 15.2.3.3 Arsenophosphates 15.2.10.4 Biintercalation 16.4.2.1.1 Bond lability in inorganic systems 15.1.3.1 Borazines safety 15.2.5.1.1 stability 15.2.5.2 synthesis 15.1.1.3.1, 15.2.5.1.1 Boron hydrides safety 15.2.6.4 synthesis 15.1.1.3.2 Boron-oxygen

oligomers 15.2.6.2 Cage polysiianes synthesis 15.2.4.4 Cage structures synthesis 15.1.1.3.2 Carborane-germane polymers synthesis 15.2.7.3 Carborane-mercury polymers synthesis 15.2.7.3 Carborane-phosphazene polymers synthesis 15.2.7.3 Carboranes chain-ring conversions 15.2.7 synthesis 15.1.1.3.2 Carborane-silane polymers synthesis 15.2.7.3 Carborane-siloxane polymers synthesis 15.2.7.2 Carborane-siloxanes synthesis 15.1.1.3.2 Carborane-stannane polymers synthesis 15.2.7.3 Catenates antimony 15.2.3.4 arsenic 15.2.3.3 phosphorus 15.2.3.2

427

428

selenium 15.2.2.2 silicon 15.2.4.1.1 sulfur 15.2.2.2 Chain-chain equilibria 15.1.3 Chain growth from cyclic oligomers 15.1.1.3.3 from monomers 15.1.1.3.3 Chain populations 15.1.3.2.1 Chain-ring distribution function 15.1.3.2.2 equilibria 15.1.3 equilibrium constants 15.1.3.2.2 interconversions of P, As, Sb, Bi 15.2.3.1 of boroxines 15.2.6.4 of carboranes 15.2.7 of cyclophosphazenes 15.2.11.2 of cyclosilanes 15.2.4.2.3 of cyclosilazanes 15.2.9.3 of cyclosiloxanes 15.2.8.3 of polyboranes 15.2.7 theory of equilibria 15.1.3.2.2 Chemie douce 16.4.8 Clatherates definitions 16.2.1 germanium host lattice 16.2.3 silicon host lattice 16.2.3 tin host lattice 16.2.3 Condensation polymerization definition 15.1.1.2 synthesis of cyclic metaborates 15.1.1.3.1 of cyclic metaphosphates 15.1.1.3.1 of phosphazenes 15.1.1.3.4 of silicones 15.1.1.3.1 Condensed phosphoric acids synthesis 15.2.10 Coordination polymers synthesis 15.2.14 Copolymer definition 15.1.1.1 Cyclic metaphosphates 15.2.10.2 Cyclodimethylgermanodimethylsilthians synthesis 15.1.3.3.4 Cyclodimethylgermthians synthesis 15.1.3.3.4 Cyclodimethylsilthans synthesis 15.1.3.3.4 Cyclolinear polymer definition 15.1.1.1 Cyclomatrix polymer definition 15.1.1.1

Subject Index Cyclophosphazanes synthesis 15.1.1.3.1 Cyclosilanes 15.2.4.1.1, 15.2.4.2.2 photolysis 15.2.4.1.2, 15.2.4.2.3 reactions 15.2.4.2.1 ring-opening 15.1.1.3.4 synthesis 15.1.1.3.1 Cyclosilazanes synthesis 15.1.1.3.1 Cyclosilthianes synthesis 15.1.1.3.1 Cyclothiazyls synthesis 15.1.1.3.1 Density of sulfur 15.2.2.2.12 Diamond synthesis 16.4.2.6 Diborane safety 15.2.5.1.2 Dielectric constant of sulfur 15.2.2.2.10 Distribution functions molecular size 15.1.3 Double-oxide notation 15.2.10 ESR of liquid sulfur 15.2.2.2.6 Electrical conductivity of liquid sulfur 15.2.2.2.14 Electrochemical intercalation 16.4.3.3 Enthalpy of scrambling 15.1.3.3.1 Entropy of scrambling 15.1.3.3.1 Equilibration definitions 15.2.8 Equilibria chain-chain 15.1.3 chain-ring 15.1.3 redistribution 15.1.3 ring-ring 15.1.3 Equilibria theory 15.1.3.2.2 Gas hydrates 16.2.2.1 Gel point to calculate ring closures 15.1.3.2.2 Germanium-terminated siloxanes synthesis 15.1.3.3.4 Grafting chemistry 16.4.8.7 Graphite anodic reduction 16.4.2.8.3 cathodic reduction 16.4.2.8.2 Graphite fluorides synthesis 16.4.2.6 Graphite oxides synthesis 16.4.2.5.2

Subject Index reactions with binary hydrides 16.4.2.5.3 fluoroacids 16.4.2.5.3 Halogens reactions with graphite 16..4.2.1.5 Halophosphoranes synthesis 15.1.1.3.1 High polymer definition 15.1.1.1 High pressure synthesis of borazines 15.2.5.1.1 Hydrohalide elimination for polymer formation 15.1.1.3.1 Intercalates cationic 16.4.8.1 graphite 16.4.2.1.1 molecular 16.4.3.4,16.4.4 of amines 16.4.3.4 of organometallics 16.4.3.4,16.4.4,16.4.5 of pyridine 16.4.3.4 In terhalogens reactions with graphite 16..4.2.1.5 Ionomers synthesis 15.2.14.3 Ladder polymers in ultraphosphates Macromolecular phosphates 15.2.10.3 Macromolecule definition 15.1.1.1 Metaborates cyclic synthesis 15.1.1.3.1 Metal-carbon polymers synthesis 15.2.13 Metal-oxygen polymers synthesis 15.2.13 Metaphosphates cyclic synthesis 15.1.1.3.1 Metaphosphoric acids synthesis 15.2.10 Methylenediphosphonate oligomers 15.2.10.4 Molecular distributions in reorganized 2.10.4 Monomer definition 15.1.1.1 Montmorillonites 16.4.7.2 NMR of As,O,-AsF, system 15.1.3.3.2 of hydrated water 16.4.3.5

429

of liquid Se 15.2.2.2.9 of Me,GeX,-Me,GeS system 15.1.3.3.2 of TaS, 16.4.3.4 to study scrambling reactions 15.1.3.3 NMR patterns of ultraphosphates 15.2.10.2 Neso systems defined 15.1.3.3.1 Neutron diffraction of Na,.,(D,O),VS, 16.4.3.5 01igomer definition 15.1.1.1 Oligomers boranamine synthesis 15.2.5.3 boron-oxygen 15.2.6.2 phosphate esters 15.2.10.1 phosphatic compounds 15.2.10.3 Orthophosphates condensations 15.2.10.1 Perbromogermanosiloxanes synthesis 15.1.3.3.4 Perchlorogermanosilthians synthesis 15.1.3.3.4 Phosphate glasses 15.2.10.1 phosphates 15.2.10.1 Phosphatic reorganized mixtures 15.2.10.1 Phosphazanes synthesis 15.1 L3.1 Phosphazenes ring-opening 15.1.1.3.4 synthesis 15.2.11.1 Phosphoric anhydride synthesis 15.2.10 Phosphorus catenates 15.2.3.2 Photolysis of cyclosilanes 15.2.4.1.2, 15.2.4.2.3 Po1yamid oximes synthesis 15.2.13.6 Pol yarenous oxyfluorides synthesis 15.1.3.3.2 Polyar ylsilsesquioxanes synthesis 15.2.13.3.4 Pol yboranes chain-ring conversions 15.2.7 synthesis 15.2.7.1 Polyborates bonding 15.2.6.1 synthesis 15.2.6.1 Polycarbonates metal synthesis 15.2.13.4 Poly(carborane-siloxanes)

430 synthesis 15.2.13.2.2 Pol ycarboslanes synthesis 15.2.13.2.2 Pol ycarbosiloxanes synthesis 15.2.13.2.2 Pol ygermanes synthesis 15.1.3.3.3 Polygermanosilthians synthesis 15.1.3.3.3 Po1ymer definition 15.1.1.1 Polymers boranamine synthesis 15.2.5.5.1 thermal stability 15.2.5.5.1 Polymetalloethers synthesis 15.2.13.5 Pol ymetallosiloxanes synthesis 15.2.13.3.2 Polymetal phosphinates synthesis 15.2.13.3.5,15.2.14.5.1 Polyoximes synthesis 15.2.13.6 Poly ox ysulfides synthesis 15.1.3.3.3 Polyphosphates 15.2.10.1 Polyphosphoric acids synthesis 15.2.10 Polysilanes synthesis 15.1.3.3.2 Pol ysilarylenes synthesis 15.2.13.2.2 Pol ysilicas synthesis 15.2.13.3.1 Poly siloxanes synthesis 15.1.3.3.4 Poly(silylary1enesiloxanes) synthesis 15.2.13.3.3 Pyrophosphates 15.2.10.3 Random statistics 15.1.3.2.1 Reaction rates group IVB 15.1.3.1 Redistribution equilibria 15.1.3 Reorganizational heat order 15.1.3.2.1 Ring-equilibration synthesis of cyclic siloxanes 15.1.1.3.1 of cyclic thiazyls 15.1.1.3.1 of silthianes 15.1.1.3.1 of sulfur rings 15.1.1.3.1 Ring-openmg polymerization definition 15.1.1.2

Subject index synthesis of cyclosilanes 15.1.1.3.4 of silazanes 15.1.1.3.4 of silicones 15.1.1.3.4 of sulfur rings 15.1.1.3.4 of thiazyls 15.1.1.3.4 Ring-polymer interconversions 15.2.1 Ring-ring equilibria 15.1.3 exchange 15.1.3.3.3 interconversions 15.2.1 of P, As, Sb, Bi 15.2.3.1 of boroxines 15.2.6.4 of carboranes 15.2.7 of cyclophosphazenes 15.2.11.2 of cyclosilanes 15.2.4.2.3 of cyclosilazanes 15.2.9.2 of cyclosiloxanes 15.2.8.3 of cyclosilthianes 15.2.9.4 of polyboranes 15.2.7 Safety borazine synthesis 15.2.5.1.1 boron hydrides 15.2.6.4 diborane 15.2.5.1.2 fluorination 16.4.2.6, 16.4.2.6.3 perchlorates 16.4.2.8.3 tetrathiazyl 15.2.12.1 Scrambling 15.1.3.3.2 enthalpy 15.1.3.3.1 entropy 15.1.3.3.1 equilibria 15.1.3 higher order 15.1.3.2.1 NMR 15.1.3.3 non-random 15.1.3.2.1, 15.1.3.3.2 theory 15.1.3.2.1 thermodynamics 15.1.3.2.1 Selenium allotropes 15.2.1 catenates 15.2.2.2 liquid NMR 15.2.2.2.9 thermal conductivity 15.2.2.2.13 ultrasonic absorption 15.2.2.2.2 specific heat 15.2.2.2.11 Semiconductor 16.4.8.3 Silane high polymers synthesis 15.2.4.5 Silazanes ring-opening 15.1.1.3.4 synthesis 15.2.9.1 Silicates 16.4.7.1 Si1icon catenates 15.2.4.1.1

Subject Index Silicones cyclic 15.2.8.1 structure 15.2.8.2 definitions 15.2.8 end-capping reactions 15.2.8.5 ring-opening 15.1.1.3.4 synthesis 15.1.1.3.1 Silthianes synthesis 15.2.9.4 Soft chemistry 16.4.8 Specific heat of Se 15.2.2.2.11 of sulfur 15.2.2.2.11 Sulfatophosphates 15.2.10.4 Sulfur allotropes 15.2.1 catenates 15.2.2.2 density 15.2.2.2.12 dielectric constant 15.2.2.2.10 electrical conductivity 15.2.2.2.14 liquid

ESR 15.2.2.2.6 Viscosity 15.2.2.2.1 ring-opening 15.1.1.3.4 specific heat 15.2.2.2.11 Thermal conductivity of liquid Se 15.2.2.2.13 Thermodynamics of liquid sulfur 15.2.2.1 ofscrambling 15.1.3.2.1, 15.1.3.3.1 of selenium rings 15.2.2.1 Thiazyls polymerization 15.2.12.1 ring-opening 15.1.1.3.4 Tromelite 15.2.10.3 Ultraphosphates crystalline 15.2.10.2 Ultrasonic absorption of liquid Se 15.2.2.2.2 van der Waals gap 16.4.3.1, 16.4.8 Vibrational spectroscopy 15.2.2.2.8 Vitreous phosphates 15.2.10.1 Zeolites 16.3.1.1, 16.4.7

43 1