Syndiotactic Polystyrene: Synthesis, Characterization, Processing, and Applications

SYNDIOTACTIC POLYSTYRENE SYNTHESIS, CHARACTERIZATION, PROCESSING, AND APPLICATIONS Edited by Jürgen Schellenberg R&D ...

Author: Jurgen Schellenberg

113 downloads 1677 Views 7MB Size Report

This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!
Report copyright / DMCA form

DOWNLOAD PDF

SYNDIOTACTIC POLYSTYRENE SYNTHESIS, CHARACTERIZATION, PROCESSING, AND APPLICATIONS

Edited by

Jürgen Schellenberg R&D Dow Central Germany Dow Olefinverbund GmbH, Schkopau, Germany

A John Wiley & Sons, Inc., Publication

SYNDIOTACTIC POLYSTYRENE

New supramolecular structure of syndiotactic polystyrene showing a “bird’snest-like” structure, obtained from a cyclohexanol solution by thermally induced phase separation (scanning electron microscope image by M. van Heeringen, Dow Terneuzen) (Van Heeringen, M., Vastenhout, B., Koopmans, R., Aerts, L. e-Polymers [2005], no. 048.)

SYNDIOTACTIC POLYSTYRENE SYNTHESIS, CHARACTERIZATION, PROCESSING, AND APPLICATIONS

Edited by

Jürgen Schellenberg R&D Dow Central Germany Dow Olefinverbund GmbH, Schkopau, Germany

A John Wiley & Sons, Inc., Publication

Cover image: Scanning electron micrographs of syndiotactic polystyrene prepared by a powder bed polymerization process showing the typical “cauliflower” structure of the powder (SEM: Dow Central Germany). Copyright © 2010 by John Wiley & Sons, Inc. All rights reserved. Published by John Wiley & Sons, Inc., Hoboken, New Jersey. Published simultaneously in Canada. No part of this publication may be reproduced, stored in a retrieval system, or transmitted in any form or by any means, electronic, mechanical, photocopying, recording, scanning, or otherwise, except as permitted under Section 107 or 108 of the 1976 United States Copyright Act, without either the prior written permission of the Publisher, or authorization through payment of the appropriate per-copy fee to the Copyright Clearance Center, Inc., 222 Rosewood Drive, Danvers, MA 01923, (978) 750-8400, fax (978) 750-4470, or on the web at www.copyright.com. Requests to the Publisher for permission should be addressed to the Permissions Department, John Wiley & Sons, Inc., 111 River Street, Hoboken, NJ 07030, (201) 748-6011, fax (201) 748-6008, or online at http://www.wiley.comigo/permission. Limit of Liability/Disclaimer of Warranty: While the publisher and author have used their best efforts in preparing this book, they make no representations or warranties with respect to the accuracy or completeness of the contents of this book and specifically disclaim any implied warranties of merchantability or fitness for a particular purpose. No warranty may be created or extended by sales representatives or written sales materials. The advice and strategies contained herein may not be suitable for your situation. You should consult with a professional where appropriate. Neither the publisher nor author shall be liable for any loss of profit or any other commercial damages, including but not limited to special, incidental, consequential, or other damages. For general information on our other products and services or for technical support, please contact our Customer Care Department within the United States at (800) 762-2974, outside the United States at (317) 572-3993 or fax (317) 572-4002. Wiley also publishes its books in a variety of electronic formats. Some content that appears in print may not be available in electronic formats. For more information about Wiley products, visit our web site at www.wiley.com. Library of Congress Cataloging-in-Publication Data: Schellenberg, Jürgen. Syndiotactic polystyrene : synthesis, characterization, processing, and applications / Jürgen Schellenberg. p. cm. Includes index. ISBN 978-0-470-28688-3 (cloth) 1. Polystyrene. 2. Microcrystalline polymers. I. Title. TP1183.M5S34 2010 668.4'233–dc22 2009020802

Printed in the United States of America. 10 9 8 7 6 5 4 3 2 1

Dedicated to Prof. Dr. habil. Joachim Ulbricht, a Pioneer in the Fields of Polymer Chemistry and Coordination Polymerization

CONTENTS

PREFACE

xvii

CONTRIBUTORS

xxi

ABOUT THE EDITOR

xxv

PART I

INTRODUCTION

1. Historical Overview and Commercialization of Syndiotactic Polystyrene

1

3

Michael Malanga, Osamu Isogai, Takeshi Yamada, Shigeo Iwasaki, and Masahiko Kuramoto

1.1 1.2 1.3 1.4 1.5

PART II

Discovery of Syndiotactic Polystyrene (SPS) Early Years of Development (1985–1989) Intense Development Years (1989–1996) Initial Commercial Launch Stage (1996–2001) Years 2001–2007

PREPARATION OF SYNDIOTACTIC POLYSTYRENE

2. Transition Metal Catalysts for Syndiotactic Polystyrene

3 5 6 12 13

15 17

Norio Tomotsu, Thomas H. Newman, Mizutomo Takeuchi, Richard Campbell Jr., and Jürgen Schellenberg

2.1 2.2

Introduction Transition Metal Compounds 2.2.1 Metals 2.2.2 Titanium Complexes 2.2.3 Molecular Weight Control 2.2.4 Supported and Heterogeneous Catalysts 2.3 Summary References

17 18 18 19 26 27 29 29 vii

viii

CONTENTS

3. Cocatalysts for the Syndiospecific Styrene Polymerization

32

Norio Tomotsu, Hiroshi Maezawa, and Thomas H. Newman

3.1 Introduction 3.2 MAO 3.3 Boron Compounds 3.4 Other Chemicals 3.5 Summary References 4. Mechanisms for Stereochemical Control in the Syndiotactic Polymerization of Styrene

32 32 36 39 40 40

42

Norio Tomotsu, Thomas H. Newman, and Richard Campbell Jr.

4.1 4.2

Introduction Insertion of the Growing Polymer Chain into the Double Bond of Styrene 4.3 Stereochemistry of the Styrene Insertion 4.4 Effects of Hydrogenation of the Catalyst 4.5 Active Site Species 4.5.1 Valence of Active Sites 4.5.2 Number of Active Sites 4.5.3 Structure of Active Sites 4.6 Theoretical Analysis of the Catalyst 4.7 Kinetic Analysis of Styrene Polymerization 4.8 Conclusions References

42 42 45 47 48 48 52 53 54 54 57 58

5. Copolymerization of Ethylene with Styrene: Design of Efficient Transition Metal Complex Catalysts

60

Kotohiro Nomura

5.1 5.2 5.3

Introduction Ethylene/Styrene Copolymers: Microstructures, Thermal Properties, and Composition Analyses Ethylene/Styrene Copolymerization Using Transition Metal Complex–Cocatalyst Systems 5.3.1 Half-Titanocenes, Cp′TiX3 5.3.2 Linked (Constrained Geometry Type) Half-Titanocenes 5.3.3 Modified Half-Titanocenes, Cp′Ti(L)X2

60 61 64 64 65 71

CONTENTS

5.3.4 5.3.5 5.3.6

Non-Cp Titanium Complexes Metallocenes Others

5.4 Summary and Outlook References

ix

79 83 85 86 87

6. Structure and Properties of Tetrabenzo[a,c,g,i]fluorenyl-Based

Titanium Catalysts

92

Rüdiger Beckhaus, Kai Schröder, and Jürgen Schellenberg

6.1 6.2 6.3

Introduction The Tbf Ligand Tbf Lithium 6.3.1 Synthesis and Characterization of Tbf Lithium 6.4 Tbf Titanium(III) Derivatives 6.4.1 Synthesis of Tbf Titanium(III) Chloride Complexes 6.4.2 Reaction of TbfTiIIICl2(THF) (VIII) with Radicals 6.5 Tbf Titanium(IV) Derivatives 6.5.1 Synthesis of Tbf Titanium Monophenoxide Complexes 6.6 Dynamic and Polymerization Behavior of Tetrabenzofluorenyl Titanium Complexes 6.6.1 Styrene Polymerization 6.7 Conclusions References

7. Rare-Earth Metal Complexes as Catalysts for Syndiospecific Styrene Polymerization

92 94 96 96 98 98 102 105 107 117 119 120 120

125

Klaus Beckerle and Jun Okuda

7.1 Introduction 7.2 Metallocene Catalysts 7.3 Constrained Geometry Catalysts 7.4 Half-Sandwich Catalysts 7.5 Nonmetallocene Catalysts 7.6 Conclusion References

125 126 129 130 134 136 136

x

CONTENTS

8. Syndiospecific Styrene Polymerization with Heterogenized Transition Metal Catalysts

140

Kyu Yong Choi

8.1 8.2

Introduction Kinetics of Syndiospecific Polymerization with Heterogeneous Metallocene Catalysts 8.2.1 Kinetic Profiles of Heterogeneous SPS Polymerization 8.2.2 Liquid Slurry Polymerization with Heterogenized Cp*Ti(OCH3)3 Catalyst 8.2.3 Modeling of Polymerization Kinetics 8.2.4 Molecular Weight Distribution of SPS with Heterogeneous Catalysts 8.3 Nascent Morphology of Syndiotactic Polystyrene 8.3.1 Physical Transitions of Reaction Mixture During Polymerization 8.3.2 Effect of Reaction Conditions on Polymer Morphology 8.4 Concluding Remarks References

PART III STRUCTURE AND FUNDAMENTAL PROPERTIES OF SYNDIOTACTIC POLYSTYRENE 9. Structure, Morphology, and Crystallization Behavior of Syndiotactic Polystyrene

140 141 141 143 145 147 149 149 151 153 153

155

157

Andrea Sorrentino and Vittoria Vittoria

9.1 9.2

9.3

Introduction Polymorphic Behavior of SPS 9.2.1 Crystallization from the Melt State 9.2.2 Crystallization from the Glassy State 9.2.3 Morphology Development in the Presence of Solvents Morphology of the Zigzag Forms 9.3.1 Crystal Structure of the α Form 9.3.2 Crystal Structure of the β Form 9.3.3 Lamellar and Spherulitic Morphology of the Zigzag Forms

157 157 159 160 163 164 164 168 170

CONTENTS

9.4 9.5

Morphology of the Mesomorphic Phases Thermodynamic and Kinetics of Crystallization 9.5.1 Thermodynamic and Kinetics of Crystallization 9.6 Melting Behavior 9.6.1 Equilibrium Melting Temperature of α and β Crystals 9.6.2 Memory Effects 9.7 Structure and Properties of the Crystallized Samples 9.7.1 Morphology of Injection Molded Samples 9.7.2 Relation between Morphology Structure, Processing, and Properties References 10. Preparation, Structure, Properties, and Applications of Co-Crystals and Nanoporous Crystalline Phases of Syndiotactic Polystyrene

xi

173 175 177 178 180 182 183 183 184 186

194

Gaetano Guerra, Alexandra Romina Albunia, and Concetta D’Aniello

10.1 10.2

Introduction Co-Crystals 10.2.1 Crystalline Structures 10.2.2 Processing and Materials 10.2.3 Characterization Studies 10.2.4 Properties and Applications 10.3 Nanoporous Crystalline Phases 10.3.1 Crystalline Structures 10.3.2 Processing and Materials 10.3.3 Characterization Studies 10.3.4 Applications 10.4 Conclusions and Perspectives 10.5 Acknowledgments References 11. Crystallization Thermodynamics and Kinetics of Syndiotactic Polystyrene

194 195 196 199 202 209 212 213 215 217 219 224 225 225

238

Tomoaki Takebe and Komei Yamasaki

11.1 11.2 11.3

Introduction Theoretical Background Equilibrium Melting Point of SPS 11.3.1 Evaluation of Spherulitic Growth Rate G

238 239 240 244

xii

CONTENTS

11.4 Analyses of Spherulitic Growth Rate G 11.5 Comparison Between SPS and IPS References

248 249 250

PART IV COMMERCIAL PROCESSES FOR MANUFACTURING OF SYNDIOTACTIC POLYSTYRENE 253 12. Processes for the Production of Syndiotactic Polystyrene

255

Masao Aida, David Habermann, Hans-Joachim Leder, and Jürgen Schellenberg

12.1 12.2 12.3 12.4

Introduction Monomer Purification Section Catalyst Section Polymerization Section 12.4.1 Continuous Stirred Tank Reactor Process 12.4.2 Continuous Fluidized Bed Reactor Process 12.4.3 Continuous Self-Cleaning Reactor Process 12.5 Styrene Stripping Section 12.6 Deactivating Section 12.7 Pelletizing Section 12.8 Blending Section 12.9 Shipping Section References

255 255 256 256 257 258 258 260 260 262 262 263 264

PART V PROPERTIES, PROCESSING, AND APPLICATIONS OF SYNDIOTACTIC POLYSTYRENE

267

13. Properties of Syndiotactic Polystyrene

269

Tomoaki Takebe, Komei Yamasaki, Keisuke Funaki, and Michael Malanga

13.1 13.2 13.3

13.4

Introduction Rheological Properties of SPS Basic Physical Mechanical Properties of SPS 13.3.1 Thermal Properties of SPS 13.3.2 Mechanical Properties of SPS Orientation of SPS and Properties of Oriented SPS 13.4.1 Properties of Uniaxially Oriented SPS 13.4.2 Properties of Biaxially Oriented SPS (BoSPS)

269 269 272 272 274 281 281 282

CONTENTS

13.5

xiii

Other Important Properties of SPS 13.5.1 Electrical Properties of SPS 13.5.2 Chemical Resistance of SPS References

286 286 287 289

14. Melt Processing of Syndiotactic Polystyrene

290

David Bank, Kevin Nichols, Harold Fowler, Jason Reese, and Gerry Billovits

14.1 14.2

Introduction Compounding 14.2.1 Introduction 14.2.2 Compounding Equipment 14.2.3 Compounding Process Conditions 14.3 Injection Molding 14.3.1 Introduction 14.3.2 General Product Design 14.3.3 Thin Wall Product Designs 14.3.4 Injection Mold Design 14.3.5 Injection Mold Melt Delivery System (Runners and Gates) 14.3.6 Venting 14.3.7 Injection Molding Cooling Cycle and Crystallinity 14.3.8 Shrinkage during the Cooling Phase 14.3.9 Injection Molding Process Set-up 14.3.10 Injection Molding Cycle 14.3.11 Special Injection Molding Cycles 14.4 Sheet and Film Extrusion 14.4.1 Introduction 14.4.2 Extrusion 14.4.3 General Extruder Design 14.4.4 Processing Parameters 14.4.5 Material Drying 14.5 Film Processing and Fabrication 14.5.1 Introduction 14.5.2 Cast Film Extrusion 14.5.3 Thermoforming 14.6 Fiber Spinning References

290 294 294 295 296 298 298 299 301 301 302 304 304 304 306 308 310 311 311 311 312 313 314 314 314 314 315 316 319

xiv

CONTENTS

15. Applications of Syndiotactic Polystyrene

321

Tom Fiola, Akihiko Okada, Masami Mihara, and Kevin Nichols

15.1 Introduction 15.2 The Performance Capabilities of SPS 15.3 Connectors for Automotive and Electronic Applications 15.4 Electronic Components: Plated and Non-Plated 15.5 Industrial and Appliance Components References 16. Blends of Syndiotactic Polystyrene with Polyamide

321 322 329 330 331 337 338

Kevin Nichols, Akihiko Okada, and Hiroki Fukui

16.1 16.2

Introduction Composition of SPS/Nylon Blends 16.2.1 Polyamides Used in SPS/Nylon Blends 16.2.2 SPS/Nylon Blend Formulations 16.2.3 SPS/Nylon Blend Composition Patents 16.2.4 SPS/Nylon Blend Compositions Described in Technical Journals 16.3 Properties of SPS/Nylon Blends 16.3.1 Mechanical Properties of SPS/Nylon Blends 16.3.2 Rheology of SPS/Nylon Blends 16.3.3 Moisture Absorption and Moisture Growth of SPS/Nylon Blends 16.3.4 Dimensional Stability of SPS/Nylon Blends 16.3.5 USCAR Performance of SPS/Nylon Blends 16.3.6 Environmental Stress Crack Resistance of SPS/Nylon Blends 16.4 Applications of SPS/Nylon Blends 16.4.1 SPS/Nylon Blend Under-the-hood Automotive Connectors 16.4.2 SPS/Nylon Blend Carpet Fibers 16.4.3 SPS/Nylon Blend Application Patents References 17. Blends of Syndiotactic Polystyrene with Polystyrenes

338 338 339 339 339 339 339 340 343 343 346 347 349 349 349 350 353 355 360

Tomoaki Takebe, Komei Yamasaki, Akihiko Okada, and Takuma Aoyama

17.1 17.2

Introduction SANS Measurements

360 361

CONTENTS

17.3 Theoretical Background 17.4 Tacticity Effect on Miscibility 17.5 Properties of Blends of SPS and APS References 18. Compatibilizers for Impact-Modified Syndiotactic Polystyrene

xv

361 363 366 370 371

Tomoaki Takebe, Akihiko Okada, and Nobuyuki Sato

18.1 18.2

Introduction Morphological Analyses of HISPS 18.2.1 SAXS Profiles of HISPS in the Crystalline State 18.2.2 Effect of Nucleators on Lamellar Orientation in HISPS 18.3 Morphology of SPS/PPO Binary Blends 18.3.1 Structural Analyses Using SAXS Technique 18.3.2 Crystallization Kinetics of SPS/PPO Blends 18.3.3 Influence of Blending PPO with Different Molecular Weights on the Morphology of HISPS 18.4 Compatibilizer Effects 18.4.1 Evaluation of Interaction Parameters 18.4.2 Evaluation of Domain Size and Interfacial Thickness References

371 372 374 375 376 377 378 380 382 383 388 393

PART VI POLYMERS BASED ON SYNDIOTACTIC POLYSTYRENES

395

19. Functionalization and Block/Graft Reactions of Syndiotactic Polystyrene Using Borane Comonomers and Chain Transfer Agents

397

T. C. Mike Chung

19.1 19.2

19.3

Introduction Functionalization of SPS via Borane Comonomers 19.2.1 Copolymerization of Styrene and B-styrene 19.2.2 Side-Chain Functionalized SPS Polymers 19.2.3 SPS Graft Copolymers Functionalization of SPS via Borane Chain Transfer Agents 19.3.1 SPS Containing a Terminal Functional Group 19.3.2 SPS Block Copolymers

397 398 398 402 406 409 410 412

xvi

CONTENTS

19.4 Summary 19.5 Acknowledgment References 20. Nanocomposites Based on Syndiotactic Polystyrene

415 415 415 417

O Ok Park and Mun Ho Kim

20.1 20.2 20.3 20.4 20.5

Introduction Polymer Nanocomposites and Microstructure Fabrication of Polymer Nanocomposites Characterization of Polymer Nanocomposites Preparation of SPS Nanocomposites 20.5.1 Effect of Alkyl Chain Aggregation in Organoclay—Bilayer versus Monolayer Arrangement 20.5.2 Improvement in the Thermal Stability of Organoclay 20.6 Properties of SPS Nanocomposites 20.6.1 Mechanical Properties 20.6.2 Crystallization Behavior 20.6.3 Dynamic Rheological Properties 20.7 Final Remarks References

INDEX

417 418 419 420 421

423 424 425 425 426 426 427 429

431

Preface

In 1953, Karl Ziegler discovered that selected transition metal compounds can be activated by aluminum alkyls and can be used as organometallic catalysts to polymerize ethylene. One year later, Giulio Natta synthesized further stereoregular polymers such as polypropylene. Both scientists were awarded the Nobel Prize for Chemistry in 1963. Their fundamental work resulted in a significant increase in research intensity and in an immediate commercialization of these polymers. After these discoveries, one of the most important achievements in coordination polymerization chemistry has been the introduction of methylaluminoxanes as a new class of aluminum alkyl activators by Sinn and Kaminsky at the end of the 1970s, prepared by controlled hydrolysis of methylaluminum alkyls. Such cocatalysts together with metallocene compounds led to an outstanding enhancement of polymerization activity and efficiency and to a considerable improvement of the molecular uniformity of the polymers obtained. Thus, methylaluminoxanes together with metallocene catalysts allowed the synthesis of highly stereoregular and stereoblock polypropylenes, ethylene copolymers with a higher content of comonomer or with higher α-olefins and other monomers, cycloolefin polymers of high crystallinity and their copolymers, as well as polyethylenes with improved rheological properties by controlled long-chain branching. As a result of these developments, syndiotactic polystyrene (SPS) was prepared by activated titanium compounds as polymerization catalysts. The first SPS was synthesized by Ishihara et al. in 1985 using a homogeneous organometallic catalytic system based on titanium compounds and methylaluminoxane as cocatalyst. The polystyrene initially obtained was a polymer with a 2-butanone-insoluble content as a measure of the syndiotacticity of 98 wt%, a weight-average molecular weight of 82,000 g/mol, and a melting temperature of about 270 °C, 40 °C higher than that of the isotactic polystyrene known since 1955. However, the most important advantage of this SPS in comparison to the isotactic polymer is the much higher crystallization rate of the polymer melt, comparable to that of polyethylene. This high crystallization rate enabled an advantageous processing of the polymer by extrusion and injection molding techniques. After the development of commercially viable transition metal catalysts with high polymerization activity, this new, highly stereoregular polymer was successfully commercialized in 1999. Further desirable properties xvii

xviii

PREFACE

of this polymer are high heat resistance based on the high melting temperature of 270 °C; excellent chemical resistance to reagents including acids, bases, oils, water, and steam; good electrical properties as insulating materials with a low dielectric constant and a low dissipation factor; low polymer density of 1.05 g/ cm3 with nearly equivalent densities in crystalline and amorphous regions; excellent processing characteristics at a very low melt viscosity; outstanding dimensional stability; and low moisture absorption. The less desirable brittleness of SPS has been overcome by reinforcement with glass fibers, altogether leading to an advantageous new polymer in the class of semicrystalline engineering thermoplastics. This book comprehensively covers all aspects of SPS, from the synthesis of this new polymer by coordination polymerization to the characterization of structure, properties, and behavior of the neat material as well as of plastic materials, including the different processing opportunities of such materials, to the widely diversified applications of polymers and blends, also considering SPS-based polymers. The Introduction gives a historical overview of SPS from the first discovery through developmental stages to the full commercialization of this polymer based on an inexpensive monomer (Chapter 1). Because the transition metal catalysts for the coordination polymerization of styrene are of high importance for the properties of the polymers, these catalysts are comprehensively covered in the section on the preparation of SPS. After an overview of the transition metal catalysts for SPS, primarily halfmetallocenes of group 4, in Chapter 2, details on cocatalysts for SPS are given in Chapter 3 followed by an overview on mechanism and kinetics of the syndiospecific styrene polymerization, including active site formation, in Chapter 4. The design of efficient transition metal complex catalysts for the copolymerization of ethylene with styrene is described in Chapter 5. The subsequent chapters include some selected groups of special transition metal catalysts, such as novel tetrabenzo[a,c,g,i]fluorenyl-based titanium catalysts as derivatives of the highly efficient hydrogenated fluorenyl catalysts (Chapter 6), rareearth metal complexes (Chapter 7), and heterogenized transition metal catalysts (Chapter 8). In the section on structure and fundamental properties of SPS, Chapter 9 summarizes the polymorphic behavior of this polymer, the structure of the different forms, and the crystallization and melting behavior. Chapter 10 describes co-crystals and nanoporous crystalline phases of SPS regarding preparation, structure, properties, and new interesting applications, for example, molecular sensors. The section concludes with Chapter 11 on selected topics of crystallization thermodynamics and kinetics of SPS. The next section on commercial processes for manufacturing of SPS describes the process for the production of this polymer in more detail (Chapter 12). The comprehensive section on properties, processing, and application of SPS discusses the rheological, mechanical, and other properties of this polymer

PREFACE

xix

(Chapter 13) and continues with melt processing, including injection molding, extrusion, films, and fibers (Chapter 14). Chapter 15 goes on to describe applications of SPS polymers themselves, followed by a discussion of blends with polyamides in Chapter 16 and with conventional polystyrenes in Chapter 17. Compatibilizers for impact-modified SPS are covered in Chapter 18. The last section on polymers based on SPS includes the different opportunities for functionalization and modification of SPS in Chapter 19 and nanocomposites based on SPS described in Chapter 20. The intent of this book is to provide the reader with comprehensive knowledge about SPS based on a historical overview on this polymer, not only by summarizing the fundamentals of this stereoregular polymer but also by describing the latest new developments in this area. However, this book cannot be exhaustive because of the very broad interest this polymer has received in academic research as well as in industry. Gratefully acknowledged is the excellent work done by all authors, as well as the successful collaboration with the publisher, John Wiley & Sons, Inc., at all stages of the preparation and publishing of this book. Jürgen Schellenberg March 2009

Contributors

Masao Aida, Process Development Center, Idemitsu Kosan Co., Ltd., Chiba, Japan Alexandra Romina Albunia, Dipartimento di Chimica, University of Salerno, Fisciano (SA), Italy Takuma Aoyama, Xarec and SPS Products, Idemitsu Chemicals USA Corporation, Southfield, MI, USA David Bank, Dow Automotive R&D, The Dow Chemical Company, Midland, MI, USA Klaus Beckerle, RWTH Aachen University, Aachen, Germany Rüdiger Beckhaus, Institute of Pure and Applied Chemistry, Carl von Ossietzky University, Oldenburg, Germany Gerry Billovits, Core R&D New Products, The Dow Chemical Company, Midland, MI, USA Richard Campbell Jr., Chemical Sciences, The Dow Chemical Company, Midland, MI, USA Kyu Yong Choi, Department of Chemical and Biomolecular Engineering, University of Maryland, College Park, MD, USA T. C. Mike Chung, Department of Materials Science and Engineering, The Pennsylvania State University, University Park, PA, USA Concetta D’Aniello, Dipartimento di Chimica, University of Salerno, Fisciano (SA), Italy Tom Fiola, Xarec and SPS Products, Idemitsu Chemicals USA Corporation, Southfield, MI, USA Harold Fowler, Ventures & Business Development, The Dow Chemical Company, Midland, MI, USA xxi

xxii

CONTRIBUTORS

Hiroki Fukui, PP Automotive Materials Division, Prime Polymer Co., Ltd., Shizuoka, Japan Keisuke Funaki, Polymer Development Center, Idemitsu Kosan Co., Ltd., Chiba, Japan Gaetano Guerra, Dipartimento di Chimica, University of Salerno, Fisciano (SA), Italy David Habermann, Information Systems Management, The Dow Chemical Company, Midland, MI, USA Osamu Isogai, Kukankoubou Co., Ltd., Chiba, Japan Shigeo Iwasaki, Technology and Engineering Department, Idemitsu Kosan Co., Ltd., Chiba, Japan Mun Ho Kim, Department of Chemical and Biomolecular Engineering, Korea Advanced Institute of Science and Technology, Daejeon, Korea Masahiko Kuramoto, Chemicals Development Center, Idemitsu Kosan Co., Ltd., Chiba, Japan Hans-Joachim Leder, R&D Dow Central Germany, Dow Olefinverbund GmbH, Schkopau, Germany Hiroshi Maezawa, Technology and Engineering Department, Idemitsu Kosan Co., Ltd., Tokyo, Japan Michael Malanga, R&D Advanced Technologies, The Dow Chemical Company, Auburn Hills, MI, USA Masami Mihara, Idemitsu Kosan Co., Ltd., Chiba, Japan Thomas H. Newman, Science Division, Delta College, University Center, MI, USA Kevin Nichols, R&D Dow Building Solutions, The Dow Chemical Company, Midland, MI, USA Kotohiro Nomura, Graduate School of Materials Science, Nara Institute of Science and Technology, Nara, Japan Akihiko Okada, Engineering Plastics Department, Idemitsu Kosan Co., Ltd., Tokyo, Japan

CONTRIBUTORS

xxiii

Jun Okuda, RWTH Aachen University, Aachen, Germany O Ok Park, Department of Chemical and Biomolecular Engineering, Korea Advanced Institute of Science and Technology, Daejeon, Korea Jason Reese, Dow Automotive R&D, The Dow Chemical Company, Midland, MI, USA Nobuyuki Sato, Engineering Plastics Department, Idemitsu Kosan Co., Ltd., Osaka, Japan Jürgen Schellenberg, R&D Dow Central Germany, Dow Olefinverbund GmbH, Schkopau, Germany Kai Schröder, Faculty of Mathematics and Natural Sciences, Molecular Inorganic Chemistry, Stratingh Institute for Chemistry, Groningen, the Netherlands Andrea Sorrentino, Chemical and Food Engineering Department, University of Salerno, Fisciano, Italy Tomoaki Takebe, Chemicals Development Center, Idemitsu Kosan Co., Ltd., Chiba, Japan Mizutomo Takeuchi, Research & Development Department, Idemitsu Kosan Co., Ltd., Tokyo, Japan Norio Tomotsu, Advanced Technology Research Laboratories, Idemitsu Kosan Co., Ltd., Chiba, Japan Vittoria Vittoria, Chemical and Food Engineering Department, University of Salerno, Fisciano, Italy Takeshi Yamada, Performance Materials Laboratories, Idemitsu Kosan Co., Ltd., Chiba, Japan Komei Yamasaki, Polymer Development Center, Idemitsu Kosan Co., Ltd., Chiba, Japan

About the Editor JÜRGEN SCHELLENBERG

Jürgen Schellenberg is a specialist in polymer research with broad interests in polystyrene and styrenic polymers. He was born in 1953 in Sangerhausen/ Germany, studied chemistry at the Technical University “Carl Schorlemmer” Leuna-Merseburg, and obtained a PhD in polymer science with a thesis on the polymerization of vinyl chloride with metal acetylacetonates. Jürgen began his career in the polystyrene department of the plastics R&D division of the Chemische Werke Buna in Schkopau in 1979. There he has worked on general purpose polystyrenes, high-impact polystyrenes, styrene-acrylonitrile copolymers, acrylonitrile-styrene-butadiene terpolymers, special purpose styrenic polymers such as toner resins, and on polyethylene and polyethylene blends. After a research stay at the Dow Chemical Company, Midland, MI, USA in 1997/1998, he continued to work on syndiotactic polystyrene (SPS) research at Dow Central Germany/Schkopau accompanied by the start-up and operation of the first commercial SPS plant worldwide, and later on worked on polypropylene and expandable polystyrene R&D too. Jürgen holds more than 70 patents and has published over 47 scientific papers including reviews.

xxv

PART I

INTRODUCTION

CHAPTER 1

Historical Overview and Commercialization of Syndiotactic Polystyrene MICHAEL MALANGA,1 OSAMU ISOGAI,2 TAKESHI YAMADA,3 SHIGEO IWASAKI,4 and MASAHIKO KURAMOTO5 1

R&D Advanced Technologies, The Dow Chemical Company, Auburn Hills, MI, USA Kukankoubou Co., Ltd., Chiba, Japan 3 Performance Materials Laboratories, Idemitsu Kosan Co., Ltd., Chiba, Japan 4 Technology & Engineering Department, Idemitsu Kosan Co., Ltd., Chiba, Japan 5 Chemicals Development Center, Idemitsu Kosan Co., Ltd., Chiba, Japan 2

1.1

DISCOVERY OF SYNDIOTACTIC POLYSTYRENE (SPS)

The polymerization of styrene monomer has been known for over 100 years and, for about 60 years, has been one of the most important polymeric materials in the world. The commercial success of atactic polystyrene is and has been based on many factors including its low cost, high clarity, good electrical properties, ability to be foamed, and its ease of polymerization. It is, of course, a completely amorphous polymer in the atactic configuration and it has a glass transition temperature (Tg) of about 100 °C above which it is easily formed into useful objects. This glass transition temperature is also one of its limitations in a practical sense since it cannot be used in applications above this temperature. This polymer was only known in the atactic configuration up until the 1950s when the use of heterogeneous coordination catalysts yielded polystyrene in the isotactic configuration. The discovery of isotactic polystyrene (IPS) gave a new dimension to this material since it now could crystallize and provide a melting point (Tm) of around 250 °C. Although it still has a Tg of 100 °C, the material will maintain its shape and may be used for many applications above this Tg and below the Tm. IPS has been the subject of several intense efforts for commercialization. Ultimately it has been unsuccessful for one primary reason that being the rate at which the polymer will crystallize is too slow under normal forming Syndiotactic Polystyrene, Edited by Jürgen Schellenberg Copyright © 2010 John Wiley & Sons, Inc.

3

4

HISTORICAL OVERVIEW AND COMMERCIALIZATION OF SYNDIOTACTIC POLYSTYRENE

processes. This makes it very difficult to use in injection molding or extrusion processes where the crystalline properties do not develop at a rate fast enough for practical use. The use of nucleating aids can increase the overall apparent rate of crystallization but not enough to overcome the relatively slow ability of IPS segments to move into its inherent helical crystal structures. Many people speculated on the possibility of creating a polystyrene molecule where the monomer units were all configured in a syndiotactic configuration. However, this polymerization had never been successfully reported, leaving many to speculate that the monomer could never be configured in this manner. In 1985, researchers at Idemitsu Kosan Central Research Laboratory were experimenting with the polymerization of styrene monomer using some of the recently discovered soluble coordination catalysts developed at the time using methylaluminoxane (MAO) as a counterion for titanium-based catalysts. These catalysts were recently developed and were being used in olefin polymerizations (primarily ethylene and propylene) with claims of high reactivity. What they discovered was that styrene monomer was polymerized by these types of catalysts but that the resulting polymer had a melting point of about 273 °C, which was well above that of the known isotactic structure Tm of 250 °C. Upon further analysis, the backbone structure of this form of polystyrene was found to be syndiotactic in configuration, and for the first time, this polymer was reported and briefly described in a paper at a meeting of the Japanese Polymer Society in August of 1986. In that paper, the structure and basic properties were reported but not the method of polymerization or the catalyst used. The report by Ishihara et al. described for the first time a new polymer from styrene monomer. What caught the interest of many people was that SPS, unlike IPS, would crystallize to greater than 50% at a rate that made it potentially practical and useful. In addition, it had a melting point some 20 °C greater than the IPS form. This was a true discovery because Idemitsu Kosan Co., Ltd. (IKC) had created a new polymer from an existing monomer. Styrene monomer is manufactured at a rate of hundreds of millions of pounds each year around the world and is done so at a relatively low cost. In a very real sense, the world was introduced to “a new trick by an old dog.” With the publication of this report by IKC researchers on the discovery of SPS, a flurry of activity was set in motion in several laboratories around the world in an effort to find the catalysts used. In a coincidence of timing, researchers at The Dow Chemical Company (TDCC) had been investigating MAO counterions for polyethylene catalyst research. Researchers there also had a renewed interest in IPS based on the ability to produce very high purity styrene monomer for anionic polymerization at a large commercial scale. By December of 1986, Dow had been able to independently replicate the IKC discovery. At that time, Dow was the largest commercial manufacturer of both styrene monomer and atactic polystyrene in the world, and a decision was made to pursue research and development of this new polymer. At the same time, an intense research and development effort was in progress at Idemitsu and this had in fact already been ongoing for almost 2 years.

EARLY YEARS OF DEVELOPMENT (1985–1989)

1.2

5

EARLY YEARS OF DEVELOPMENT (1985–1989)

In 1988, unbeknownst to the rest of the world, Idemitsu and Dow entered into a joint development agreement to facilitate the more rapid commercial introduction of SPS. Idemitsu had already applied for many patents around the world including basic composition of matter for SPS, the polymerization process, the catalysts, and many of the applications for this new material. Dow had also submitted patent applications in many of the same areas, but these were generally predated by the IKC patent applications. Dow had, however, endeavored to scale up the polymerization process and was already producing SPS at the scale of several hundred pounds per batch in a 500-gal pilot plant reactor. Based on the strengths, interests, and positions of both companies, an agreement was reached to share research and development results in a combined effort to accelerate this development. This agreement included the sharing of all jointly developed future intellectual properties. At the same time, a commercial agreement was put in place to allow both companies to share the value created. Dow would have an exclusive license from Idemitsu on all the discovery patents for SPS that predated the agreement. The excitement at both companies revolved around the fact that this was a new semicrystalline polymer with a melting point that made it competitive in that regard with polyethylene terephthalate (PET), polybutylene terephthalate (PBT), nylon 6 and 6,6, poly phenylenesulfide (PPS), and even liquid crystalline polymers (LCP) while at the same time bringing the chemical resistance and dielectric properties of polystyrene. Because the crystallization rate was relatively high, it could be injection molded and could still develop high crystallinity directly out of the molds. It was quickly realized that these apparent crystallization rates could be enhanced with the use of added nucleating agents. With glass fiber, rubber modification, and ignition resistance additives, it was anticipated that SPS would become an alternative engineering plastic with its own advantages. At the same time, early work in film and fiber extrusion showed that SPS could undergo strain induced crystallization with good oriented strength and modulus. This opened up the prospects of applications in those markets as well. Clear films were possible if the crystalline phase was kept small enough not to scatter light. The key commercial prospects stemmed from the combination of • • • • •

high heat resistance, good chemical and electrical properties, some unique optical properties as a film, ease of formability, and a low cost, available monomer as the only raw material.

If the catalyst and polymerization process could be developed in such a way that the manufacturing costs were kept down, it was anticipated that SPS could

6


take its place in the plastics market rapidly as a new engineering plastic material.

1.3

INTENSE DEVELOPMENT YEARS (1989–1996)

The initial catalysts used in the discovery of SPS involved the use of MAO made by the careful reaction of water and trimethyl aluminum as a counterion and cyclopentadienyl titanium trichloride as the active site. This proved to be a very effective catalyst system for coordinating and inserting the monomer in a syndiotactic configuration but was relatively inefficient in terms of yield. This would leave a large amount of alumina and titanium oxide in the final product that needed to be removed (deashed) to purify the polymer. Titanium alkoxides and alkyls were also found to be effective and yields could be improved but still left enough residual ash in the polymer that it required a deashing purification step. This deashing was generally accomplished by washing the powder from polymer with aqueous caustic to dissolve out the alumina and then with repeated water extraction to remove the caustic residues prior to drying. Although small ampoule polymerizations could take advantage of the improved kinetics afforded by carrying out the conversion in 100% styrene monomer, it became quickly apparent that this is a difficult system to scale up. The SPS polymer will quickly precipitate from the monomer as it crystallizes and after 5%–10% conversion, it becomes a gelled mass that will solidify to a plug if left to continue to convert. The fact that the monomer still swells the polymer creates a sticky and difficult to agitate mass that overwhelms most reactor agitators. For these reasons, the first larger-scale polymerizations were conducted as slurry polymerizations using a long-chain alkyl solvent such as isooctane. In this process, the polymer precipitates as fine particles (being very insoluble in the alkane) and can readily be agitated and pumped out of reactors after completion of polymerization as a slurry of SPS particles in the nonsolvent. The slurry polymerization also afforded a very good means to control and remove the heat of polymerization by vaporization and condensation of the cooled solvent back into the reactor. Dow used this process to make the first 2000 lbs of SPS in multipurpose slurry reactor pilot plants. The SPS needed to be filtered from the alkane, dried, and then deashed with aqueous base and water extraction and then further devolatilized to produce a pure SPS homopolymer. Idemitsu and Dow together explored a number of alternative process technologies for the SPS polymerization. The kinetics of the catalysts being developed really required that one develop a process where the relative local concentration of monomer was nearly 100%. This would allow for the high yields per unit of catalyst needed to eliminate the residual catalyst removal steps, to simplify the process, and to reduce cost. The problems with these systems were that the heat of polymerization needed to be controlled to obtain the molecular weights and molecular weight distributions desired in the


7

product and that the polymer would inevitably foul agitator blades and reactor walls as it progressed through the conversion curve in batch systems. In the late 1980s and early 1990s, the two companies were running pilot plant facilities with some form of reactors that had continuously wiped walls and agitator blades. In one form, these were extruder-like reactors that only carried out the polymerization through the first 10%–20% of the total conversion and then fed a monomer-swollen powder into a larger agitated reactor to complete the conversion and to deal with the remaining heat evolution through the reactor walls. These systems worked well and allowed for the required market development quantities with millions of pounds produced over several years. This process, however, still suffered from difficulties of heat removal in the first reactor, and scale-up required high capital costs because of the expensive extruder prereactor. Styrene monomer purification and the effects of trace impurities on catalyst deactivation were also an important part of the early development of the process technology for SPS. Much of the technology for oxygen and water removal to less than 1 ppm was leveraged from years of research and development on anionic polystyrene. These impurities and the removal of them on an industrial scale were well known. In addition to those obvious impurities for air/water sensitive catalysts, there are a number of unsaturated impurities and oxidation impurities (oxides, aldehydes, and ketones) that can form in the manufacture and storage of styrene monomer and can have a detrimental effect on the soluble coordination catalysts used for SPS polymerization. Dow and Idemitsu isolated and identified the effects of two key impurities, indene and phenyl acetylene in the monomer. A number of effective treatments were put in place to reduce these to very low levels. Improvements in yield and molecular weight control were realized through the total purification of the monomer and any solvents used. This was an important part of the process development for a commercially viable manufacturing plant. By 1993, it was apparent that a reduction in the capital and operating cost and an improvement in plant reliability and uptime would require modification of the polymerization reactor systems from what was operating in the pilot facilities at both Idemitsu and Dow. In fact, what was required was an entirely new to the world polymerization process specifically designed to the requirements and characteristics of the SPS polymerization. SPS polymerization is unique and therefore requires a new polymerization process. The polymer is insoluble in the monomer and will precipitate at very low conversion (less than 1%–2%). The kinetics of polymerization at the temperatures that the catalysts are stable and the desired molecular weights can be achieved are highly favored in 100% concentration (pure monomer). Furthermore, the polymer is not very soluble in any suitable polymerization solvents below 110 °C. These limitations make solution polymerization impractical on a commercial scale. An additional restraint is that the polymerization must occur in the liquid phase. No vapor-phase activity has been shown for styrene with these catalyst systems. At any rate, styrene monomer would be very difficult and expensive to keep in the vapor form at these low tempera-

8


tures, requiring very high vacuum levels. Batch polymerization is possible, but because of the heat of polymerization and the fact that the polymer forms the very “sticky” gel phase described previously, continuously wiped agitator and wall reactors are required. Batch or bulk polymerization processes of this sort are also limited by heat removal capabilities. Styrene monomer polymerizes very exothermically with a heat evolution of 17.8 kcal/mole. Controlling this heat throughout the reacting mass is essential to making the molecular weight desired and maintaining catalyst activity. Polymerization temperature must be kept at ∼50–80 °C depending on the desired molecular weight. Emulsion and suspension polymerization systems are not practical due to the need to completely exclude water. Although a suspension is possible in a fluorinated nonsolvent for styrene, the cost of such solvents and the purification of recycle streams again make it economically impractical on a commercial scale. The breakthrough in process thinking came by considering a fluidized powder bed polymerization reaction system where the average conversion in the reactor was kept at a steady-state high level. This required a mechanical fluidization of the reacting powder mass and a continuous addition of monomer and catalyst into the reactor with continuous removal of polymer powder out the bottom of the reactor. In addition, one can control and remove the heat of polymerization through the use of a low-boiling nonreactive solvent continuously added to the reacting powder mass, vaporized, externally condensed, and recycled to the reactor. One further requirement of this kind of polymerization process is that in some way, new particles of polymer powder must be produced in the reactor or continuously added as “seed” to maintain a steady state of surface area for the polymerization. Process research uncovered that this seeding can be accomplished through the action of friction on particles and interparticle collisions and does not require additional seeding once the reaction is under way. In the years between 1993 and 1995, Idemitsu and Dow process engineers and chemists worked together to modify and to perfect this new polymerization process, which became the basis for the design of the future manufacturing plants. This was another historically important milestone for SPS commercialization. No other polymer in the world is manufactured in a process quite the same as this, and this development allowed for a cost-effective process to be scaled up to commercial-scale plants. A closely analogous system is the vapor-phase fluidized beds used for polypropylene (PP) manufacturing, but in that case, the monomer can be kept in the vapor phase and does not need to be added as a liquid. This also allows one to use the propylene monomer vapor as a means to remove the heat of polymerization. PP also requires the addition of “seed” heterogeneous catalyst particles. Another close analogy is the precipitation mass polymerization system for polyvinyl chloride (PVC). In that case, the polymer does precipitate from the monomer, but it does not go through a self-adhering “sticky” stage. This allows


9

for seed particles to be generated outside the main reactor and then to be added along with more monomers and initiators with minimized reactor fouling. In the case of PVC, higher temperatures can also be tolerated and no other solvents need to be added as vinyl chloride can be vaporized to remove heat. In this same time frame, there was continuing development of the catalyst systems to improve efficiency and to increase rate and yield. At Dow and at Idemitsu, as well as in several other research laboratories, the development of soluble metallocene catalysts for olefin polymerization was occurring at the same time that SPS development was under way. The activation of these metallocene-based catalysts with MAO or with other anionic counterions allows for highly efficient and selective olefin polymerization. By careful manipulations of the ligand structure, the selectivity and efficacy of these metallocenes became a significant breakthrough in the catalysis for SPS as well. In addition, the optimization of the MAO structure and the use of important activators and cocatalysts all contributed to tremendous improvements in catalyst technology for this polymerization. The importance of these catalyst improvements for the SPS process cannot be overstated. Those improvements in efficiency allowed for the elimination of a deashing process step and thereby a significant reduction in the capital and operating cost of a commercial plant. Concurrent with this rapid advancement in catalyst and process development, the product and application development for SPS was also in progress. It was realized early on that the commercial viability of SPS as a hightemperature engineering plastic would rely on the effective use of the crystalline phase of the material. The glass transition temperature (Tg) of the amorphous phase of SPS is approximately 100 °C, which is identical to that of the atactic amorphous polystyrene used throughout the world. For that reason, the practical use of SPS at temperatures above 100 °C requires that the modulus drop be mitigated through the use of reinforcing fillers such as glass or carbon fiber. This is the same material science used for other engineering polymers such as PBT or nylon where reinforcing fibers are added to bridge the crystalline phase and to maintain the modulus and strength of the material above the Tg. Glass fiber reinforcement of all these semicrystalline polymers requires good adhesion between the polymer matrix and the fiber surface. The pure polystyrene backbone does not offer significant chemical bonding chemistry opportunities with its relatively inert phenyl ring and hydrocarbon backbone. To overcome this in a practical sense requires the correct silane surface additives for the glass fibers and a compatible polymer phase with the SPS that will also chemically bond with that silane coupling agent. Idemitsu and Dow developed several successful technical solutions to this and then optimized around the use of a proprietary modified polymer additive during the extrusion compounding step. This system enabled the effective reinforcement required and was a key development leading to a number of products from 15% to 45% glass fiber content for various applications.

10


Other important product developments that were worked on during this time included: •

•

•

nucleating agents to improve the apparent rate of crystallization and to facilitate better injection molding products thermal and oxidative stabilizers to allow for processing at temperatures above the melting point up to ∼320 °C and for long-term application durability ignition-resistant additives for applications requiring that the plastic meet certain regulations with regard to flame and ignition tests.

An additional need in some applications was for improved impact resistance in both unfilled and filled grades. Unmodified atactic polystyrene (APS) itself is a brittle polymer. There are several ways in which it is modified to improve the impact strength and to allow its range of application to be expanded. High-impact polystyrene (HIPS) and acrylonitrile–butadiene–styrene (ABS) copolymer are just two examples for APS. Below its glass transition temperature, SPS brittle fracture is very similar to APS. Research to improve SPS toughness followed similar strategies of adding impact modifiers that allowed for absorption of energy on impact without initiating large cracks that would propagate rapidly and cause failure of parts. In general, this involved the addition of compatible polymers, copolymers, or graft copolymers with glass transition temperatures below zero. One complication for SPS over APS in this regard is the upper use temperatures required. APS is only useful to a point just below Tg since it will lose modulus and therefore dimensional properties above Tg. In that case, butadiene-based or other unsaturated rubber copolymers can be used to toughen it. For SPS, researchers were faced with the fact that as an engineering plastic, it would be used well above Tg to temperatures where oxidative degradation of these unsaturated rubber materials would be an issue in long-term use. It was therefore important for researchers to develop toughened grades with polymer additives that were resistant to thermal oxidation, and one very good way to accomplish this is with saturated styrene ethylenebutylene styrene (SEBS) block copolymers. These copolymers are used already as tougheners for some APS systems when improved oxidative stability is needed. They are highly compatible with the amorphous regions of the SPS matrix and they act to absorb energy through some crazing and crack blunting mechanisms. Researchers at Idemitsu also took advantage of the miscibility of APS and SPS to make PS/SPS blends (using HIPS as part or all of the polystyrene (PS) phase) to create a balanced blend of heat resistance, chemical resistance, and impact properties for some applications. The development of all these grades required extrusion melt compounding of the SPS matrix and the various additives. This required process research and development around the effective compounding of SPS above its melting


11

point of 270 °C without thermal degradation of the polymer. Compounding screw designs, addition sequences, die designs, and cooling and pelletizing equipment all had to be developed specifically to the needs of SPS. Almost all of this information is know-how and trade secret to the compounding manufacturing process, but again is important in a historical sense to developing the technology required to provide commercial grades. The market development efforts from 1990 to 1995 for these grades of compounded SPS revolved around applications requiring heat, chemical and electrical properties combined with ease of manufacture, and a unique property of dimensional stability for a semicrystalline plastic material. This last property is due to the fact that the density of the crystalline and the amorphous phases of SPS are almost identical. This has the practical effect that the development of the crystalline phase during cooling does not impart warp or bow in fabricated parts to nearly the extent of most other polymers in this class. The major market areas that were targeted during this development phase included: • • • • •

electrical and electronics, automotive, appliance, packaging, and specialty films and fibers.

Another historical development involved the need to develop film and fiber grade products with improved clarity and strength. In this case, it is important to develop products that take advantage of the high crystalline melting point but can be fabricated in such a way that strain applied at temperatures between Tg and Tm induces that crystallization. The pure homopolymer of SPS has an apparent crystallization rate that is high enough that it becomes very difficult on a practical application basis to quench it from the melt phase and to keep it uniformly amorphous. Research in this case revolved around adding comonomers that would slow down the crystallization rate enough without complete loss of crystallinity and without too much reduction in the melting point. In the early composition of matter, process and product patents that IKC had filed, reference and claims to all types of substituted styrene homopolymers and copolymers in the syndiotactic configuration had been made. These became the practical basis for the development of film and fiber grade products with improved (in this case reduced) crystallization rates from the melt. The incorporation of small amounts of alkyl-substituted styrene monomers led to a new line of products for these markets and applications. The almost complete random copolymer of para-methyl styrene and styrene with these catalysts is the basis of this technology. Once it was discovered that this was effective, it was only a matter of optimizing the melt crystallization rate with

12


the melting point and the achievable strain-induced crystallinity for the film and fiber applications. Further research showed that the process and catalyst for such systems are perfectly compatible so these copolymers could be manufactured in the same plants as the SPS homopolymer with little modification. 1.4

INITIAL COMMERCIAL LAUNCH STAGE (1996–2001)

The design and construction of the first commercial plant for SPS was done in Japan by the Idemitsu Petrochemical Company over the years 1994–1996 with plant start-up in the later part of 1996. This plant is capable of producing ∼5000 MT/year of polymer and is located in Chiba, Japan. With the start-up of this first plant and the completion of the license agreement between Idemitsu and Dow, the real launch and sales of SPS began in 1997. Idemitsu launched the SPS product line under the trade name of Xarec primarily in Japan and the rest of Asia. Dow formally launched the Questra line of SPS products about 1 year later, concentrating mainly in North America and in Europe. The new product lines included a number of glass-filled, ignition-resistant, mineral-filled, blended, and neat homopolymer and copolymer grades. The uniqueness of SPS over existing engineering plastics is in the combination of properties it brings together. This combination includes: • • • • • • •

heat resistance, chemical resistance, hydrolytic stability, ease of injection molding and other melt fabrication, dimensional stability out of the mold and with increasing temperature, good electrical properties, and lower density than competitive engineering plastics.

This is the special blend of properties that was brought into the market with the introduction of Xarec and Questra. The markets described earlier were then explored and validated with these products, and applications that have developed in those markets have been varied, numerous, and continue to expand (Table 1.1). Idemitsu also introduced plating grade systems with zero filler or glass fiber/ mineral fillers as well as pure homopolymers and copolymers for certain film applications. Recently a new series of Xarec products based on PS/SPS alloys has been introduced, aimed at taking advantage of the chemical resistance enhancement gained with the addition of the semicrystalline SPS phase in the APS. Since there is complete compatibility between these polymers, these alloys are

13 TABLE 1.1

Explored Markets with Syndiotactic Polystyrene

Automotive Appliances Electrical/electronic Water supply

Electrical connectors, electrical components, sensors and module cases, water pumps, lighting, mirror brackets Microwave oven parts, vacuum cleaner components, dishwashers, rice cookers, lighting components Connectors, motors, coil and sensor components, switches, power terminals, wiring components Pump housings, piping components

readily compounded. This has expanded the use of SPS products into more commodity applications. These products are too low in SPS to take full advantage of the melting point in terms of heat resistance. However, they offer improvements in chemical resistance over standard PS grades that make them useful in bathroom components, air conditioner drip pans, and refrigerator components. They have also found application in some packaging applications when chemical resistance to oils, detergents, etc. is required. These alloys also offer the advantage of being completely suitable for recycle as polystyrene material. The Dow Chemical Company also designed and constructed a full-scale production plant that went online in 1999 in Schkopau, Germany and had a name plate capacity of nearly 36,000 MT/year. The process technology and catalyst technology for this plant were similar although not identical to the Idemitsu plant. Dow developed the Questra product in line with the Xarec products: glassfilled, ignition-resistant, and toughened grades with properties and application developments parallel to Xarec. Toughened, glass-filled grades were also offered with long glass fiber for improved strength and modulus. Dow developed and commercialized a technically successful SPS/nylon alloy series that combined the best attributes of nylon 6,6 and SPS. Mainly, these blends offer significantly improved toughness (impact resistance) of the nylon while maintaining some of the chemical resistance, hydrolytic stability, and dimensional stability of the SPS.

1.5

YEARS 2001–2007

In 2005, Dow announced it was exiting from the Questra business, citing slower than anticipated growth and profitability. This exit involved the decision to shut down the plant in Germany and to stop production. IKC continued with its strong commitment to this new polymer from a business perspective, announcing capital improvements and modernization of the plant in Chiba, Japan completed in 2006. It is expected that this polymer will continue to find more and more uses in the decades to come. This book will explore much of the technology that has led SPS to find its place in the world.

PART II

PREPARATION OF SYNDIOTACTIC POLYSTYRENE

CHAPTER 2

Transition Metal Catalysts for Syndiotactic Polystyrene NORIO TOMOTSU,1 THOMAS H. NEWMAN,2 MIZUTOMO TAKEUCHI,3 RICHARD CAMPBELL JR.,4 and JÜRGEN SCHELLENBERG5 1

Advanced Technology Research Laboratories, Idemitsu Kosan Co., Ltd., Chiba, Japan Science Division, Delta College, University Center, MI, USA 3 Research & Development Department, Idemitsu Kosan Co., Ltd., Tokyo, Japan 4 Chemical Sciences, The Dow Chemical Company, Midland, MI, USA 5 R&D Dow Central Germany, Dow Olefinverbund GmbH, Schkopau, Germany 2

2.1

INTRODUCTION

Polystyrene is an inexpensive and rigid plastic. Polystyrene is a kind of vinyl polymer. Structurally, it is a long hydrocarbon chain with a phenyl group attached to every other carbon atom. Usually, polystyrene is produced by a radical or anionic polymerization starting from the monomer styrene. These polymers are atactic and amorphous and consequently are not crystallizable. One of the most important achievements in the field of synthetic polymer chemistry is the discovery of the coordination polymerization of polyethylene by Ziegler in 1953 [1]. After Ziegler’s discovery, Natta succeeded in the polymerization of propylene and other α-olefins with stereoregularities. In 1955, Natta and Pino successfully prepared isotactic polystyrene (IPS) using titanium tetrachloride together with alkylaluminum compounds [2]. IPS is a semicrystalline polymer with a high melting point of about 240 °C [3]. Several catalysts have been investigated to commercialize this polymer, but the crystallization rate of IPS is too slow to be practical in injection molding processes. In the field of the transition metal catalyzed coordination polymerization, the stereoregularly controlled polymerization of vinyl monomers was dramatically improved by Sinn and Kaminsky [4]. They discovered the combination of a homogeneous metallocene and methylaluminoxane (MAO) acting as a

Syndiotactic Polystyrene, Edited by Jürgen Schellenberg Copyright © 2010 John Wiley & Sons, Inc.

17

18

TRANSITION METAL CATALYSTS FOR SYNDIOTACTIC POLYSTYRENE

high-performance catalyst. In contrast with the traditional Ziegler–Natta catalysts, the metallocene ligands allow monomer insertions in a very homogeneous manner. Due to this structural control, metallocene catalysts can be easily modified to control polymer stereoregularity, molecular weight, and molecular weight distribution. In 1985, Ishihara et al. were successful in the synthesis of syndiotactic polystyrene (SPS) using cylopentadienyltitanium compounds with MAO [5]. This is the first known case of the syndiospecific polymerization of styrene. In the following chapter, an attempt is made to review the history of the development of catalysts for the syndiospecific polymerization of styrene.

2.2 2.2.1

TRANSITION METAL COMPOUNDS Metals

The syndiotacticity of SPS results from the homogeneous coordination polymerization mechanism. Many metal complexes were examined, but only group 4 transition metal complexes or rare earth metal complexes with cocatalysts such as MAO, pentafluorophenyl borate, or borane derivatives allow the synthesis of SPS with high activities [6–10]. Some results are summarized in Table 2.1. Yang, Cha, and Shen examined rare earth coordination catalysts [9]. The Nd(naph)3/Al(iso-Bu)3 catalyst system was found to produce syndiotactic-rich polystyrene. They proposed that the catalytically active species might be an ionic complex because the addition of CCl4 increased the catalytic activity. The high performance of catalysts using rare earth metal complexes is summarized in Chapter 7. Among the group 4 transition metal complexes, titanium complexes showed higher activities than zirconium and hafnium complexes. It could be related to a less electrophilic character of the metal during an interaction with the monomer. Some cyclopentadienyl zirconium complexes like Cp*ZrCl3 or Cp*ZrF3 give SPSs with lower activity and lower stereospecificity than titanium complexes, whereas others like Cp*Zr(CH3)3 and CpZrCl3, or the biscyclopentadienyl complexes Cp2ZrCl2 and Si(CH3)2Cp2ZrCl2, lead to atactic polymers [11]. Many noncyclopentadienyl zirconium complexes are known, for example, Zr(On–C3H7)4, Zr(Oi–C3H7)4, Zr(C7H8)2, (C6H5COCHCOC6H5)2ZrCl2, ((CH3)3CCOCHCOC(CH3)3)2ZrCl2, ((CH3)3CCOCHCOC(CH3)3)2ZrCl2, (C2B9H11)Zr(N(C2H5)2)2(NH(C2H5)2), (acac)2ZrCl2, and Zr(CH2C6H5)4, which polymerize styrene to syndiotactic polymers, but the activities are low and the stereospecificities of the polymers formed are lower than those of titanium complexes. Contrary to the titanium and zirconium complexes, the hafnium compounds investigated, like Cp2HfCl2 and Cp*Hf(CH3)3, only provide atactic polymers.

TRANSITION METAL COMPOUNDS

TABLE 2.1

19

Polymerization of Styrene Using Metal Compounds with MAO

Catalyst TiCl4 TiBr4 Ti(OCH3)4 Ti(OC2H5)4 CpTiCl3 Cp2TiCl2 C5(CH3)2TiCl2 C5(CH3)2TiClH Ti(η-C6H6)2 Ti(η-CH3C6H6)2 Ti(η-(CH3)2C6H6)2 Ti(η-(CH3)3C6H6)2 Ti(acac)2Cl2 Ti(N(C2H5)2)4 ZrCl4 CpZrCl3 Cp2ZrCl2 Cp2HfCl2 Cp2VCl2 Nb(OC2H5)5 Ta(OC2H5)5 Cr(acac)3 Co(acac)3 Ni(acac)2 Ni(acac)2

[A1] mol

Conversion (wt %)

Stereospecificity

0.04 0.04 0.04 0.04 0.015 0.03 0.03 0.03 0.03 0.025 0.025 0.025 0.025 0.04 0.04 0.04 0.04 0.04 0.04 0.04 0.04 0.04 0.01 0.01 0.01 0.04

4.1 2.1 3.8 9.5 68.2 99.2 1.0 2.0 8.8 5.4 5.9 5.7 6.0 0.4 0.4 0.4 1.3 1.3 0.7 0.7 0.2 0.1 1.8a 1.8a 80.8a 31.0b

Syndiotactic Syndiotactic Syndiotactic Syndiotactic Syndiotactic Syndiotactic Syndiotactic Syndiotactic Syndiotactic Syndiotactic Syndiotactic Syndiotactic Syndiotactic Syndiotactic Syndiotactic Atactic Syndiotactic Atactic Atactic Atactic Atactic Atactic Atactic Atactic Atactic Isotactic

Polymerization conditions: metal compounds (5×10−5 mol), styrene (23 ml), and toluene (100 ml) at 50 °C for 2 h [6,7]. a Metal compounds (2.5×10−5 mol), styrene (50 ml), and toluene (100 ml) at 50 °C for 2 h. b Metal compounds (3.3×10−5 mol), styrene (10 ml), and toluene (10 ml) at room temperature for 20 h [8].

2.2.2

Titanium Complexes

The titanium complexes show higher polymerization activities and result in polymers with higher syndiotacticities than the other metal complexes. Initial evaluations of various titanium compounds with MAO were published [5,7,10]. Titanium halide compounds (e.g., TiCl4, TiBr4, CpTiCl3, Cp*TiCl3) and titanium compounds without halogen atoms (e.g., Ti(OC2H5)4, Ti(OC4H9)4, and Ti(N(C2H5)2)4) can produce SPS. Not only Ti(IV) but also Ti(III) compounds, such as CpTiCl2, together with MAO as cocatalyst, lead to the formation of SPS. Zambelli, Oliva, and Pellecchia [7] reported that Ti(II) (e.g., Ti(Ph)2) also could produce SPS, but Ti(bipy)3, formally Ti(0), gave atactic polystyrene. However, as shown in Table 2.1, Ti(0) compounds can also produce SPS [6].

20


Pyrazolylborate complexes of titanium [12] have been studied for SPS polymerization. Similarities between the cyclopentadienyl ligand and the hydridotris(pyrazolyl)borate ligand have been noted for transition metal complexes. But the catalytic activities of pyrazolylborate complexes are much lower than those of the analogous pentamethylcyclopentadienyl complexes. This ligand may donate too much electron density to the titanium. 2.2.2.1 Variation of the π-Bonded Ligand The catalytic activity was found to vary according to the ligands at the titanium. The influence of monoor bis-cyclopentadienyl substitutions of the transition metal complex on the catalytic activities was investigated. Among the SPS-producing catalysts, titanium complexes with one cyclopentadienyl ligand yield the highest activities for SPS (Table 2.2). The polymerization activities of titanium complexes with one cyclopentadienyl ligand have been reported [12,13]. The catalytic activities increase with the number of methyl groups at the Cp ligand. This is parallel to an increase of the molecular weight of the polymers. The chain propagation reaction is accelerated by the electron-releasing properties of the substituents. Hafner and Okuda [14] and Gassman, Campbell, and Macomber [15] showed the relationship between catalytic activities and 49Ti-nuclear magnetic resonance (NMR) chemical shifts. 49Ti-NMR peaks are shifted to the lower magnetic field by introducing an electron-donating substituent at the cyclopentadienyl ligand. Yokota et al. [18] also examined the catalyst activities of titanium complexes with one cyclopentadienyl ligand and the 49Ti-NMR chemical shifts (Table 2.3). The data indicate that substituents at the cyclopentadienyl ligand TABLE 2.2

Catalyst Performance of Ti Complexes

Catalyst TiCl4 TiBr4 Ti(OCH3)4 Ti(NEt2)4 Ti(acac)2Cl2 CpTiCl3 CpTi(III)Cl2 CpTi(III)Cl2 Ti(η-C6H6)2 Ti(η-CH3C6H6)2 Ti(η-(CH3)3C6H6)2 Cp2TiCl2 CH3TiCl3 Cp*TiCl3

mmol

[Al]/[Ti]

Conversion wt %

Polymerization Conditions

0.05 0.05 0.05 0.02 0.01 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05

40 40 40 10 40 30 30 30 25 25 25 30 30 45

4.1 2.1 3.8 0.4 0.4 99.2 44.1 44.1 5.4 5.9 5.9 1.0 1.0 100

1 1 1 3 1 2 1 1 2 2 2 2 1 2

Polymerization conditions: styrene/toluene (ml/ml): 1, 180/100; 2, 23/50; 3, 50/100; temperature: 50 °C; time: 2 h.


TABLE 2.3 Ligands

21

Catalytic Activities of Titanium Compounds with Cyclopentadienyl

Complex CpTiCl3 (MrCp) TiCl3 (1,2,4-MeCp) TiCl3 (Me5Cp) TiCl3 (EtMe4Cp) TiCl3

Relative Activitya

Mw

Mw/Mn

100 110 110 156 180

62,000 63,000 63,000 750,000 870,000

2.2 2.2 2.2 2.2 2.2

49

Ti Chemical Shiftb (ppm) −394 −332 −95.3 −103.28

a

Catalytic activity of CpTiCl3 as 100. TiCl4 as standard. Mw=weight average molecular weight; Mn=number average molecular weight. b

TABLE 2.4 Ligands

Catalytic Activities of Titanium Compounds with Cyclopentadienyl

Complex (Me5Cp) TiCl3 (EtMe4Cp) TiCl3 (n-PrMe4Cp) TiCl3 (n-BuMe4Cp) TiCl3 (i-PrMe4Cp) TiCl3 (cyc-HexMe4Cp) TiCl3 (AdaCH2Me4Cp) TiCl3

Activity (kg/gTi)a

Mw

Mw/Mn

170 180 47 107 62 47 54

750,000 870,000 945,000 906,000 406,000 559,000 678,000

2.2 2.2 2.2 2.2 2.2 2.2 2.2

49

Ti Chemical Shiftb (ppm) −95.3 −103.28 −102.30 −101.00

a

Polymerization conditions: Ti : MAO :TIBA (molar ratios) = 1 : 200 : 200 at 70 °C for 1 h. TiCl4 as standard.

b

with electron-releasing properties increase the polymerization activities. Ready, Chien, and Rausch [16] and Kim, Koo, and Do [17] reported similar results using indenyl–titanium complexes. Yokota examined tetramethylcyclopentadienyl ligands with different substituents. Me5CpTiCl3, (Cp*TiCl3), EtMe4CpTiCl3, n-PrMe4CpTiCl3, and n-BuMe4CpTiCl3 showed almost the same 49Ti-NMR chemical shifts. These titanium complexes have a similar state of the titanium, but the catalytic activities of n-PrMe4CpTiCl3, i-PrMe4CpTiCl3, and n-BuMe4CpTiCl3 are lower than those of Me5CpTiCl3 and EtMe4CpTiCl3 [18]. Moreover, the catalytic activities of titanium compounds with bulky substituents like cyc-HexMe4CpTiCl3 and AdaCH2Me4CpTiCl3 are lower than those of Me5CpTiCl3 (Table 2.4). This result suggests the stabilization of the active site by electron-releasing substituents, and bulky substituents decrease the catalytic activities. In this correlation, the catalytic activities of titanium compounds with bulky substituents at the cyclopentadienyl ligand show lower activities than expected from the chemical shifts of 49Ti-NMR. The electron densities of titanium compounds can also be observed when the titanium complexes possess the same

22


Catalytic activity (kgSPS/gTi)

300

Et Me 4 Cp

250

Me 5Cp

200 Me 4Cp

150 1.2.4-Me 3Cp

100 50

(SiMe 3 )2 Cp

t-Bu2 Cp

0 4.04

4.06

4.08

CNMe 4 Cp Cp

4.10

4.12

Chemical shift of MeO (ppm)

Figure 2.1 Relationship between the chemical shift of the methoxy group and the catalytic activities.

H8

H9

H4

H25

H15

H24

H3

H23

H14 H13

H10

C15

C11

C9

H17

C7

H5 C8

C2 H12

H7

C5 C4

C14

C12

C3 H11

H6 C6 H1

H16

C10 C1

H19 H21

H22

C13

H2

H20 H18 T11 CL3

CL2

Figure 2.2

CL1

X-ray crystal structure of (C5Et5)TiCl3.

sigma-bonded groups. The linear relationship between the chemical shifts of titanium compounds containing methoxy groups in 1H-NMR and the catalytic activities is shown in Figure 2.1 [19]. This relationship also indicates that bulky ligands reduce the catalytic activities. Yokota compared Et5CpTiCl3 and Me5CpTiCl3. One ethyl group of the cyclopentadienyl ligand turns to the titanium atom and reduces the catalytic activity of the complex (Fig. 2.2). In this case, molecular weights of the polymers formed with bulky ligand titanium


23

compounds are lower than those with titanium compounds that show a linear correlation between activities and electron densities at the titanium. The coordination of the monomer to the titanium may also be obstructed by bulky substituents. Ready et al. [20] observed that indenyltitaniumtrichloride, IndTiCl3, is a significantly more active catalyst than CpTiCl3. But Tomotsu et al. compared the two catalysts and found that the catalytic activity of IndTiCl3 is lower than that of CpTiCl3 [13]. The difference in polymerization conditions may account for the observed differences in the catalyst performances. Furthermore, Chien [21] investigated the influence of aromatic substituents at the indenyl ligands. The results suggested that benzindene stabilized the active catalytic species more than the phenyl substitution at the C5 ring for the indenyl ligand. Yokota found that 2-methylbenzindenyltitanium and pentamethylcyclopentadienyltitanium compounds had almost the same catalytic activity [18]. The polymerization activities for several ansa–titanocene complexes have been reported by Miyashita, Mabika, and Suzuki [22]. The catalytic activity increases by decreasing the bite angle, the angle of the Cp centroid–Ti–Cp centroid (Table 2.5). The activities of ansa–titanocene complexes are lower than those of the monocyclopentadienyl complexes. These results indicated that highly active titanium compounds exhibit one cylopentadienyl ligand with electron-donative substituents, and bulky substituents reduce the catalytic activities. From this point of view, 1,2,3-trimethyl-tetrahydroindenyltitanium compounds were examined, resulting in high catalytic activities (Table 2.6). The effect of methyl groups as substituents in tetrahydroindenyltitanium compounds is shown in Figure 2.3. The catalytic activities and the stabilities of the compounds are changed by the substituents. 500

kg/gTi/h

400

300

200 1,2-Me2[65]Ti(OMe)3 1,3-Me2[65]Ti(OMe)3

100

1,2,3-Me3[65]Ti(OMe)3 0 40

50

60

70

80

Polymerization temperature (°C)

Figure 2.3 Relationship between substitution in tetrahydroindenyltitanium compounds and catalytic activities.

24


TABLE 2.5

Effects of the Bite Angle of Cp Ligands on Catalyst Performance Bite Angle (°)

Activity (g/g-Ti/h)

SPS/Total PS (%)

Tacticity (rrrr %)

131 137

21 57

11 73

94 94

Ti Cl Cl

121

573

96

99

Ti Me Me

121

1037

98

99

Cp2TiCl2 Cp*2 TiCl 2

Me2Si

Ti Cl Cl

128

669

98

99

Me2Si

Ti Me Me

128

957

98

99

131

38

11

95

Ti Cl Cl

PS=polystyrene.

TABLE 2.6

Side Ring Effects of Cyclopentadienyl Ligands of Titanium Compounds

Complex

Cl

a

Relative Activity

Ti Cl Cl

49

Ti Chemical Shifta

100

−95.3

Cl

Ti Cl Cl

122

−80.7

Cl

Ti Cl Cl

42

−60.9

Use 49TiCl4 as standard.


362/39600

397/55400

390/559000

400/509000

297/5260000

340/632000

297/452000

240/2380000

25

Figure 2.4 Catalytic activities of titanium compounds with different ring sizes and substituents. (The numbers are the catalytic activity (kg/[gTi × h]) and the Mw of SPS; polymerization conditions: styrene/MAO/TIBA/Ti (molar ratios) = 26300/200/200/1 at 60 °C for 1 h.)

The effect of the ring size on the catalytic activity is examined in Figure 2.4. The catalytic activities of titanium compounds with a six-membered ring have the highest activities, and the methyl group substitution at the cyclopentadienyl ligand does not change the catalytic activities. However, the molecular weight of the SPS is high when titanium complexes with three methyl groups are used. Novel titanium complexes with corannulenyl-based ligands were also synthesized and investigated [23]. A cyclopentadienyl ligand with two six-membered rings, the 1,2,3,4,5,6,7,8-octahydrofluorenyl ligand, can be synthesized by the hydrogenation of fluorene [24]. 1,2,3,4,5,6,7,8-Octahydrofluorenyltitanium compounds were examined for the styrene polymerization, and it was found that the catalytic activities are high enough and the molecular weight of the polymers produced is between those of Cp*Ti(OMe)3 and Me4CpTi(OMe)3 [25,26]. 2.2.2.2 σ-Bonded Ligands The polymerization activities in the presence of Cp*TiR3 compounds, in which R is the alkoxide and chloride ligands, which are activated by MAO, are as follows [13], in order of decreasing catalytic activity: Cp*Ti(Oi-Pr)3 and Cp*Ti(OMe)3 > Cp*Ti(OPh)3 > Cp*Ti(OC6H4CH3)3 > Cp*TiCl3 > Cp*Ti(Ot-Bu)3 > Cp*Ti(Oi-C3HF6)3. The chloride ligand and the electron withdrawing substituted alkoxide ligands decrease the conversion as does the bulky tert-butoxide ligand. The methoxide, isopropoxide, phenoxide, and p-methylphenoxide ligands are all similar in terms of conversion. The addition of triisobutylaluminum (TIBA) to these titanium compounds increases the catalytic activities, and the catalytic activities adjust among the different Cp*TiOR3 complexes (Table 2.7).

26


TABLE 2.7 Catalytic Activities of Titanium Catalysts with Different Alkoxy Groups Complex Cp*Ti(OEt)3 Cp*Ti(Oø)3 Cp*TiMe3 Cp*Ti(CH2ø)3

Relative Activitya

Mw

Mw/Mn

440 450 580 520

520,000 560,000 590,000 550,000

2.5 2.4 2.4 2.1

a

Catalytic activity of CpTi(OMe)3 without TIBA as 100.

Cp*Ti(OMe)3

Cp*TiCl 3

Cp*Ti(OPr) 3

3315

3340

3365 G

Figure 2.5 ESR spectrum of mixtures of MAO, TIBA, and Cp*Ti(OR)3.

The electron spin resonance (ESR) spectra of the toluene mixtures of the different titanium compounds, MAO, and TIBA are the same (Fig. 2.5). The active site of the syndiospecific styrene polymerization is the same after the reaction of the titanium compounds with TIBA. 2.2.3

Molecular Weight Control

The molecular weight of SPS can primarily be controlled by the polymerization temperature, which sharply affects the rate of chain transfer via β-hydride elimination. As the polymerization temperature is increased, the SPS molecular weight decreases significantly. However, substituents at the cyclopentadienyl ring of the monometallocene titanium complexes can also affect the SPS molecular weight. A dramatic decrease in the molecular weight of the SPS produced with CpTiCl3 is observed relative to that with Cp*TiCl3. This would


27

indicate that β-hydride elimination occurs more readily for CpTiCl3. The Cp* ligand apparently stabilizes the active center and retards β-hydride elimination. 2.2.4

Supported and Heterogeneous Catalysts

Mixtures of highly isotactic and syndiotactic polystyrenes are mainly obtained using heterogeneous titanium compounds such as TiCl3 and TiCl4 supported on Mg compounds [27] (Table 2.8). There are two types of polymers arising from two different active sites. The SPS fraction increases by increasing the molar ratio of Al to Ti. After the supported catalyst was washed by toluene, the catalyst performance was examined. The toluene-soluble portion of the catalyst was found to produce SPS and the insoluble portion of the catalyst was found to result in IPS. Syndiospecific polystyrene is also produced using SiO2-supported Ti(OC4H9)4 with MAO, and a SiO2-supported reacted mixture of Ti(OC4H9)4 and MAO (Table 2.9) [28]. The syndiotacticity of the polymers with both catalysts was almost 100%. The catalytic activity is independent of the Al/Ti molar ratio for this supported catalyst. The heterogeneous –SiO(Ti(OC4H9)3) species is more stable against reduction by MAO than the active species in the soluble system. CpTiCl3 or Cp*TiCl3 supported on Al2O3 with TIBA affords mixtures of IPS and SPS [29]. Two possible initiating sites are thought to be present on the catalyst surface. The first one is formed by the reaction of CpTiCl3 with

TABLE 2.8

Polymerization of Styrene Using Ti Compounds with MAO

Catalyst TiCl3(AA) TiCl3(Solvay) Mg(OEt)2/EB/ TiCl4

TiCl4

Ti(OEt)4

(mmol)

Al/Ti

Conversion (%)

Stereospecificity

1.0 0.2 1.0 0.2 2.0

100 1000 20 1000 50

8.2 2.0 1.9 0.9 2.9

0.02

500

1.1

0.02

1000

1.4

10 40 500 10 50 500

7.2 0.4 0.7 0.3 2.5 0.9

Isotactic + Syndiotactic Isotactic + Syndiotactic Isotactic + Syndiotactic Isotactic + Syndiotactic Isotactic (84) + Syndiotactic (16) Isotactic (12) + Syndiotactic (88) Isotactic (10) + Syndiotactic (90) Isotactic Isotactic + Syndiotactic Syndiotactic Atactic Syndiotactic Syndiotactic

40 5 0.2 2 2 0.2

Polymerization conditions: styrene, 50 ml; toluene, 100 ml; 50 °C; 2 h.

28

0.034 0.034 0.034 0.034 0.100 0.100 0.100 0.100 0.360 0.270 0.140

Amount (g)

— — — — 1.00 1.00 1.00 1.00 0.616 0.616 0.616

Ti — — — — — — — — 16.1 11.0 14.2

Al

Amount of Metals on SiO2(mmol/g) Ti

mmol

0.5 1.0 2.0 3.0 1.0 2.0 3.0 4.0 5.8 3.0 2.0

MAO 5 10 20 30 10 20 30 40 26 18 23

Al/Ti

Polymerization Conditions

0.10 0.10 0.10 0.10 0.10 0.10 0.10 0.10 0.22 0.17 0.088

Polymerization of Styrene Using Supported Catalysts with MAO

Ti(OBu)4 Ti(OBu)4 Ti(OBu)4 Ti(OBu)4 Ti(OBu)4/SiO2 Ti(OBu)4/SiO2 Ti(OBu)4/SiO2 Ti(OBu)4/SiO2 Ti(OBu)4/MAO/SiO2 Ti(OBu)4/MAO/SiO2 Ti(OBu)4/MAO/SiO2

Catalyst

TABLE 2.9

47 30 860 716 595 1800 2190 2100 3450 10,222 27,028

Polymer Yield (mg/mmolTi)

REFERENCES

29

hydroxyl groups on the surface of Al2O3 and is responsible for isospecific polymerization [30]. The second one is formed by the reaction of CpTiCl3 with Lewis acid sites and leads to syndiospecific polymerization.

2.3

SUMMARY

The different transition metals for the syndiospecific polymerization of styrene were summarized. Compounds of titanium with one cyclopentadienyl ligand show a high performance for the SPS production. Transition metals are stabilized by cyclopentadienyl ligands with electron-releasing substituents, and bulky substituents decrease the catalytic activities. σ-Bonded groups at the transition metal complex are substituted by other groups by MAO or TIBA and showed comparable catalytic activities in the syndiospecific styrene polymerization [31].

REFERENCES 1. Boor, J. Jr. Ziegler-Natta Catalysts and Polymerization, Academic Press, New York, 1979. 2. Natta, G., Pino, P., Mantica, E., Danusso, F., Mazzanti, G., Peraldo, M. Stereospecific polymerization of α-olefins. Chim. Ind., 38, 124–127 (1956). 3. Natta, G., Danusso, F., Sianesi, D. Stereospecific polymerization and isotactic polymers of vinyl aromatic monomers. Makromol. Chem., 28, 253–261 (1958). 4. Sinn, H., Kaminsky, W. Ziegler-Natta catalysis. Adv. Organomet. Chem., 18, 99– 149 (1980). 5. (a) Ishihara, N., Seimiya, T., Kuramoto, M., Uoi, M. Synthesis and properties of polystyrene with new stereoregularity. Polym. Prepr. Jpn., 35, 240 (1986). (b) Ishihara, N., Seimiya, T., Kuramoto, M., Uoi, M. Crystalline syndiotactic polystyrene. Macromolecules, 19, 2464–2465 (1986). (c) Ishihara, N., Kuramoto, M., Uoi, M. Japanese Patent 62187708 (to Idemitsu Kosan Co. Ltd.), 1986. (d) Ishihara, N., Kuramoto, M., Uoi, M. Japanese Patent 62104818 (to Idemitsu Kosan Co. Ltd.), 1986. (e) Ishihara, N., Kuramoto, M., Uoi, M. European Patent 210,615 A2 (to Idemitsu Kosan Co. Ltd.), 1987.(f) Ishihara, N., Kuramoto, M., Uoi, M. European Patent 224,097 A1 (to Idemitsu Kosan Co. Ltd.), 1987. (g) Ishihara, N., Kuramoto, M., Uoi, M. U.S. Patent 4,680,353 (to Idemitsu Kosan Co. Ltd.), 1987. (h) Ishihara, N., Kuramoto, M., Uoi, M. Stereospecific polymerization of styrene giving the syndiotactic polymer. Macromolecules, 21, 3356-3360 (1988). 6. Ishihara, N. Transition metal-catalyzed olefin polymerization. Diss., Oxford, 1990. 7. Zambelli, A., Oliva, L., Pellecchia, C. Soluble catalysts for syndiotactic polymerization of styrene. Macromolecules, 22, 2129–2130 (1989). 8. Longo, P., Grassi, A., Oliva, L., Ammendola, P. Some carbon-13 NMR evidence on isotactic polymerization of styrene. Makromol. Chem., 191, 237–242 (1990).

30


9. Yang, M., Cha, C., Shen, Z. Polymerization of styrene by rare earth coordination catalysts. Polym. J., 22, 919–923 (1990). 10. Hou, Z., Wakatsuki, Y. Recent developments in organolanthanide polymerization catalysts. Coord. Chem. Rev., 231, 1–22 (2002). 11. Longo, P., Proto, A., Oliva, L. Zirconium catalysts for the syndiotactic polymerization of styrene. Macromol. Rapid Commun., 15, 151–154 (1994). 12. Newman, T. H., Campbell, R. E., Malanga, M. T. Metcon ’93, Houston, TX, May 26–29, 1993, Abstracts, 315–324. 13. Tomotsu, N., Kuramoto, M., Takeuchi, M., Maezawa, H. The catalyst for syndiotactic-specific polymerization of styrene. In Metallocenes ’96, Skillmann, N.J. (ed.), Scotland Business Research, Duesseldorf, 1996, pp. 179–196. 14. Hafner, A., Okuda, J. Titanium NMR data for some titanium half-sandwich complexes bearing substituted cyclopentadienyl ligands. Organometallics, 12, 949–950 (1993). 15. Gassman, P. G., Campbell, W. H., Macomber, D. W. An unusual relationship between titanium-49 chemical shift and Ti(2p3/2) binding energy. The use of titanium-49 NMR in evaluating the electronic effect of methyl substitution on the cyclopentadienyl ligand. Organometallics, 3, 385–387 (1984). 16. Ready, T. E., Chien, J. C. W., Rausch, M. D. Alkyl-substituted indenyl titanium precursors for syndiospecific Ziegler-Natta polymerization of styrene. J. Organomet. Chem., 519, 21–28 (1996). 17. Kim, Y., Koo, B. H., Do, Y. Synthesis and polymerization behavior of various substituted indenyl titanium complexes as catalysts for syndiotactic polystyrene. J. Organomet. Chem., 527, 155–161 (1997). 18. Yokota, K., Inoue, T., Naganuma, H., et al. Syndiospecific polymerization of styrene. In Metalorganic Catalysts for Synthesis and Polymerization, Kaminsky, W. (ed.), Springer, Berlin, 1999, pp. 435–445. 19. Tomotsu, N., Ishihara, N. Recent development of catalysts for syndiospecific polymerization of styrene. In Science and Technology in Catalysis, Soga K. (ed.), Kodansha, 1998, pp. 269–276. 20. Ready, T. E., Day, R. O., Chien, J. C. W., Rausch, M. D. (η5-indenyl)trichlorotitanium. An improved syndiotactic polymerization catalyst for styrene. Macromolecules, 26, 5822–5823 (1993). 21. Chien, J. C. W. New and improved catalysts for syndiotactic polystyrene. In Metallocenes ’96, Skillmann, N.J. (ed.), Scotland Business Research, Duesseldorf, 1996, pp. 223–237. 22. Miyashita, A., Mabika, M., Suzuki, T. Mechanistic study on syndiotactic polymerization of styrene catalyzed by titanocene catalyst. Proceedings of the International Symposium on Synthetic, Structural and Industrial Aspects of Stereospecific Polymerization, Milan, Italy, June 6–10, 1994. 23. Chin, M., Schellenberg, J. Coordination polymerization with the novel η5C20H17Ti(OiPr)3 complex. Eur. Polym. J., 43, 2165–2169 (2007). 24. Newman, T. H., Klosin, J., Nickias, P. N., U.S. Patent 5 670 680 (to Dow Chemical Co.), 1997. 25. Nickias, P. N., Borodychuk, K. K., Newman, T. H. U.S. Patent 5 536 797 (to Dow Chemical Co.), 1996.

REFERENCES

31

26. (a) Schellenberg, J. Effect of catalyst transition metal and ancillary ligand on syndiospecific polymerization of styrene. Eur. Polym. J., 42, 487–494 (2006). (b) Schellenberg, J. New multinuclear half-titanocene catalysts in the syndiospecific polymerization of styrene. Eur. Polym. J., 40, 2259–2267 (2004). (c) Schellenberg, J. The syndiospecific polymerization of styrene in the presence of fluorine-containing half-sandwich metallocenes. J. Polym. Sci. A Polym. Chem., 38, 2428–2439 (2000). (d) Schellenberg, J., Newman, T. H. Effect of phenylsilane on the syndiospecific polymerization of styrene in the presence of half-sandwich metallocenes. Eur. Polym. J., 37, 1733–1739 (2001). (e) Schellenberg, J. Influence of the catalyst on monomer insertion in the syndiospecific copolymerization of styrene and paramethylstyrene. J. Polym. Sci. A Polym. Chem., 43, 2061–2067 (2005). 27. Soga, K., Monoi, T. Polymerization of styrene with Mg(OH)xCl2−x (x = 0-2)supported Ti(OBu)4 catalysts combined with methylaluminoxane. Macromolecules, 23, 1558–1560 (1990). 28. Soga, K., Nakatani, H. Syndiotactic polymerization of styrene with supported Kaminsky-Sinn catalysts. Macromolecules, 23, 957–959 (1990). 29. Soga, K., Koide, R., Uozumi, T. Syndiotactic polymerization of styrene with alumina supported Cp(*)TiCl3 (Cp: cyclopentadienyl, Cp(*): 1,2,3,4,5-pentamethylcyclopentadienyl) catalyst activated by trialkylaluminums. Macromol. Chem. Rapid Commun., 14, 511–514 (1993). 30. Soga, K., Uozumi, T., Yanagihara, H., Shiono, T. Polymerization of styrene with heterogeneous Ziegler-Natta catalysts. Macromol. Chem. Rapid Commun., 11, 229–234 (1990). 31. For comprehensive reviews, see (a) Schellenberg, J., Tomotsu, N. Syndiotactic polystyrene catalysts and polymerization. Prog. Polym. Sci., 27, 1925–1982 (2002). (b) Schellenberg, J. Recent transition metal catalysts for syndiotactic polystyrene. Progress Polym. Sci., 34, 688–718 (2009).

CHAPTER 3

Cocatalysts for the Syndiospecific Styrene Polymerization NORIO TOMOTSU,1 HIROSHI MAEZAWA,2 and THOMAS H. NEWMAN3 1

Advanced Technology Research Laboratories, Idemitsu Kosan Co., Ltd., Chiba, Japan Technology & Engineering Department, Idemitsu Kosan Co., Ltd., Tokyo, Japan 3 Science Division, Delta College, University Center, MI, USA 2

3.1

INTRODUCTION

As described in Chapter 2, Ishihara et al. succeeded in the synthesis of syndiotactic polystyrene (SPS). By using cylopentadienyltitanium compounds and methylaluminoxane (MAO), the reacted materials act as catalyst for the syndiospecific polymerization of styrene. MAO is the special cocatalyst that activates the corresponding transition metal complex. After the discovery of MAO by Kaminsky, many chemicals had been examined as cocatalysts. For an SPS production catalyst, Campbell found that boron compounds based on tris(pentafluorophenyl)boron and its derivatives are suitable as cocatalysts in the syndiotactic polymerization of styrene [1]. In the following, an attempt is made to review the development of cocatalysts, MAO, as well as borane and borate compounds in the syndiospecific polymerization of styrene mediated by transition metal catalysts, including a further activation of these complexes by other compounds.

3.2

MAO

MAO is commonly used as an activator of transition metal catalysts in the syndiotactic polymerization of styrene. MAO acts as a reductant and as an alkylating reagent of the transition metal. Many different aluminum alkyls were examined for the syndiospecific polymerization of styrene (see Table 3.1) [2]. A rapid alkylation of the metallocene by MAO takes place and the active species arises from a ligand transfer reaction between the metallocene alkyls and MAO. Syndiotactic Polystyrene, Edited by Jürgen Schellenberg Copyright © 2010 John Wiley & Sons, Inc.

32

MAO

33

TABLE 3.1 Polymerization of Styrene Using CpTiCl3 with Various Organoaluminum Compounds Organoaluminum (mmol/dm3) — TMA(0.05) TEA(0.05) TMA(0.4) + H2O(0.4) TEA(0.4) + H2O(0.4) MAO(0.4) MAO(0.2) + TMA(0.2) MAO(0.2) + TEA(0.2) MAO(0.2) + TIBA(0.2) Me2AlOAlMe2(0.6) MAO(Mw < 500)(0.4) MAO(Mw > 500)(0.4)

Yield (g)

Conversion (wt %)

Stereospecificity

0.8 0.1 0.2 17.6 0.8 14.9 7.1 0.3 15.5 0.1 Trace 14.9

0.5 0.1 0.1 10.8 0.5 9.2 4.4 0.2 9.5 0.1 — 9.2

Atactic Atactic Atactic Syndiotactic Syndiotactic Syndiotactic Syndiotactic Syndiotactic Syndiotactic Atactic Syndiotactic Syndiotactic

Polymerization conditions: CpTiCl3, 5 × 10−4 mol/l; toluene, 0.1 l; styrene, 8.7 mol/l at 50 °C for 2 h.

Al

O

CH3

Figure 3.1 Various proposed structures of MAO.

MAO can be considered as an oligomer of an aluminum alkyl. Many models of MAO were proposed (Fig. 3.1), but the final detailed structure has not been clarified yet. MAO as the cocatalyst activates the transition metal compounds, resulting in the formation of the active site for the polymerization reaction. From an analytical point of view, MAO might exist in a linear, cyclic, or cage form (Fig. 3.1). The role of MAO for the syndiospecific polymerization of styrene was examined by Miyashita [3]. He used titanocene complexes with bridged cyclopentadienyl (Cp) rings as catalysts. MAOs of different molecular weights were prepared by the distillation of usual MAO, and the effects of the molecular weight of the MAO on the catalytic activity were examined. It was found that Me(Al(Me)O)15AlMe2 showed the highest polymerization activity, and a large amount of MAO as cocatalyst was required. The structure of the reacted compound between the titanocene complex and MAO was analyzed by 13C-

34

COCATALYSTS FOR THE SYNDIOSPECIFIC STYRENE POLYMERIZATION

Ti

CH 3 Cl C Al O H2

H2 C

CH3 Al CH 3

Ti

+

1.5

CH 2

O

Al Al

Me

O n

Me

Figure 3.2

Structure of the active site proposed by Miyashita.

0 –0.4 0.4 Chemical shift (ppm) (a) MAO in toluene

Figure 3.3

0

–0.5 –1.0 –1.5 Chemical shift (ppm)

(b) MAO in toluene and dioxane 1

H-NMR spectrum of MAO.

and 1H-nuclear magnetic resonance (NMR), and the structure of the active site was determined as shown in Figure 3.2. Recently, it has been shown by investigations at very low amounts of MAO molecularly comparable to the transition metal complex and using a halfsandwich metallocene catalyst that MAO as a cage of six monomeric MAO units seems to be more preferred in the active site with regard to the polymerization activity in contrast to MAO consisting of a cage of 12 monomeric units [4]. Miyashita, Nabika, and Suzuki also examined the molar electric conductivity of the reacted compound between the titanocene and MAO [5]. The electric conductivity was found as 0.006 S · cm2/mole in toluene, and it was concluded that the active site for polymerization has the structure of a zwitterionic Ti cation center. MAO is known also to contain trimethylaluminum (TMA), both in a form coordinated with MAO and as free TMA. Figure 3.3a shows the typical 1HNMR spectrum of MAO in toluene [6]. The amount of TMA can be calculated by a detailed peak analysis of the spectrum. The sharpest peak at high field species represents TMA. By the addition of a small amount of dioxane to the toluene solution of MAO, dioxane coordinates with TMA and changes the spectrum of the MAO solu-

MAO

35

Relative catalytic activity (%)

100

80

60

40

20

0

0

10

20

30

40

50

60

TMA contents (%)

Figure 3.4

Effect of TMA on the catalytic activity of Cp*Ti(OMe)3.

Relative catalyst activity (%)

100

80

60

40

20

0 100

200

300 400 Mw of MAO

500

600

Figure 3.5 Effect of the molecular weight (Mw) of MAO on the catalytic activity of Cp*Ti(OMe)3.

tion with TMA. The amount of TMA in MAO can be obtained from the peak area ratio of the sharp TMA peak and the broad MAO peak. The catalytic activity of Cp*Ti(OMe)3 with MAO samples containing different amounts of TMA is shown in Figure 3.4. Clearly, the addition of TMA decreases the catalytic activity of the Cp*Ti(OMe)3 complex. The average molecular weight of MAO can be measured by a cryoscopic method. The MAO solution in toluene is dried up, dissolved in dioxane, and subsequently the temperature of the freezing point can be measured. A small amount of TMA strongly coordinated with the MAO is a contamination to this solution and can cause some error to the molecular weight of the MAO, but the optimum molecular weight of MAO is around 400 g/mol (Fig. 3.5).

36


Modified aluminoxanes sometimes were used as cocatalysts in the syndiospecific polymerization of styrene as well; however, they do not seem to allow the high polymerization activities reached by the MAO cocatalysts.

3.3

BORON COMPOUNDS

Perfluorophenyl borate derivatives have also been used as cocatalysts in SPS polymerizations [1]. Various catalyst systems for SPS using borates or boranes were proposed. B(C6F5)3 [1,7], [NR1R2R3H][B(C6F5)4] [8], [NR1R2R3R4] [B(C6F5)4] [9], and [Ph3C][B(C6F5)4] [10] were used as cocatalysts for the polymerization of styrene. In case of borate as cocatalyst, the catalytic activity of the titanium complex with a pentamethylcyclopentadienyl ligand is high, but a titanium complex with a cyclopentadienyl ligand without any substituents is not active for the syndiospecific styrene polymerization. The reason is that the reaction product of the borate and the cycopentadienyltitanium compound is unstable. The stability of the active site with the borate compound is lower in comparison to that with MAO. The reaction of CH2(Cp)2Ti(Me)2 with dimethylanilinium tetrakis(pentafluorophenyl)borate or tris(pentafluorophenyl)borane in an equimolar mixture has been examined by Miyashita, Nabika, and Suzuki [11]. Two types of methylene bis(cyclopentadienyl)titanium ion complexes were isolated (see Fig. 3.6). These complexes were active in the polymerization of styrene, but only atactic polystyrene was formed. Baird et al. [12] examined the catalytic activity of Cp*TiMe3 with boranes. The styrene polymer received was a mixture of atactic polystyrene and SPS. Examining the effect of the polarity of the solvent in the polymerization by using CH2Cl2 and 1,2-C2H4Cl2, it was found that polar solvents increased the yield of the atactic polystyrene.

Ti

Ti

CH3 [PhNHMe ][B(C F ) ] + 2 6 5 4 CH3

CH3 B(C F ) + 6 5 3 CH3

–C H 4

Ti

– PhNMe2

CH3 B(C6F5)4

Ti

CH3 MeB(C6F5)3

Figure 3.6 Reaction between methylene bis(cyclopentadienyl)titaniumdimethyl and boron compounds.

BORON COMPOUNDS

TABLE 3.2 Activity

37

Effects of Fluorine of the Borate on Catalytic

Borate Compounds

Catalytic Activity (kg/gTi)

[NMe2PhH][B(C6H5)4] [NMe2PhH][B(C6H4F)4] [NMe2PhH][B(2,4-F2C6H3)4] [NMe2PhH][B(1,3,4-F3C6H2)4] [NMe2PhH][B(C6F5)4] [NMe2PhH][B(3-(CF3)C6F4)4]

0 0 5 10 250 20

Polymerization conditions: styrene, 10 ml; Cp*TiMe3, 5 × 10−7 mol; borate, 5 × 10−7 mol; TIBA, 3 × 10−6 mol; polymerization temperature: 70 °C; time 4 h.

Borate compounds active in the syndiospecific polymerization of styrene possess a tetraphenylborate anion. The catalytic activity of Cp*Ti(OMe)3 with dimethylanilinium borate and triisobutylaluminum (TIBA) are summarized in Table 3.2. Borates with different numbers and positions of fluorine substitution were examined, and the substitution in the 3,4,5 positions of the phenyl group increases the catalyst activity. The highest catalytic activity was observed using tetrapentafluorophenylborate. The effect of using a borate compound together with a small amount of TIBA as cocatalyst for the polymerization of styrene to SPS was examined by Campbell [1], Tomotsu [13], and Kucht et al. [14]. TIBA was found to be not only a good scavenger of impurities in styrene monomer, but also a component to increase the number of the active sites as well as of the syndiotacticity of the resulting styrene polymers. Pellecchia et al. [15] also examined the catalytic activity of several organometallic derivatives with similar boranes. They observed that the catalytic activity of Cp*TiMe3 with dimethylaniliniumtetrakis(pentafluorophenyl) borate was lower than that with tris(pentafluorophenyl)borane. It was proposed that the free amine coordinates with the active site and interferes with the polymerization reaction. The reaction of the titanium compounds and the borate forms the active site. The equations are as follows: Cp*TiR 3 + [R 4′ N][ B (C6 F5 )4 ] → [Cp*TiR 2 ] [ B (C 6 F5 )4 ] + R 3′ N + R ′-R

(3.1)

Cp*TiR 3 + [R 3′ C ][ B (C6 F5 )4 ] → [Cp*TiR 2 ] [ B (C 6 F5 )4 ] + R 3′ CR

(3.2)

+

−

+

−

Cp*TiR 3 + [R 3′ NH ][ B (C 6 F5 )4 ] → [ Cp*TiR 2 ] [B (C 6 F5 )4 ] + R 3′ N + R-H. (3.3) +

−

The effects of ammonium ions on the catalyst activity (Eq. 3.1) are summarized in Table 3.3 [16]. Ammonium borate compounds with a lower pKa, such as [2-CN-pyridine(N)Me][B(C6F5)4] (pKa = −0.3), showed a higher activity than those with a

38


TABLE 3.3 Ammonium Borate as Cocatalyst Borate Compounds 1

2

3

4

Catalytic Activity (kg/gTi) 1

2

3

[NR R R R ]

Pka of NR R R

NMe3Ph PyMe 4-CN-PyMe 3-CN-PyMe 2-CN-PyMe

5.1 5.2 1.9 1.0 −0.3

2.0 2.0 2.8 31.8 40.5

Polymerization conditions: styrene, 10 ml; Cp*TiMe3, 5 × 10−7 mol; borate, 5 × 10−7 mol; TIBA, 3 × 10−6 mol; polymerization temperature: 70 °C; time: 4 h.

300

[(MeOPh )3C] +

SPS activity (kgSPS/gTi)

250 200

[(MeOPh )2Ph C] + 150 100

[Ph 3 C] +

50 0 0

20

40

60 80 TIBA/Ti

100

120

Figure 3.7 Carbenium borate as cocatalyst (polymerization conditions: styrene, 10 ml; Cp*Ti(OMe)3, 5 × 10−7 mol; borate, 5 × 10−7 mol; TIBA, 3 × 10−6 mol; polymerization temperature: 70 °C; time: 4 h).

higher pKa. The by-products of the active site formation reaction are thought to coordinate strongly with the active site. On the other hand, the by-product of the reaction between titanium compounds and [Ph3C][B(C6F5)4] is Ph3CR and does not coordinate with the active polymerization site (Eq. 3.2). In this case, however, [Ph3C][B(C6F5)4] reacts with TIBA and is decomposed. [Ph(PhOMe)2C][B(C6F5)4] or [(PhOMe)3C][B(C6F5)4] are more stable compounds and do not react with aluminum alkyls. A lower reactivity against aluminum alkyls results in an increase of the apparent catalyst activity (Fig. 3.7). The decomposition of the borate by TIBA is observed by 1H-NMR, and the excess amount of borate increases the catalyst activity.

OTHER CHEMICALS

3.4

39

OTHER CHEMICALS

MAO and boron compounds are expensive chemicals that increase the cost of the SPS production. However, the cocatalyst is an indispensable compound for the titanium complexes used as styrene polymerization catalysts. The roles of the cocatalyst are supposed to reduce the valence of the titanium compounds by forming the precursor of the active site, to activate the precursor, and subsequently to stabilize the active site by weak coordination. TIBA reacts with the transition metal and reduces Ti(IV) to Ti(III), and this may be the precursor of the active site. Yabunouchi found that the catalytic activity in the styrene polymerization increased by the addition of a small amount of ((R1)3CO)n–Al–(R2)3−n to the styrene monomer, even if the amount of MAO is decreased [17]. One of these compounds is [(C6H5)3CO](i–C4H9)2Al (Fig. 3.8). The catalytic activitiy was increased more than two times by the addition of a small amount of [(C6H5)3CO](i–C4H9)2Al to a mixture consisting of Ti : MAO :TIBA = 1 : 50 : 25 in molar ratio. These compounds can be used for every titanium compound with MAO as a cocatalyst. This is a kind of alkylaluminum, but the effect on the catalytic activity is different from usual alkylaluminums like TMA, TIBA, or trioctylaluminum. Such a chemical is also applicable for borane compounds as cocatalyst. In the catalyst system with borate compounds, the catalytic activity is also increased by the addition of [(C6H5)3CO](i–C4H9)2Al.

90 80 50:25:1 50:25:1 (additive) 110:25:1

Conversion (%)

70 60 50 40 30 20 10 0

0

50

100

150

200

250 Time (min)

Figure 3.8 Effects of the additive [(C6H5)3CO](i–C4H9)2Al (polymerization conditions: styrene, 10 ml; Cp*Ti(OMe)3, 5 × 10−7 mol; polymerization temperature: 70 °C; mixing ratios of MAO and TIBA: ⵧ, 50 : 25 : 1 with [(C6H5)3CO](i–C4H9)2Al in styrene; Δ, 110 : 25 : 1 without [(C6H5)3CO](i–C4H9)2Al in styrene; 䊊, 50 : 25 : 1 without [(C6H5)3CO](i–C4H9)2Al in styrene).

40


This aluminum compound does not change the syndiotacticity of the polymer produced. The detailed mechanism of increasing the efficiency of MAO and borate cocatalysts has not been clarified yet. 3.5

SUMMARY

The cocatalysts for the syndiospecific polymerization of styrene were summarized. MAO and borate or borane compounds are useful cocatalysts for the syndiotactic styrene polymerization. There is an optimum molecular weight of MAO with regard to the polymerization activity of the transition metal complex, whereas TMA as an impurity in MAO reduces the activity of the catalyst complex [18]. The performance of MAO and borane compounds as cocatalysts can successfully be enhanced by the addition of selected new chemicals.

REFERENCES 1. Campbell, R. E. European Patent 421659 (to Dow Chemical Co.), 1990; U.S. Patent 5066741 (to Dow Chemical Co.), 1991. 2. Ishihara, N., Kuramoto, M. Syntheses and properties of syndiotactic polystyrene. Stud. Surf. Sci. Catal., 89, 339–350 (1994). 3. Miyashita, A. Catalyst for styrene polymerization. Polym. Prepr. Jpn., 42, 2295 (1993). 4. Schellenberg, J. Coordination polymerization at very low amounts of methylaluminoxane as cocatalyst. Eur. Polym. J., 41, 3026–3030 (2005). 5. Miyashita, A., Nabika, M., Suzuki, T. Bridged cyclopentadienyl titanium complexes for styrene polymerization. Abstract. 40th Symp. Organometallic Chem. Jpn., 46 (1993). 6. Tomotsu, N. Catalyst for syndiospecific polymerization of styrene. Polym. Prepr. Jpn., 42, 919 (1993). 7. Takeuchi, M. Japanese Patent 4-366108 (to Idemitsu Kosan Co. Ltd.), 1990. 8. Tomotsu, N. Japanese Patent 4-249504 (to Idemitsu Kosan Co. Ltd.), 1990. 9. Takeuchi, M. Japanese Patent 4-366109 (to Idemitsu Kosan Co. Ltd.), 1990. 10. Takeuchi, M., Japanese Patent 5-186527 (to Idemitsu Kosan Co. Ltd.), 1990. 11. Miyashita, A., Nabika, M., Suzuki, T. Mechanism of styrene polymerization. Abstract. 41st Symp. Organometallic Chem. Jpn. 46 (1994). 12. Quyoum, R., Wang, Q., Tudoret, M., Baird, M. η5-C5Me5TiMe3B(C6F5)3: A carbocationic olefin polymerization initiator masquerading as a Ziegler-Natta catalyst. J. Am. Chem. Soc., 116, 6435–6436 (1994). 13. Tomotsu, N. Japanese Patent 2-415574 (to Idemitsu Kosan Co. Ltd.), 1990. 14. Kucht, H., Kucht, A., Chien, J. C. W., Rausch, M. D. (η5-Pentamethylcyclopentadienyl)trimethyltitanium as a precursor for the syndiospecific polymerization of styrene. Appl. Organomet. Chem., 8, 393–396 (1994).

REFERENCES

41

15. Pellecchia, C., Longo, P., Proto, A., Zambelli, A. Novel aluminoxane-free catalysts for syndiotactic-specific polymerization of styrene. Macromol. Chem. Rapid Commun., 13, 265–268 (1992). 16. Naganuma, S. Cocatalyst as borate compounds for styrene polymerization. The Chemical Society of Japan’s 73rd Meeting (1997). 17. Yabunouchi, N. Japanese Patent 1-2000256421 (to Idemitsu Kosan Co. Ltd.), 1999; U.S. Patent 6825294 (to Idemitsu Kosan Co. Ltd.), 1999. 18. For comprehensive reviews, see (a) Po, R., Cardi, N. Synthesis of syndiotactic polystyrene: Reaction mechanisms and catalysis. Prog. Polym. Sci., 21, 47–88 (1996). (b) Schellenberg, J., Tomotsu, N. Syndiotactic polystyrene catalysts and polymerization. Prog. Polym. Sci., 27, 1925–1982 (2002). (c) Schellenberg, J. Recent transition metal catalysts for syndiotactic polystyrene. Progress Polym. Sci., 34, 688–718 (2009).

CHAPTER 4

Mechanisms for Stereochemical Control in the Syndiotactic Polymerization of Styrene NORIO TOMOTSU,1 THOMAS H. NEWMAN,2 and RICHARD CAMPBELL JR.3 1

Advanced Technology Research Laboratories, Idemitsu Kosan Co., Ltd., Chiba, Japan Science Division, Delta College, University Center, MI, USA 3 Chemical Sciences, The Dow Chemical Company, Midland, MI, USA 2

4.1

INTRODUCTION

Catalysts for the syndiospecific polymerization of styrene are typically group 4 metal compounds, especially titanium compounds, with MAO or borate derivatives as the cocatalyst [1–3]. Catalysts for the syndiospecific polymerization of propylene should have complex ligands to control the coordination and the insertion of the monomer. However, catalysts for the syndiospecific polymerization of styrene do not need special ligands. This chapter will summarize studies that clarify the nature of the active center and will explain the mechanisms of polymerization and of stereochemical control.

4.2 INSERTION OF THE GROWING POLYMER CHAIN INTO THE DOUBLE BOND OF STYRENE It is known that both the isotactic and syndiotactic polymerizations of propylene with Ziegler–Natta catalysts occur by the cis opening of the double bond. Figure 4.1 shows the relationship between two kinds of copolymers obtained from deuterium-substituted styrene monomer and the conformation of their vicinal protons. As shown in Figure 4.1, the cis opening polymerization should give a trans conformation for copolymer A, as was obtained from the copolymerization of cis-β-dl-styrene and α,β,β-d3-styrene, and a gauche conformation for copolySyndiotactic Polystyrene, Edited by Jürgen Schellenberg Copyright © 2010 John Wiley & Sons, Inc.

42

43

INSERTION OF THE GROWING POLYMER CHAIN INTO THE DOUBLE BOND OF STYRENE

H

H

H

D

D H

Gauche

H

H

D

D +

D

trans opening

D Copolymer A cis opening

H

Figure 4.1

Copolymer B

D H +

D

D D

trans opening

H

H D

H

cis opening

trans

D H

Conformation of a deuterated styrene polymer.

mer B, as was obtained from the copolymerization of trans β-dl-styrene and α,β,β-d3-styrene. In contrast, when the trans opening occurs in copolymerization, a gauche conformation should be observed for copolymer A and a trans conformation for copolymer B. The catalyst used for the syndiotactic polymerization was CpTiCl3/MAO [4]. Two types of copolymers were prepared. Copolymer A was prepared with cis-β-dl-styrene (15 mol %) and α,β,β-d3-styrene (85 mol %). Copolymer B was prepared with trans-β-dl-styrene (15 mol %) and α,β,β-d3-styrene (85 mol %). Figure 4.2 shows the phenyl C1 resonances of copolymers A and B. The very sharp signals at δ = 145.10 ppm are assigned to the racemic pentad (rrrr) configuration. The syndiotacticity, directly measured from the relative peak areas, was 85% and 80%, respectively, showing that these copolymers have a high degree of stereoregularity. The 1H-nuclear magnetic resonance (NMR) spectra of copolymers A and B have been determined, and the resonances are assignable to the methine and methylene protons in the main chain region as shown in Figure 4.3. The NMR data are summarized in Table 4.1. From the X-ray diffraction measurements by Zambelli, Giongo, and Natta [5], the structure of syndiotactic polystyrene (SPS) and of the main chain of the syndiotactic polymer is a statistically trans–trans conformation. SPS crystallizes in a planar zigzag (trans–trans) conformation and, consequently, the dihedral angle between the two vicinal protons must be trans (180 °) or gauche (60 °). The JH–H coupling constant of a trans conformation examined by Karplus [6] is larger than that of a gauche conformation. The 1H-NMR spectra show that the coupling constant of copolymer A is larger than that of copolymer B. Therefore, copolymer A has a trans (Jt = 9.28 Hz) conformation and copolymer B is gauche (Jg = 5.37 Hz). Independently, Longo et al. [7] also observed a trans (Jt = 9.0 Hz) conformation from a copolymer of type A, which they obtained using the tetrabenzyltitanium/MAO catalyst system.

44

THE SYNDIOTACTIC POLYMERIZATION OF STYRENE

Copolymer A cis-β-dl-styrene and α,β,β-d3-styrene

Copolymer B trans-β-dl-styrene α,β,β-d3-styrene

147

146

145 Chemical shift (ppm)

144

143

Figure 4.2 100-MHz 13C-NMR spectra of the phenyl C1 region of copolymers A and B.

Copolymer A cis-β-dl-styrene and α,β,β-d3-styrene

Copolymer B trans-β-dl-styrene α,β,β-d3-styrene

2.5

2.0 1.5 Chemical shift (ppm)

1.0

Figure 4.3 400-MHz 1H-NMR spectra of methine and methylene regions of copolymers A and B.

STEREOCHEMISTRY OF THE STYRENE INSERTION

TABLE 4.1 Copolymer

Copolymer A Copolymer B

45

Chemical Shifts and Coupling Constants of Copolymers A and B Chemical Shift (ppm) Methylene

Methine

1.40 1.41

1.94 1.94

Coupling Constant (Hz)

9.28 5.37

It was concluded that the double-bond opening mechanism in syndiospecific polymerizations of styrene is the cis opening. This is the same opening mechanism as for the syndiospecific polymerization of propylene using the VCl4/Al(C2H5)2Cl catalyst system at low temperatures [8].

4.3 STEREOCHEMISTRY OF THE STYRENE INSERTION The regiochemistry of the insertion can, in principle, proceed in two ways, either primary (Eq. 4.1) or secondary (Eq. 4.2): M-Pn + CH 2 = CHR → M-CH 2 − CHR-Pn

(4.1)

M-Pn + CHR = CH 2 → M-CHR− CH 2 -Pn.

(4.2)

The polymerization of propylene and/or higher α-olefins with heterogeneous Ziegler–Natta catalysts proceeds by a primary insertion with occasional errors [9]. Vanadium complexes produce syndiotactic polypropylene by secondary insertion [10]. The phenyl group of styrene is a bulky group and is an electron-withdrawing substituent. Styrene behaves differently from an alkyl-olefin with respect to its polymerization reaction. The possible insertion process for styrene into a metal–methyl bond, which is presumed to arise from the methylation of the titanium by MAO in syndiospecific catalyst systems, may be formulated as a primary insertion (Eq. 4.3) or as a secondary insertion (Eq. 4.4): M-CH 3 + CH 2 = CHPh → M-CH 2 − CHPh-CH 3

(4.3)

M-CH 3 + CHPh = CH 2 → M-CHPh − C 2 H 5 .

(4.4)

n-Propylbenzene and ethylbenzene (EtC6H5) are the products after the methanolysis of the syndiospecific polymerization reaction mixture [11]. n-Propylbenzene (n-PrC6H5) is the compound formed after the insertion of styrene into the Ti–CH3 bond proceeding by a secondary (2-1 addition) process, and ethylbenzene is the product after the insertion of styrene into the Ti–H bond. The titanium hydride arises from the β-hydride elimination of the growing polymer chains.

46


*

50

Figure 4.4

40

*

30

20 Chemical shift (ppm)

75.5-MHz 13C-NMR spectrum of low-molecular-weight SPS in CDCl3.

The 13C-NMR spectrum of the low-molecular-weight SPS formed in the reaction shows additional resonances in the region of the saturated carbons at δ = 37.0 and 21.0 ppm. These resonances can be attributed to C(1), C(2) of C(1)H(C6H6)C(2)H3 (Fig. 4.4). These groups arise from the reinitiation via secondary insertion on the M–H bonds that would give the observed 1-phenylethyl end groups. These end groups could also arise from the primary insertion into M–CH3 bonds, but this possibility can be disregarded, since a secondary insertion has been observed by gas chromatography, and there is no evidence (NMR) for headto-head arranged monomer units. Furthermore, it is observed that the amount of ethylbenzene is larger than those of n-PrC6H5. This fact suggests that polymer chains are mainly initiated via M–H bonds and are terminated by β-hydride elimination. Zambelli et al. [12] independently studied the stereochemistry of the styrene insertion catalyzed by the Ti(benzyl)4/MAO system. The 13C-NMR spectrum of the polymer obtained showed that the chain end of the polymer was consistent with a secondary insertion mechanism both in the initiation steps and in the propagation steps even when the former occurs via M–H bonds. They also showed that most of the polymer chains were initiated by M–H bonds. Newman and Malanga [13] examined the characteristic backbone error of syndiotactic styrene polymerization to differentiate chain-end control over site control mechanisms. The syndiotacticity of the polymer produced by the CpTi(OMe)3/MAO system is lower than those of Cp*Ti(OMe)3, especially at higher polymerization temperatures. Both catalysts caused rrmr type errors and a lack of rmmr pentads. These results show that the steric effects of the Cp ligands increase the stereoselectivity and the mechanism of SPS proceeds via a chain-end control. Ute, Takahashi, and Hatada [14] examined the supercritical fluid chromatography of the polymer made by CpTiCl3 with MAO. They separated the polymer by the molecular weight (Fig. 4.5). The end groups of the polymer were -CH=CHPh and CH3-CHPh- units. They concluded that the polymeriza-

EFFECTS OF HYDROGENATION OF THE CATALYST

n = 20

47

n = 25 n = 30

n = 15

0

5

10

15

20

25

Elution time (min)

Figure 4.5

Supercritical fluid chromatography of SPS.

tion proceeds via insertion of the monomer to the Ti–H bond and β-hydride elimination must occur. They also observed that there were many errors in stereoregularity in the low-molecular-weight SPS.

4.4

EFFECTS OF HYDROGENATION OF THE CATALYST

As in the previous chapter, triisobutylaluminum (TIBA) is a good activation agent for styrene polymerization. Cp*TiR 3 + TIBA + MAO → Cp*TiR ( iBu ) → Cp*TiR ( H ) + ( CH 3 )2 C =CH 2 Hydrogenated titanium complexes may be the active site. If this hypothesis is correct, hydrogen also increases the overall catalytic activity. The effects of hydrogen pressure on the catalytic activity and the molecular weight of SPS are summarized in Table 4.2 [15]. Hydrogen increases the catalytic activity as well as the polydispersity Mw/ Mn. The change in the molecular weight distribution might be the effect of the lack of uniformity of the polymerization system or the result of the formation of new types of active sites. The addition of hydrogen increases the catalytic activity, and these results support the hypothesis that the polymerization proceeds via a Ti–H complex. The molecular weight of the polymer in the presence of hydrogen is lower and the molecular weight distribution is broadened. Ethylbenzene, which is the hydrogenated compound of styrene, is also observed after the polymerization of styrene. Marks reported that phenylsilane C6H5SiH3 acts as a chain transfer agent for homogeneous olefin polymerization catalysts [16]. Newman found that C6H5SiH3 increased the catalytic activity in the styrene polymerization [17]. The polymerization conversions with time are shown in Table 4.3. The data

48


TABLE 4.2

Effects of Hydrogen on Catalytic Activities Relative Activitya

Hydrogen Pressure None 0.1 0.5 1.0

100 160 210 220

a

Catalytic activity without hydrogen as 100.

TABLE 4.3 Comparison of Polymerization Results with and without Phenylsilane (20 ml, Added to 10 ml of Styrene Monomer) Temperature/Time 50 °C/30 min 60 °C/30 min 70 °C/30 min 90 °C/30 min 90 °C/60 min

Blank (%)

With PhSiH3 (%)

63.0 56.4 48.5 31.1 34.4

84.7 82.4 72.5 56.2 87.6

indicate an increased catalytic activity in the presence of phenylsilane as well as an increased lifetime of the catalyst at high polymerization temperatures. A comparison of the SPS molecular weight obtained in the presence of phenylsilane and without phenylsilane shows that phenylsilane acts as a chain transfer agent and has a greater effect at lower polymerization temperatures. In all cases, narrow molecular weight distributions were obtained.

4.5 4.5.1

ACTIVE SITE SPECIES Valence of Active Sites

In the polymerization of α-olefins, it is generally accepted that homogeneous catalysts based on group 4 metallocenes with MAO consist of cationic complexes. The catalyst for the styrene polymerization is prepared by the mixing of a compound with MAO. The catalytic activity of Cp*Ti(OMe)3 is increased by the mixing time under room temperature, and the intensity of the electron spin resonance (ESR) spectrum (g=1.998) also showed an increase of the trivalent titanium species (Fig. 4.6). The polymerization activity of a titanium catalyst increases with an increase in the molar ratio of MAO to Ti. Ti(III) is considered to be the active species in the SPS polymerization. The 1 H-NMR spectrum of the reacted materials between the titanium compounds and MAO does not show the titanium compound peaks. This also supports the trivalent titanium as the active site. Newman and Malanga [18] synthesized

ACTIVE SITE SPECIES

Figure 4.6

49

ESR spectrum of Cp*TiCl3/MAO/TIBA.

TABLE 4.4 Effect of TIBA on the Catalytic Activities of Titanium (IV), Cp*Ti(OMe)3, and Titanium (III), Cp*Ti(OMe)2 Catalyst Cp*Ti(OMe)2

Cp*Ti(OMe)3

TIBA/Ti

Conversion (%)

0 6 10 50 0 6 10 50

58 62 64 66 30 47 53 62

Cp*Ti(OMe)2 via the reduction of Cp*Ti(OMe)3 with t-butyllithium and characterized the structure of Cp*Ti(OMe)2 by X-ray crystallography. The complex is dimeric with two bridging methoxide groups, two terminal methoxide groups, and two Cp* rings trans to each other with respect to the Ti–Ti bond. The two halves of the dimer are related by a twofold symmetry axis. Figure 4.7 shows a comparison of the percent conversion for Cp*Ti(OMe)2 and Cp*Ti(OMe)3 with varying MAO ratio in the MAO-activated system. The catalytic activity of this complex with smaller amounts of MAO is almost the same as that of Cp*Ti(OMe)3 at high amounts of MAO. Table 4.4 shows a comparison of the conversion for Cp*Ti(OMe)2 and Cp*Ti(OMe)3 with the addition of TIBA in the MAO cocatalyst system. The trend of an increasing conversion with the TIBA content is observed for Cp*Ti(OMe)3, and a relatively low conversion is observed without TIBA. However, for Cp*Ti(OMe)2, high conversions are obtained even without TIBA, and the conversion is higher than those for Cp*Ti(OMe)3. This suggests

50


70

Percent conversion

60 50 40 30 20

Cp*Ti(OMe)2

10

Cp*Ti(OMe)3

0 0

25

50

75

100

125

150

175

200

MAO : Ti Ratio

Figure 4.7 Catalytic activity of titanium (IV), Cp*Ti(OMe)3, and titanium (III), Cp*Ti(OMe)2, at different MAO ratios.

that the function of the aluminum alkyl is the reduction of the Ti(IV) species, while the MAO cocatalyst generates the active Ti(III) site. Thus, the Ti(III) complex requires less aluminum alkyl overall, that is, less MAO and no TIBA. The effect of the molar ratio of MAO to Ti on the catalytic activity in case of the catalyst system CpTiCl3 and MAO was investigated [15]. The activity increases with an increasing molar ratio of MAO to Ti. The amount of the cationic Ti(III) species measured by ESR increases with the increasing ratio of MAO to Ti as well. This suggests that MAO acts as a reducing agent for Ti(IV) to Ti(III). Ti(III) might be an active species for the synthesis of SPS. Some evidence reported in the literature suggests that the active species promoting the syndiotactic polymerization of styrene is a Ti(III) complex bearing an η5 anionic ligand (Cp or Cp*) and the growing polymer chain [19]. Chien, Salajka, and Dong [20] asserted that the titanium oxidation state for the active catalytic species was Ti(III), both in CpTiX3/MAO and in TiX4/MAO systems. They found that in the Ti(CH2C6H5)4/MAO system, titanium had a distribution of oxidation states, with Ti(IV):Ti(III):Ti(II) of about 53 : 27 : 20. The addition of styrene caused a change in this ratio to 36 : 48 : 16. The ESR spectra are also consistent with the almost quantitative formation of such Ti(III) complexes. Analogous studies were carried out by Dall’occo et al. [21]. They found that in the absence of styrene, titanium had mainly a trivalent oxidation state. ESR measurements revealed the formation of an unidentified new Ti(III) species, possibly containing a Ti–H bond, suggesting a certain degree of βhydrogen abstraction of the bonded polymer.

ACTIVE SITE SPECIES

51

TABLE 4.5 Polymerization Activities of Cp*TiCl3 with MAO and an Alkylation Reagent Reagent None Al(CH3)3 Al(C2H5)3 Al(n–C4H9)3 Al(i–C4H9)3 Al(n–C8H17)3 Al(C2H5)2(OC2H5) Zn(C2H5)2

Relative Activitya

Mw

100 13 23 76 560 100 140 48

750,000 64,000 84,000 570,000 580,000 670,000 870,000 130,000

a

Catalytic activity without alkylation reagent as 100.

40

300

200 Conversion

20

Mw 100

Mw of SPS/10,000

Conversion (%)

30

10

0

0

1

2

3

4

5

0

TIBA/MAO

Figure 4.8

Effect of TIBA on the catalytic activities of Cp*Ti(OMe)3 with MAO.

Titanium compounds are most likely reduced before the active site formation. Metal–alkyl compounds are both alkylating and reducing reagents for the titanium. The effects of reductants on the catalytic activity were evaluated and the data are summarized in Table 4.5 [15]. Very strong reduction reagents like Al(CH3)3 and Al(C3H5)3 reduce catalytic activity. In this case, the titanium compound may be reduced to Ti(II) or Ti(I) by these reagents. But the catalytic activity is increased by the addition of TIBA. The details of the effects of TIBA were evaluated and are shown in Figure 4.8 [15]. TIBA increases the catalytic activity and reduces the molecular weight of the polymer. This compound most likely reacts with the titanium complex to reduce it from Ti(IV) to Ti(III). Moreover, TIBA reacts with metal–alkyl bonds and therefore acts as a chain transfer agent during the polymerization.

52

4.5.2


Number of Active Sites

The polydispersity Mw/Mn produced by the Cp*Ti(OMe)3 catalyst with MAO is always found to be about 2. These results show that there is only one active site and that chain transfer occurs during the polymerization. The number and the kind of active sites were clarified by the stopped-flow method, which was tried for heterogeneous Ziegler–Natta catalysts by Terano [22]. This method can synthesize the polymer without a chain transfer reaction. The polymerization proceeds like a “living polymerization,” and the number of the polymer chains indicates the active sites and the polymer chain length indicates the chain propagation rate. p-Methylstyrene (PMS) was used in this experiment to avoid plugging during polymerization due to crystallization [23]. The results are shown in Figure 4.9. The Mw/Mn of the polymer was found from 1.05 to 1.15. This indicates that a single active site is responsible for the polymerization of the styrene. The number of polymer chains is the same as the number of active sites. The active site number is about half of those of the added titanium compound. By the addition of hydrogen into PMS, the number of chains increased without any broadening of the molecular weight distribution. Hydrogen did not act as a chain transfer agent just after the polymerization started. Ethylmethylbenzene was observed as a by-product after the polymerization with hydrogen. The amount of ethylmethylbenzene is almost equimolar to the hydrogen added. These results show that about all of the titanium compounds turn to active sites by the reaction of MAO and TIBA, but about half of it becomes dormant sites. An active site reacts with the monomer and produces a polymer chain. The dormant site reacts with hydrogen and produces ethylmethylbenzene and titanium hydride compounds. These reactions are very fast and are repeated

Chain/Ti

1.0

With hydrogen

0.8

0.6

0.4 Without hydrogen 0.2

0.0

0.2

0.4 0.6 Time (s)

0.8

1.0

Figure 4.9 Stopped-flow polymerization of p-methylstyrene by Cp*Ti(OMe)3 with MAO.

ACTIVE SITE SPECIES

(b)

(a)

53

(c)

Figure 4.10 Proposed structure of the active site (eliminate hydrogen) and the dormant site for the styrene polymerization. (a) Active site, (b) dormant site by error coordination of monomer, and (c) dormant site by polymer chain rotation. OMe or R Cp (a) (b) (a) (b) (c) (c) (d)

(d)

Figure 4.11 XANES and EXAFS of (a) Cp*Ti(OMe)3, (b) Cp*Ti(OMe)3 + TIBA, (c) Cp*Ti(OMe)3 + TIBA + MAO, and (d) Cp*Ti(OMe)3 + TIBA + [NMe2PhH] [B(C6F5)3].

until all hydrogen turn to ethylmethylbenzene. All titanium compounds are converted to active sites during these reactions. The structure of the dormant sites might be an irregular coordination of the monomer (Fig. 4.10b) or a change of the direction of the monomer coordination (Fig. 4.10c). The polymerization reaction may be stopped after an irregular coordination of styrene. This also supports the chain-end controlled mechanism of stereospecificity. 4.5.3

Structure of Active Sites

Cp*Ti(OMe)3, reacted mixtures of Cp*Ti(OMe)3 and TIBA, and reacted mixtures of Cp*Ti(OMe)3, TIBA, and MAO are examined by X-ray absorption near-edge structure (XANES) and extended X-ray absorption fine structure (EXAFS) analysis. Figure 4.11 shows the results of these measurements [23].

54


A sharp peak was observed in XANES for Cp*Ti(OMe)3, but it disappeared by the addition of TIBA, MAO, or borate. TIBA, MAO, or borate changes the coordination structure of the titanium. The position of the edge is shifted to the left, showing that the alkylaluminum reduces the valence of titanium from +4 to +3. EXAFS shows that the electron density of the titanium has been reduced by TIBA and MAO or borate and that the cyclopentadienyl ligand came closer to the titanium. Therefore, the structures of the active sites formed using either MAO or borate are probably almost the same. The signal inside the cyclopentadienyl ligand decreased by the addition of cocatalysts. The –OMe group may change to hydrogen. 4.6

THEORETICAL ANALYSIS OF THE CATALYST

The mechanism of the CpTi-catalyzed styrene polymerization reaction was studied theoretically using the B3LYP density functional method. Three cases of the polymerization were examined, that is, CpTi(CH3)(CH(C6H5)CH3)+, CpTi(CH3)(CH(C6H5)CH3), and CpTi(CH(C6H5)CH3)+. All intermediate and transition state structures were optimized without any symmetry constraints. The systematic vibrational analysis was carried out for the identification of the number of imaginary frequencies, and the determination of zero-point energy and Gibbs free energy was performed only for some important intermediates and the migratory insertion transition state of the chain initiation reaction. The coordination of the monomer with Ti(III) is more stable than that with Ti(IV) because the coordination of the vinyl group is fortified by the back donation from the d-orbital to the vinyl group of styrene. The coordination of styrene with the Ti(III) cation is too stable to an insertion reaction. The activation energy of the reaction of styrene and CpTi(CH3) (CH(C6H5)CH3) is 9 kcal/mol and that of styrene and CpTi(CH(C6H5)CH3)+ is 28 kcal/mol. The active site is a neutral trivalent titanium instead of a trivalent titanium cation with a coordinated phenyl group. MAO or borate are existing near the active site and neutralize the active site to avoid the coordination of the phenyl group. 4.7

KINETIC ANALYSIS OF STYRENE POLYMERIZATION

The polymerization proceeds by a normal coordination polymerization: Active site formation Chain propagation Chain transfer

C → C* (C*: active site) C* → C*–M → C*–P C*–P → C* + P C*–P + AlR → C* + Al–P C*–P + H2 → C*–H + P–H

40

Conversion (%)

Conversion (%)

KINETIC ANALYSIS OF STYRENE POLYMERIZATION

1h

30 15 min

40 30

20

20

10

10

0

1 2 Catalyst concentration

3

55

0

20

40

60

80

Monomer concentration (%)

Figure 4.12 Effects of catalyst and monomer concentrations on catalytic activities.

The effect of the catalyst and monomer concentrations on the polymerization was examined. The reaction rate increased proportional to the catalyst concentration and the monomer concentration (Fig. 4.12). On the other hand, the decay of the polymerization reaction rate is too fast to explain it as a first-order reaction. The time–conversion curves are fitted as a second-order reaction, but it does not explain the effects of the catalyst concentration. These results indicate that this polymerization proceeds by a single site catalyst under different morphological conditions and/or under variable monomer concentrations, for example, a polymerization in the crystalline polymer and in the amorphous polymer state. From these results, the polymerization reaction can be described by the following equations. d [M ] = −k p1 [ M ][C1* ] − k p 2 [ M ][C2* ] dθ d [Cn* ] = −kdn [C *] dθ ⎤ ⎡ k p1[C1* ]0 k [C * ] conv. (1 − exp ( −kd 1θ)) − p 2 2 0 (1 − exp ( −kd 2θ ))⎥ = 1 − exp ⎢ − 100 kd 1 kd 2 ⎥⎦ ⎢⎣ [M] θ [C1*] [C2*] kp1 kp2

monomer concentration polymerization time concentration of active site in higher monomer concentration concentration of active site in lower monomer concentration chain propagation rate constant of active site 1, including effects of monomer concentration around the active site chain propagation rate constant of active site 2, including effects of monomer concentration around the active site

56


kd1 kd2

decay rate constant of active site 1 (polymerization in monomer) decay rate constant of active site 2 (polymerization in crystallized polystyrene)

The polymerization rate constants are measured by an adjustment of the equations above to the experimental polymerization results. The activation energies of chain propagation and decay for active sites 1 and 2 were the same

( ) ( )[

894 k p1[C1* ]0 = 0.502 ∗ exp − ∗ [ Ti ] t 894 k p2 [C2* ]0 = 0.050 ∗ exp − ∗ Ti ] t 5300 ⎞ kd1 [C1* ]0 = 699000 ∗ exp ⎛ − ⎝ t ⎠ 5300 ⎞ kd 2 [C2* ]0 = 49400 ∗ exp ⎛ − ⎝ t ⎠ because this catalyst system is a single site. t [Ti]

polymerization temperature titanium concentration

The results of the fitting of calculation and experimental polymerization are shown in Figure 4.13. The chain transfer reaction was also examined by these equations, and the comparison of calculated

40 Catalyst conc. = 3

40

80

Catalyst conc. = 2

20

Conversion (%)

Conversion (%)

60

30 20 60 10

Catalyst conc. = 1 0

0

70

60

120 180 Time (min)

240

0

60

120 180 Time (min)

240

Figure 4.13 Estimations from the kinetic analysis. Symbols are experimental data and lines are estimations from the kinetic analysis.

CONCLUSIONS

57

2000

Mw/1000

1500

1000

500

0 50

60

70 80 Temperature (°C)

90

Figure 4.14 Estimation of molecular weights of polymers produced by Cp*Ti(OMe)3 with MAO and TIBA.

Mn =

104 ∗ rp rt

rt = kt [C *] + ka1 [C *][TIBA] + ka 2 [C *][ MAO] predictions and experimental results is shown in Figure 4.14. Mn rp rt kt ka1 ka2

number average molecular weight polymerization rate chain transfer rate termination rate constant chain transfer reaction rate constant of TIBA chain transfer reaction rate constant of MAO. kt = 1.19 × 1010 × exp ( − 8157 t ) ka1 = 1.64 × 10 7 × exp ( − 6860 t ) ka2 = 1.19 × 108 × exp ( − 8615 t )

MAO and TIBA act as chain transfer agents, whereas the β-hydride elimination was not the main reaction for chain transfer with this catalyst system. 4.8

CONCLUSIONS

The analysis of experimental data and the theoretical and kinetic analyses are summarized and the mechanisms of the syndiospecific styrene polymerization are clarified.

58


1. The chain initiation reaction starts from the coordination of styrene to the metal center followed by the insertion of styrene into the metal–alkyl moiety. 2. Ti(III) compounds are the active sites of the polymerization. 3. Detailed mechanisms of the reactions of the SPS polymerization are clarified.

REFERENCES 1. Ishihara, N., Seimiya, T., Kuramoto, M., Uoi, M. Synthesis and properties of polystyrene with new stereoregularity. Polym. Prepr. Jpn., 35, 240 (1986). 2. Zambelli, A., Oliva, L., Pellecchia, C. Soluble catalysts for syndiotactic polymerization of styrene. Macromolecules, 22, 2129 (1989). 3. Campbell, R.E. U.S. Patent 5 066 741 (to Dow Chemical Co.), 1991. 4. Tomotsu, N., Ishihara, N., Newman, T. H., Malanga, M. T. Syndiospecific polymerization of styrene. J. Mol. Catal. A, 128, 167 (1998). 5. Zambelli, A., Giongo, M., Natta, G. Polymerization of propylene to syndiotactic polymer. IV. Addition to the double bond. Makromol. Chem., 112, 183 (1968). 6. (a) Karplus, M. Contact electron-spin coupling of nuclear magnetic moments. J. Chem. Phys., 30, 11 (1959). (b) Karplus, M. Vicinal proton coupling in nuclear magnetic resonance. J. Am. Chem. Soc., 85, 2870 (1963). 7. Longo, P., Grassi, A., Proto, A., Ammendola, P. Syndiotactic polymerization of styrene: mode of addition to the double bond. Macromolecules, 21, 24 (1988). 8. Doi, Y., Ueki, S., Tamura, S., Nagahara S., Keii, T. A kinetic study of “living” coordination polymerization of propene with the soluble V(acac)3-Al(i-C4H9)2Cl system. Polymer, 23, 258 (1982). 9. (a) Longi, P., Mazzanti, G., Roggero, A., Lachi, A. M. High temperature anionic stereospecific polymerization of propylene. Makromol. Chem., 61, 63 (1963). (b) Natta, G., Pino, P., Mantica, E., Danusso, F., Mazzanti, G., Peraldo, M. Stereospecific polymerization of α-olefins. Chim. Ind.(Milan), 38, 124 (1956). (c) Zambelli, A., Locatelli, P., Rigamonti, E. Carbon-13 nuclear magnetic resonance analysis of tail-to-tail monomeric units and of saturated end groups in polypropylene. Macromolecules, 12, 156 (1979). (d) Zambelli, A., Sacchi, M. C., Locatelli, P., Zann, G. Isotactic polymerization of α-olefins: stereoregulation for different reactive chain ends. Macromolecules, 15, 211 (1982). (e) Sacchi, M. C., Tritto, I., Locatelli, P., Ferro, D. R. Stereochemical aspects of the insertion of the first monomeric unit in the stereospecific polymerization of racemic 1-alkenes. Makromol. Chem. Rapid Commun., 5, 731 (1984). 10. (a) Takegami, T., Suzuki, T. A mechanism of the syndiotactic polymerization of α-olefin. Bull. Chem. Soc. Jpn., 42, 848 (1969). (b) Suzuki, T. Takegami, T., The insertion reactions of olefins into metal-ethyl bonds in the vanadium compoundsethyl aluminum Ziegler catalyst system. A proposed mechanism for the syndiotactic polymerization of α-olefin. Bull. Chem. Soc. Jpn., 43, 1484 (1970). (c) Locatelli,

REFERENCES

11. 12.

13. 14. 15.

16.

17. 18.

19.

20. 21.

22. 23.

59

P. Sacchi, M. C., Rigamonti, E., Zambelli, A. Syndiotactic polymerization of propene: regiospecificity of the initiation step. Macromolecules, 17, 123 (1984). (d) Zambelli, A., Tosi, C., Sacchi, M. C. Polymerization of propylene to syndiotactic polymer. VI. Monomer insertion. Macromolecules, 5, 649 (1972). Ishihara, N. Transition metal-catalyzed olefin polymerization. Diss., Oxford, 1990. (a) Zambelli, A., Longo, P., Pellecchia, C., Grassi, A. Synthesis of highly syndiotactic polystyrene with organometallic catalysts and monomer insertion. Macromolecules, 20, 649 (1987). (b) Pellecchia, C., Longo, P., Grassi, A., Ammendola, P., Zambelli, A. Makromol. Chem. Rapid Commun., 8, 277 (1987). Newman, T. H., Malanga, M. T. In 4th SPSJ International Polymer Conference, Yokohama, Japan, November 29–December 3, 1993, p. 27. Ute, K., Takahashi, T., Hatada, K. Fractionation of syndiotactic polystyrene with preparative supercritical fluid chromatography. Polym. Prepr. Jpn., 43, 168 (1994). Tomotsu, N., Kuramoto, M., Takeuchi, M., Maezawa, H. The catalyst for syndiotactic-specific polymerization of styrene. In Metallocenes ’96, Skillmann, N.J. (ed.), Scotland Business Research, Duesseldorf, 1996, p. 179. Fu, P., Marks, T.J. Silane as chain transfer agents in metallocene-mediated olefin polymerization. Facile in situ catalytic synthesis of silyl-terminated polyolefins. J. Am. Chem. Soc., 117, 10747 (1995). Newman, T. H., Borodychuk, K.K. U.S. Patent 6 355 745 (to Dow Chemical Co.), 2000. Newman, T. H., Malanga, M. T. Syndiotactic polystyrene polymerization results using a titanium(III) complex, CpTi(OMe)2 and implications to the mechanism of polymerization. J. Macromol. Sci.-Pure Appl. Chem. A, 34, 1921 (1997). Zambelli, A., Pellecchia, C., Oliva, L., Longo, P., Grassi, A. Catalysts for syndiotactic-specific polymerization of styrene: a tentative interpretation of some experimental data. Makromol. Chem. 192, 223 (1991). Chien, J. C. W., Salajka, Z., Dong, S. Syndiospecific polymerization of styrene. 3. Catalyst structure. Macromolecules, 25, 3199 (1992). Zambelli, A., Pellecchia, C., Oliva, L., Longo, P., Grassi, A. Catalysts for syndiotactic-specific polymerization of styrene: A tentative interpretation of some experimental data. Makromol. Chem., 192, 223 (1991). Mori, H., Terano, M. Stopped-flow techniques in olefin polymerization. Trends Polym. Sci., 5(10), 314–321 (1997). Scheirs, J., Priddy, D., eds. Modern Styrenic Polymers, John Wiley & Sons Inc., New York, 2003.

CHAPTER 5

Copolymerization of Ethylene with Styrene: Design of Efficient Transition Metal Complex Catalysts KOTOHIRO NOMURA Graduate School of Materials Science, Nara Institute of Science and Technology, Nara, Japan

5.1

INTRODUCTION

Precise control over macromolecular structures is a central goal in synthetic polymer chemistry, and copolymerization is an important method that usually makes it possible to alter physical, mechanical, and electronic properties by varying the ratio of the individual components. Considerable effort has been devoted to establishing a new synthetic strategy for the precise placement of a chemical functionality, and the design and synthesis of efficient transition metal complex catalysts for precise olefin polymerization have been a subject of extensive studies [1–6]. Ethylene/styrene copolymers, which cannot be prepared by free radical or ordinary Ziegler–Natta processes [7], have interested chemists because of their promising properties [8]. The introduction of styrene into the polyethylene (PE) backbone results in drastic changes in both the viscoelastic behavior and the thermomechanical properties of the polymeric material [8b], since the crystallizability of PE chains is gradually inhibited by the incorporation of styrene. The copolymers (ethylene–styrene interpolymers [ESI]) range from semicrystalline to amorphous materials, depending on the styrene content [8a]. Therefore, these copolymers can become effective blend compatibilizers for polystyrene (PS)/PE blends and also have potential in foam, film, and sheet applications. Half-titanocenes such as Cp*TiF3, Cp*Ti(OMe)3, and IndTiCl3 are efficient catalyst precursors for syndiospecific styrene polymerization, as described above [9–11]. However, these catalyst precursors showed low catalytic activities and the resultant polymers in the ethylene/styrene copolymerization afforded a mixture of PE, syndiotactic polystyrene (SPS), and the copolymer Syndiotactic Polystyrene, Edited by Jürgen Schellenberg Copyright © 2010 John Wiley & Sons, Inc.

60

ETHYLENE/STYRENE COPOLYMERS

61

[12], as described below. In contrast, nonbridged modified half-titanocenes of the type Cp′Ti(L)X2 (Cp′ = cyclopentadienyl group; L = anionic ligand such as O-2,6-iPr2C6H3, N=CtBu2, N=PR3, etc.; X = halogen, alkyl) are effective not only for syndiospecific styrene polymerization but also for ethylene/styrene copolymerization to exclusively give poly(ethylene-co-styrene)s [6a,13–15]. Linked half-titanocenes (so-called “constrained geometry type”) are also effective for ethylene/styrene copolymerization [2,16–19], although these complexes have exhibited extremely low catalytic activities for styrene polymerization [16a,19a]. Certain metallocenes and group 4 transition metal complexes, so-called post metallocenes, are also known to be effective for copolymerization [20–23]. In this chapter, reported examples of the copolymerization of ethylene with styrene, especially using group 3 and group 4 transition metal (Sc, Ti, Zr, Hf) complex catalysts, and present mechanistic descriptions are summarized [24].

5.2 ETHYLENE/STYRENE COPOLYMERS: MICROSTRUCTURES, THERMAL PROPERTIES, AND COMPOSITION ANALYSES Poly(ethylene-co-styrene)s are generally obtained by the copolymerization of ethylene with styrene in the presence of catalysts composed of a transition metal complex and a cocatalyst (methylaluminoxane [MAO], methyl isobutyl aluminoxane [modified MAO {MMAO}], borate such as B(C6F5)3, Ph3CB(C6F5)4, [PhN(H)Me2] [B(C6F5)4)]. In most cases, the resultant copolymer (E/S copolymer) should be isolated from a mixture of atactic polystyrene (APS, prepared by the cocatalyst itself), PE, and SPS. Such isolation generally requires two steps, and the process is important for confirming the product distribution and composition, even though the resultant polymer was the E/S copolymer exclusively. APS was separated from the mixture (E/S copolymer, PE, SPS) by extracting with boiling acetone (or centrifugation), and the E/S copolymer could be extracted with tetrahydrofuran (THF)-soluble fractions [25]. Each fraction should be carefully checked by examining nuclear magnetic resonance (NMR) spectra and differential scanning calorimetry (DSC) thermograms to confirm that the separation procedure was complete. Figure 5.1 shows typical 13C NMR spectra (methylene and methine regions) of the copolymers (THF-soluble fraction) prepared by [Me2Si(C5Me4)(NtBu)] TiCl2, (1,2,3-Me3C5H2)TiCl2(O-2,6-iPr2C6H3) catalysts in the presence of a MAO cocatalyst [13b]. Table 5.1 also summarizes the assignments of resonances for poly(ethylene-co-styrene) in the 13C NMR spectrum based on the distortionless enhancement by polarization transfer (DEPT) spectrum and data reported previously [12–18], and monomer sequences in the copolymer are shown in Scheme 5.1. As described below, the microstructures for the resultant poly(ethylene-co-styrene)s depend on the catalysts used. As shown in Figure 5.2, the glass transition temperature (Tg) as measured by DSC increased with an increase in the styrene content (−8.1 to 58.3 °C). This is because, as

62

COPOLYMERIZATION OF ETHYLENE WITH STYRENE

Sγδ, Sδδ, Sγγ

Sβδ, Sβγ

Sαγ, Sαδ

Sββ

Tδδ, Tγδ, Tγγ

Sαβ

(a)

Sγδ, Sδδ, Sγγ Sβδ, Sβγ Sαγ, Sαδ

Tδδ, Tγδ, Tγγ

Tβδ

Sαα

50 ppm (b)

Sαβ

Sββ

Sαβ

Tββ

40

30

20

Figure 5.1 13C NMR spectra (in CDCl3 at 60 °C) of poly(ethylene-co-styrene)s (methylene and methine regions, THF-soluble fraction) [13b]. (a) catalyst = [Me2Si(C5Me4) (NtBu)]TiCl2; styrene content, 32.7 mol %; (b) catalyst = (1,2,3-Me3C5H2)TiCl2(O-2,6i Pr2C6H3); styrene content, 38.8 mol %. The peak at 30.5 ppm is due to the impurity (2,6-di-tert-butyl-p-cresol) as the added stabilizing agent.

described above, the crystallizability of PE chains is gradually inhibited by the incorporation of styrene and the copolymers are semicrystalline to amorphous materials, depending on the styrene content [8a]. It is very important to prove that the resultant E/S copolymers possessed a single composition, since the resultant (co)polymers (using CpTiCl3–MAO catalyst, etc.) may consist of a mixture of separate polymer fractions (not a

ETHYLENE/STYRENE COPOLYMERS

TABLE 5.1 Carbon Sαα

63

13

C NMR Chemical Shift for Poly(ethylene-co-styrene)[13b] ppm 43.6–45.2

Sαβ Sαγ Sαδ

36.6

Sβγ Sβδ Sγγ Sγδ Sδδ Tββ Tβδ Tγγ Tγδ Tδδ

27.5 27.5 29.7 29.7 29.7

ppm ESSE SSSS m SaESSS

43.8 44.5 34.3 35.3–35.8

ESE SSSE SES m

SSSE ESSEE

ppm SSSE

44.3

r

35.0

36.9 37.4 25.2

ESSE

37.9

SES r

25.4

41.0 43.5

SSSS SSSEE

40.7 43.2

45.2–46.7 45.2–46.7 45.2–46.7

a

Styrene insertion with regioirregularity (inversion).

“Tail to tail” and/ or SES

Scheme 5.1

“Head to tail”

Monomer sequences in poly(ethylene-co-styrene).

block copolymer, etc.) even though the gel permeation chromatography (GPC) trace for the resultant polymer showed a unimodal molecular weight distribution. While the results of DSC thermograms may not fulfill this purpose, the confirmation process should be important. Some previous reports have discussed the microstructures and/or styrene contents (styrene incorporation) based on the fact that the resultant copolymers contained APS/SPS or PE. Cross-fractionation chromatography (CFC) is one of the most powerful


Glass transition temperature, Tg (°C)

64

60

40

20

0

–20 10

20

30

40

50

60

70

80

Styrene content (mol %)

Figure 5.2 Plots of glass transition temperature (by DSC thermograms) versus styrene content (mol % estimated by 1H NMR spectra) in poly(ethylene-co-styrene)s prepared by (1,2,3-Me3C5H2)TiCl2(O-2,6-iPr2C6H3) and [Me2Si(C5Me4)(NR)]TiCl2 [R = tBu, Cy]–MAO catalyst systems [13b].

tools for analyzing comonomer distribution, especially in ethylene/α-olefin copolymer, and the copolymer sample can be used under similar conditions. However, the confirmation of whether or not APS is still contaminated (remained) may not be clear enough with these procedures (1H and 13C NMR, DSC, GPC, CFC), since the present sample is eluted even at low temperature in the CFC method. In this sense, the use of GPC/FT-IR (GPC combined with Fourier transform–infrared [FT-IR]), which is a powerful tool for analyzing comonomer distribution, especially for ethylene/α-olefin copolymers [26], should be effective for analyzing the styrene distribution in the E/S copolymer for each molecular weight.

5.3 ETHYLENE/STYRENE COPOLYMERIZATION USING TRANSITION METAL COMPLEX–COCATALYST SYSTEMS 5.3.1

Half-Titanocenes, Cp′TiX3

Half-titanocenes such as Cp*TiF3, Cp*Ti(OMe)3, and IndTiCl3 are efficient catalyst precursors for syndiospecific styrene polymerization, as described in the Introduction [9–11]. However, these catalyst precursors showed low catalytic activities for olefin polymerization, and the resultant polymers in the ethylene/styrene (co)polymerization afforded a mixture of PE, SPS, and the copolymer (E/S copolymer) [12]. Longo reported that the CpTiCl3–MAO catalyst system afforded ethylene/ styrene copolymers (E/S copolymer) with SPS inserted sequences, and the

ETHYLENE/STYRENE COPOLYMERIZATION

65

TABLE 5.2 Ethylene/Styrene Copolymerization Using the Cp*Ti(CH2Ph)3– B(C6F5)3–AliBu3 Catalyst System[12b]a Temperature

0 25 50 50 50 50 75

Styrene (mol/l)

1.1 0.8 0.4 0.65 1.2 2.0 0.5

Yield (g)b

0.3 0.45 0.35 0.53 0.49 1.0 0.35

Activityc

48.0 72.0 56.0 84.8 78.4 160.0 56.0

Composition (wt %)d PE

E/S

SPS

>90 63 33 12 6 8 29

22 58 63 72 20 41

15 9 25 22 72 30

Conditions: Cp*Ti(CH2Ph)3/B(C6F5)3/AliBu3 = 25/25/25 μmol; ethylene, 1 atm; toluene + styrene, total of 26 ml. b After removal of APS. c Activity in kg-polymer/mol-Ti·h. d Estimated from 13C NMR spectra. a

microstructure and the composition strongly depended on the [Al]/[Ti] ratio [12a]. However, Aaltonen and Seppälä later reported that the same catalytic system only gave a mixture of PE and SPS, even under closely similar conditions [27]. The product distribution as well as the catalytic activity in (co) polymerization using a CpTiX3–MAO catalyst system is, indeed, very sensitive to the anionic donor ligand (X = Cl, CH2Ph, OCH3, OCH2Ph, etc.), the reaction conditions, the nature of MAO (MAO or MMAO, removal of AlMe3, etc.) [12c,d], and the exact mixing sequence and precontact time, which usually leads to poor reproducibility. Pellecchia et al. reported that Cp*Ti(CH2Ph)3–B(C6F5)3 afforded E/S copolymers including PE and SPS, and the distributions were dependent upon the polymerization temperature and the pretreatment procedure employed (Table 5.2) [12b]. The resultant copolymer possessed an alternating sequence, and no resonances ascribed to styrene repeating units (Sαβ, Sαα) were seen. However, as shown in Table 5.2, it seems impossible to find suitable conditions for the exclusive preparation of E/S copolymers, and selective co-oligomerization proceeded in the presence of Cp′TiCl3 (Cp′ = Cp, Cp*, indenyl)–B(C6F5)3 (MAO) catalyst systems under certain conditions [28]. 5.3.2

Linked (Constrained Geometry Type) Half-Titanocenes

Linked half-titanocenes (so-called constrained geometry-type titanium complexes) like [Me2Si(C5Me4)(NtBu)]TiCl2 have also shown to be efficient catalyst precursors [2,16–19], although they have shown extremely low activity for syndiospecific styrene polymerization [16a,19a]. It has also been reported that styrene incorporation by linked Cp–amide Ti catalyst (constrained geometry

66


Ti catalyst cocatalyst

+

Me 2Si

Ti N t Bu

1a

Me2Si Cl Cl

Ti N t Bu

x

1-x

SiMe3

Me2Si Cl Cl

2

Ti N t Bu

Cl Cl

3

Ph Me2Si

Me2Si Me2Si Ti Ti Ph Ti Me Me Me N N N Me Me Me t Bu t t Bu Bu 5 1b 4

Scheme 5.2 Selective list of linked half-titanocenes employed for ethylene/styrene copolymerization [16,17].

catalyst [CGC]) systems is invariably 50 mol %) [19a]. They assumed that the arene ring of the last-inserted styrene may preferentially coordinate with the adjacent Ti


69

Scheme 5.3

Ti

P

Ti P

Ti

P

2,1insertion

Ti

Ti Ti

P

Ti

P 1,2insertion

P

P

Ti

P

Ph

Ti

P

Ti

Ti

P

Ti P

P

Scheme 5.4

center in the bimetallic Ti2 (Scheme 5.4), thus reducing coordinative saturation at the polymerization site and accelerating homopolymerization [19a,b]. The coordinated arene rings can, in principle, participate in several types of multimetallic/enchainment-altering interaction.

70

+ B1 + B2 + B1 + B1 + B1 + B1 + B1

5/25 5/25 5/25 10/50 20/40 30/30 60/0

Styrene/Toluene (ml/ml) — — — 1.0 1.0 1.0 1.0

Ethylene (atm) 3.0 3.0 3.0 1.0 1.5 0.5 0.5

Time (h) 2.7 104 112 259 194 384 312

Activityb

b

a

— — — 21.7 31.4 35.8 39.6

Tgc (°C ) 11.96 1.04 0.80 35.8 47.1 43.8 47.9

Mwd × 10 −4

Styrene Polymerization and Ethylene/Styrene Copolymerization Catalyzed by Ti2[19]a

Conditions: Ti1 (10 μmol) or Ti2 (5 μmol) + B1 (10 μmol) or B2 (5 μmol); 20 °C; ethylene, 1.0 atm. Activity in kg-polymer/mol-Ti·h. c By DSC thermograms. d GPC data versus polystyrene standards. e Estimated by 13C NMR spectra.

Ti1 Ti2 Ti2 Ti2 Ti2 Ti2 Ti2

Catalyst

TABLE 5.5

1.84 1.44 1.47 1.82 1.33 2.40 1.72

Mw Mnd

100 100 100 39 50 66 76

Styrene (mol %)e


5.3.3

71

Modified Half-Titanocenes, Cp′ Ti(L)X2

5.3.3.1 Copolymerization Using Cp′ Ti(L)X2–Cocatalyst Systems As described above, Cp′TiX3–cocatalyst systems exhibited notable catalytic activities for syndiospecific styrene polymerization [9–11], but showed low catalytic activities for olefin polymerization; the resultant polymers in the ethylene/ styrene (co)polymerization afforded a mixture of PE, SPS, and the copolymer [12]. As related examples, Cp*TiMe3 exhibited catalytic activities for both ethylene polymerization and styrene polymerization in the presence of MAO or B(C6F5)3, and the resultant polystyrene possessed both atactic (via a cationic mechanism) and syndiotactic (via a coordination insertion mechanism) stereoregularity, and the ratios were dependent upon the polymerization temperature [29–31]; this catalyst also polymerizes isobutene via a carbocationic mechanism rather than a coordination insertion mechanism [31]. In contrast, nonbridged half-titanocenes of the type Cp′Ti(L)X2 (Cp′ = cyclopentadienyl group; L = anionic ligand such as O-2,6-iPr2C6H3, N=CtBu2, and N=PR3; X = halogen, alkyl) have been known to exhibit unique characteristics as olefin polymerization catalysts [6,13–15,32–35]. In particular, Cp′TiCl2(OAr) (OAr = O-2,6-iPr2C6H3) not only showed high catalytic activities for ethylene (α-olefin) polymerization [32] but also exhibited efficient comonomer incorporation for the copolymerization of ethylene with α-olefin [33], styrene [13], cyclic olefin [34], and disubstituted-α-olefin [35] in the presence of cocatalysts. Both cyclopentadienyl and anionic donor ligands strongly affect the catalytic activity and the comonomer incorporation in this catalysis [6,13,32–35], and the desired catalysts for the desired (co)polymerization can be tuned simply by modifying these ligands [6]. (Aryloxo)(cyclopentadienyl)titanium complexes, Cp′TiCl2(OAr) (Cp′ = 1,2,3Me3C5H2 (6), 1,3-Me2C5H3 (7), tert-BuC5H4 (8), etc.; Ar = 2,6-iPr2C6H3), exhibited remarkable catalytic activities for both syndiospecific styrene polymerization and ethylene/styrene copolymerization in the presence of MAO (Scheme 5.5) [13a,b], although the Cp* analogue (9) exhibited notable catalytic activity for olefin (ethylene, α-olefin) polymerization [6,32]. Poly(ethylene-co-styrene) s were obtained exclusively without PE and/or SPS as by-products (Table 5.6) [13a,b]. The resultant copolymers possessed not only relatively high molecular weights with unimodal molecular weight distributions but also a single composition as confirmed by DSC thermograms, CFC, and GPC/FT-IR [13b]. The catalytic activities of 6–8 decreased slightly with an increase in the styrene concentration, whereas the styrene contents in the copolymers increased upon increasing the [S]/[E] initial feed molar ratios. In one case, a styrene content of up to 73.6 mol % could be achieved [13b]. As shown in Figure 5.3, styrene incorporations with aryloxo analogues are more efficient than those with [Me2Si(C5Me4)(NtBu)]TiCl2 (1a). Thus, the present catalysis provides an efficient synthesis of copolymer with high styrene content, especially higher than 50 mol %, in a random manner.

72


Cp'TiCl2(OAr)

Ti

cocatalyst

O

cocatalyst

O

Cl 7 Cl

t Bu

Cp'TiCl2(OAr)

+

Ti

Cl 6 Cl

x

1-x

Ti O

Ti

Cl 8 Cl

O

Cl 9 Cl

Scheme 5.5 Effective catalyst precursors for random ethylene/styrene copolymerization [13].

Styrene in copolymer (mol %)

80

60

40

20

0 0

5 10 15 [Styrene]/[ethylene] molar ratio

20

Figure 5.3 Plots of styrene (mol %) in copolymer versus [styrene]/[ethylene] molar ratio for copolymerization of ethylene with styrene (Table 5.3). Catalyst: Cp′TiCl2(O2,6-iPr2C6H3) [Cp′ = 1,2,3-Me3C5H2 (6, 䉫), 1,2,4-Me3C5H2 (ⵧ), 1,3-Me2C5H3 (7, 䉬), tert-Bu (8, 䊏)]; [Me2Si(C5Me4)(NtBu)]TiCl2 (1a, •) [13b].

An analysis of the microstructure of the resultant poly(ethylene-co-styrene) s by 13C NMR spectroscopy (methylene and methine regions, Fig. 5.1; monomer sequences and assignments of the resonances are summarized in Scheme 5.1 and Table 5.1, respectively) indicated that the resultant copolymer prepared by 6 possesses resonances ascribed to two or three styrene units connected via head-to-tail coupling (at δ = 40.7–41.0 ppm [Tββ], 43.1–45.0 ppm [Sαα and Tβδ], respectively), in addition to the resonances ascribed to the tail-to-tail coupling of a styrene unit or a head-to-head coupling bridged by an intervening ethyl-


73

TABLE 5.6 Ethylene/Styrene Copolymerization by Cp′TiCl2(OAr) (Cp′ = 1,2,3Me3C5H2 (6), 1,3-Me2C5H3 (7), tert-BuC5H4 (8); OAr = O-2,6-iPr2C6H3] or [Me2Si(C5Me4)(NtBu)]TiCl2 (1a)–MAO Catalyst Systems[13b]a Complexes

(1,2,3Me3C5H2) TiCl2(OAr) (6) (1,2,3Me3C5H2) TiCl2(OAr) (6) (1,2,3Me3C5H2) TiCl2(OAr) (6) (1,3-Me2C5H3) TiCl2(OAr) (7) (1,3-Me2C5H3) TiCl2(OAr) (7) (1,3-Me2C5H3) TiCl2(OAr) (7) (tert-BuC5H4) TiCl2(OAr) (8) (tert-BuC5H4) TiCl2(OAr) (9) [Me2Si(C5Me4) (NtBu)] TiCl2 (1a)

Styrene (ml)

Activityb

3

THF Soluble (E/S Copolymer) Content (wt %)

Mwd × 10 −4

Mw Mnd

Styrene (mol %)e

4100

99.1

17.0

1.6

26.0

5

3070

98.3

11.0

1.7

38.8

10

2720

97.8

6.6

1.6

51.2

3

3670

97.1

6.4

1.8

32.3

5

4280

98.2

6.0

2.1

38.5

10

4140

98.2

3.7

1.6

49.0

5

2180

99.6

5.9

1.7

37.6

10

1840

98.7

3.5

2.2

51.2

10

5630

99.6

18.0

1.8

32.7

c

Reaction conditions: catalyst, 1.0 μmol (2 μmol/ml toluene); ethylene, 4 atm; total volume of toluene and styrene = 30 ml; MAO white solid (Al/Ti = 2000, molar ratio); 25 °C; 10 min. b Activity (kg-polymer/mol-Ti·h), polymer yield in acetone-insoluble fraction. c Percentage of content in copolymer based on polymer obtained. d GPC data in o-dichlorobenzene versus polystyrene standards. e Styrene content (mol %) in copolymer by 1H NMR (1,1,2,2-C2D2Cl4). a

74


50

[Tββ]/[Ttotal] in %

40

30 Bernoullian 20

10

0 0

20 40 60 Styrene content (mol %)

80

❍

Figure 5.4 Plots of [Tββ]/[Ttotal] ratio versus styrene in the copolymer determined by 13 C NMR spectra (in CDCl3 at 60 °C, methylene and methine regions, THF-soluble fraction). [Tββ]/[Ttotal] value (in percent) based on Bernoullian mode (•) could be calculated as (styrene content)2 × 100. Catalyst: Cp′TiCl2(O-2,6-iPr2C6H3) [Cp′ = 1,2,3Me3C5H2 (6, 䊏), 1,3-Me2C5H3 (7, ), tert-BuC5H4 (8, ⵧ)]; [Me2Si(C5Me4)(NtBu)]TiCl2 (1a, 䉬) [13b].

ene unit (Sαβ at δ = 34.3 and 35.1 ppm). This is especially interesting in contrast to the results with the linked half-titanocene (1a) (Fig. 5.1a). The ratios of [Tββ]/[Ttotal] and [Sαβ]/[Sαα + Sββ] were chosen to evaluate the frequency as well as the regioselectivity of styrene insertion. If the insertion of styrene follows a Bernoullian mode in a regioselective manner, [Tββ] should be calculated as [St]3, and the value of [Tββ]/[Ttotal] should be as shown in Figure 5.4 (marked in •). The [Tββ]/[Ttotal] value with the linked half-titanocene (1a) should be zero (marked with 䉬), since the resultant copolymer does not contain a head-to-tail sequence (especially Tββ) for three styrene repeat units. All of the copolymer samples prepared with the Cp′-aryloxo analogues (6–8) showed lower [Tββ]/[Ttotal] ratios than those estimated based on a Bernoullian mode, and these results assume that styrene is less reactive for insertion into propagating chain ends than ethylene. This can be easily explained because, in the cases of ethylene/α-olefin copolymerization, not only with ordinary metallocenes and linked half-titanocenes but also with Cp′-aryloxo analogues, copolymerization proceeds in a random manner and the monomer sequences do not follow a Bernoullian model but rather a first-order Markov model [33c,36]. In addition, this is also because the structural features of the catalyst, in particular the steric bulk of the ligand, bite angle, configuration, and con-


Ti

N

Ti Cl X Cl R P N 3 11 X 10

Ti N

N N

R = Cy(cyclohexyl) X = Cl,Me

75

CH2Ph CH2Ph 12

Scheme 5.6 Other modified half-titanocenes employed for ethylene/styrene copolymerization [14,15].

formation, do influence the coordination and/or insertion of monomers in transition metal-catalyzed coordination polymerization reactions in most cases, and this is a distinct difference from conventional radical and ionic polymerization reactions [36,37]. Copolymerization with the use of other modified half-titanocenes has also been reported (Scheme 5.6). Ethylene/styrene copolymerization by Cp*TiCl2(N=CtBu2) (10) took place in a living manner in the presence of a MAO cocatalyst, although the homopolymerization of ethylene and styrene did not proceed in a living manner [14a]. No styrene repeating units were observed in the resultant copolymers, which suggests that a certain degree of styrene insertion inhibits chain transfer in this catalysis. In contrast, the Cp analogue showed negligible catalytic activity under the same conditions [14a]; only the above set showed living copolymerization [14b]. The living nature was maintained under various conditions (Al/Ti molar ratios, ethylene pressure, styrene concentrations, temperature). Copolymerization with CpTiX2(N=PCy3) (12, X = Cl, Me)–cocatalyst (MAO, borates) systems proceeded with remarkable catalytic activities (at 60–90 °C; ethylene, 70 psi (4.76 atm); [S]/[E] = 12); however, styrene incorporation seemed less efficient than with either aryloxo analogues (6–8) or linked half-titanocenes (1a) (styrene content: 33.4–61.4 wt % [99

>99

>99

>99 >98 Trace

40

55

70

25 40 25

40

55

Cl (13)

Cl (13)

69.8

30.2

18.4

Trace Trace 13.2

Trace

Trace

Trace

Trace

SPS

260

280

504 660 250

1260

1110

790

396

Activityc

31.9 34.3 >99.0f — >99.0f — >99.0f —

12.2

10.4

9.3

7.4

Styrene Content (mol %)d

9.28 9.79 5.85 0.29 5.07 0.31 3.56 0.21

16.3

19.7

14.4

9.5

Mnj × 10 −4

1.62 1.5 1.26 2.69 1.31 1.75 1.48 1.77

1.57

1.31

1.28

1.18

Mw Mnj

a

Conditions: catalyst, 2.0 μmol; MAO (prepared by removing AlMe3 and toluene from PMAO), 3.0 mmol; ethylene, 6 atm; styrene, 10 ml; styrene + toluene, total of 30 ml; 10 min. b Based on a mixture of PE, SPS, and copolymer (acetone-insoluble fraction). c Activity in kg-polymer/mol-Ti·h. d Styrene content (mol %) estimated by 1H NMR. e GPC data in o-dichlorobenzene versus polystyrene standards. f Confirmed by GPC/FT-IR, 13C NMR spectra, and DSC thermograms.

Trace

81.6

Trace

>99

25

N=C Bu2 (10) N=CtBu2 (10) N=CtBu2 (10) N=CtBu2 (10) OAr (9) OAr (9) Cl (13)

Trace

PE

t

Composition (%)b

E-S

Temperature (°C)

Catalyst L

TABLE 5.7 Copolymerization of Ethylene with Styrene by Cp*TiCl2(L) (L = N CtBu2 (10), O-2,6-iPr2C6H3 (OAr, 9), Cl (13))–MAO Catalyst Systems[14b]a

78


TABLE 5.8 Ethylene/Styrene Copolymerization by Cp*TiMe2(O-2,6-iPr2C6H3) (14), Cp*TiMe3 (15)–MAO or [PhN(H)Me2][B(C6F5)4] (AFPB) Catalyst Systems[14b]a Catalyst (μmol)

Time (min)

14/MAO (2.0) 14/AFPB (5.0) 15/MAO (2.0) 15/AFPB (10.0)

10

Composition (%)b E-S

Activityc

Styrene Content (mol %)d

Mne × 10 −4

Mw Mne

PE

SPS

99

Trace

Trace

519.0

30.5

5.34

2.05

20

99

Trace

Trace

79.2

46.7

2.81

2.16

10

Trace

68.6

31.4

366.0

20

Trace

99

Trace

>99.0 — Trace

8.66 0.81 0.5

1.37 2.33 3.53

43.8

a Conditions: catalyst, 2.0 μmol; MAO, 3.0 mmol, or AliBu3, 1.0 mmol (borate system); [C6H5NH(CH3)2] [B(C6F5)4] (AFPB), 2.0 μmol; styrene, 10.0 ml; toluene, 20 ml; 10 min. b Based on a mixture of PE, SPS, and copolymer (acetone-insoluble fraction). c Activity in kg-polymer/mol-Ti·h. d Styrene content (mol %) estimated by 1H NMR. e GPC data in o-dichlorobenzene versus polystyrene standards.

(a) Introduction of ethylene after styrene polymerization Ethylene

Ti catalyst Ph

n

MAO Syndiospecific styrene polymerization

Ph

x Ph

6 atm

Ph

Syndiotactic polystyrene (SPS)

1-x

Poly(ethyleneco-styrene)

(b) Removal of ethylene after ethylene/styrene copolymerization Ethylene

Ti catalyst

+

x Ph

MAO Ph

Copolymerization of ethylene with styrene

1-x

Poly(ethyleneco-styrene) Trace amount

removal

Poly(ethyleneco-styrene)

Ti O Two-step (co)polymerization by half-titanocenes

Cl Cl

Ti O

Cl Cl

Scheme 5.7 Two-step (co)polymerization of ethylene with styrene using Cp′TiCl2 (O-2,6-iPr2C6H3)–MAO catalyst systems [41].


O

Ti

O

R

Ph

Ph

79

Ti R

Ph

n

x

Ph

Ph

1-x

Scheme 5.8

tion did not proceed when ethylene was removed from the reaction mixture of ethylene/styrene copolymerization (likely due to oxidation upon exposure to ethylene) [41]. Based on findings in the literature [14b,39], the cationic Ti(IV) species, [Cp′Ti(L)R]+, likely plays a role in copolymerization, and the active species containing an anionic ancillary donor ligand (assumed to be neutral Ti(III), Cp′Ti(L)R) proposed by Tomotsu and Ishihara [11b,c] plays a role in syndiospecific styrene polymerization (Scheme 5.8) [41]. These results should also explain the reported finding that the catalytic activities and molecular weight of the resultant SPS in styrene polymerization using Cp′Ti(L) X2–cocatalyst systems were highly dependent upon the anionic donor ligand (L), regardless of kind of the cocatalyst used [39]. These proposals are in contrast to the hypothesis that cationic Ti(III) species, [Cp′Ti(R)(styrene)]+, play a role as the catalytically active species for the styrene polymerization using Cp’TiX3′[40,42–44]. This hypothesis should help to explain why polystyrene structures in the resultant copolymers prepared with Cp′TiCl2(L)–MAO catalysts are atactic [13–15]. 5.3.4

Non-Cp Titanium Complexes

Ethylene/styrene copolymerization using titanium 2,2′-thiobis(phenolate) complex (16a) was first reported by Miyatake, Mizunuma, and Kakugo at Sumitomo Chemical Co. [45], and they assumed a presence of two catalytically active species (for the production of alternating copolymer and SPS) from a single catalyst precursor in the presence of MAO [45]. Okuda later carefully conducted copolymerization using a series of titanium bis(phenolate) complexes (16–19) shown in Scheme 5.9 and concluded that these systems afforded random copolymers containing a trace amount of SPS (in certain polymerization runs) [46]. The bridging unit between bis(phenolate) moieties plays a decisive role with respect to styrene incorporation and the catalyst activity; steric hindrance at the active center was responsible for reduced styrene incorporation, probably because the ligand was forced into a conformation that

80


O S Ti X X O 16

O

O

Ti X X O 17

Ti Cl Cl O 18

O S=O

i Ti O Pr i O Pr

O 19

X = OiPr (a), Cl (b)

Scheme 5.9 TABLE 5.9 Copolymerization of Ethylene with Styrene Using Different Bis(phenolate) Titanium Complexes (16-19)–MAO Catalyst Systems[46]a Complex 16a 16a 16a 16a 16a 16b 17a 17b 18 19

[S]/[E]b

Activityc

Mwd × 10 −4

2.5 5.0 10.0 16.9 31.4 5.0 5.0 5.0 5.0 5.0

216 144 144 68 41 109 1.0 1.5 1.0 13

2.5 2.1 1.4 6.0 4.0 — — — — —

Mw Mnd 3.6 3.4 2.4 3.2 3.5 — — — — —

Styrene (mol %)e 2.4 5.5 11.7 19.8 35.6 6.0 35.2 36.4 Mixturef 10.0

Conditions: complex, 20 μmol; ethylene, 2.56 bar; styrene + toluene, total of 200 ml; Al/Ti = 1000 (molar ratio); 60 °C; 1 h. b Initial feed molar ratio. c Activity in kg-polymer/mol-Ti·h. d GPC data versus polyethylene standards. e Styrene content in E/S copolymer estimated by 1H NMR spectra. f Mixture of homopolymers. a

effectively shields against styrene coordination. Catalysts are rapidly deactivated after a high initial activity, and styrene incorporation was found to be proportional to the styrene concentration. However, very high styrene/ethylene molar ratios are required to achieve high styrene incorporation [46]. Therefore, it seems likely that the findings described in the former report [45] may be the result of the use of certain specified conditions [46] (Table 5.9). Okuda reported copolymerization using {dithiabutanediyl-2,2′-bis(4,6-ditert-butylphenoxy)}titanium complex (20) in the presence of MAO (Table 5.10, Scheme 5.10) [47]. The catalytic activities were moderate and the resultant copolymers possessed uniform molecular weight distributions [47]. Although styrene polymerization with 20 afforded isotactic polystyrene [48],


81

TABLE 5.10 Ethylene/Styrene Copolymerization Using Dichloro{1,4dithiabutanediyl-2,2′-bis(4,6-di-tert-butylphenoxy)}Titanium Complex (20)–MAO Catalyst System[46]a [Styrene] (mol/l) 0.25 0.5 1.0 0.55 1.0 2.0 2.2g 4.0 7.9

[S]/[E]b

Yield (g)

Activityc

Mnd × 10 −3

Mw Mnd

Tge (°C)

Styrene Contentf (mol %)e

1.25 2.50 5.0 2.75 5.0 10.0 11.0 20.0 39.5

3.4 5.6 40.1 6.6 10.8 17.6 2.6 7.0 1.6

300 280 490 1300 390 440 1600 360 200

8.6 10.6 10.3 8.8 13.4 15.7 12.1 12.6 9.8

2.4 2.3 2.9 2.3 2.3 2.3 2.2 2.8 3.5

–22.5 –0.3 3.7 4.5 11.0 23.6 30.5 31.5 37.0

23 35 38 40 42 49 55 59 68

Conditions: [Ti], 8.0 × 10−5 mol/l; Al/Ti = 1500; [ethylene] = 0.2 mol/l; 40 °C. Initial feed molar ratio. c Activity in kg-polymer/mol-Ti·h. d GPC data versus polystyrene standards. e By DSC thermograms. f Estimated by 1H NMR spectra. g Polymerization at 25 °C. a

b

20

R

n

R

MAO

O

Isotactic +

S S

20 MAO

O x

1-x

R

20 MAO

n

Scheme 5.10

m

Cl Cl

20 R

+

Ti

n

82


Styrene in copolymer (mol %)

80 By 6 60 By 20 40 By 1a 20

0 0

10 20 30 40 [Styrene]/[ethylene] molar ratio

50

Figure 5.5 Plots of styrene (mol %) in copolymer versus [styrene]/[ethylene] molar ratio for copolymerization of ethylene with styrene using dichloro{1,4-dithiabutanediyl2,2′-bis(4,6-di-tert-butylphenoxy)}titanium complex (20) (Table 5.10, marked in •) [47a]. Plots marked with 䉫 and 䉬 are the results with (1,2,3-Me3C5H2)TiCl2(O-2,6i Pr2C6H3) (6) and [Me2Si(C5Me4)(NtBu)]TiCl2 (1a), respectively, for comparison [13b].

the resultant copolymer possessed alternating sequences (isolated and alternating styrene inserted units) with very high regioselective styrene insertion, and resonances ascribed to Sαβ, Tβδ, etc. were observed to very low extents [47]. The styrene contents in the copolymers exceeded 50 mol % under high [S]/[E] molar ratios, but the resultant copolymer possessed head-to-tail styrene repeat units (SSS) and isolated ethylene units (SSESS) in addition to alternating sequences [47]. As shown in Figure 5.5, 20 showed better styrene incorporation at low [S]/[E] molar ratios, but it seems very difficult to obtain copolymers with high styrene contents, probably because the resultant copolymer possessed alternating rather than random microstructures. This trend may also explain why the glass transition temperatures in copolymers with high styrene contents were rather low. The copolymerization of styrene with small amounts of ethylene afforded isotactic polystyrene containing isolated ethylene units [49]. Styrene polymerization in the presence of propylene with 20–MAO catalyst afforded multiblock copolymers containing long isotactic styrene sequences interrupted by short isotactic propylene strings (Scheme 5.10). An analysis of the microstructure by 13C NMR spectrometry showed that the opposite regiochemistry of insertion of the two monomers is retained in copolymerization to give tail-to-tail and head-to-head linkages between the homopolymer blocks [50]. Since propylene polymerization gave oligomers containing predominantly unsaturated chain ends, which reflect a primary (1,2) regiochemistry of insertion [50], this method was applied to the synthesis of olefin- and diolefin-terminated isotactic polystyrenes including novel vinyl-terminated

83


Me 2C Zr Zr Cl Cl Cl Cl 21

22

Me2C Zr Cl Cl 25

Zr

Cl

Ph Cl

Cl

23

Me2C Zr Cl Cl 26

Cl

Ph Zr

Me2C Zr Cl Cl

Cl

28

24

Zr

Cl

Me 2Si

27

Ti N t Bu

1a

Cl Cl

Me2C Zr Cl Cl 29

Scheme 5.11

macromonomers [51]. The molecular weight of isotactic polystyrene was controlled by varying the olefin/styrene molar ratio without sacrificing high catalyst activities [51]. 5.3.5

Metallocenes

Although ordinary metallocenes are inactive for styrene polymerization, certain metallocenes can be effective for ethylene/styrene copolymerization [20–23]. Oliva et al. reported that the regiospecificity of styrene insertion into the Zr–13CH3 bonds is predominantly secondary in ethylene/styrene copolymerization using the [Me(Ph)C(fluorenyl)(Cp)]ZrCl2–MAO catalyst system [20]. Oliva later presented that some styrene units can be introduced in an isotactic polypropylene sequence (1,2-insertion) by using a small amount of ethylene to reactivate the catalyst after the styrene insertion (2,1-insertion) in propylene/styrene copolymerization using a rac-[Et(indenyl)2]ZrCl2–MAO catalyst system [21]. Arai, Ohtsu, and Suzuki reported that copolymerization with a rac[Me2C(indenyl)2]ZrCl2 (26)–MAO catalyst system afforded products with high molecular weight, “stereoregular and Bernoullian copolymers,” with moderate catalytic activities (Table 5.11, Scheme 5.11). [Me2C(benzindenyl)2] ZrCl2 (28) and [Me2C(3-cyclopenta[c]phenanthryl)]ZrCl2 (29) were also effective in terms of both activity and styrene incorporation and gave the copolymer with isotactic stereoregularity. The copolymer consists of Et-Et, St-Et, and head-to-tail St-St sequences and has a highly isotactic Et-St alternating sequence. It possesses a melting point of 80–110 °C due to the isotactic alternating sequence [22a]. These complexes also afforded isotactic polystyrene, although the activity was somewhat low [22b]. Both the small bite angle and the C2 symmetry of the catalyst structure have been proposed to be prerequisites for both high activity and efficient styrene incorporation, since complexes that possess a Cs symmetry, like [Ph2C(fluorenyl)

84

10/16 10/16 10/16 10/16 10/16 10/26 60/20 60/20 800/4000 1800/3000 4000/800 4000/800 4000/800 1500/3300 4000/800

21 (23) 22 (8.4) 23 (8.4) 24 (8.4) 25 (8.4) 26 (8.4) 26 (8.4) 26 (8.4) 26 (8.4)e 26 (21)f 26 (84)e 26 (84)e 24 (164)e 1a (21)f 27 (84)e

0.6 0.6 0.6 0.6 0.6 0.6 0.2 0.15 1.1 0.6 0.6 0.2 0.4 1.1 0.2

Ethylene (MPa) 1.3 9.2 7.3 2.2 9.1 7.5 3.4 3.0 816 800 1660 1220 286 550 386

Yield (g) 57 1100 870 260 1100 890 400 400 32000 9000 4900 2100 440 10000 770

Activityb

b

Conditions: 50 °C, 1 h (large scale experiment 2.5–7.0 h), Al/M = 1000 (molar ratio). Activity in kg-polymer/mol-M·h. c Estimated by 1H NMR spectra. d Tacticity of Et-St alternating sequence. e Al/M = 10,000. f Al/M = 4000.

a

Styrene/Toluene (ml/ml) — 1.7 3.1 — 5.7 14.0 8.1 16.6 17.2 16.0 14.8 15.5 50.2 19.0 5.0

Mw × 10−4

Ethylene/Styrene Copolymerization Using a Series of Metallocenes[22a]a

Complex (μmol)

TABLE 5.11

— 1.9 1.5 — 2.1 2.2 1.9 2.1 2.0 1.8 2.0 1.8 2.7 1.6 2.0

Mw/Mn 0 13.3 16.2 14.2 27.2 39.1 49.4 52.0 12.8 28.0 43.5 49.3 21.1 13.0 9.0

Styrene Content (mol %)c

— 0.5 0.6 0.6 0.85 >0.95 — — — — — — — — —

Et-Std (mm)


85

(Cp)]ZrCl2, exhibited both extremely low activity and inefficient styrene incorporation in the ethylene/styrene copolymerization. Izzo, Napoli, and Oliva presented later that the regiochemistry for styrene insertion by the rac[H2C(3-R-1-indenyl)2]ZrCl2–MAO catalyst changed from secondary to primary with an increase in the steric bulk of R: H < CH3 < CH2CH3 < CH(C H3)2 < C(CH3)3 [23]. High isospecificity was observed with rac-[H2C(3-tert-Bu1-indenyl)2]ZrCl2, whereas unsubstituted or ethyl-substituted analogues showed some syndiospecificity [23]. Based on these results, the oxidation state for the catalytically active species for styrene polymerization with this series of catalyst precursors is proposed to be cationic Zr(IV), which is somewhat different from that proposed for polymerization using half-titanocene complex catalysts [9–11]. More recently, a stereorigid meso-[norbornane-7,7-bis(indenyl)]titanium chloride showed higher activity than the other ordinary ansa-metallocenes (Zr and Ti) in the copolymerization in the presence of MAO to give copolymers with homogeneous chemical compositions [52]. It was suggested that inclusion of the bi-cycle hinders rotation of the bridge carbon atom–indenyl bonds, resulting in a remarkable conformational stability that remains fixed throughout the catalytic process. While the styrene contents in the copolymers were quite modest (4 mol %), the molecular weights were high (Mn up to 171,000). 5.3.6

Others

Half-zirconocene and hafnocene were generally inactive in both styrene homopolymerization and copolymerization with ethylene. Zr and Hf complexes containing phosphorus pendants, [{(2,4,6-Me3C6H2)P(H)CH2CH2} C5H4]M(CH2Ph)3, [(CH2CH2)P(C5H4)(2,4,6-Me3C6H2)]Zr(CH2Ph)2 (M = Zr, Hf), showed moderate catalytic activities (32–108 kg-polymer/mol-M·h) for copolymerization at 50–80 °C [53]. However, the resultant polymers possessed broad molecular weight distributions, although the styrene contents were higher than 50 mol % in some cases (31–87 mol %). Polymeric Sm(II) complexes containing Cp* and anionic donor ligand of type, Cp*Sm(ER){Cp*K(THF)n}(THF)m [ER = O-2,6-iPr2C6H3, O-2,6-tBu2-4Me-C6H2, S-2,4,6-iPr3C6H2, NH-2,4,6-tBu3C6H2: n = 1 or 2, m = 0 or 1], not only polymerize styrene and ethylene but also copolymerize them into block styrene–ethylene copolymers in the presence of both monomers (toluene, 25 °C, 1 atm) [54]. In copolymerization, polystyrene was also present as a by-product, and the ratio of homopolymer/copolymer was dependent upon the anionic donor ligand (ER) employed [54]. Sequential copolymerization experiments suggest that copolymerization is initiated by the polymerization of ethylene followed by the successive incorporation of styrene. A cationic scandium complex, [(C5Me4SiMe3)Sc(CH2SiMe3)]+[B(C6F5)4]−, showed remarkable catalytic activity for syndiospecific styrene polymerization, and copolymerization with ethylene afforded multiblock copolymers

86


TABLE 5.12 Copolymerization of Ethylene with Styrene Using (C5Me4SiMe3)Sc(CH2SiMe3)2(THF)–[Ph3C][B(C6F5)] Catalyst[55]a Styrene (mmol) 0 21 10 21 31 41

Ethylene (atm)

Yield (g)

Activityb

Styrene Content (mol %)c

Mnd × 10 −4

Mw Mnd

1 0 1 1 1 1

0.55 0.45 0.40 0.79 0.92 1.62

786 643 600 1120 1310 2310

0 100 13 56 65 87

17.23 6.04 7.92 11.13 16.26 15.09

1.72 1.41 1.14 1.19 1.17 1.26

Conditions: toluene, 50 ml; Sc, 21 μmol; [Sc]/[B] = 1/1 (mol/mol). Activity in kg-polymer/(mol-Sc·h). c Estimated by 1H NMR spectra in C2D2Cl4 at 120 °C. d GPC data versus polystyrene standards. a

b

consisting of SPS and PE with low Mw/Mn values (Table 5.12) [55]. The styrene contents in the copolymers could be controlled by varying the ethylene/styrene feed ratios, and random copolymers were not obtained even under high styrene/ethylene molar ratios (probably due to the significant differences in monomer reactivity [rE and rS values] or differences in propagating species). Although the mechanistic details are not yet clear, the results should be important as limited examples of the synthesis of copolymers containing SPS sequences.

5.4

SUMMARY AND OUTLOOK

The reports regarding ethylene/styrene copolymerization using group 3 and group 4 transition metal complex catalysts have been reviewed, and the microstructures, thermal properties, and composition analyses of the resultant copolymers have also been briefly introduced. Promising results have been reported regarding copolymerization using linked (so-called constrained geometry type) half-titanocenes and modified half-titanocenes (Cp′Ti(L)X2, L: anionic ancillary donor ligands), affording random copolymer with various styrene contents, and both the catalytic activities and the styrene incorporation are highly affected by both the cyclopentadienyl fragments and the anionic donor ligands employed. The efficient synthesis of random copolymers could be achieved by using half-titanocenes containing aryloxo ligands. Cationic Ti(IV) species play an important role as catalytically active species in copolymerization, whereas (neutral and/or cationic) Ti(III) species play roles in syndiospecific styrene polymerization, and these findings may suggest why the polystyrene structure in random copolymers prepared with half-titanocenes was atactic. Certain metallocenes such as rac-[Me2C(indenyl)2]ZrCl2 also gave copolymers, and both the bite angle and the symmetry are important factors in

REFERENCES

87

both the activity and the styrene incorporation. {Dithiabutanediyl-2,2′bis(phenolate)}titanium complex was effective for the synthesis of multiblock copolymers containing an isotactic polystyrene microstructure; [(C5Me4SiMe3) Sc(CH2SiMe3)]+[B(C6F5)4]− was effective for the synthesis of multiblock copolymers containing an SPS microstructure. Various copolymers, which differ with regard to their compositions, microstructures, and properties, have been prepared by using several types of early transition metal catalysts, and these unique characteristics are dependent upon the nature of the complex catalysts used. The resultant copolymers possess unique properties as new polyolefin materials. The development of other new polyolefin materials (e.g., amphiphilic) may follow the use of this synthetic technique, which can be applied not only to styrene but also to styrene containing reactive functionalities (in combination with other polymerization techniques such as ATRP, RAFT, and ROP).

REFERENCES 1. (a) Brintzinger, H. H., Fischer, D., Mülhaupt, R., Rieger, B., Waymouth, R. M. Angew. Chem. Int. Ed. Engl., 34, 1143 (1995). (b) Kaminsky, W. Macromol. Chem. Phys., 197, 3903 (1996). (c) Kaminsky, W., Arndt, M. Adv. Polym. Sci., 127, 143 (1997). (d) Suhm, J., Heinemann, J., Wörner, C., Müller, P., Stricker, F., Kressler, J., Okuda, J., Mülhaupt, R. Macromol. Symp., 129, 1 (1998). 2. McKnight, A. L., Waymouth, R. M. Chem. Rev., 98, 2587 (1998). 3. (a) Britovsek, G. J. P., Gibson, V. C., Wass, D. F. Angew. Chem. Int. Ed. Engl., 38, 429 (1999). (b) Gibson, V. C., Spitzmesser, S. K. Chem. Rev., 103, 283 (2003). (c) Bolton, P. D., Mountford, P. Adv. Synth. Catal., 347, 355 (2005). 4. Gladysz, J. A., ed. Frontiers in metal-catalyzed polymerization (special issue). Chem. Rev., 100(4), 1167–1682 (2000). For example, (a) Ittel, S. D., Johnson, L. K., Brookhart, M. Chem. Rev., 100, 1169 (2000). (b) Alt, H. G., Köppl, A. Chem. Rev., 100, 1205 (2000). (c) Chen, E. Y-X., Marks, T. J. Chem. Rev., 100, 1391 (2000). 5. (a) Coates, G. W., Hustad, P. D., Reinartz, S. Angew. Chem. Int. Ed., 41, 2236 (2002). (b) Domski, G. J., Rose, J. M., Coates, G. W., Bolig, A. D., Brookhart, M. Prog. Polym. Sci., 32, 30 (2007). 6. (a) Nomura, K., Liu, J., Padmanabhan, S., Kitiyanan, B. J. Mol. Catal. A, 267, 1 (2007). (b) Stephan, D. W. Organometallics, 24, 2548 (2005). 7. (a) Soga, K., Lee, D. H., Yanagihara, H. Polym. Bull., 20, 237 (1988). (b) Mani, P., Burns, C. M. Macromolecules, 24, 5476 (1991). (c) Cheung, Y. W., Guest, M. J. In Modern Styrenic Polymers: Polystyrenes and Styrenic Copolymers, Scheirs, J., Priddy, D. B. (eds.), John Wiley & Sons Ltd., Chichester, 2003, p. 605. (d) Guest, M. J., Cheung, Y. W., Diehl, C. F., Hoenig, S. M. In Metallocene-Based Polyolefins, Scheirs, J., Kaminsky, W. (eds.), John Wiley & Sons Ltd., 2000, vol. 2, p. 271. 8. For example, (a) Chen, H., Guest, M. J., Chum, S., Hiltner, A., Baer, E. J. Appl. Polym. Sci., 70, 109 (1998). (b) Chum, P. S., Kruper, W. J., Guest, M. J. Adv. Mater.,

88

9.

10.

11.

12.

13.


12, 1759 (2000). (c) Cheung, Y. W., Guest, M. J. J. Polym. Sci. Part B: Polym. Phys., 38, 2976 (2000). Selected examples: (a) Ishihara, N., Seimiya, T., Kuramoto, M., Uoi, M. Polym. Prepr. Jpn., 35, 240 (1986). (b) Ishihara, N., Seimiya, T., Kuramoto, M., Uoi, M. Macromolecules, 19, 2464 (1986). (c) Ishihara, N., Seimiya, T., Kuramoto, M., Uoi, M. Macromolecules, 21, 3356 (1988). (d) Newman, T. H., Campbell, R. E., Malanga, M. T. In Metcon ’93, 1993, p. 315. Selected examples: (a) Zambelli, A., Longo, P., Pellecchia, C., Grassi, A. Macromolecules, 20, 2035 (1987). (b) Zambelli, A., Oliva, L., Pellecchia, C. Macromolecules, 22, 2129 (1989). (c) Ready, T. E., Day, R. O., Chien, J. C. W., Rausch, M. D. Macromolecules, 26, 5822 (1993). (d) Chien, J. C. W. In Metallocenes ’96, 1996, p. 223. (e) Kaminsky, W., Lenk, S., Scholz, V., Roesky, H. W., Herzog, A. Macromolecules, 30, 7647 (1997). (f) Wu, Q., Ye, Z., Lin, S. Macromol. Chem. Phys., 198, 1823 (1997). Reviews: (a) Tomotsu, N., Kuramoto, M., Takeuchi, M., Maezawa, H. In Metallocenes ’96, 1996, p. 179. (b) Po, R., Cardi, N. Synthesis of syndiotactic polystyrene: Reaction mechanism and catalysis. Prog. Polym. Sci., 21, 47 (1996). (c) Tomotsu, N., Ishihara, N. Catal. Surv. Jpn., 1, 97 (1997). (d) Tomotsu, N., Ishihara, N., Newman, T. H., Malanga, M. T. J. Mol. Catal. A, 128, 167 (1998). (e) Pellechia, C., Grassi, A. Top. Catal., 7, 125 (1999). (f) Schellenberg, J. Prog. Polym. Sci., 34, 688 (2009) (g) Schellenberg, J., Tomotsu, N. Syndiotactic polystyrene catalysts and polymerization. Prog. Polym. Sci., 27, 1925 (2002). Examples in ethylene/styrene copolymerization: (a) Longo, P., Grassi, A., Oliva, L. Makromol. Chem., 191, 2387 (1990). (b) Pellecchia, C., Pappalardo, D., D’Arco, M., Zambelli, A. Macromolecules, 29, 1158 (1996). (c) Oliva, L., Mazza, S., Longo, P. Macromol. Chem. Phys., 197, 3115 (1996). (d) Xu, G., Lin, S. Macromolecules, 30, 685 (1997). (e) Lee, D. H., Yoon, K. B., Kim, H. J., Woo, S. S., Noh, S. K. J. Appl. Polym. Sci., 67, 2187 (1998). (a) Nomura, K., Komatsu, T., Imanishi, Y. Macromolecules, 33, 8122 (2000). (b) Nomura, K., Okumura, H., Komatsu, T., Naga, N. Macromolecules, 34, 5388 (2002). (c) Nomura, K., Zhang, H., Byun, D.-J. J. Polym. Sci. Part A: Polym. Chem., 46, 4162 (2008).

14. (a) Zhang, H., Nomura, K. J. Am. Chem. Soc., 127, 9364 (2005). (b) Zhang, H., Nomura, K. Macromolecules, 39, 5266 (2006). 15. (a) Kretschmer, W. P., Dijkhuis, C., Meetsma, A., Hessen, B., Teuben, J. H. Chem. Commun., 608 (2002). (b) Wang, Q., Lam, P., Zhang, Z., Yamashita, G., Fan, L. US 6579961 B1, 2003. 16. (a) Stevens, J. C., Timmers, F. J., Wilson, D. R., Schmidt, G. F., Nickias, P. N., Rosen, R. K., Knight, G. W., Lai, S. Y. EP 0416815 A2, 1991. (b) Timmers, F. J. USP 6670432 B1, 2003. (c) Arriola, D. J., Bokota, M., Campbell, R. E. Jr., Klosin, J., LaPointe, R. E., Redwine, O. D., Shankar, R. B., Timmers, F. J., Abboud, K. A. J. Am. Chem. Soc., 129, 7065 (2007). 17. (a) Sernetz, F. G., Mülhaupt, R., Waymouth, R. M. Macromol. Chem. Phys., 197, 1071 (1996). (b) Sernetz, F. G., Mülhaupt, R., Amor, F., Eberle, T., Okuda, J. J. Polym. Sci. Part A: Polym. Chem., 35, 1571 (1997). (c) Sukhova, T. A., Panin, A. N., Babkina, O. N., Bravaya, N. M. J. Polym. Sci. Part A: Polym. Chem., 37, 1083 (1999).

REFERENCES

89

18. (a) Xu, G. Macromolecules, 31, 2395 (1998). (b) Kamigaito, M., Lal, T. K., Waymouth, R. M. J. Polym. Sci. Part A: Polym. Chem., 38, 4649 (2000). (c) Nomura, K., Okumura, H., Komatsu, T., Naga, N., Imanishi, Y. J. Mol. Catal. A: Chem., 190, 225 (2002). 19. (a) Guo, N., Li, L., Marks, T. J. J. Am. Chem. Soc., 126, 6542 (2004). (b) Li, H., Marks, T. J. Proc. Natl. Acad. Sci. U.S.A., 103, 15295 (2006). (c) Guo, N., Stern, C. L., Marks, T. J. J. Am. Chem. Soc., 130, 2246 (2008). 20. Oliva, L., Caporaso, L., Pellecchia, C., Zambelli, A. Macromolecules, 28, 4665 (1995). 21. (a) Caporaso, L., Izzo, L., Oliva, L. Macromolecules, 32, 7329 (1999). (b) Caporaso, L., Izzo, L., Zappile, S., Oliva, L. Macromolecules, 33, 7275 (2000). 22. (a) Arai, T., Ohtsu, T., Suzuki, S. Macromol. Rapid Commun., 19, 327 (1998). (b) Arai, T., Suzuki, S., Ohtsu, T. In Olefin Polymerization: Emerging Frontiers, ACS Symposium Series 749, Arjunan P. (ed.), American Chemical Society, Washington, DC, 2000, p. 66. (c) Arai, T., Ohtsu, T., Nakajima, M. Polym. Prepr. Jpn., 48, 1666 (1999). 23. Izzo, L., Napoli, M., Oliva, L. Macromolecules, 36, 9340 (2003). 24. (a) Braunschweig, H., Breitling, F. M. Coord. Chem. Rev., 250, 2691 (2006). (b) Rodrigues, A.-S., Carpentier, J.-F. Coord. Chem. Rev., 252, 2137 (2008). 25. These procedures are described in Reference 13a (supporting information) and 13b. 26. For GPC/FT-IR analysis: Housaki, T., Satoh, K., Nishikida, K., Morimoto, M. Makromol. Chem. Rapid Commun., 9, 525 (1988). Styrene contents in the copolymers as determined by GPC/FT-IR may be somewhat different from those estimated by 1H NMR spectra due to the standard samples in the calibration curve for the analysis of styrene contents chosen for these analyses [13,14]. 27. Aaltonen, P., Seppälä, J. Eur. Polym. J., 30, 683 (1994). 28. (a) Pellecchia, C., Mazzeo, M., Gruer, G.-J. Macromol. Rapid Commun., 20, 337 (1999). (b) Pellecchia, C., Pappalardo, D., Oliva, L., Mazzeo, M., Gruter, G.-J. Macromolecules, 33, 2807 (2000). 29. (a) Ewart, S. W., Baird, M. C. Topics in Catalysis, 7, 1 (1999). (b) Baird, M. C. Chem. Rev., 100, 1471 (2000). 30. For example, (a) Gillis, D. J., Tudoret, M.-J., Baird, M. C. J. Am. Chem. Soc., 115, 2543 (1993). (b) Quyoum, R., Wang, Q., Tudoret, M.-J., Baird, M. C. J. Am. Chem. Soc., 116, 6435 (1994). (c) Wang, Q., Quyoum, R., Gillis, D. J., Tudoret, M.-J., Jeremic, D., Hunter, B. K., Baird, M. C. Organometallics, 15, 693 (1996). 31. In this catalyst system with Cp*TiMe2(μ-Me)B(C6F5)3 generated from Cp*TiMe3 and B(C6F5)3, styrene polymerization took place in both cationic and coordination insertion manners to give atactic polystyrene and syndiotactic polystyrene, respectively. The cationic polymerization of isobutene also took place, and 1-hexene polymerization afforded polymer with a broad molecular weight distribution. Since the styrene homopolymerization also took place only with borate and MAO in a cationic manner, it might be difficult to estimate the catalytic activity with a titanium catalyst.

90


32. Nomura, K., Naga, N., Miki, M., Yanagi, K., Imai, A. Organometallics, 17, 2152 (1998). 33. (a) Nomura, K., Naga, N., Miki, M., Yanagi, K. Macromolecules, 31, 7588 (1998). (b) Nomura, K., Oya, K., Komatsu, T., Imanishi, Y. Macromolecules, 33, 3187 (2000). (c) Nomura, K., Oya, K., Imanishi, Y. J. Mol. Catal. A, 174, 127 (2001). 34. (a) Nomura, K., Tsubota, M., Fujiki, M. Macromolecules, 36, 3797 (2003). (b) Wang, W., Tanaka, T., Tsubota, M., Fujiki, M., Yamanaka, S., Nomura, K. Adv. Synth. Catal., 347, 433 (2005). (c) Wang, W., Fujiki, M., Nomura, K. J. Am. Chem. Soc., 127, 4582 (2005). 35. (a) Nomura, K., Itagaki, K., Fujiki, M. Macromolecules, 38, 2053 (2005). (b) Itagaki, K., Fujiki, M., Nomura, K. Macromolecules, 40, 6489 (2007). 36. For example, Fink, G., Richter, J. W. In Polymer Handbook, 4th edn., Briandrup, J., Immergut, E. H., Grulle, E. A. (eds.), John Wiley & Sons Inc., New York, 1999, vol. 2, p. 329. 37. Related examples in ethylene/1-hexene copolymerization: (a) Reybuck, S. E., Meyer, A., Waymouth, R. M. Macromolecules, 35, 637 (2002). (b) Nomura, K., Fujita, K., Fujiki, M. J. Mol. Catal. A, 220, 133 (2004). 38. According to the described experimental procedures (including the results regarding Tg values of the resultant polymers) [15b], it is not yet clear whether or not the styrene contents reported here may include atactic/syndiotactic polystyrene. The Mw/Mn values in certain polymerization runs also seemed somewhat broad. 39. (a) Byun, D.-J., Fudo, A., Tanaka, A., Fujiki, M., Nomura, K. Macromolecules, 37, 5520 (2004). (b) Nomura, K., Fujita, K., Fujiki, M. Catal. Commun., 5, 413 (2004). (c) Nomura, K., Tanaka, A., Katao, S. J. Mol. Catal. A, 254, 197 (2006). 40. Mahanthappa, M. K., Waymouth, R. M. J. Am. Chem. Soc., 123, 12093 (2001). 41. Zhang, H., Byun, D.-J., Nomura, K. Dalton Trans., 1802 (2007). 42. Examples of mechanistic studies for styrene polymerization [40], (a) Grassi, A., Zambelli, A., Laschi, F. Organometallics, 15, 480 (1996). (b) Minieri, G., Corradini, P., Guerra, G., Zambelli, A., Cavallo, L. Macromolecules, 34, 5379 (2001). 43. Formation of Ti(III) in the presence of MAO and oxidation into Ti(IV) by propylene: Volkis, V., Lisovskii, A., Tumanskii, B., Shuster, M., Eisen, M. S. Organometallics, 25, 2656 (2006). In this manuscript, some Ti(III) intermediates, [η3-PhC(NSiMe3)2]2TiCl2 in the presence of MAO, characterized by ESR, are not active in the polymerization but are cleanly oxidized upon exposure to propylene to afford Ti(IV) species. 44. Under certain conditions, especially with extensively high [S]/[E] molar ratios (under low ethylene pressure), exclusive formation of the catalytically active species in the copolymerization seemed difficult; the copolymerization afforded a mixture of SPS and the copolymer [13c]. 45. Miyatake, T., Mizunuma, K., Kakugo, M. Macromol. Symp., 66, 203 (1993). 46. (a) Fokken, S., Spaniol, T. P., Okuda, J., Sernetz, F. G., Mülhaupt, R. Organometallics, 16, 4240 (1997). (b) Sernetz, F. G., Mülhaupt, R., Fokken, S., Okuda, J. Macromolecules, 30, 1562 (1997). 47. Capacchione, C., Proto, A., Ebeling, H., Mülhaupt, R., Okuda, J. J. Polym. Sci. Part A: Polym. Chem., 44, 1908 (2006).

REFERENCES

91

48. (a) Capacchione, C., Proto, A., Ebeling, H., Mülhaupt, R., Möller, K., Spaniol, T. P., Okuda, J. J. Am. Chem. Soc., 125, 4964 (2003). (b) Capacchione, C., Proto, A., Ebeling, H., Mülhaupt, R., Möller, K., Manivannan, R., Spaniol, T. P., Okuda, J. J. Mol. Catal. A, 213, 137 (2004). (c) Beckerle, K., Capacchione, C., Ebeling, H., Manivannan, R., Mülhaupt, R., Proto, A., Spaniol, T. P., Okuda, J. J. Organomet. Chem., 689, 4636 (2004). (d) Capacchione, C., Manivannan, R., Barone, M., Beckerle, K., Centore, R., Oliva, L., Proto, A., Tuzi, A., Spaniol, T. P., Okuda, J. Organometallics, 24, 2971 (2005). (e) Beckerle, K., Manivannan, R., Spaniol, T. P., Okuda, J. Organometallics, 25, 3019 (2006). 49. Capacchione, C., D’Acunzi, M., Motta, O., Oliva, L., Proto, A., Okuda, J. Macromol. Chem. Phys., 205, 370 (2004). 50. Capacchione, C., De Carlo, F., Zannoni, C., Okuda, J., Proto, A. Macromolecules, 37, 8918 (2004). 51. Gall, B. T., Pelascini, F., Ebeling, H., Beckerle, K., Okuda, J., Mülhaupt, R. Macromolecules, 41, 1627 (2008). 52. Lobon-Poo, M. J., Barcina, O., Martinez, A. G., Exposito, M. T., Vega, J. F., Martinez-Salazar, J., Reyes, M. L. Macromolecules, 39, 7479 (2006). 53. Ishiyama, T., Miyoshi, K., Nakazawa, H. J. Mol. Catal. A, 221, 41 (2004). 54. (a) Hou, Z., Tezuka, H., Zhang, Y., Yamazaki, H., Wakatsuki, Y. Macromolecules, 31, 8650 (1998). (b) Hou, Z., Zhang, Y., Tezuka, H., Xie, P., Tardif, O., Koizumi, T.-A., Yamazaki, H., Wakatsuki, Y. J. Am. Chem. Soc., 122, 10533 (2000). 55. Luo, Y., Baldamus, J., Hou, Z. J. Am. Chem. Soc., 126, 13910 (2004).

CHAPTER 6

Structure and Properties of Tetrabenzo[a,c,g,i ]fluorenyl-Based Titanium Catalysts RÜDIGER BECKHAUS,1 KAI SCHRÖDER,2 and JÜRGEN SCHELLENBERG3 1

Institute of Pure and Applied Chemistry, Carl von Ossietzky University, Oldenburg, Germany Faculty of Mathematics and Natural Sciences, Molecular Inorganic Chemistry, Stratingh Institute for Chemistry, Groningen, the Netherlands 3 R&D Dow Central Germany, Dow Olefinverbund GmbH, Schkopau, Germany 2

6.1

INTRODUCTION

The cyclopentadienid anion and its derivatives are one of the widely used ligands in coordination as well as in organometallic chemistry [1]. It has been recognized that the replacement of hydrogen atoms on the cyclopentadienyl ring by substituents influences the physical and chemical properties of the corresponding metal complexes [2]. In addition to the well-known permethylated ligand Cp*, a broad range of other Cp substituents are used to prepare sterically demanding Cp ligands. On one hand, bulky Cp ligands become available by pentasubstitution employing i-propyl [3], benzyl [4], phenyl [5], as well as ferrocenyl ligands directly [6] or indirectly attached to the five-membered ring [7]. On the other hand, more bulky Cp ligands become available by using fused aromatic ring systems like indenyl [8], fluorenyl [9], or corannulenes [10]. By modification of the steric and electronic requirements, the complex properties can be tuned in an excellent way, for example, in order to obtain better solubilities, thermal stabilities, or to avoid subsequent reactions. Particularly, the fine tuning of the sterical and electronical properties of these complexes is of interest in order to optimize properties in catalytic applications such as the stereoregular polymerization of olefins [11], hydroamination, or hydroformylation [12]. Particularly, half-sandwich titanium(IV) complexes are useful in the syndiotactic polymerization of styrene [13]. It has been shown that monocyclopentadienyl titanium derivatives show increasing catalytic activities with the increasing sterical size of the Syndiotactic Polystyrene, Edited by Jürgen Schellenberg Copyright © 2010 John Wiley & Sons, Inc.

92

INTRODUCTION

93

Me Ti

Ti Cl

Cl

Cl

Cl I

Ti Cl

MeO

Cl

OMe OMe III

II

Scheme 6.1

η5

(a) M

η3

(b) M

M

η1

(c) M

M

η5

(d) M

Scheme 6.2

spectator ligands. In such a way, titanium complexes with 2-phenyl-cyclopenta[l] phenanthrene [14] (II), 2-methyl-benz[e]indene [15] (I), or 1,2,3,4,5,6,7,8-octahydrofluorene [16] (III) are important catalysts for styrene polymerization (Scheme 6.1). Particularly, benzofused Cp derivatives such as indene and fluorene had serious impact on the development of catalyst precursors for stereospecific olefin polymerization [17]. However, fusing aromatic ring systems to the cyclopentadienyl ring can sometimes cause problems with regard to the bonding situation between metal and ligand because the tendency of the benzofused ring to preserve its aromatic six-electron system leads to a haptotropic shift as depicted in Scheme 6.2. Ring slippage can occur on all Cp-type ligand systems and this greatly adds to their chemical variety [18]; however, for the abovementioned reason, this is more likely to take place with (b) indenyl [19] and (c) fluorenyl ligands (Flu) [20]. Especially the fluorenyl moiety usually not only refuses to form η5 complexes even in its ansa-bridged specimen [21] but often dimerizes in a redox

94

TETRABENZO[a,c,g,i ]FLUORENYL-BASED TITANIUM CATALYSTS

Ti X

Ti

X

X

X

X X

Scheme 6.3

reaction with the metal [22]. As a natural consequence, only very few unsubstituted fluorenyl transition metal complexes are known such as FluMn(CO)3 [23] and FluTi(OiPr)2 [24], for instance. By further adding benzofused rings to the fluorenyl moiety complex, stability significantly improves as by forming the anionic five-membered ring, none of the six-electron ring systems is disturbed like in the (d) tetrabenzo[a,c,g,i] fluorenyl (Tbf) ligand [25], which is also depicted in Scheme 6.2. Originally developed as a protecting anchor group for the synthesis of oligonucleotides [26], polypeptides [27], and proteins [28] it proved to be a useful spectator ligand in half-sandwich complexes intended for syndiospecific styrene polymerization [25]. Furthermore, due to the sterical repulsion between the benzofused rings, a slight helical distortion, which makes the Tbf moiety intrinsically chiral, can be expected (see Scheme 6.3). But due to a very low racemization barrier in solution, only half the signal set can be observed via nuclear magnetic resonance (NMR) [25]. In this chapter, the organometallic chemistry of the Tbf ligand will be summarized.

6.2

THE Tbf LIGAND

Instigated by these developments, there is interest in developing a nearly planar Cp ligand system with an increasing number of annulated benzene rings and subsequently a trial to synthesize a novel type of titanium complexes bearing the tetrabenzo[a,c,g,i]fluorenid (Tbf) as ligand. The corresponding protonated hydrocarbon 8bH-tetrabenzo[a,c,g,i]fluorene (TbfH, IVa) is easily available starting from 9-bromophenanthrene in a classical procedure with slight modifications to the protocol (Scheme 6.4) [25,29]. The tetrabenzo[a,c,g,i]fluorene moiety has not been widely used in organometallic chemistry so far. By thermal heating or by reacting IVa with bases (NEt3 in CH2Cl2), 17H-tetrabenzo[a,c,g,i]fluorene (IVb) is formed (Scheme 6.5). The 17H-isomer IVb is the thermodynamically preferred one, compared to TbfH (IVa) as the kinetic one [30]. The former can be easily obtained as needle-shaped monoclinic crystals, suitable for X-ray diffraction [25]. 17H-Tetrabenzo[a,c,g,i]fluorene (IVb) crystallizes in space group P2/c with four molecules per unit cell (Fig. 6.1).

THE Tbf LIGAND

95

H OH F3CCOOH, H2CCl 2

Br Mg, HCOOCH 3, THF

70%

64%

H

IVa

Scheme 6.4 H

. H

H

Base or Δ H IVa

IVb

Scheme 6.5

Figure 6.1 Oak Ridge thermal ellipsoid plot (ORTEP) plot of the solid-state molecular structure of 17H-tetrabenzo[a,c,g,i]fluorene (IVb). All protons are freely refined; thermal ellipsoids are drawn on a 50% probability level. Selected bond lengths (Å) and angles (degree): C(1)–C(2) 1.5022(16), C(2)–C(3) 1.3718(16), C(3)–C(4) 1.4809(16), C(4)–C(5) 1.3719(15), C(1)–C(5) 1.4971(16), C(17)–C(3)–C(4)–C(18) 22.3(2).

96

6.3


Tbf LITHIUM

6.3.1

Synthesis and Characterization of Tbf Lithium

TbfH (IV) [25] can be deprotonated with n-butyllithium in THF to give TbfLi(THF)4 (V) as blocky yellow crystals when stored at −80 °C (Scheme 6.6). The substance is soluble in THF, in aromatic solvents, and in cyclohexane and exhibits sensitivity to air and moisture, but is not pyrophoric in contrast to common lithium cyclopentadienides. In the case of V, a low melting point without decomposition of 81 °C is found. The crystals obtained are suitable for X-ray structure analysis; V crystallizes in space group P-1; the triclinic unit cell contains four separated ion pairs. As to be seen in the ORTEP plot (Fig. 6.2), the tetrabenzo[a,c,g,i]fluorenide anion is not planar but exhibits a slightly helical distortion due to the steric repulsion of the aromatic ring protons. These results raise the question whether Tbf lithium also forms separated ion pairs in the solid state when other coordinating solvents such as dimethoxyethane (DME) are used. Therefore, Tbf lithium was prepared by deprotonation of TbfH by n-butyllithium in toluene. The white precipitate was filtered and redissolved in DME (Scheme 6.6). Cooling the resulting yellow solution to −80 °C yields blocky yellow crystals (melting point 98 °C), suitable for X-ray diffraction. Like V TbfLi(DME)3 (VI) crystallizes in space group P-1, the triclinic unit cell contains 0.25 equiv of DME in addition to four separated ion pairs of VI (Fig. 6.3). Like in the case of V, also for VI, two symmetry independent molecules are found in the unit cell. However, the structural data are identical with respect to the standard deviation; for that reason, only one molecule is discussed. The tetrabenzofluorenid complexes V and VI represent rare examples of η0-coordinated carbanions employing “simple” Lewis bases, like THF or DME, instead of crown ethers for coordinative saturation of the cation. NMR experiments were performed on VI in order to inspect if the compound also exists as separated ion pairs in solution, especially in nonpolar

IV

n-BuLi,solvent, RT, 2 h

Li(solvent)x

–Bu-H

Solvent: THF, x = 4

V

Solvent: DME, x = 3

VI

Scheme 6.6

Tbf LITHIUM

97

Figure 6.2 ORTEP plot of the solid-state molecular structure of V. Hydrogen atoms are omitted for clarity; thermal ellipsoids are drawn on a 50% probability level. Selected bond lengths (Å) and angles (degree): C(1)–C(2) 1.404(3), C(1)–C(5) 1.406(3), C(2)–C(3) 1.431(3), C(3)–C(4) 1.431(3), C(4)–C(5) 1.425(3), C(17)–C(3)–C(4)–C(29) 30.4(4).

Figure 6.3 ORTEP plot of the solid-state molecular structure of VI. Hydrogen atoms are omitted for clarity; thermal ellipsoids are drawn on a 50% probability level. Selected bond lengths (Å) and angles (degree): C(1)–C(2) 1.403(2), C(1)–C(5) 1.397(2), C(2)–C(3) 1.425(2), C(3)–C(4) 1.432(2), C(4)–C(5) 1.437(2), C(17)–C(3)–C(4)–C(18) 22.2(3).

98


TABLE 6.1 Selected 1H (ppm, C6D6, 500.13 MHz, 300 K) NMR Data of Compounds V and VI 2 15

3

1 16 17

14 27 26

13

28

12

25

18 23 22

29 24

19

4 20

21 5

9 11

8

10

6 7

H-1/H-16 H-2/H-15 H-3/H-14 H-4/H-13 H-5/H-12 H-6/H-11 H-7/H-10 H-8/H-9 H-17

V

VI

8.48 7.53 7.40 8.59 8.60 7.35 7.48 9.30 7.690

8.49 7.55 7.41 8.60 8.62 7.36 7.48 9.28 7.670

solvents like C6D6. Nuclear Overhauser effect (NOE) measurements do not reveal any coupling between the protons of the DME molecules coordinated to the lithium cation and the protons of the Tbf anion. Therefore, it is to assume that even in nonpolar solvents, separated ion pairs are formed due to the strong delocalization of the negative charge on the anion aromatic ring system. Characteristic 1H NMR shifts of V and VI are given in Table 6.1. 6.4

Tbf TITANIUM(III) DERIVATIVES

Directly reacting TbfLi(THF)4 (V) with TiCl4(THF)2 ended up in the reduction of the metal center and in the coinstantaneous oxidative coupling of the ligand. As a natural consequence, the synthesis of Tbf titanium trichloride via a silylated tetrabenzo[a,c,g,i]fluorene derivative was approached, but it did not yield any identifiable products so far [25]. For these reasons, the more redox stable Ti(III) complexes were prepared first [31]. 6.4.1

Synthesis of Tbf Titanium(III) Chloride Complexes

Reacting TbfLi(THF)4 (V) with TiCl3(THF)3 (VII) in toluene formed a green solution that was left for crystallization after 1 min of stirring (Scheme 6.7).


99

+TiCl3(THF)3 (VII) V

–LiCl

Ti

Cl

O Cl

VIII

Scheme 6.7

Figure 6.4 ORTEP plot of the solid-state molecular structure of VIII. Hydrogen atoms are omitted for clarity; thermal ellipsoids are drawn at the 50% probability level. Selected bond lengths (Å) and angles (degree): Ti(1)–Ct 2.0340, Ti(1)–Cl(1) 2.3006(5), Ti(1)–Cl(2) 2.3115(5), Ti(1)–O(1) 2.0536(12), Cl(1)–Ti(1)–Cl(2) 103.19(3), Cl(1)– Ti(1)–O(1) 96.81(4), Cl(2)–Ti(1)–O(1) 92.99(4), C17–C3–C4–C19 18.486(290).

After 2 days, dark green crystals of TbfTiCl2(THF) (VIII) were afforded in acceptable yield (64%), which were suitable for X-ray diffraction. The compound proved to be extremely air and moisture sensitive and was hardly soluble in organic solvents with the exception of THF. In solid state, the compound decomposes without prior melting at 120 °C. Compound VIII crystallizes in a monoclinic system; the space group is C2/c with eight molecules per unit (Fig. 6.4). The distance between Ti(1) and Ct(1) measures 2.0340 Å, which is considerably longer than the Ti–Ct distance in the analogue cyclopentadienyl titanium complex CpTiCl2(THF) (2.014 Å) (Table 6.2), which, to our best knowledge, is the only other example of a monomeric Cp-type titanium(III) chloride compound that could be isolated [32,33]. Bis THF adducts, as described for CpTiCl2(THF)2, are not found in the case of VIII [32].

100


TABLE 6.2 Comparison of Selected Bond Length and Angles between Compound VIII and CpTiCl2(THF) TbfTiCl2(THF) (VIII)

CpTiCl2(THF)

Bond Length (Å) Ti(1)–Ct Ti(1)–Cl(1) Ti(1)–Cl(2) Ti(1)–O(1)

2.0340 2.3006(5) 2.3115(5) 2.0536(12)

2.014 2.310(2) 2.339(3) 2.065(4)

Bond Angle (degree) Cl(1)–Ti(1)–Cl(2) Cl(1)–Ti(1)–O(1) Cl(2)–Ti(1)–O(1)

TiCl 3(THF)3 (VII)

103.19(3) 96.81(4) 92.99(4)

105.6(1) 93.7(1) 94.4(1)

1.TbfLi(THF)4 (V),toluene 2.CCl 4 Ti

–LiCl

Cl

Cl Cl

IX Scheme 6.8

In order to synthesize a Tbf titanium(IV) species from compound VIII, oxidative chlorination with CCl4 promised to be successful. Reacting TiCl3(THF)3 with TbfLi(THF)4 in toluene with subsequent extraction and thus simultaneous oxidative chlorination using CCl4 afforded TbfTiCl3 (IX) in an acceptable overall yield of 66% (Scheme 6.8). The dark purple compound is quite insensible to air and moisture. In contrast to its moderate solubility in dichloromethane and THF, IX is poorly soluble in toluene and is completely insoluble in diethylether and aliphatic solvents; therefore, crystals of IX suitable for X-ray diffraction could easily be obtained by layering a solution in dichloromethane with an equal amount of n-hexane. The dark violet needle-shaped crystals contain four molecules of IX per unit cell along with 2 equiv of solvent. The crystal system is monoclinic with space group P21/n. As shown in Figure 6.5, the titanium atom is coordinated in a pseudotetrahedral fashion with the Tbf moiety in the apical position, with a Ct–Ti(1) distance measuring 2.0357 Å, which is about the same distance as in TbfTiCl2(THF) within accuracy of measurement. The bond lengths between titanium and the three chloride atoms amount to an average out at 2.23 Å. Table 6.3 gives a comparative survey of the most important structural data of compound IX and similar organometallic complexes like CpTiCl3 [34] and IndTiCl3 [35]. It can be easily seen that all Ti–Cl bond lengths and Cl–Ti–Cl


101

Figure 6.5 ORTEP plot of the solid-state molecular structure of IX. Hydrogen atoms are omitted for clarity; thermal ellipsoids are drawn at the 50% probability level. Selected bond lengths (Å) and angles (degree): Ti(1)–Ct 2.0357, Ti(1)–Cl(1) 2.2303(11), Ti(1)–Cl(2) 2.2173(11), Ti(1)–Cl(3) 2.2512(11), Cl(1)–Ti(1)–Cl(2) 103.07(4), Cl(2)– Ti(1)–Cl(3) 106.57(4), Cl(1)–Ti(1)–Cl(3) 102.15(5), C18–C4–C3–C17 15.710(626). TABLE 6.3 Comparison of Selected Bond Length and Angles between Compound IX, CpTiCl3, and IndTiCl3 TbfTiCl3 (IX)

CpTiCl3

IndTiCl3

Bond Length (Å) Ti(1)–Ct Ti(1)–Cl(1) Ti(1)–Cl(2) Ti(1)–Cl(3)

2.0357 2.2303(11) 2.2173(11) 2.2512(11)

2.01 2.201(5) 2.248(5) 2.221(2)

2.032 2.2248(8) 2.2319(8) 2.2355(8)

Bond Angle (degree) Cl(1)–Ti(1)–Cl(2) Cl(2)–Ti(1)–Cl(3) Cl(1)–Ti(1)–Cl(3)

103.07(4) 106.57(4) 102.15(5)

102.2(2) 104.1(2) 102.3(3)

102.44(3) 104.09(3) 103.32(3)

angles are almost alike in all three complexes, thus leaving the Ti–Ct distance in CpTiCl3 with 2.01 Å to be the only major difference from IX and IndTiCl3 in whose molecular structures the Ti–Ct distances amount to 2.03 Å. These crystallographic data and the dark violet color suggest that compound IX is electronically quite similar to IndTiCl3.

102


6.4.2

Reaction of TbfTiIIICl2(THF) (VIII) with Radicals

Another possible synthetic route to introduce ligands into an organometallic complex is the reaction of one of its low-valent species with radicals. Thus, an O- or N-centered radical is reduced to a monoanionic ligand while the metal center is oxidized. However, some applications of the radical-like properties of titanium(III) species are known in the literature. In that way, the free radical-like properties of CpTiIIICl2 derivatives are observed in reactions, for example, with quinones [36] as well as with azo and diazo compounds [32], whereas the intermediately generated Cp2TiIIICl reagent is well introduced in organic reactions [37]. A variety of persistent nitroxyl radicals such as TEMPO (2,2,6,6-tetramethylpiperidine-N-oxyl) or di-tert-butyl nitroxide have been used to synthesize an array of different metal complexes, including s-block metalloids [38], early transition metals [39], and lanthanoids [40]. As depicted in Scheme 6.9, compound VIII was synthesized in situ and subsequently reacted with TEMPO at ambient temperature to form a burgundy red reaction mixture. Filtration and cooling of the solution to −10 °C afforded crystals of TbfTiCl2TEMPO (X) in 51% yield. Compound X is well soluble with the exception of aliphatic solvents and crystallizes in the triclinic space group P-1 with two molecules of X and one molecule of toluene per unit cell (Fig. 6.6). Not only persistent radicals can be reacted with low-valent metal complexes but also those that are generated in situ by the thermal cleavage of peroxides. To our best knowledge, so far, only samarium [41] and ytterbium [42] compounds were reported into which a tert-butoxide moiety was introduced by reduction of di-tert-butylperoxide by the metal center. Here, the first Cp-type titanium monoalkoxide complex synthesized by reacting a titanium(III) species with di-tert-butylperoxide is reported. As depicted in Scheme 6.10 the titanium(III) compound VIII was prepared in situ and half an equivalent di-tert-butylperoxide was added. Heating the reaction mixture up fast until reflux affords a deep red suspension, which was taken to dryness. The residue was extracted with dichloromethane and was filtered. Layering the solution with n-hexane afforded deep red crystals of

TiCl 3(THF)3 (VII)

1. TbfLi(THF) 4, toluene 2. TEMPO, toluene Ti

–LiCl O TEMPO:

N

Cl

Cl O N

X

Scheme 6.9


103

Figure 6.6 ORTEP plot of the solid-state molecular structure of X. Hydrogen atoms are omitted for clarity; thermal ellipsoids are drawn at the 50% probability level. Selected bond lengths (Å) and angles (degree): Ti(1)–Ct 2.0875, Ti(1)–Cl(1) 2.2940(5), Ti(1)–Cl(2) 2.2789(6).

TiCl3(THF) 3

VII

1. TbfLi(THF)4, toluene 2. 0.5 equiv t BuOOt Bu, toluene –LiCl

Ti Cl

Cl O t

IX

KOt Bu, toluene

Bu

XI

–KCl

Scheme 6.10

104


Figure 6.7 ORTEP plot of the solid-state molecular structure of XI. Hydrogen atoms are omitted for clarity; thermal ellipsoids are drawn at the 50% probability level. Selected bond lengths (Å) and angles (degree): Ti(1)–Ct 2.0683, Ti(1)–Cl(1) 2.2808(5), Ti(1)–Cl(2) 2.2817(5), Ti(1)–O(1) 1.7414(11), Ti(1)–O(1)–C(30) 165.38(12), Cl(1)– Ti(1)–O(1) 103.22(4), Cl(2)–Ti(1)–O(1) 100.25(4), Cl(1)–Ti(1)–Cl(2) 103.44(2), C18–C4–C3–C17 19.489(298).

TbfTiCl2(OtBu) (XI) in 10% yield. This is strongly associated with thermal decay of the monomeric titanium(III) species at those elevated temperatures. Higher yields are obtained by reacting IX with potassium tert-butoxide in a toluene suspension. Workup of the reaction mixture was identical and afforded crystals of XI in 27% yield. Compound XI is well soluble in chlorinated solvents, in THF, and, to a lesser extent, in aromatic solvents. It crystallizes in the monoclinic space group P21/c with four molecules per unit cell. The Ti–Ct distance measures 2.0683 Å, and the Ti–O–C bond angle of 165.38 ° (Fig. 6.7) is very similar to the one in TbfTiCl2(OiPr) reported previously [25]. Common with Tbf titanium compounds are intermolecular face-to-face πstacking interactions [43] in solid state between the Tbf ligands, resulting in the abovementioned low solubility of these compounds. This phenomenon is to be found in all solid-state structures discussed so far and is most intense in the solid-state structure of compound IX (see Fig. 6.8), in which the distance between the centroids of the five-membered ring and one of the six-membered rings amounts to 3.470 Å, being indicative of strong face-to-face π interactions. This correlates with the compound’s solubility, which is lowest by comparison.

105

3.470

3.470

Tbf TITANIUM(IV) DERIVATIVES

Figure 6.8 Intermolecular face-to-face π-stacking interactions (3.470 Å) in solid state between the tetrabenzo[a,c,g,i]fluorenyl ligands on the example of compound IX.

In the solid-state structures of all other compounds, this phenomenon occurs to a lesser extent, but being more soluble, too. 6.5


One of the standard procedures for the synthesis of half-sandwich cyclopentadienyltitanium trichlorides involves the use of silylated cyclopentadienyl [44] derivatives in cases where the use of lithiated compounds leads to reduction of the metal center [22]. As mentioned, directly reacting of TbfLi(THF)4 (V) with TiCl4(THF)2 ended up in the reduction of the metal center and in the coinstantaneous oxidative coupling of the ligand, leading to the well-crystallizing Tbf–Tbf hydrocarbon coupled in the C17 position (Scheme 6.11). The low solubility of XII prohibits NMR measurements, whereas in the electron impact (EI) mass spectra, the molecular peak can be observed (m/z: 730.2). Bis(17-tetrabenzo[a,c,g,i]fluorene) (XII) is poorly soluble in any organic solvent and proves to be highly temperature resistant for an aromatic compound as it decomposes without melting at 363 °C. The compound crystallizes from dichloromethane in space group P21/n; the monoclinic cell unit contains four molecules of (Tbf)2 along with 4 equiv of dichloromethane (Fig. 6.9). A extraordinary long C–C single bond between C1 and C30 (1.615(2) Å) is

106


V

TiCl 4 –TiCl 3, –LiCl

XII Scheme 6.11

Figure 6.9 ORTEP plot of the solid-state molecular structure of XII. Hydrogen atoms and one molecule H2CCl2 are omitted for clarity; thermal ellipsoids are drawn on a 50% probability level. Selected bond lengths (Å) and angles (degree): C(1)–C(30) 1.615(2), C(1)–C(2) 1.509(2), C(1)–C(5) 1.512(2), C(2)–C(3) 1.366(2), C(3)–C(4) 1.474(2), C(4)–C(5) 1.370(2), C(17)–C(3)–C(4)–C(29) 19.3.

found, nearly identical to the corresponding C–C bond in a comparable compound described by Malaba, Tessier, and Youngs [45]. A nearly perfect faceto-face π-π stacking of the C5/C24 and C31/C41 rings is observed (distance of the ring centers 3422 Å) [43].


107

Cl + V, toluene Ti L L:

Oi

O i Pr

Ti

–LiCl

L

L

Pr XIII

L: O i Pr

O i Pr

L

XIV

Scheme 6.12

However, as also alternative reactions of Tbf-SiMe3 and TiCl4 failed, resulting only in unidentified products, an alternative synthetic path was forced to be choosen, employing titanium alkoxides in order to increase the electron density on the metal center [46]. Thus, for fluorenyltitanium complexes, an sufficient thermal stability is observed [24]. Reacting a solution of V in toluene with ClTi(OiPr)3 (XIII) at ambient temperature and subsequent crystallization from n-hexane (−60 °C) yield yellow crystals of XIV (Scheme 6.12), which can be isolated in acceptable yields (68%). The Tbf complex XIV melts at 135 °C and shows good solubilities in common organic solvents. Of high diagnostic value is the 1H NMR shift of H17 in XIV (δ 7.6 ppm). The other signals are in the expected range; further details are given later on. The crystals obtained are also suitable for X-ray diffraction (Fig. 6.10). The orthorhombic crystals contained eight molecules of compound XIV in the unit cell (space group Pbca). The distance Ti(1)–Ct(1) measures 2.120 Å and is nearly identical to the Ti–Ct distance in (η5-Flu)(η1-Flu)Ti(OiPr)2 (Flu: fluorenyl, 2.122 Å) [24], but is longer as in CpTiCl3 (2.01 Å)34 or in (η5-Ind)TiCl3 (2.032 Å) [35]. Two of the OiPr groups show short Ti–O distances (av 1.77 Å), whereas a larger value of 1.823(2) Å is found for the third group. This corresponds with short C–O distances (av 1.39 Å) and large Ti–O–C angles (av 167 °) for the first two alkoxide ligands, indicative of higher Ti–O bond orders. On the other hand, a shorter C(7)–O(3) bond (1.319(9) Å) as well as a smaller Ti(1)– O(3)–C(7) angle (137.4(5)°) is found. Being stable as a solid compound, XIV shows some decomposition, mainly more in polar solvents than in nonpolar ones. 6.5.1

Synthesis of Tbf Titanium Monophenoxide Complexes

Standard synthetic procedures for the synthesis of Cp-type titanium monophenoxide complexes from their trichloride precursors involve the use of either lithium phenoxides [47] or phenols in conjunction with an organic base in order to absorb the HCl released [48]. As depicted in Scheme 6.13, compound IX was reacted with lithium phenoxides in a toluene slurry to form an array of Tbf titanium monophenoxides of the general formula TbfTiCl2OAr. Generally, working up the reaction

108


Figure 6.10 ORTEP plot of the solid-state molecular structure of XIV. Hydrogen atoms are omitted for clarity; thermal ellipsoids are drawn on a 50% probability level. Selected bond lengths (Å) and angles (degree): Ti(1)–Ct(1) 2.120, Ti(1)–O(1) 1.775(8), Ti(1)–O(2) 1.768(8), Ti(1)–O(3) 1.823(2), O(1)–C(1) 1.396(3), O(2)–C(4 °) 1.394(3), O(3)–C(7) 1.319(9), C(10)–C(11) 1.435(3), C(10)vC(14) 1.417(4), C(11)–C(12) 1.418(4), C(12)–C(13) 1.409(4), Ti(1)vO(1)–C(1) 165.0(5), Ti(1)–O(2)–C(4) 169.2(2), Ti(1)– O(3)–C(7) 137.4(5), C(15)–C(10)–C(11)–C(27) 22.4(4).

LiOAr, toluene Ti Cl

–LiCl

Cl Cl

IX

Ti Cl

Cl OAr

XV–XIX t

Ar = 4- BuC6H4 (XV), 2,6-Me2C6H3 (XVI), 2,4,6-Me3C6H2 (XVII) 2,6- iPr2C6H3 (XVIII), 2,6-Ph2C6H3 (XIX)

Scheme 6.13

mixtures included evaporation of the solvent with subsequent extraction of the residue with dichloromethane with the exception of TbfTiCl2(2,6Ph2C6H3O) (XIX), which was filtered directly in order to remove LiCl. All solutions obtained were layered with n-hexane to induce crystallization of the product complexes. In this way, crystals suitable for X-ray diffraction are obtained on a regular basis (Table 6.4).

109


TABLE 6.4 XV–XIX

Comparison of Selected Bond Length and Angles between Compounds XV

XVI6

XVII

XVIII

XIX

Bond Length (Å) Ti(1)–Ct Ti(1)–Cl(1) Ti(1)–Cl(2) Ti(1)–O(1)

2.0665 2.274(3) 2.257(3) 1.782(6)

2.0701 2.2766(16) 2.2637(17) 1.779(3)

2.0589 2.2508(6) 2.2614(7) 1.7737(16)

2.0544 2.2467(6) 2.2545(7) 1.7751(14)

2.0554 2.2509(6) 2.2570(7) 1.7913(15)

Bond Angle (degree) Ti(1)–O(1)–C Cl(1)–Ti(1)–Cl(2)

161.1(5) 102.95(10)

159.7(3) 102.32(7)

158.94(13) 103.45(3)

159.17(13) 104.45(2)

154.61(12) 104.08(3)

Figure 6.11 ORTEP plot of the solid-state molecular structure of XV. Hydrogen atoms are omitted for clarity; thermal ellipsoids are drawn at the 50% probability level. Selected bond lengths (Å) and angles (degree): Ti(1)–Ct 2.0665, Ti(1)–Cl(1) 2.274(3), Ti(1)–Cl(2) 2.257(3), Ti(1)vO(1) 1.782(6), Ti(1)–O(1)–C(30) 161.1(5), O(1)–Ti(1)– Cl(1) 101.6(2), O(1)–Ti(1)–Cl(2) 105.1(2), Cl(1)–Ti(1)–Cl(2) 102.95(10), C18–C4– C3–C17 −13.722(1594).

TbfTiCl2(4-tBuC6H4O) (XV) is well soluble in dichloromethane and THF, is of low solubility in toluene, and is completely insoluble in aliphatic solvents. It could be obtained in 23% yield as deep red crystals with two molecules per unit cell in the triclinic crystal system P-1. The Ti(1)–Ct distances measure 2.0665 Å; the Ti(1)–Cl bond lengths amount to an average 2.266 Å (Fig. 6.11). The distance between centers of the benzofused tetrabenzo[a,c,g,i]fluorene ring C6–C11 and the phenoxide ring C30–C35 measures 3.841 Å, and therefore it is to assume that face-to-face π-stacking interactions take place [43].

110


Figure 6.12 ORTEP plot of the solid-state molecular structure of XVI. Hydrogen atoms are omitted for clarity; thermal ellipsoids are drawn at the 50% probability level. Selected bond lengths (Å) and angles (degree): Ti(1)–Ct 2.0701, Ti(1)–Cl(1) 2.2766(16), Ti(1)–Cl(2) 2.2637(17), Ti(1)–O(1) 1.779(3), Ti(1)–O(1)–C(30) 159.7(3), Cl(1)– Ti(1)–O(1) 105.97(12), Cl(2)–Ti(1)–O(2) 100.72(12), Cl(1)–Ti(1)–Cl(2) 102.32(7), C29–C5–C1–C6 12.825(946).

TbfTiCl2(2,6-Me2C6H3O) (XVI) is well soluble in dichloromethane and THF, is of low solubility in toluene, and is completely insoluble in aliphatic solvents. It could be isolated in 21% yield as red crystals with four molecules per unit cell in the monoclinic crystal system P21/c. As listed in Table 6.3, the Ti–Ct distance measures 2.0701 Å, and the Ti–Cl bond lengths amount to an average of 2.270 Å (Fig. 6.12). Like in compound XV also in XVI it is to assume that face-to-face π-stacking interactions take place between the benzofused ring C12–C17 and the phenoxide ring C30–C35 as the distance between its respective centroids measures 3.62 Å. Interestingly, in contrast to complex XVI, TbfTiCl2(2,4,6-Me3C6H2O) (XVII) does not show any π–π interactions between the phenoxide and any of the benzofused rings, although they only differ by one methyl group in the para-position of the phenoxide ligand. It is not known if this is due to a slightly different electronic structure or a phenomenon of different packing in the crystal. As regards solubility and other physical properties, compound XVII is very similar to complex XVI. It crystallizes in the monoclinic space group P21/n with four molecules per crystal cell. The Ti–Ct bond length measures 2.0589 Å and is thus slightly shorter than in XVI, whereas the Ti–O bond length and the Ti–O–C angle are almost identical (Fig. 6.13).


111

Figure 6.13 ORTEP plot of the solid-state molecular structure of XVII. Hydrogen atoms are omitted for clarity; thermal ellipsoids are drawn at the 50% probability level. Selected bond lengths (Å) and angles (degree): Ti(1)–Ct 2.0589, Ti(1)–Cl(1) 2.2508(6), Ti(1)–Cl(2) 2.2614(7), Ti(1)–O(1) 1.7737(16), Ti(1)–O(1)–C(30) 158.94(13), Cl(1)– Ti(1)–O(1) 102.79(5), Cl(2)–Ti(1)–O(1) 103.36(5), Cl(1)–Ti(1)–Cl(2) 103.45(3), C18– C4–C3–C17 12.510(412).

Like all other Tbf titanium monophenoxide complexes mentioned so far, TbfTiCl2(2,6-iPr2C6H3O) (XVIII) is well soluble in dichloromethane and in THF, is soluble to a lesser extent in aromatic solvents, and is completely insoluble in aliphatic solvents. It can be obtained in 34% yield as red crystals in the monoclinic space group P21/n with four molecules per unit cell (Fig. 6.14). In contrast to the excellent catalyst precursor Cp*TiCl2(2,6-iPr2C6H3O)36, the Ti–O–C bond angle in compound XVIII exhibits only a relatively small value of 159.17 °. As opposed to the TbfTiCl2(OAr) complexes XV–XVIII, TbfTiCl2(2,6Ph2C6H3O) (XIX) is well soluble in aromatic solvents. Its red crystals can be obtained from a toluene solution in 34% yield, and the compound crystallizes in the triclinic space group P-1 with two molecules of XIX and one molecule of toluene per unit cell (Fig. 6.15).

112


Figure 6.14 ORTEP plot of the solid-state molecular structure of XVIII. Hydrogen atoms are omitted for clarity; thermal ellipsoids are drawn at the 50% probability level. Selected bond lengths (Å) and angles (degree): Ti(1)–Ct 2.0544, Ti(1)–Cl(1) 2.2467(6), Ti(1)–Cl(2) 2.2545(7), Ti(1)–O(1) 1.7751(14), Ti(1)–O(1)–C(30) 159.17(13), Cl(1)– Ti(1)–O(1) 102.28(5), Cl(2)–Ti(1)–O(1) 102.57(5), Cl(1)–Ti(1)–Cl(2) 104.45(2), C18– C4–C3–C17 14.966(406).

By reacting Cl2Ti(OPri)2 (XX) with V in toluene, the Tbf complex XXI can be isolated as red crystals (m.p. 208 C). Suitable crystals for X-ray structure determination are obtained by layering the mother liquor with n-hexane (Fig. 6.16). Being space group P21/c, the cell unit contains four molecules. The decrease of electron density on the titanium center leads to a significantly shortened Ti(1)–Ct(1) distance of 2.049 Å as opposed to compound XIV and therefore to a tighter bond between titanium and spectator ligand. The Ti(1)– O(1) distance (1.733(3) Å) is indicative of an expected dπ–pπ interaction, leading to a large Ti–O–C angle of 173.25(1) . As opposed to XIV, compound XXI is stable in all solvents. The preparation of Tbf titanium phenoxides involved the development of a suitable type of chlorotitanium aryloxides as starting material since most complexes of the general formula ClTi(OAr)3Xn (Ar = any phenyl, X = Lewis


113

Figure 6.15 ORTEP plot of the solid-state molecular structure of XIX. Hydrogen atoms are omitted for clarity; thermal ellipsoids are drawn at the 50% probability level. Selected bond lengths (Å) and angles (degree): Ti(1)–Ct 2.0554, Ti(1)–Cl(1) 2.2509(6), Ti(1)–Cl(2) 2.2570(7), Ti(1)–O(1) 1.7913(15), Ti(1)–O(1)–C(30) 154.61(12), Cl(1)– Ti(1)–O(1) 103.98(5), Cl(2)–Ti(1)–O(1) 102.78(5), Cl(1)–Ti(1)–Cl(2) 104.08(3), C18– C4–C3–C17 −10.506(390).

base, n = 0–2) are often only obtained as highly viscous oils or gums and would thus be quite inconvenient to handle [49]. In comparison to the corresponding aliphatic alkoxides, titanium aryloxides are more Lewis acidic due to a σn → π* interaction between the lone pair of the oxygen atom and an antibonding orbital of the aryl ring system [50]. They can easily be reacted with Lewis bases to form adducts [51,52]. In the search for a convenient starting material for the preparation of TbfTi(OAr)3 complexes, TiCl4 or Ti(OEt)4 were refluxed with different phenols (phenol, 4-methylphenol, and 4-tert-butylphenol) in toluene (Scheme 6.14). After evaporation of all volatiles, the resulting residues were redissolved in THF and were layered with n-hexane to give red to orange crystalline compounds {ClTi(OAr)3(THF)}2 (XXII–XXIV). The nearly air stable and crystalline aryloxides XXII (m.p. 94 ° C), XXIII (m.p. 109 ° C), and XXIV (m.p. 125 ° C) are highly soluble in common solvents with the exception of n-hexane. They are fully characterized, including single-crystal X-ray diffraction [25]. Reacting the chlorotitanium triphenoxides XXII, XXIII, and XXIV with V in cyclohexane or toluene at ambient temperature yields mononuclear Tbf

114


Figure 6.16 ORTEP plot of the solid-state molecular structure of XXI. Hydrogen atoms are omitted for clarity; thermal ellipsoids are drawn on a 50% probability level. Selected bond lengths (Å) and angles (degree): Ti(1)–Ct(1) 2.049, Ti(1)–O(1) 1.733(3), Ti(1)–Cl(1) 2.284(9), Ti(1)–Cl(2) 2.285(7), O(1)–C(1) 1.426(3), C(4)–C(5) 1.417(3), C(4)–C(8) 1.446(3), C(5)–C(6) 1.422(3), C(6)–C(7) 1.416(3), C(7)–C(8) 1.423(3), Ti(1)– O(1)–C(1) 173.25(1), C(9)–C(4)–C(8)–C(32) 21.3(4).

THF

TiX4

1. ArOH, toluene, reflux 2. THF

ArO

Ti ArO Cl

X = Cl, OEt

Ar O

Cl OAr

Ti O Ar

OAr THF

Ar = C6H5 (XXII), 4-MeC6H4 (XXIII ) 4-t BuC6H4 (XXIV) Scheme 6.14

titanium triaryloxides TbfTi(OAr)3 (Ar = C6H5 [XXV], 4-MeC6H4 [XXVI], 4-tBuC6H4 [XXVII]) (Scheme 6.15). These Tbf titanium aryloxides are obtained as red crystalline solids (m.p. 153 ° C [XXV], 154 ° C [XXVI], 173 ° C [XXVII]) showing acceptable solubilities in common solvents with the exception of n-hexane. Layering the mother liquors with n-hexane yielded crystals suitable for X-ray diffraction of compounds XXV and XXVI (Figs. 6.17 and 6.18). Any


THF ArO

Ar O

Ti ArO Cl

Cl OAr

Ti O Ar

115

OAr

2 TbfLi(THF)4, (V) toluene –2 LiCl

2

Ti ArO

THF

OAr

OAr

Ar = C6H5 (XXV) 4-MeC6H4 (XXVI) 4-t BuC6H4 (XXVII)

Ar = C6H5 (XXII) 4-MeC6H4 (XXII) 4-t BuC6H4 (XXIV)

Scheme 6.15

Figure 6.17 ORTEP plot of the solid-state molecular structure of XXV. Hydrogen atoms are omitted for clarity; thermal ellipsoids are drawn on a 50% probability level. Selected bond lengths (Å) and angles (degree): Ti(1)–Ct(1) 2.069, Ti(1)–O(1) 1.851(6), Ti(1)–O(2) 1.799(3), Ti(1)–O(3) 1.806(9), O(1)–C(30) 1.358(8), O(2)–C(36) 1.348(7), O(3)–C(42) 1.356(3), C(1)–C(2) 1.402(3), C(1)–C(5) 1.424(4), C(2)–C(3) 1.438(3), C(3)–C(4) 1.452(4), C(4)–C(5) 1.431(3), Ti(1)–O(1)–C(30) 142.76(1), Ti(1)–O(2)–C(36) 166.11(1), Ti(1)–O(3)–C(42) 161.52(1), C(17)–C(3)–C(4)–C(18) 11.7(5).

116


Figure 6.18 ORTEP plot of the solid-state molecular structure of XXVI. Hydrogen atoms are omitted for clarity; thermal ellipsoids are drawn on a 50% probability level. Selected bond lengths (Å) and angles (degree): Ti(1)–Ct(1) 2.067, Ti(1)–O(1) 1.829(9), Ti(1)–O(2) 1.844(5), Ti(1)–O(3) 1.800(0), O(1)–C(30) 1.353(2), O(2)–C(37) 1.359(2), O(3)–C(44) 1.361(2), C(1)–C(2) 1.431(2), C(1)–C(5) 1.418(3), C(2)–C(3) 1.437(3), C(3)–C(4) 1.433(3), C(4)–C(5) 1.417(3), Ti(1)–O(1)–C(30) 150.10(1), Ti(1)–O(2)–C(37) 141.72(1), Ti(1)–O(3)–C(44) 160.28(1), C(18)–C(3)–C(19)–C(17) 13.3(4).

attempts to obtain crystals of compound XXVII that would have been suitable for X-ray diffraction have not been successful so far. Compound XXV crystallizes in the monoclinic space group P21/n with four molecules per unit cell, whereas XXVI crystallizes in the triclinic space group P-1 with two molecules per unit cell. The Tbf complexes XXV and XXVI are characterized by the formation of a tetrahedral coordinated titanium center. The Ti(1)–Ct distances in XXV (2.069 Å) and XXVI (2.067 Å) are shorter than those in XIV (2.120 Å). Similar to XIV, two shorter (av 1.80 Å) and one longer (1.851(6) Å) Ti(1)–O(1) bonds are found also for XXV. The short distances correlate also as in XIV to the Ti–O–C angles. For XXVI, three different Ti–O bond lengths are found. Short contacts of the aromatic rings (center of C(18)/C(23) and C(42)/C(47) 3.78 Å) in XXV are indicative again of a face-to-face π–π stacking [43].

DYNAMIC AND POLYMERIZATION BEHAVIOR

117

6.6 DYNAMIC AND POLYMERIZATION BEHAVIOR OF TETRABENZOFLUORENYL TITANIUM COMPLEXES Due to the helical shape of the ligand, all η5-Tbf titanium complexes exhibit C1 symmetry (Scheme 6.16) giving two optical isomers (A and B), both of which crystallize as racemates, leading to centrosymmetric space groups. NMR spectroscopy reveals that in solution, both isomers rapidly interconvert as instead of a twofold signal set, which would be indicative of two isomers, only half the signal set can be observed at ambient temperature. Thus, temperature-dependent NMR experiments were conducted in order to determine the energetic barrier for interconversion of both isomers (Fig. 6.19).

Ti

Ti X

X

X

X

A

X X

B

Scheme 6.16 H17

H8/H9

300K 293K 283K 273K 263K 253K 243K 233K 223K 213K 203K

8.50

8.00

7.50

ppm (t1)

Figure 6.19 Variable-temperature 1H NMR spectrum (300.13 MHz, CDCl3) of XXVI. Only the range relevant for the tetrabenzo[a,c,g,i]fluorenyl ligand is shown.

118


TABLE 6.5 Selected 1H NMR Data (500.13 MHz, 300K) of Compounds XIV, XXI, and XXV–XXVII (Numbering Scheme in Accordance to Table 6.1) Solvent

H-1/H-16 H-2/H-15 H-3/H-14 H-4/H-13 H-5/H-12 H-6/H-11 H-7/H-10 H-8/H-9 H-17

XIV

XXI

XXV

XXVI

XXVII

C6D6

CDCl3

CDCl3

CDCl3

CDCl3

8.24 7.47 * 8.41 8.43 * * 9.14 7.59

8.39 * * 8.60 8.63 * 7.60 8.85 8.01

8.14 7.60 7.53 8.38 8.41 7.48 7.56 8.81 7.80

8.16 7.63 7.56 8.43 8.45 7.50 7.59 8.83 7.80

8.14 7.58 7.51 8.33 8.36 7.48 7.54 8.81 7.79

*Signals could not be clearly assigned.

A sample of compound XXVI, which was chosen due to its good solubility and stability, was dissolved in CDCl3, and measurements were taken within a temperature range of 300–203K in the attempt to “freeze in” the flipping movement of the benzofused rings. Unfortunately, a complete stoppage and thus a twofold signal set could not be achieved. The spectra only show what seems to be a deceleration of the interconversion movement as the doublet for the H-8/H-9 protons, which cause the ligands to twist helically due to their steric repulsion, is shifted upfield with decreasing temperature. When comparing the NMR data of the different η5-Tbf titanium complexes (Table 6.5), one difference between the compounds measured in CDCl3 is striking: Whereas the proton signals of H-1/H-16, H-4/H-13, H-5/H-12, and H-17 for the triaryloxide derivatives all fall in the same range, the abovementioned proton signals of compound XXI are shifted downfield for a roundabout 0.2 ppm. Most interestingly, this shift correlates with different Ti–Ct bond lengths and twist angles of the helically shaped ligand found in X-ray crystal structures (Table 6.6). The torsion angle between the carbon atoms denoted C25-C24-C23-C22 can be taken as indicative of the ligand helical twisting. Interestingly, this torsion angle is almost identical in 17H-Tbf (IVb), in the anion of VI, and in the complexes XIV and XXI despite their different electronic situations. In the triaryloxide complexes XXV and XXVI, the torsion angle declines from an average of 22 ° (VI, XIV, XXI) to 11.7(5) ° (XXV) and 13.3(4)° (XXVI), respectively. Comparison of bond lengths in the Tbf moiety reveals that the C–C distances of the five-membered ring are strongly influenced by the charge of the molecule. In the Li salts V and VI, the C17–C18 and C17–C29 bonds (av 1.40 Å, numbering in accordance to Table 6.1) are found to be shorter as in the neutral compounds IVb and XII (av 1.50 Å). A similar shortening is observed for the titanium complexes XIV, XXI, XXV, and XXVI (av 1.41 Å). The C24–C29 as well as the C18–C23 bond is elongated comparing the neutral

DYNAMIC AND POLYMERIZATION BEHAVIOR

119

TABLE 6.6 Comparison of Selected Bond Lengths (Å) and Torsion Angles (Degree) (Numbering Scheme in Accordance to Table 6.1)

C25–C24– C23-C22 (degree) Ct–Ti (Å)

IVb

V

VI

XIV

XXI

XXV

XXVI

22.3(2)

30.4(4)

22.2(3)

22.4(4)

21.3(4)

11.7(5)

13.3(4)

—

—

—

2.120

2.049

2.069

2.067

(av 1.37 Å), ionic, as well as titanium derivatives (av 1.43 Å). Also, the C23– C24 distances are influenced as expected, for the neutral derivatives (1.40 Å [av]), the lithium salts (1.43 Å [av]), and the titanium complexes (1.44 Å [av]) are observed. The other C–C distances are less influenced by electronic effects. However, an increasing aromatic character of the annulated rings is found comparing IVb and XII with the lithium as well as with the titanium complexes. In this way, the C–C distances C20–C21, C26–C27 (x); C18–C19, C28– C29 (y); and C23–C22, C24–C25 (z) of the central six-membered rings increase significantly from the neutral to the lithium up to the titanium Tbf derivatives (x, av: neutral 1.45, Li 1.46, Ti 1.47 Å; y, av: neutral 1.42, Li 1.44, Ti 1.45 Å; z, av: neutral 1.45, Li 1.45, Ti 1.46 Å). 6.6.1

Styrene Polymerization

To assess the capability of the complexes as catalysts in polymerization reactions, the coordination polymerization of styrene as a monomer was investigated in the presence of the η5-Tbf titanium complexes. Methylaluminoxane (MAO) was selected as a cocatalyst and was used in relatively low concentrations with regard to the transition metal corresponding to a molar ratio of MAO/Ti = 110/1. In addition, triisobutylaluminum (TIBA) was used as a strong reducing agent for the titanium compound and was added in a molar ratio of TIBA/Ti = 25/1. The polymerizations were initiated at a high molar ratio of styrene/Ti = 700,000/1 by the addition of the catalyst premix solution to the styrene monomer at 30° C. The polymerization results (Table 6.7) indicate a significant activity of all η5-Tbf titanium complexes as catalysts in the coordination polymerization of styrene with MAO as cocatalyst. However, the polymerization activity of the different complexes varies depending on the kind of ancillary ligands in addition to the η5-Tbf group. The activity increases in the following order: XXI (1420-kg SPS/(mol Ti × mol styrene × h) < XXVII (3400) < XXVI (3740) < XIV (6040) < XXV (6720). On the other hand, the stereospecificity of the complexes as catalysts in the coordination polymerization of styrene can be determined by investigating the melting points of the polystyrenes received by differential scanning calorimetry (DSC) (Table 6.7). All polymer samples obtained exhibited high melting points, indicating syndiotactic polymers. Furthermore, based on the melting points, the polystyrenes show a very high degree of syndiotacticity comparable to about 100%

120


TABLE 6.7 110/25/1) Run No. 1 2 3 4 5 6 7 8

Polymerization of Styrene (30° C, styrene/MAO/TIBA/Ti = 700,000/ Catalyst

Polymerization Time (min)

Activitya

Melting Pointb (°C)

XXI XIV XXV XXV XXVI XXVI XXVII XXVII

120 120 60 120 60 120 60 120

1420 6040 6720 4720 3740 2000 3400 2000

267.2 268.0 266.2 265.6 266.4 266.2 266.0 267.0

Activity in kg SPS/(mol Ti × mol styrene × h). Melting point of the polymer.

a

b

of the rrrrr-hexad fraction via 13C NMR, in comparison to, for example, syndiotactic polystyrene with a melting point of 257 C and a rrrrr-hexad fraction of 89% received with CpTiCl3 as catalyst [53]. 6.7

CONCLUSIONS

The efficient syntheses of IVa allows an effective access to the organometallic chemistry of the Tbf moieties. Using the remarkable η0-coordinated lithium salts V and VI, monosubstituted titanium(IV) alkoxides and aryloxides become available. The Tbf moiety is coordinated as a sterically demanding Cp derivative in an η5 fashion to the titanium centers. The formal benzannulation of the fluorenyl system, as present in the Tbf titanium complexes, leads to large mono Cp complexes of high thermal stability compared to η5-fluorenyl titanium derivatives. Due to its helical twist, the η5-coordinated Tbf ligands show C1 symmetry in the solid state but seem to interconverse in solution. Additionally, the titanium complexes were subjected to polymerization experiments of styrene to some extent. REFERENCES 1. Jutzi, P., Edelmann, F., Bercaw, J. E., Beckhaus, R., Negishi, E., Royo, P., Okuda, J., Halterman, R. L., Janiak, C., Hoveyda, A. H., Togni, A., Manners, I. In Metallocenes, Halterman, R. L.; Togni, A. (ed.), Wiley-VCH, Weinheim, 1998, pp. 1–832. 2. Janiak, C., Schumann, H. Bulky or supracyclopentadienyl derivatives in organometallic chemistry. Adv. Organomet. Chem., 33, 291–393 (1991). 3. Buchholz, D., Gloaguen, B., Fillaut, J. L., Cotrait, M., Astruc, D. Mono- and bis (pentaisopropylcyclopentadienyl) cobalt and rhodium sandwich complexes and other decabranched cyclopentadienyl complexes. Chem. Eur. J., 1, 374–381 (1995).

REFERENCES

121

4. Císarová, I., Varga, V., Horácek, M., Kubista, J., Mach, K. Exceptionally symmetric crystal structure of (pentabenzylcyclopentadienyl)-(cyclooctatetraene) titanium(III). Collect. Czech. Chem. Commun., 69, 2036–2044 (2004). 5. An, J., Urrieta, L., Williams, R., Tikkanen, W., Bau, R., Yousufuddin, M. Piano stool complexes containing the bulky pentaphenylcyclopentadienyl(C5Ph5) ligand: Preparation, characterization and X-ray structure of C5Ph5Zr(N(CH3)2)3 (I). J. Organomet. Chem., 690, 4376–4380 (2005). 6. Yu, Y., Bond, A. D., Leonard, P. W., Vollhardt, P. C., Whitener, G. D. Syntheses, structures, and reactivity of radial oligocyclopentadienyl metal complexes: Penta(ferrocenyl)cyclopentadienyl and congeners. Angew. Chem., 118, 1826–1831 (2006). 7. Carella, A., Rapenne, G., Launay, J. P. Design and synthesis of the active part of a potential molecular motor. New. J. Chem., 29, 288–290 (2005). 8. Cadierno, V., Diez, J., Pilar Gamasa, M., Gimeno, J., Lastra, E. Indenyl complexes of group 8 metals. Coord. Chem. Rev., 193–195, 147–205 (1999). 9. Alt, H. G., Samuel, E. Fluorenyl complexes of zirconium and hafnium as catalysts for olefin polymerization. Chem. Soc. Rev., 27, 323–329 (1998). 10. Chin, R. M., Jarosh, M. S., Russell, J. D., Lachicotte, R. J. Synthesis and characterization of a zirconium complex containing a corannulene-based ligand. Organometallics, 21, 2027–2029 (2002). 11. Möhring, P. C., Coville, N. J. Group 4 metallocene polymerisation catalysts: Quantification of ring substituent steric effects. Coord. Chem. Rev., 250, 18–35 (2006). 12. Müller, T.E.; Beller, M. In Transition Metals for Organic Synthesis, Beller, M.; Bolm, C. (ed.), Wiley-VCH, Weinheim, 1998, pp. 316–330. 13. Schellenberg, J., Tomotsu, N. Syndiotactic polystyrene catalysts and polymerization. Prog. Polym. Sci., 27, 1925–1982 (2002). 14. Schneider, N., Prosenc, M. H., Brintzinger, H. H. Cyclopentadienyl[1] phenanthrene titanium trichloride derivatives: Syntheses, crystal structure and properties as catalysts for styrene polymerization. J. Organomet. Chem., 545–546, 291–295 (1997). 15. Xu, G., Cheng, D. Syndiospecific polymerization of styrene with half-sandwich titanocene catalysts. Influence of ligand pattern on polymerization behavior. Macromolecules, 33, 2825–2831 (2000). 16. Schellenberg, J. The syndiospecific polymerization of styrene in the presence of fluorine-containing half-sandwich metallocenes. J. Polym. Sci. Part A: Polym. Chem., 38, 2428–2439 (2000). 17. Kaminsky, W. Highly active metallocene catalysts for olefin polymerization. J. Chem. Soc. Dalton Trans., 1413–1418 (1998). 18. O’Connor, J. M., Casey, C. P. Ring-slippage chemistry of transition metal cyclopentadienyl and indenyl complexes. Chem. Rev., 87, 307–318 (1987). 19. Veiros, L. F. Substituent effects on haptotropic rearrangements of bis(indenyl) zirconium sandwich complexes. Organometallics, 25, 2266–2273 (2006). 20. Kirillov, E., Saillard, J. Y., Carpentier, J. F. Groups 2 and 3 metal complexes incorporating fluorenyl ligands. Coord. Chem. Rev., 249, 1221–1248 (2005).

122


21. Dahlmann, M., Fröhlich, R., Erker, G. Reaction of a zirconocene(butadiene) borate-betaine single-component Ziegler catalyst with trimethylphosphane. Eur. J. Inorg. Chem., 1789–1793 (2000). 22. Knjazhanski, S. Y., Moreno, G., Cadenas, G., Belsky, V. K., Bulychev, B. M. Preparation of bifluorenes via the synthesis and thermal decomposition of fluorenyltitanium(IV) trichlorides. Molecular and crystal structure of 9,9′-bis(trimethylsilyl)-bi-9,9′-fluorene. Tetrahedron, 55, 1639–1646 (1999). 23. Decken, A., MacKay, A. J., Brown, M. J., Bottomley, F. Synthesis, characterization, and thermochemistry of (η1-C13H9)Mn(CO)5 and (η5-C13H9)Mn(CO)3. Organometallics, 21, 2006–2009 (2002). 24. Knjazhanski, S. Y., Cadenas, G., Garcia, M., Perez, C. M., Nifant’ev, I. E., Kashulin, I. A., Ivchenko, P. V., Lyssenko, K. A. (Fluorenyl)titanium triisopropoxide and bis(fluorenyl)titanium diisopropoxide: A facile synthesis, molecular structure, and catalytic activity in styrene polymerization. Organometallics, 21, 3094–3099 (2002). 25. Schröder, K., Lützen, A., Haase, D., Saak, W., Beckhaus, R., Wichmann, S., Schellenberg, J. Tetrabenzo[a,c,g,i]fluorenyllithium and η5-tetrabenzo[a,c,g,i] fluorenyltitanium complexes. Organometallics, 25, 3824–3836 (2006). 26. Ramage, R., Wahl, F. O. 4-(17-Tetrabenzo[a,c,g,i]fluorenylmethyl)-4′,4″dimethoxytrityl chloride: A hydrophobic 5′-protecting group for the separation of synthetic oligonucleotides. Tetrahedron Lett., 34, 7133–7136 (1993). 27. Ramage, R., Raphy, G. Design of an affinity-based Nα-amino protecting group for peptide synthesis: Tetrabenzo[a,c,g,i]fluorenyl-17-methyl urethanes (Tbfmoc). Tetrahedron Lett., 33, 385–388 (1992). 28. Brown, A. R., Irving, S. L., Ramage, R., Raphy, G. (17-Tetrabenzo[a,c,g,i] fluorenyl)methylchloroformate TbmocCl) a reagent for the rapid and efficient purification of synthetic peptides and proteins. Tetrahedron, 51, 11815–11830 (1995). 29. Hay, A. M., Hobbs-Dewitt, S., MacDonald, A. A., Ramage, R. Use of tetrabenzo[a,c,g,i]fluorene as an anchor group for the solid/solution phase synthesis of the quinolone antibacterial, ciprofloxacin. Synthesis, 1979–1985 (1999). 30. Stien, D., Gastaldi, S. Design of polyaromatic hydrocarbon-supported tin reagents: A new family of tin reagents easily removable from reaction mixtures. J. Org. Chem., 69, 4464–4470 (2004). 31. Schröder, K., Haase, D., Saak, W., Beckhaus, R., Kretschmer, W. P., Lützen, A. Tetrabenzo[a,c,g,i]fluorenyltitanium(III) and -(IV) complexes: Syntheses, reactions and catalytic application. Organometallics, 27, 1859–1868 (2008). 32. Gambarotta, S., Floriani, C., Chiesi-Villa, A., Guastini, C. Cyclopentadienyldichlorotitanium(III): A free-radical-like reagent for reducing azo (N:N) multiple bonds in azo and diazo compounds. J. Am. Chem. Soc., 105, 7295–7301 (1983). 33. Cozak, D., Melnik, M. Titanium organometallic compounds: Analysis and classification of crystallografic data. Coord. Chem. Rev., 74, 53–99 (1986). 34. Engelhardt, L. M., Papasergio, R. I., Raston, C. L., White, A. H. Crystal structures of trichloro(Cp)Ti- and -zirconium(IV). Organometallics, 3, 18–20 (1984). 35. Ready, T. E., Day, R. O., Chien, J. C. W., Rausch, M. D. (η5-Indenyl) trichlorotitanium. An improved syndiotactic polymerization catalyst for styrene. Macromolecules, 26, 5822–5823 (1993).

REFERENCES

123

36. Nomura, K., Naga, N., Miki, M., Yanagi, K., Imai, A. Synthesis of various nonbridged titanium(IV) cyclopentadienyl-aryloxy complexes of the type CpTi(OAr)X2 and their use in the catalysis of alkene polymerization. Important roles of substituents on both aryloxy and cyclopentadienyl groups. Organometallics, 17, 2152–2154 (1998). 37. Daasbjerg, K., Svith, H., Grimme, S., Gerenkamp, M., Mück-Lichtenfeld, C., Gansäuer, A., Barchuk, A., Keller, F. Elucidation of the mechanism of titanocenemediated epoxide opening by a combined experimental and theoretical approach. Angew. Chem., 118, 2095–2098 (2006). 38. Forbes, G. C., Kennedy, A. R., Mulvey, R. E., Rodger, P. J. A. TEMPO: A novel chameleonic ligand for s-block metal amide chemistry. Chem. Commun., 1400– 1401 (2001). 39. Mahanthappa, M. K., Huang, K. W., Cole, A. P., Waymouth, R. M. Synthesis and molecular structure of titanium complexes containing a reduced TEMPO radical. Chem. Commun., 502–503 (2002). 40. Evans, W. J., Perotti, J. M., Doedens, R. J., Ziller, J. W. The tetramethylpiperidinyl -1-oxide anion (TMPO-) as a ligand in lanthanide chemistry: Synthesis of the per(TMPO-) complex [(ONC5H6Me4)2Sm(m-ONC5H6Me4)]2. Chem. Commun., 2326–2327 (2001). 41. Barbier-Baudry, D., Heiner, S., Kubicki, M. M., Vigier, E., Visseaux, M., Hafid, A. An easy synthetic route to heteroleptic samarium monoalkoxides for ringopening polymerization initiators. Molecular structures of [(C5HiPr4)SmI(THF)2]2, SmI2Ot-Bu(THF)4, and (C4Me4P)2SmOt-Bu(THF). Organometallics, 20, 4207– 4210 (2001). 42. Berg, D. J., Andersen, R. A., Zalkin, A. Electron-transfer chemistry of (Me5C5)2Yb: Cleavage of diorganoperoxide and related chalcogenides to give (Me5C5)2Yb(ER) (L) (E = O, S, Se, or Te; L = a Lewis base). Crystal structure of (Me5C5)2Yb(TePh) (NH3). Organometallics, 7, 1858–1863 (1988). 43. Janiak, C. A critical account on–stacking in metal complexes with aromatic nitrogen-containing ligands. J. Chem. Soc. Dalton Trans., 3885–3896 (2000). 44. Llinas, G. H., Mena, M., Palacios, F., Royo, P., Serrano, R. Cp*SiMe3 as a mild and effective reagent for transfer of the Cp* ring: An improved route to Cp*MX3 of the group 4 elements. J. Organomet. Chem., 340, 37–40 (1988). 45. Malaba, D., Tessier, C. A., Youngs, W. J. Metal-promoted oxidative π-conjugation and coupling of substituted-fulvalene-type ligands: Synthesis and crystal structures of π-conjugated C24H12(SiMe3)2 and the dimer [C24H13(SiMe3)2]2. Organometallics, 15, 2918–2922 (1996). 46. Kucht, A., Kucht, H., Barry, S., Chien, J. C. W., Rausch, M. D. New syndiospecific catalysis for styrene polymerisation. Organometallics, 12, 3075–3078 (1993). 47. Nomura, K., Naga, N., Miki, M., Yanagi, K. Olefin polymerization by (cyclopentadienyl)(aryloxy)titanium(IV) complexes–cocatalysts systems. Macromolecules, 31, 7588–7597 (1998). 48. Sturla, S. J., Buchwald, S. L. Monocyclopentadienyltitanium aryloxide complexes: Preparation, characterization, and application in cyclization reactions. Organometallics, 21, 739–748 (2002). 49. Bradley, D., Mehrotra, R. C., Rothwell, I. Alkoxo and Aryloxo Derivatives of Metals, Academic Press, London, 2001.

124


50. Nielson, A. J., Schwerdtfeger, P., Waters, J. M. Trichloro monophenoxide complexes of titanium(IV). J. Chem. Soc. Dalton Trans., 529–537 (2000). 51. Gupta, R., Singh, A., Mehrotra, R. C. Synthesis, characterization and reactions of some titanium aryloxides. Indian J. Chem., 29A, 596–598 (1990). 52. Malhotra, K. C., Sharma, N., Bhatt, S. S., Chaudhry, S. C. Preparation and characterization of monochlorotris(4-tert-butylphenoxo)titanium(IV). Polyhedron, 11, 2065–2068 (1992). 53. Schellenberg, J. Melting properties of syndiotactic polystyrenes and effect of hydrogen on molecular weight distribution. Macrom. Mat. Eng., 290, 675–680 (2005).

CHAPTER 7

Rare-Earth Metal Complexes as Catalysts for Syndiospecific Styrene Polymerization KLAUS BECKERLE and JUN OKUDA RWTH Aachen University, Aachen, Germany

7.1

INTRODUCTION

Syndiotactic polystyrene (SDS) is synthesized commonly by titanium complexes that furnish as putative active catalyst a cationic trivalent species of the general formula [Ti(L)R]+ A− (L = neutral or monoanionic ligand; R = alkyl, hydride; A = weakly coordinating ligand). Most commonly, titanium halfsandwich complexes of the type [Ti(η5-C5R5′)R3] are used as precursors leading to the active species [Ti(η5-C5R′5)R]+ after activation. Isostructural species containing a group 3 metal center (including lanthanides), [Ln(η5-C5R′5)R]+, remained unknown until the recent discovery of the scandium catalyst [Sc(η5C5Me4SiMe3)(CH2SiMe3)]+ [1]. Apart from the unpaired electron of the Ti(III) center and possibly of f-electrons in the case of the lanthanides, such group 3 metal complexes should be comparatively active in syndiospecific styrene polymerization, as the steric requirements for the stereoregulation can be regarded as similar. On the other hand, the larger size of the group 3 metal centers will clearly have consequences on the choice of a suitable ligand. Early reports on styrene polymerization catalyzed by group 3 compounds included ionic compounds such as [Eu(CH3CN)(BF4)3]x [2], metallocene complexes such as [(η5-tBuC5H4)2LnMe]2 (Ln = Pr, Nd, Gd) [3], and binary systems (e.g., Gd(O2CR)3/Al(iBu)3) [4], but were generally not very active and/or not stereoselective. This chapter summarizes the current knowledge on group 3 metal catalysts for syndiospecific polymerization. It is organized according to the nature of the ligand system used and should provide some systematic insight into this emerging class of polymerization catalysts.


125

126

SYNDIOSPECIFIC STYRENE POLYMERIZATION

7.2

METALLOCENE CATALYSTS

In contrast to the large number of publications on group 4-based catalysts for syndiospecific styrene polymerization, reports on analogous rare-earth metal catalysts are still limited [5]. Early investigations on neodymium-based sandwich complexes (1–3) activated by methylaluminoxane (MAO) (Fig. 7.1) have shown that these systems are promising candidates for styrene polymerization catalysis with the resulting polymers syndiotactically enriched. The selectivity as well as the activity of these systems are moderate [6]. Recently, Carpentier et al. published highly syndiospecific ansametallocene catalysts (4–13, Fig. 7.2) bearing a labile allyl ligand [7,8]. The active allyl complexes are accessible in two steps by salt elimination reactions. In the first step, the lithiated ligand is reacted with the trivalent metal chloride yielding monochloro complexes, dimeric chloro complexes, or ate complexes depending on the exact nature of the ligand. These compounds in turn give the target complexes upon addition of the allyl Grignard reagent C3H5MgCl. Crystal structure analysis revealed that the fluorenyl ring in these complexes is bonded to the metal in an η3 mode with an additional tetrahydrofuran (THF) molecule coordinated, forming a 16-electron species. It is interesting that this electron count is maintained by switching to η5 coordination in case of the closely related thf-free complexes 12 and 13 bearing an allyl ligand with bulky substituents [5,9]. In spite of the lack of a donor molecule, these compounds are not more efficient catalysts than the thf-containing species.

Nd Cl

Nd Cl

1

2

Nd Cl

3

Figure 7.1 Lanthanide-based sandwich complexes for generation of syndiotactically enriched polystyrene.

SiMe3 Ln

Ln

Si

Ln

O

O

Ln O SiMe3

4 5 6 7

Ln = Y Ln = La Ln = Nd Ln = Sm

Figure 7.2

8 Ln = Y 9 Ln = Nd

10 Ln = Y 11 Ln = Nd

12 Ln = Y 13 Ln = Nd

ansa-Metallocenes for syndiospecific styrene polymerization.

METALLOCENE CATALYSTS

127

(a) Generation of the active species

–THF

Nd O

+PhCHCH2

Nd

–PhCHCH2

+THF

Nd Ph

(b) Polymerization +n PhCHCH 2 Nd

Nd Ph

Ph n/2

Scheme 7.1 Proposed mechanism for styrene polymerization by ansa-lanthanoidocene catalysts.

Allyl complexes 4–11 behave as single-component catalysts for styrene under mild conditions. They show activity in bulk polymerization at temperatures as low as 20 °C. Although polymerization rates are considerably lower than for highly active titanium-based catalysts (activities up to 105 kg SPS/mol [Ti]h were reported) [10], the polymers obtained have a melting temperature of 260 °C, showing high stereoselectivity. Activity increases considerably at elevated temperature and reaches values above 1733kg SPS/mol[Ln] h for bulk polymerization at 120 °C. While the active species in titanium-based catalysts as well as in cationic group 3 complexes (see below) are isolobal cations with a general composition of [CpsubstM(III)R]+ (with Cpsubst representing a substituted cyclopentadienyl ligand and R standing for any alkyl chain), in an ansa-lanthanidocene system, polymerization is only possible if a metal– ligand bond is cleaved [8,11]. Since there is no experimental data supporting this unlikely mechanism, Carpentier and coworkers suggest dissociation of coordinated thf for the formation of the active species. Subsequently, a styrene molecule takes up the free space in the unsaturated coordination sphere of the rare-earth metal (Scheme 7.1). The variable temperature nuclear magnetic resonance (NMR) data, showing the flexibility of the allyl ligand, and the matrix-assisted laser desorption/ionization time-of-flight (MALDI-TOF) analysis, showing formation of polystyrene with an allyl end group, support this mechanism. The syndiospecificity of these catalyst systems even at temperatures above 100 °C is high. Polymers with rrrr pentad abundance of ≥99% have been reported. 13C NMR analysis of the stereoerrors in the polymer revealed the presence of rrrrrr, rrmrrr, rmrrrr, and mrrrrr heptads, typical for chain-end control during the polymerization [11]. Even for a less stereoregular polymer obtained at 120 °C polymerization temperature, the relative intensities of 0.74 for the major (rrrrrr) signal and 0.09 for each of the other heptads

128


were detected. The role of chain-end control is further emphasized by the syndiotacticity of polystyrenes prepared using C1 symmetric compounds of the same catalyst family and by the polymerization using neutral allyl complexes (vide infra). Although the lower symmetry can be expected to discriminate one of the reaction sites involved in syndiotactic polymerization and therefore to reduce selectivity, the polymers are found to contain over 99% rr triads. In addition to highly stereoselective homopolymerization, copolymerization of styrene with dienes and ethylene was achieved, opening up a route to interesting new materials that might offer a method to overcome the inherent problems of syndiotactic polystyrene as a bulk material (i.e., high brittleness, low solubility, problems in processing due to high melting temperature) [12]. While copolymerization catalysts for olefins had been reported before, the resulting materials either contained atactic polystyrene, too little styrene, or too short polystrene blocks to retain the desired rheological properties of SPS. Employing the neodymium-based catalyst 6, copolymers with ethylene that incorporate long SPS blocks and isolated ethylene units at polymer compositions of 40–99 mol % styrene are accessible. The melting temperature of the material is strongly decreased at ethylene incorporation as low as 2 mol % (Tm = 220 °C vs. 270 °C for the homopolymer). Similar observations were reported for copolymerization with isoprene. Again, the polymers contain isolated comonomer units or very short oligoisoprene sequences that strongly influence melting and glass transition temperatures. This type of catalysts also offers access to low-molecular-weight SPS by chain transfer to magnesium compounds. Starting from a dimeric neodymium complex, the addition of dibutyl magnesium leads to the formation of a bimetallic alkyl complex in equilibrium with the monomeric active species and free magnesium alkyl [13] (Scheme 7.2).

+(MgR 2)n

Cl Nd Cl

Nd

R Nd

–MgCl 2 +MgR 2

Scheme 7.2

+MgR 2

Ph R

Nd

R –MgR 2

–MgR 2 Ph

Mg R

n/2

+n PhCHCH2

Nd

R

Proposed mechanism for the preparation of syndiotactic oligostyrenes.

CONSTRAINED GEOMETRY CATALYSTS

129

R Ln R (THF)n 14 Ln = Y, R = H (n = 1) 15 Ln = Y, R = SiMe3 (n = 0) 16 Ln = Nd, R = H (n = 1) 17 Ln = Nd, R = SiMe3 (n = 0)

Figure 7.3

ansa-Metallocene complexes for isospecific styrene polymerization.

Activity is far lower under these conditions than for the polymerization with the analogous allyl compound, but soluble oligomers of syndiotactic polystyrene are formed. Molecular weights range between 1600 and 6500 g/mol at narrow to moderate molecular weight distributions (Mw/Mn = 1.3–2.5). Changing the symmetry of the complex to C2 (14–17, Fig. 7.3) results in the switch of the stereoselectivity of the polymerization from SPS to isotactic PS along with switch from chain end to catalyst site control [14].

7.3

CONSTRAINED GEOMETRY CATALYSTS

Group 3 compounds are isoelectronic and often isostructural to analogous group 4 cations with the same ligands and can therefore be expected to be active single-component catalysts. Investigations into the reactivity of linked amido-cyclopentadienyl (CpA)-based rare-earth metal compounds (“constrained geometry catalysts”), however, have shown that this picture is somewhat too simple. Neutral CpA complexes were reported to be neither active nor stereoselective in styrene polymerization. This is partly due to the Lewis acidity of the group 3 metals, which leads to dimerization and the coordination of Lewis bases such as thf. Dimeric hydride complexes of rare-earth metals bearing a CpA ligand (18–22) are, by comparison, inactive in styrene polymerization. After the insertion of a single α-olefin unit into the metal hydrogen bond, these complexes can be isolated as thf-free species. Without the coordinated base, the catalyst efficiency of these compounds is very low; on the other hand, an excess of thf deactivates the system. It has been shown that these compounds initiate polymerization of styrene (Scheme 7.3). The polymers show a syndiotactically enriched microstructure with around 70%– 80% r dyads and little dependence on polymerization temperature (varying from 0 to 75 °C). Concerning activity, an increase with the metal radius from lutetium to terbium was observed, with activities staying comparatively low. The yttrium complex, for example, converts roughly 120 equiv of styrene in 24 h [15].

130


CMe3

Me2Si

H

Ln N

H

THF N Ln

THF

Me3C 18 19 20 21 22

SiMe2

1.)

nBu

nBu

n 2.)

Ln = Y Ln = Tb Ln = Er Ln = Yb Ln = Lu

Scheme 7.3 Activation of Cp-amido yttrium hydride complexes for styrene polymerization.

7.4

HALF-SANDWICH CATALYSTS

Recently, Luo, Baldamus, and Hou published a series of half-sandwich complexes that are highly active and syndiospecific for styrene polymerization upon activation by a borate (Fig. 7.4) [16]. Interestingly, a closely related lanthanum complex (23) bearing an η5-C5Me5 ligand without the trimethylsilyl substituent (vide infra) was among the first rare-earth metal compounds reported to be active in styrene polymerization albeit with only moderate activity (80% conversion for [St]/[Ln] = 100 at 50 °C after 24 h), yielding predominantly atactic polymer [17]. In compounds 24–27 with a η5-C5Me4SiMe3 ligand, one of the two trimethylsilylmethyl ligands is abstracted in the activation step (Scheme 7.4). A single THF molecule remains coordinated, which does not inhibit the polymerization reaction. As in the case of CpA complexes, the neutral compounds are not active in styrene polymerization at room temperature. The addition of 1 equiv of [Ph3C][B(C6F5)3] leads to the formation of an active cationic complex. Among the metals used, scandium forms by far the most active complexes, showing activities up to 1.36 × 104 kg SPS/mol[Sc]h, which are considerably higher than those found for metallocene systems, but still not as high as for some titanium-based catalysts (vide supra). These catalysts also allow control of molecular weights by the monomer/ catalyst ratio. A nearly linear increase of molecular weight with the number of monomer equivalents is observed, while the molecular weight distribution remains narrow (Mw/Mn ≈ 1.43–1.5). The stereoregularity is high in all cases (rrrr > 99%), and no formation of atactic by-product is observed. Similar to the compounds published by Carpentier (vide supra), they are able to incorporate substantial amounts of comonomers, specifically dienes and ethylene. The compositions of these copolymers can be adjusted in the whole range from 0% to 100% styrene content for the copolymerization with ethylene [12b,18]. Copolymerization with isoprene can be achieved using the scandium complex 24. Statistical as well as block copolymers are accessible. The acces-

131


SiMe3 La Me3SiH2C

Ln

THF Me3SiH2C

CH2SiMe3

24 25 26 27

23

CH2SiMe3

Me3SiH2C Me3SiH2C

L

29 R = Me; L = THF 30 R = Ph; L = 1,4-dioxane

Ln THF

Me3SiH2C

37 Ln = Sc 38 Ln = Y 39 Ln = Lu

Cl THF

28

Me2 Si

Me2 Si

Sc

Me3SiH2C

Sc thf Cl

Ln = Sc Ln = Y Ln = Gd Ln = Lu

SiMe3

Me2 Si

THF

SiMe3

Ln THF

Me3SiH2C

Ln

Me3SiH2C

31 Ln = Sc 32 Ln = Y 33 Ln = Lu

34 Ln = Sc 35 Ln = Y 36 Ln = Lu

Me2 Si

F Me3SiH2C

Ln

Me3SiH2C

O THF

40 Ln = Sc 41 Ln = Y 42 Ln = Lu

N THF

P Ln Me3SiH2C

43 44 45 46

THF

CH2SiMe3 Ln = Sc Ln = Y Ln = Gd Ln = Lu

N N

Sc N

47 Ln = Sc 48 Ln = Y 49 Ln = La

Figure 7.4

Group 3 sandwich complexes for styrene polymerization.

132


SiMe3 Sc Me3SiH2C

THF

CH2SiMe3

n [Ph3C][B(C6F5)4] (Sc/B = 1:1) toluene, 25 °C

n/2

Scheme 7.4 Activation of scandium half-sandwich complexes.

sible structures are styrene/isoprene AB block copolymers as well as ABA and multiblock polymers with stereoregular polystyrene blocks. The isoprene segments are stereoirregular and contain both 1,4- and 3,4-inserted units [19]. A series of related scandium complexes with substituted Cp ligands 31–42 were tested in styrene polymerization [20]. Active catalysts of this type typically give highly stereospecific polystyrenes. In general, rare-earth metals other than scandium are significantly less active. These systems are extremely sensitive toward impurities, most probably due to the high Lewis acidity of the cationic metal centers. Scavengers such as MAO or aluminum alkyls can be added to avoid deactivation by impurities, but their use is limited by the tendency of group 3 as well as group 13 compounds to form dimers or clusters that effectively block the active centers. Similarly, the analogous dibenzyl complex 29 was shown to give syndiotactic polystyrene, but with significantly lower activity. The corresponding dichloro compound 28 is virtually inactive under activation conditions typical for the [Ti(η5-C5Me5)Cl3] catalyst system [21]. Further development of this family of compounds includes solvent-free lanthanide complexes of a mono(phospholyl) ligand (43–46) [22]. When prepared in Lewis basic solvents, these compounds incorporate one or more equivalents of thf or pyridine. Solvent coordination as well as formation of dimers can be avoided by introducing an additional donor function in the alkyl ligands (specifically by using the o-N,N-dimethylaminobenzyl ligand), allowing for a protected coordination sphere in the monomeric species. The scandium complex shows high polymerization activity, much like the original catalyst precursor 24 with slightly broader molecular weight distribution. For the larger metals yttrium and samarium, lower activities are observed again. The most recent addition to this family of catalysts includes substituted pyrrolyl ligands [23]. Depending on the substituents, the corresponding bis(oN,N-dimethylaminobenzyl) complexes show different coordination behaviors. While a di-tert-butyl-substituted pyrrolyl is η5 coordinated like a Cp ring, the tetramethyl derivative adopts an η1-bonding mode. This change in bonding mode has drastic consequences for polymerization. The η5-coordinated pyrrolyl complex 47 is active in syndiospecific styrene polymerization. With activity at 3300kg SPS/mol[Sc]h, it is not as active as 24. No activity at all was


133

N N

Sc

[B(C6F5)4]

50

Figure 7.5

Cationic complex used as a model for the active species.

observed for the compound with an η1 bond under comparable conditions. These investigations give a clearer picture of the active species. The bonding mode of the ring is not fluxional, as the two structurally characterized complexes give distinct polymerization results. On the other hand, an analogous scandium complex bearing a tetramethyl Cp ligand is a suitable precatalyst for styrene polymerization, ruling out the ring substitution pattern (tetramethyl vs. di-tert-butyl) as the cause for the different polymerization behavior. Further experimental proof for the cationic active species with an η5coordinated ring and an alkyl ligand was given by employing the isolated and structurally characterized cationic complex 50 (with the borate anion used in the activation process, Fig. 7.5). Virtually the same polymerization behavior is found when activation of the neutral complex is performed. The addition of 1 equiv of 1,2-dimethoxyethane (DME) to that complex results in a welldefined compound but renders the cation inactive as the DME coordinates to the scandium and efficiently blocks the active center. Apart from the half-sandwich alkyl complexes, a neodymium-based borohydride of the composition (η5-C5Me5)Nd(BH4)2(THF)2 was employed in styrene polymerization in combination with n-butylethylmagnesium as a chain transfer agent [24]. The resulting polymers are not highly syndiotactic (with a syndioselectivity of ca. 85% reported), but the molecular weight distributions are very narrow (around Mw/Mn = 1.2). Since experimental data show the growth of two polymer chains per magnesium, it might represent a step toward the chain shuttling polymerization of SPS [25]. In addition to trivalent group 3 metal-based catalysts for styrene polymerization, ytterbium(II) complexes were used as catalysts [26]. The halfsandwich complex 51 (Fig. 7.6) gives a syndiotactically enriched polymer (r = 82%; rr = 67%) with significantly lower syndiotacticity than the analogous calcium complex (r = 91.2%; rr = 83.1%). This effect can be explained by a faster inversion of the chain end on the ytterbium center as compared to calcium leading to the formation of stereoerrors (Scheme 7.5). This inversion in competition with syndiospecific insertion of styrene determines the overall microstructure.

134


Me2N

H

* SiMe3

Yb

O

Me3Si

51

Figure 7.6 Ytterbium(II) half-sandwich complex.

Ph

Ph

Ph H

Ph Yb

*

Ph

+

Ph

Ph

Ph

Ph

*

Insertion

Yb

Inversion

Ph

Ph

Ph Ph

*

Yb

Scheme 7.5

7.5

Ph

+

H

Ph

Ph

Ph

Insertion

Ph

*

Yb

Inversion of the asymmetric chain end [26].

NONMETALLOCENE CATALYSTS

Only very few rare-earth-based nonmetallocene compounds were used as styrene polymerization catalysts (Fig. 7.7). Harder investigated the polymerization behavior of homoleptic ytterbium(II) complexes, in comparison to isostructural calcium compounds [26]. The use of homoleptic ytterbium(II) complex 52 gives polymers of significantly higher tacticity (r = 94.9%; rr = 90.0% at −20 °C), contradicting the assumption that a spectator ligand is necessary for stereoselective homogeneous single-site catalysts. The stereoerrors identified in the product (i.e., rrmrrr, rmrrrr, and mrrrrr heptads) are the ones expected for a chain-end control mechanism, as it was observed for the rare-earth metal allyl complexes. The analogous homoleptic complex containing a larger samarium(II) shows a surprisingly different polymerization behavior, although the crystal structures of the two compounds are similar. Experimental data give some insight into a unique polymerization mechanism involving the transfer of two electrons from two samarium centers to a styrene molecule, resulting in a binuclear species closely related to a binuclear samarium styrene complex that was characterized by X-ray crystallography [27]. Another notable samarium(II)-based catalyst system is Sm(OAr)2(THF)3 (53). While it is inactive as a catalyst for styrene polymerization at 0.1 MPa,

NONMETALLOCENE CATALYSTS

Me2N O Me3Si

H

* SiMe3

Yb

O

* H

NMe2

135

O THF

Sm

THF THF

O

52 53

(Me3Si)2N

N(SiMe3)2

NH NH [Li(dioxane)]

Ln

HN HN

H Ln

Ln H

NH NH

HN HN

(Me3Si)2N

N(SiMe3)2

54 Ln = Sm 55 Ln = Nd 56 Ln = Yb 57 Ln = Lu Figure 7.7 Nonmetallocene complexes used in the syndiotactic polymerization of styrene.

activity increases at elevated pressure, yielding a basically atactic polymer at 100 MPa. Interestingly, an increase in stereoselectivity up to ca. 60% r dyads occurs at pressures exceeding 500 MPa, suggesting a steric effect in the formation of syndiotactic sequences [28]. Neutral and anionic samarium and neodymium species 54 and 55 with allyl ligands were reported to give syndiotactic-rich polystyrene [29]. Although the presence of lithium allyl in equilibrium suggests an anionic polymerization mechanism, the formation of a syndiotactic-rich product indicates the participation of rare-earth metals in the polymerization. Recently, hydride complexes supported by guadinate ligands 56 and 57 were found to be active in styrene polymerization, but only for the smaller lanthanides [30,31]. Both the lutetium and the ytterbium complexes show low conversion rates (the Lu compound converts 100 equiv within 3 days); the products are highly syndiotactic, and the polymer produced by the ytterbium compound was reported to have a high melting temperature of 289–293 °C (Mn = 90,000 g/mol, Mw/Mn = 2.6).

136

7.6


CONCLUSION

The introduction of group 3 metal complexes as catalyst precursors for syndiospecific styrene polymerization has resulted in the following new features: 1. Cationic group 3 metal catalysts of the type [Ln(η5-C5Me4R′)R]+, which are isostructural with the trivalent titanium system [Ti(η5-C5Me4R′)R]+, clearly indicate that the structure of the active species as well as the stereoregulating mechanism are related, supporting the original proposal by Zambelli et al. [32]. 2. Group 3 metal catalysts tend to be more suitable for the efficient copolymerization of styrene with, for example, ethylene, offering the possibility of resolving the practical problem of brittleness of SPS and the high melting temperature of pure SPS. 3. A variety of new ligand architectures are conceivable for group 3 metal centers, expanding the potential to control the polymerization behavior broadly, coupled with the choice of the suitable metal size (radius for trivalent ions of coordination number 6), ranging from the smallest scandium (0.885 Å) to the largest lanthanum (1.172 Å). With the recent advent of truly homogeneous isospecific styrene polymerization catalysts [14,33], new perspectives for the stereoselective polymerization of styrene, such as stereoblock polystyrene, appear to be within reach.

REFERENCES 1. (a) Hou, Z., Wakatsuki, Y. Recent developments in organolanthanide polymerization catalysts. Coord. Chem. Rev., 231, 1–22 (2002). (b) Yasuda, H. Organo rare earth metal catalysis for the living polymerizations of polar and nonpolar monomers. Top. Organomet. Chem., 2, 255–283 (1999). (c) Arndt, S., Okuda, J. Mono(cyclopentadienyl) complexes of the rare-earth metals. Chem. Rev., 102, 1953–1976 (2002). 2. Thomas, R. R., Chebolu, V., Sen, A. Chemistry of weakly solvated lanthanidemetal cations. Synthesis, characterization, and catalytic chemistry of [Eu(CH3CN)3(BF4)3]x. J. Am. Chem. Soc., 108, 4096–4103 (1986). 3. Cheng, Y. X., Shen, Q. Styrene polymerization by organolanthanide complexes [(ButCp)2LnCH3]2 (Ln = praseodymium, neodymium, gadolinium). Chin. Chem. Lett., 4, 743–744 (1993). 4. Kobayashi, E., Kaita, S., Aoshima, S., Furukawa, J. Polymerization of butadiene and styrene with gadolinium tricarboxylate catalyst: Effect of ligand on the catalytic activity for homo- and copolymerizations. J. Polym. Sci. Part A: Polym. Chem., 32, 1195–1198 (1994).

REFERENCES

137

5. Rodrigues, A.-S., Kirillov, E., Carpentier, J.-F. Group 3 and 4 single-site catalysts for stereospecific polymerization of styrene. Coord. Chem. Rev., 252, 2115–2136 (2008). 6. Cui, L., Ba, X., Teng, H., Ying, L., Li, K., Jin, Y. Preliminary investigations on polymerization catalysts composed of lanthanocene and methylaluminoxane. Polym. Bull., 40, 729–734 (1998). 7. Kirillov, E., Lehmann, C. W., Razavi, A., Carpentier, J.-F. Highly syndiospecific polymerization of styrene catalyzed by allyl lanthanide complexes. J. Am. Chem. Soc., 126, 12240–12241 (2004). 8. Rodrigues, A.-S., Kirillov, E., Lehmann, C. W., Roisnel, T., Vuillemin, B., Razavi, A., Carpentier, J.-F. Allyl ansa-lanthanidocenes: Single-component, single-site catalysts for controlled syndiospecific styrene and styrene-ethylene (co)polymerization. Chem. Eur. J., 13, 5548–5565 (2007). 9. For the coordination chemistry of fluorenyl ligands in rare-earth metal complexes, see Kirillov, E., Saillard, J.-Y., Carpentier, J.-F. Group 2 and 3 metal complexes incorporating fluorenyl ligands. Coord. Chem. Rev., 249, 1221–1248 (2005). 10. Schellenberg, J., Tomotsu, N. Syndiotactic polystyrene catalysts and polymerization. Prog. Polym. Sci., 27, 1925–1985 (2002). 11. Feil, F., Harder, S. New stereochemical assignments of 13C NMR signals for predominantly syndiotactic polystyrene. Macromolecules, 36, 3446–3448 (2003). 12. (a) Rodrigues, A.-S., Kirillov, E., Vuillemin, B., Razavi, A., Carpentier, J.-F. Stereocontrolled styrene-isoprene copolymerization and styrene-ethylene-isoprene terpolymerization with a single-component allyl ansa-neodymocene catalyst. Polymer, 49, 2039–2045 (2008). (b) Rodrigues, A.-S., Carpentier, J.-F. Group 3 and 4 single-site catalysts for styrene-ethylene styrene-α-olefin copolymerization. Coord. Chem. Rev., 252, 2137–2154 (2008). 13. Rodrigues, A.-S., Kirillov, E., Vuillemin, B., Razavi, A., Carpentier, J.-F. Binary ansa-lanthanidocenes/dialkylmagnesium systems versus single-component catalyst: Controlled synthesis of end-capped syndiotactic oligostyrenes. J. Mol. Cat., 273, 87–91 (2007). 14. Rodrigues, A.-S., Kirillov, E., Roisnel, T., Razavi, A., Vuillemin, B., Carpentier, J.-F. Highly isospecific styrene polymerization catalyzed by single-component bridged bis(indenyl)allyl yttrium and neodymium complexes. Angew. Chem. Int. Ed., 46, 7240–7243 (2007). 15. (a) Hultzsch, K.C., Spaniol, T. P., Okuda, J. Half-sandwich alkyl and hydride complexes of yttrium: Convenient synthesis and polymerization catalysis of polar monomers. Angew. Chem. Int. Ed., 38, 227–229 (1999). (b) Hultzsch, K.C., Voth, P., Beckerle, K., Spaniol, T. P., Okuda, J. Single-component polymerization catalysts for ethylene and styrene: Synthesis, characterization, and reactivity of alkyl and hydrido yttrium complexes containing a linked amido-cyclopentadienyl ligand. Organometallics, 19, 228–243 (2000). (c) Okuda, J., Arndt, S., Beckerle, K., Hultzsch, K. C., Voth, P., Spaniol, T. P. Rare earth metal-based catalysts for the polymerization of nonpolar and polar monomers. Pure Appl. Chem., 73, 351–354 (2001). (d) Voth, P., Arndt, S., Spaniol, T. P., Okuda, J., Ackerman, L. J., Green, M. L. H. Dimeric n-alkyl complexes of rare-earth metals supported by a linked amido-cyclopentadienyl ligand: Evidence for β-agostic bonding in bridging n-alkyl ligands and its role in styrene polymerization. Organometallics, 22, 65–76 (2003).

138

16.

17.

18.

19. 20.

21.

22.

23.

24.

25.

26. 27.


(e) Okuda, J. Rare earth metal complexes that contain linked amido-cyclopentadienyl ligands: ansa-Metallocene mimics and “constrained geometry” catalysts. Dalton Trans., 12, 2367–2378 (2003). (f) Beckerle, K. Seltenerdkomplexe als Einkomponentenkatalysatoren für die Polymerisation polarer und unpolarer Monomere. Diss., Mainz, 2003. Luo, Y., Baldamus, J., Hou, Z. Scandium half-metallocene-catalyzed syndiospecific styrene polymerization and styrene-ethylene copolymerization: Unprecedented incorporation of syndiotactic styrene-styrene sequences in styrene-ethylene copolymers. J. Am. Chem. Soc., 126, 13910–13911 (2004). Tanaka, K., Furo, M., Ihara, I., Yasuda, H. Unique dual function of La(C5Me5) [CH(SiMe3)2](THF) for polymerizations of both nonpolar and polar monomers. J. Polym. Sci. Part A: Polym. Chem., 39, 1382–1390 (2001). Hou, Z., Luo, Y., Li, X. Cationic rare earth metal alkyls as novel catalysts for olefin polymerization and copolymerization. J. Organometallic Chem., 691, 3114–3121 (2006). Zhang, H., Luo, Y., Hou, Z. Scandium-catalyzed syndiospecific copolymerization of styrene with isoprene. Macromolecules, 41, 1064–1066 (2008). (a) Hitzbleck, J., Beckerle, K., Okuda, J., Halbach, T., Mülhaupt, R. Syndiospecific polymerization catalysts for styrene based on rare earth metal half-sandwich complexes. Macromol. Symp., 236, 23–29 (2006). (b) Hitzbleck, J., Okuda, J. Synthesis, characterization, and polymerization activity of the scandium half-sandwich complex [Sc(η5-C5Me4{SiMe2(C6F5)})(CH2SiMe3)2(THF)]. Z. Anorg. Allg. Chem., 632, 1947–1949 (2006). (c) Hitzbleck, J., Okuda, J. Ring-opening of a furyl group appended to the cyclopentadienyl ligand in rare-earth metal half-sandwich complexes. Organometallics, 26, 3227–3235 (2007). Hitzbleck, J., Beckerle, K., Okuda, J. Half-sandwich dibenzyl complexes of scandium: Synthesis, structure, and styrene polymerization activity. J. Organomet. Chem., 692, 4702–4707 (2007). Jaroschik, F., Shima, T., Li, X., Mori, K., Ricard, L., Le Goff, X.-F., Nief, F., Hou, Z. Synthesis, characterization, and reactivity of mono(phosphyl)lanthanoid(III) bis(dimethylaminobenzyl) complexes. Organometallics, 26, 5654–5660 (2007). Nishiura, M., Mashiko, T., Hou, Z. Synthesis and styrene polymerization catalysis of η5- and η1-pyrrolyl-ligated cationic rare earth metal aminobenzyl complexes. Chem. Commun., 17, 2019–2021 (2008). Zinck, P., Valente, A., Mortreux, A., Visseaux, M. In situ generated halflanthanidocene based catalysts for the controlled oligomerisation of styrene: Selectivity, block copolymerisation and chain transfer. Polymer, 48, 4609–4614 (2007). For chain shuttling and chain transfer reactions, see (a) Arriola, D. J., Carnahan, E. M., Hustad, P. D., Kuhlman, R. L., Wenzel, T. T. Catalytic production of olefin block copolymers via chain shuttling polymerization. Science, 312, 714–719 (2006). (b) Kempe, R. How to polymerize ethylene in a highly controlled fashion? Chem. Eur. J., 13, 2764–2773 (2007). Harder, S. The chemistry of CaII and YbII: Astoundingly similar but not equal! Angew. Chem. Int. Ed., 20, 2714–2718 (2004). Evans, W. J., Ulibarri, T. A., Ziller, J. W. Reactivity of (C5Me5)2Sm with arylsubstituted alkenes: Synthesis and structure of a bimetallic styrene complex that contains an η2-arene lanthanide interaction. J. Am. Chem. Soc., 112, 219–223 (1990).

REFERENCES

139

28. (a) Zhang, Y., Hou, Z., Wakatsuki, Y. Polymerization of styrene by divalent organolanthanide catalysts under high pressure. Macromolecules, 32, 939–941 (1999). (b) For synthesis and structure of the complex, see Hou, Z., Fujita, A., Yoshimura, T., Jesorka, A., Zhang, Y., Yamazaki, H., Wakatsuki, Y. Heteroleptic lanthanide complexes with aryloxide ligands. Synthesis and structural characterization of divalent and trivalent samarium aryloxide/halide and aryloxide/cyclopentadienide complexes. Inorg. Chem., 35, 7190–7195 (1996). 29. Baudry-Barbier, D., Camus, E., Dormond, A., Visseaux, M. Homogeneous organolanthanide catalysts for the selective polymerization of styrene without aluminium cocatalysts. Appl. Organomet. Chem., 13, 813–817 (1999). 30. Trifonov, A. A., Skvortsov, G. G., Lyubov, D. M., Skorodumova, N. A., Fukin, G. K., Baranov, E. V., Glushakova, V. N. Postmetallocene lanthanide-hydride chemistry: A new family of complexes [{Ln{(Me3Si)2NC(NiPr)2}2(μ-H)}2] (Ln = Y, Nd, Sm, Gd, Yb) supported by guanidinate ligands—Synthesis, structure, and catalytic activity in olefin polymerization. Chem. Eur. J., 12, 5320–5327 (2006). 31. For bimetallic guadinato complexes generating atactic polystyrene, see Luo, Y., Yao, Y., Shen, Q. [(SiMe3)2NC(NiPr)2]2Ln(μ-Me)2Li(TMEDA) (Ln = Nd, Yb) as effective single-component initiators for styrene polymerization. Macromolecules, 35, 8670–8671 (2002). 32. (a) Zambelli, A., Oliva, L., Pellechia, C., Cinquina, P. Preliminary kinetic investigation on syndiotactic polymerization of styrene. Macromolecules, 22, 1642–1645 (1989). (b) Zambelli, A., Pellecchia, C., Oliva, L., Longo, P., Grassi, A. Catalysts for syndiotactic-specific polymerization of styrene: A tentative interpretation of some experimental data. Makromol. Chem., 192, 223–231 (1991). 33. Capacchione, C., Proto, A., Ebeling, H., Mülhaupt, R., Möller, K., Spaniol, T. P., Okuda, J. Ancillary ligand effect on single-site styrene polymerization: Isospecificity of group 4 metal bis(phenolate) catalysts. J. Am. Chem. Soc., 125, 4964–4965 (2003).

CHAPTER 8

Syndiospecific Styrene Polymerization with Heterogenized Transition Metal Catalysts KYU YONG CHOI Department of Chemical and Biomolecular Engineering, University of Maryland, College Park, MD, USA

8.1

INTRODUCTION

Syndiotactic polystyrene (SPS) can be readily polymerized using homogeneous or heterogeneous metallocene catalysts, based on group 4 metal compounds, especially titanium compounds like TiCl4, CpTiCl3, and Cp*Ti(OCH3)3 with methyl aluminoxane (MAO) as cocatalyst [1–3]. The recent developments of transition metal catalysts and reaction mechanisms are discussed in earlier chapters. This chapter will be focused on the quantitative aspects of SPS polymerization kinetics and related physical and chemical phenomena. SPS does not dissolve in its own monomer (styrene) or common hydrocarbon solvents; SPS precipitates from the liquid phase as soon as polymerization begins. Therefore, the polymerization of styrene is heterogeneous in nature. In addition, as the amount of the solid phase increases with monomer conversion, the reaction mixture develops into a thermoreversible gel that is not a covalently cross-linked gel but a physical gel formed by the strong intermolecular interactions between the syndiotactic polymer and monomer/solvent molecules. Once SPS gel is formed, mechanically agitating the reaction mixture becomes impossible by conventional means. Even with a homogeneous (or soluble) catalyst, the catalytically active sites are buried or embedded in the matrix of precipitated solid polymer as soon as the reaction begins, and the polymerization proceeds heterogeneously. The fundamental question of how the polymerization kinetics and polymer properties are affected by the heterogeneity of the polymerization system can be asked.


140

KINETICS OF SYNDIOSPECIFIC POLYMERIZATION

141

For the study of intrinsic kinetics of catalytic polymerization of styrene, one can use a very dilute solution to avoid the physical effects on the reaction kinetics caused by the global gelation and potential monomer mass transfer resistance in the solid phase. However, for commercial production of SPS, a reasonably high total solid content (e.g., >30–40 wt%) should be obtainable without operational difficulties of a polymerization reactor. From this aspect, there is a question of whether the polymerization kinetics for the extremely dilute polymerization system can be directly applicable to the reaction system with high solid content. In this chapter, the analysis of the heterogeneous styrene polymerization kinetics as well as the morphological changes of the SPS precipitates or particles during the polymerization will be presented.

8.2 KINETICS OF SYNDIOSPECIFIC POLYMERIZATION WITH HETEROGENEOUS METALLOCENE CATALYSTS For the development of industrial SPS synthesis process, a quantitative understanding of the polymerization kinetics is crucial. One needs to develop an understanding of how the catalyst and reaction conditions affect the rate of polymerization and the properties of polymers. Although the syndiotacticity of the polymer is primarily determined by the nature of the catalyst being employed, other important polymer properties (e.g., molecular weight, molecular weight distribution, and particle morphology) are strongly influenced by the reaction conditions for a given catalyst system. Like in α-olefin (e.g., ethylene and propylene) polymerization over transition metal catalysts (e.g., Ziegler-Natta catalysts and metallocene catalysts), the specific kinetics of polymerization are dependent on the nature of the specific catalyst being employed for the polymerization. However, many of these polymerization catalysts exhibit similar overall kinetic behaviors, and a standard kinetic modeling framework, if available, can be easily modified or tailored for a given catalyst system. The kinetic model helps better understand the polymerization behavior of a given catalyst but it can also be used to design and optimize the polymerization process systems for industrial production. In what follows, it shall be illustrated how a quantitative kinetic model for heterogeneous polymerization of styrene can be developed.

8.2.1

Kinetic Profiles of Heterogeneous SPS Polymerization

In SPS polymerization over a homogeneous metallocene catalyst such as Cp*Ti(OCH3)3/MAO, precipitated polymer microparticles agglomerate and become a wet-cake-like gel material [4]. The entire gel phase is pseudohomogeneous and since styrene monomer is uniformly distributed in the gel phase where active catalytic sites are present, polymerization can be driven to

142

HETEROGENIZED TRANSITION METAL CATALYSTS

Yield (g)

Cp*Ti(OMe) 3/MAO

Second injection of styrene

First injection of styrene

Time (min)

Figure 8.1 SPS yield profiles with sequential injections of styrene into a reaction vial at 70 °C with Cp*Ti(OCH3)3/MAO catalyst.

a very high conversion. In styrene polymerization with n-heptane as a diluent, for example, it has been observed that the overall polymerization kinetics do not follow the simple first-order kinetics with respect to monomer concentration [4]. Also, the polymerization rate is very high at the beginning of polymerization but it rapidly decays with reaction time as monomer concentration decreases and the catalyst deactivation continues. However, the rapid rate decay does not seem to be simply caused by the catalyst site deactivation because, if additional styrene is added to a reaction mixture, the polymerization rate is quickly recovered, indicating that catalytic sites are still quite active. Figure 8.1 shows the SPS yield profile with homogeneous Cp*Ti(OMe)3/ MAO catalyst system for a 10-ml initial reaction volume at 70 °C in n-heptane in a small glass vial reactor (20 ml volume). The polymer yield rapidly increases initially and levels off at about 96% conversion and the resulting polymer becomes a wet-cake-like material. Then, as 1 ml of styrene is injected into the reactor at 60 min, the polymerization activity is quickly recovered. At 120 min, another 1 ml of styrene is injected and again, the yield increases. It is also interesting to observe in Figure 8.1 that initial polymerization rate is larger than the polymerization rate after the first injection of styrene. After the second styrene injection, the polymerization rate becomes smaller. The rate-declining effect is probably caused by the intrinsic catalytic site deactivation. The experimental results shown in Figure 8.1 indicate that the kinetics of SPS polymerization, particularly the rate decay phenomena, are also affected by the presence of the solid phase. The polymerization rate is directly dependent on the monomer concentration at the catalyst site, but since the catalyst sites are embedded in the solid phase (gel phase), it is likely that the actual monomer concentration at the catalyst sites


143

([M]s) might be lower than the monomer concentration in the bulk interstices of the gel phase ([M]b) (i.e., [M]s = f([M]b)). 8.2.2 Liquid Slurry Polymerization with Heterogenized Cp*Ti(OCH3)3 Catalyst The homogeneous metallocene catalyst can be heterogeneized by supporting the catalytically active component onto a solid phase. For example, the catalyst can be embedded into an SPS phase in a two-stage polymerization: styrene is polymerized with a homogeneous catalyst to a very low conversion to allow the catalyst to be embedded into the SPS matrix. Then, the polymer particles containing the active catalyst site become the precursor for the subsequent main polymerization [5]. The embedded catalyst can be recovered as solid particles or it can be used directly as a wet slurry. Figure 8.2a shows the scanning electron microscopy-wavelength dispersive spectroscopy (SEM-WDS) image of aluminum distribution map of the embedded catalyst particles. Here, red-white zones represent the region of high-aluminum concentration in MAO (bottom, left) and green-blue regions (gray) represent the thin polymer matrix surrounding the catalyst sites. The image indicates that a particle contains one or two highly concentrated regions of titanium catalyst complexed with MAO. Figure 8.2b is a close-up SEM image of the resulting SPS particle. It is interesting to see that the surface of the polymer particle is covered with fibrils of about 50 nm. The surface texture of the SPS particle shown in Figure 8.2b is drastically different from that of α-olefin polymers synthesized over transition metal catalysts. Cp*Ti(OCH3)3 catalyst can also be supported onto silica particles and activated by MAO for slurry-phase polymerization of styrene [6,7]. Figure 8.3 illustrates the polymerization rate profiles with time and the initial polymerization rates at four different monomer concentrations at 70 °C.

(a)

(b)

Figure 8.2 SEM-WDS element mapping of embedded Cp*Ti(OCH3)3/MAO catalyst particles. (See color insert.)

144


4

4

Rp [mol/l hr]

3

Rp0 [mol/l hr]

[M]b0 = 0.81 mol/l [M]b0 = 2.03 mol/l [M]b0 = 3.24 mol/l [M]b0 = 4.86 mol/l

2 1 0 0

70oC

3 2 1

20

40

60 80 Time (min)

100

120

(a)

0

0

1

2 3 [M]b0 [mol/l]

4

5

(b)

Figure 8.3 Effect of monomer concentration on polymerization rate–time profiles and initial polymerization rates (Rp0) with silica-supported Cp*Ti(OCH3)3/MAO catalyst.

Figure 8.3b shows that the initial polymerization rate (Rp0) is not linearly dependent on the monomer concentration (here, the initial polymerization rates were estimated by extrapolating the polymerization rate data to t = 0). The initial rate levels off for the initial monomer concentration larger than about 2.0 mol/l, possibly because of a rapid buildup of polymer around the silica-supported catalyst particles. The initial polymerization rate profile can be fitted experimentally using the following Langmuir-type equation where it is assumed that the monomer sorption equilibrium is established between the solid phase and the liquid phase:

[ M ]s =

K1 [ M ]b 1 + K 2 [ M ]b

(8.1)

where K1 and K2 represent the parameters to be determined experimentally. Then, we can express the overall polymerization rate as follows: Rp = k p [ M ]s [C *] =

k p K1 [ M ]b k p′ [ M ]b [C *] ≡ [C *] 1 + K 2 [ M ]b 1 + K 2 [ M ]b

(8.2)

where [M]s is the monomer concentration at the catalytic site, and [C*] is the active catalyst concentration. In the above equation, k p′ is the effective propagation rate constant (= kpK1). Notice that at very low bulk phase monomer concentration or for very small K2 value, the polymerization rate becomes linearly dependent on the bulk phase monomer concentration ([M]b). Such situation can be realized when one performs the styrene polymerization experiments at very low monomer concentrations. The above equation was tested for Cp*Ti(OCH3)3/MAO catalyst and the estimated rate constants are k p′ = 8.15 × 10 3 l mol hr, K2 = 0.47 l/mol at 70 °C [6].


8.2.3

145

Modeling of Polymerization Kinetics

A polymerization kinetic model is a very useful tool to quantify the polymerization rate phenomena and to predict the resulting polymer properties. To develop a kinetic model for SPS polymerization, the following kinetic scheme that has been well accepted for many metallocene catalyzed styrene polymerization systems [6,8,9] is employed. Catalyst site activation: a C0* + MAO ⎯k⎯ →C*

(8.3)

Propagation: k

p → P1 C * + M ⎯⎯

k

p → Pn+1 Pn + M ⎯⎯

(8.4)

Chain transfer to monomer: tM Pn + M ⎯k⎯ ⎯ → Mn + P1

(8.5)

β-hydrogen elimination: k

tβ Pn ⎯⎯ → Mn + C *

(8.6)

Catalyst site deactivation: d → D* C * ⎯k⎯ d → Mn + D* Pn ⎯k⎯

(8.7)

where C0* is the potent catalyst site; C* is the activated catalyst site; Pn and Mn are the live and dead polymer chains of length n; M is the monomer; and D* is the deactivated catalyst site. kj represents the reaction rate constant for each corresponding reaction. It is assumed that catalyst activation reaction (Eq. 8.3 is very fast. In the above kinetic scheme, the chain transfer to alkyls is ignored and it is assumed that β-hydrogen elimination and monomer chain transfer are the main chain transfer reactions. The basic mass balance equations for various species in the polymerization system are represented by the following equations. d [C *] = −kd [C *] − k p [C *][ M ]s + ktβ λ P 0 dt

(8.8)

d [ M ]s = −k p [ P ][ M ]s − ktM [ P ][ M ] ≈ −k p [ P ][ M ]s dt

(8.9)

146


d [ P1 ] = k p [C *][ M ]s − k p [ P1 ][ M ]s − ktM [ P1 ][ M ]s + ktM λ P 0 [ M ]s − ktβ [ P1 ] − kd [ P1 ] dt (8.10) d [ Pn ] = k p ([ Pn−1 ] − [ Pn ])[ M ]s − ktM [ Pn ][M ]s − ktβ [ Pn ] − kd [ Pn ] ( n ≥ 2 ) dt d [ Mn ] = kd [ Pn ] + ktβ [ Pn ] + ktM [ Pn ][ M ]s ( n ≥ 2 ) dt

(8.11) (8.12)

To calculate the molecular weight averages, the polymer molecular moment equations can be derived with the k-th molecular weight moments of live and ∞

∞

n =1

n=1

dead polymers defined as λ Pk ≡ ∑ nk[ Pn ] and λ Mk ≡ ∑ nk[ Mn ], respectively. The first three leading molecular weight moment equations are derived as follows. Zero-order moments: dλ P 0 = k p [C *][ M ]s − ktβ λ P 0 − kd λ P 0 dt

(8.13)

dλ M 0 = ktβ λ P 0 + kd λ P 0 + ktM λ P 0 [ M ]s dt

(8.14)

First-order moments: dλ P 1 = k p [C *][ M ]s + k pλ P 0 [ M ]s + ktM [ M ]s( λ P 0 − λ P 1 ) − ktβ λ P 1 − kd λ P 1 dt dλ M 1 = ktβ λ P 1 + ktM λ P 1 [ M ]s + kd λ P 1 dt

(8.15) (8.16)

Second-order moments: dλ P 2 = k p [C *][ M ]s + k p [ M ]s( 2λ P 1 + λ P 0 ) − ktβ λ P 2 + ktM [ M ]s( λ P 0 − λ P 2 ) − kd λ P 2 dt (8.17) dλ M 2 = ktβ λ P 2 + ktM λ P 2 [ M ]s + kd λ P 2 dt

(8.18)

[P] is the total live polymer concentration. Number-average and weight-average molecular weights are calculated using the following equations:


147

Mn =

λ P1 + λM1 λ ( mw )sty ≈ M 1 ( mw )sty λP0 + λM0 λM0

(8.19)

Mw =

λP2 + λM2 λ ( mw )sty ≈ M 2 ( mw )sty λ P1 + λM1 λM1

(8.20)

where (mw)sty represents the molecular weight of styrene. Notice that in Equations 8.19 and 8.20, the contributions of live polymers to overall molecular weight averages are ignored because the concentrations of live polymers are far smaller than the concentration of dead polymers. Also, in the above kinetic model, it is assumed that the catalyst is a single-site catalyst. The single-site kinetic model presented in the above can be easily modified for a real polymerization process where site heterogeneity often occurs when supported onto a catalyst support material because of nonuniform surface structure of the catalyst support.

8.2.4 Molecular Weight Distribution of SPS with Heterogeneous Catalysts With an assumption of single-site catalytic polymerization mechanism, the instantaneous number-average degree of polymerization is represented by the following equation: Xn =

Rp k p [ M ]s [ P ] = Rt + Rd ktM [ M ]s [ P ] + ktβ [ P ] + kd [ P ]

(8.21)

where Rp is the chain propagation rate, Rt is the total chain transfer rate, and Rd is the site deactivation rate. [P] represents the total active site concentration ∞

(i.e., [ P ] = [C *] + ∑ [ Pn ]). Using Equation 8.2, Equation 8.21 can be rearranged n=1

to ktβ + kd 1 K 2( ktβ + kd ) ktβ + kd 1 1 k k′ = tM + = tM + + X n kp k p [ M ]s k p′ k p′ k p′ [ M ]b

(8.22)

where ktM ′ = ktM K1. The reaction rate constants in Equation 8.22 can be esti− mated by plotting 1/X n versus [M]b [6]. It has been shown that the single-site kinetic model with these rate constants provide satisfactory predictions of number-average molecular weight (it is expected because the rate constants − were derived from M n data) but the predicted weight-average molecular weight deviated from the experimental data. The discrepancy between the model-calculated and experimentally measured SPS molecular weight distributions (MWDs) can be attributed to the presence of multiple active sites or

148


the monomer transport resistance diffusing from the bulk liquid phase to the active sites in the solid phase. To see the effect of site heterogeneity, the singlesite kinetic model can be modified by assuming the presence of two catalytic sites of different chain transfer activities. The site heterogeneity with a “singlesite catalyst” when it is supported onto a solid surface has been frequently reported in α-olefin polymerization [10–12] and SPS polymerization [3]. It is believed that the broadening of MWD in SPS is, like in α-olefin polymerization, caused primarily by the presence of multiple active sites of different activity and selectivity. Monomer diffusion resistance may not be as strong as the site heterogeneity. When a metallocene catalyst is supported onto a silica by forming a complex with MAO that is already anchored onto a silica surface, it is likely that the activity of the catalyst will be influenced by the heterogeneity of the silica-MAO complex, causing the site heterogeneity [13]. A silica surface is known to have different types of surface structures represented by single (isolated) silanols, silanediols (geminal), H-bonded vicinal silanols (vicinals), etc. [14]. The concentrations of surface hydroxyl groups that may affect the catalyst reactivity are dependent upon the calcination temperature [15]. For example, when a silica gel is calcined at 250–300 °C or above, geminal groups exist only in limited amount, and single silanol and vicinal groups exist almost 50% each [16,17]. If the main catalyst component is supported onto the surface hydroxyl groups of different structures, it is quite possible that each catalyst site can exhibit a different polymerization activity. In the two-site model, it is assumed that the catalytic sites have the same polymerization activity (propagation activity) but they differ in their chain transfer capabilities. The two-site model is the simplest of the multisite model and its main advantage is that the number of adjustable parameters is minimal. The weight chain length distribution function of the two-site model takes the following form: X w = φ1 xτ 12 x exp ( − τ 1 x ) + φ 2 xτ 22 x exp ( − τ 2 x )

(8.23)

where φi is the weight fraction of active site i and the parameter τi is defined Rt ,i + Rd ,i as τ i = . Notice that for each site, the polymer chain length distribution R p,i is represented by the most probable distribution. With the propagation and deactivation rate constants fixed for each site, τi is changed by adjusting the termination rate constants (ktβ, and ktM ′ ) and the weight fraction of each active site φi to fit the experimentally measured MWD data. Figure 8.4 illustrates the model-predicted and the experimentally measured MWD profiles of SPS synthesized at two different initial monomer concentrations. Although some discrepancies in very low molecular weight regions and very high molecular weight regions are present, the overall MWD profiles are adequately simulated by the two-site model.

NASCENT MORPHOLOGY OF SYNDIOTACTIC POLYSTYRENE

149

0.07 [M]b0 = 0.81 mol/l 0.06

[M]b0 = 4.86

Weight fraction

0.05 0.04 0.03 0.02 0.01 0.00 101

102

103 Chain length

104

105

Figure 8.4 Molecular weight distributions for two different initial monomer concentrations (symbols—experimental data, lines—model simulations), t = 30 min.

8.3


8.3.1 Physical Transitions of Reaction Mixture During Polymerization In a heterogeneous SPS polymerization at a typical reaction temperature (40–90 °C) with a silica-supported catalyst, the reaction mixture undergoes a series of physical changes as SPS particles precipitate out from the liquid phase. Initially, the total solid content (TSC) in the reaction mixture is very small and it is a clear liquid (Fig. 8.5a). As TSC increases, the reaction mixture becomes turbid and the precipitated particles agglomerate to soft aggregates (Fig. 8.5b). The active catalyst particles become embedded in the polymer phase and they continue to polymerize styrene and the apparent density of the aggregates increases. These soft polymer aggregates are broken up by mechanical agitation and become smaller (Fig. 8.5c). As styrene conversion increases, the SPS particles absorb unreacted styrene and diluent, and the reaction mixture becomes a soft wet cakelike material (Fig. 8.5d). Interestingly, from this point, a separate liquid phase disappears (it is all absorbed in the solid phase) and the reaction mixture appears to be a lightly wet or dry powders (Fig. 8.5e). Figure 8.5f illustrates the interior of the reactor after the completion of the polymerization experiment. Notice that there is no liquid phase and the reaction mixture is a fairly dry and soft aggregate of polymer particles.

150


(a)

(b)

(c)

(d)

(e)

(f)

Figure 8.5 Physical changes of reaction mixture with silica-supported metallocene catalyst in n-heptane with time (a→e); f—final reaction mixture. (See color insert.)

The change in the slurry phase volume in SPS polymerization can be calculated using a simple mass balance model. We first consider a mass balance for the entire slurry phase: dVslurry 1 dWM 1 dWsPS ⎛ 1 1 ⎞ = + =⎜ − V R ⎝ ρsPS ρM ⎟⎠ slurry p dt ρM dt ρsPS dt

(8.24)

where Vslurry is the total slurry volume; WM is the weight of monomer; WsPS is the weight of the polymer; ρM is the monomer density; ρsPS is the polymer density; and Rp is the rate of polymerization. The liquid phase volume is calculated by VL = Vslurry − φWsPS −

WsPS ρsPS

(8.25)

where φ (l/g-SPS) is the amount of liquid absorbed per gram of SPS. Equation 8.24 is then solved with the kinetic model equations to estimate the reaction time when separate liquid phase disappears and the reaction mixture becomes a wet cake-like material. Figure 8.6 illustrates the volume fraction profiles of liquid phase decreasing with the increase in the total solid content for the initial monomer concentrations of 0.81 mol/l and 4.86 mol/l.


151

Liquid phase volume fraction

1 0.8

[M]b0= 0.81 mol/l

0.6 0.4 4.86 mol/l 0.2 0

0

0.05

0.1

0.15

0.2

0.25

Total solid content (g/g)

Figure 8.6 Liquid phase volume fraction versus total solid content in SPS polymerization over silica-supported Cp*Ti(OCH3)3/MAO catalyst at 70 °C.

At high initial monomer concentrations, the separate liquid phase disappears after about 15%–16% of TSC is reached. It is interesting to observe that even for the vastly different initial monomer concentrations, the liquid phase volume fraction decreases with the increase in the solid phase almost linearly. At lower monomer concentrations, the larger content of inert diluent (e.g., n-heptane) tends to give a slightly larger liquid phase volume fraction than at higher monomer concentration cases. 8.3.2

Effect of Reaction Conditions on Polymer Morphology

SPS is known to exhibit complex polymorphic behavior of having four different crystalline forms (i.e., α, β, γ, δ forms) with all-trans planar zigzag (TTTT) and 21-helix (TTGG) conformations. The SPS in solution or after absorption of solvent molecules takes the TTGG conformation, which is believed to be responsible for the thermoreversible gelation of SPS [18–20]. The δ-form SPS is a nanoporous metastable polymorph including centrosymmetric crystalline cavities of about 120–160 Å3 [21]. The recent morphological analysis of SPS particles synthesized over the silica-supported Cp*Ti(OCH3)3/MAO catalyst system shows the nanofibrillar morphology with extensive silica particle fragmentation. Figure 8.7a,b shows the scanning electron microscopy (SEM) images of silica gel catalyst support particles (average pore diameter of 20 nm) and polymerized SPS particles obtained in n-heptane slurry polymerization, respectively. It is clearly seen that original silica particles grow with the encapsulation of SPS by the polymerization and that the fully grown SPS particles maintain the shape characteristics of the original silica catalyst particles. It is an interesting result because as shown in Figure 8.7c,d, the SPS particles show a fibrillar

152


100 mm

(b)

(a)

1mm (c)

100 mm

500 nm (d)

Figure 8.7 SEM images of silica gel particles (a), polymerized SPS particles (b), and SPS nanofibrils (c and d).

morphology that is quite different from the multigrain-type morphology commonly observed in α-olefin polymerization processes over heterogeneous transition metal catalysts. Figure 8.7c shows that the particle surface is covered with heavily entangled and long nanofibrils of 30–50 nm diameter. Figure 8.7d indicates that some polymer nanofibrils are as thin as 10–15 nm and some nanofibrils aggregate to larger diameter bundles. Based on the experimental observations of SPS particle morphologies discussed above, the following mechanism is proposed for the formation and growth of polymer particles with heterogeneous catalysts. First, monomer and solvent penetrate into silica gel pores and monomer polymerizes at the surface of silica micrograins. Then, the SPS formed at the surface crystallizes and grows as a thin nanofibril. The nanofibrils from the neighboring active catalyst sites self-assemble to a bundle of nanofibrils of larger diameter. As these bundles of nanofibrils fill up the void space in the particle, a buildup of hydraulic pressure occurs within the catalyst pores, eventually leading to the disintegration of the primary silica particles. Since the polymer grows in fibrillar form, the disintegrated particles are held together and grow as monomer is further polymerized. It is believed that the fibrillar morphology offers very

REFERENCES

153

small mass transfer resistance for the monomer diffusion and the entire polymer particle grows with a uniform interior structure created by the uniform sized SPS nanofibrils. The entire polymer particle continues to expand as more polymer nanofibrils are produced, leading to shape replication.

8.4

CONCLUDING REMARKS

In this chapter, the kinetic aspects of styrene polymerization to SPS with a heterogeneous metallocene catalyst has been discussed with Cp*Ti(OCH3)3/ MAO/n-heptane as a model system. The SPS polymerization is characterized by a series of significant macroscopic phase changes during polymerization and by the development of complex particle morphology. The SPS molecular weight distribution with the metallocene catalyst is not adequately described by a single-site kinetic model because of site heterogeneity in the silica-supported catalyst. The proposed two-site model has been found to be an effective alternative to the simple single-site model. The experimental identification of the exact nature of site heterogeneity may have to be investigated further in future research. The quantitative kinetic model presented in this chapter provides a basic framework for the quantitative analysis of the polymerization kinetics and for the design of a slurry phase polymerization process. It is interesting to note that although the SPS particle morphology, represented by fibrillar morphology, differs from that of similar polymerization processes such as α-olefin polymerization, the particle shape replication also occurs in the SPS process. Finally, it is noted that recently, ultrahigh molecular weight SPS was synthesized with Cp*Ti(OCH3)3/MAO catalyst in a silica nanotube reactor (SNTR) system [22]. It was suggested that the chain transfer reaction is significantly reduced by the geometric confinement of the reaction space that limits the mobility of rapidly crystallizing SPS.

REFERENCES 1. Po, R., Cardi, N. Synthesis of syndiotactic polystyrene: Reaction mechanisms and catalysis. Prog. Polym. Sci., 21, 47–88 (1996). 2. Tomotsu, N., Ishihara, N., Newman, T. H., Malanga, M. T. Syndiospecific polymerization of styrene. J. Mol. Catal. A: Chem., 128, 167–190 (1998). 3. Schellenberg, J., Tomotsu, N. Syndiotactic polystyrene catalysts and polymerization. Prog. Polym. Sci., 27, 1925–1982 (2002). 4. Choi, K. Y., Chung, J. S., Woo, B. G., Hong, M. H. Kinetics of slurry phase polymerization of styrene with pentamethyl cyclopentadienyl titanium trimethoxide and methyl aluminoxane. I. Reaction rate analysis. J. Appl. Polym. Sci., 88, 2132–2137 (2003). 5. Chung, J. S., Woo, B. G., Choi, K. Y. Syndiospecific polymerization of styrene with embedded metallocene catalysts. Macromol. Symp., 206, 375–382 (2004).

154


6. Han, J. J., Lee, H. W., Yoon, W. J., Choi, K. Y. Rate and molecular weight distribution modeling of syndiospecific styrene polymerization over silica-supported metallocene catalyst. Polymer, 48, 6519–6531 (2007). 7. Han, J. J., Yoon, W. J., Lee, H. W., Choi, K. Y. Nascent morphology of syndiotactic polystyrene synthesized over silica-supported metallocene catalyst. Polymer, 49, 4141–4149 (2008). 8. Kaminsky, W., Arrowsmith, D., Strübel, C. Polymerization of styrene with supported half-sandwich complexes. J. Polym. Sci. Polym. Chem., 37, 2959–2968 (1999). 9. Schellenberg, J. The syndiospecific polymerization of styrene in the presence of fluorine-containing half-sandwich metallocenes. J. Polym. Sci. Polym. Chem., 38, 2428–2439 (2000). 10. Frauenrath, H., Keul, H., Höcker, K. H. Deviation from single-site behavior in zirconocene/MAO catalyst systems, 1. Macromol. Chem. Phys., 202(18), 3543–3550 (2001). 11. Frauenrath, H., Keul, H., Höcker, H. Deviation from single-site behavior in zirconocene/MAO catalyst systems, 2. Macromol. Chem. Phys., 202(18), 3551–3559 (2001). 12. Kou, B., McAuley, K. B., Hsu, C. C., Bacon, D. W., Yao, K. Z. Mathematical model and parameter estimation for gas-phase ethylene homopolymerization with supported metallocene catalyst. Ind. Eng. Chem. Res., 44(8), 2428–2442 (2005). 13. Ciardelli, F., Altomare, A., Michelotti, M. From homogeneous to supported metallocene catalysts. Catal. Today., 41(1–3), 149–157 (1998). 14. Flörke, O. W. et al. Silica. In Ullman’s Encyclopedia of Industrial Chemistry, Arpe, H. J. (ed.), VCH Publishers, Weinheim, 1993, A23, pp 614–620. 15. van Grieken, R., Calleja, G., Serrano, D., Martos, C., Melgares, A., Suarez, I. Ethylene polymerization over supported MAO/(nBuCp)(2)ZrCl2 catalysts: Influence of support properties. Polym. React. Eng., 11(1), 17–32 (2003). 16. Haukka, S., Lakomaa, E. L., Root, A. An Ir and NMR study of the chemisorption of TiCl4 on silica. J. Phys. Chem., 97(19), 5085–5094 (1993). 17. Zhuravlev, L. T. The surface chemistry of amorphous silica. Zhuravlev model. Colloids Surf. A., 173(1–3), 1–38 (2000). 18. Prasad, A., Mandelkern, L. Thermoreversible gelation of syndiotactic polystyrene. Macromolecules, 23, 5041–5043 (1990). 19. Daniel, C., Dammer, C., Guenet, J. M. On the definition of thermoreversible gels— The case of syndiotactic polystyrene. Polymer, 35(19), 4243–4246 (1994). 20. Tashiro, K., Yoshioka, A. Molecular mechanism of solvent-induced crystallization of syndiotactic polystyrene glass. 2. Detection of enhanced motion of the amorphous chains in the induction period of crystallization. Macromolecules, 35(2), 410–414 (2002). 21. Reverchon, E., Guerra, G., Venditto, V. Regeneration of nanoporous crystalline syndiotactic polystyrene by supercritical CO2. J. Appl. Polym. Sci., 74(8), 2077– 2082 (1999). 22. Choi, K. Y., Han, J. J., He, B., Lee, S. B. Ultrahigh molecular weight syndiotactic polystyrene nanofibrils in silica nanotube reactors. J. Amer. Chem. Soc., 130, 3920– 3926 (2008).

PART III

STRUCTURE AND FUNDAMENTAL PROPERTIES OF SYNDIOTACTIC POLYSTYRENE

CHAPTER 9

Structure, Morphology, and Crystallization Behavior of Syndiotactic Polystyrene ANDREA SORRENTINO and VITTORIA VITTORIA Chemical and Food Engineering Department, University of Salerno, Fisciano, Italy

9.1

INTRODUCTION

Since syndiotactic polystyrene (SPS) was first synthesized in the 1980s, it has stimulated interest from both technological and theoretical points of view. From the technological point of view, SPS presents various desirable properties such as high melting temperature, low dielectric constant, low permeability to gas, and good mechanical properties. At the same time, the very complex polymorphic behavior and the extreme sensitivity to the processing conditions spurred many investigations on this polymer. The great number of papers and patents that appeared on this polymer in the last years are the main results of these studies. The principal objective of these studies was the crystallization behavior of SPS, the structure of the ordered forms, and the properties of SPS with the respect to the processing conditions. The possibility of controlling the conditions for obtaining controlled structures indeed is particularly interesting in view of the different physical properties that the various polymorphic structures show. In this chapter, an attempt is made to synthesize the principal founding on this polymer and analyze it in terms of structural organization and crystal morphology.

9.2

POLYMORPHIC BEHAVIOR OF SPS

Syndiotactic polystyrene can crystallize in many crystal forms as well nonequilibrium structures depending on the thermo-mechanical processing conditions


157

158

CRYSTALLIZATION BEHAVIOR OF SYNDIOTACTIC POLYSTYRENE

[1–14]. The most stable α and β forms are characterized by chains in transplanar (zigzag) conformation, whereas γ and δ forms are characterized by chains in helical conformation. The α and β forms can be obtained principally by melt crystallization or by annealing at a proper temperature, whereas the γ and δ forms can be obtained through solvent treatments of the amorphous or α form. The polymorphism of SPS is further complicated by the presence of structural disorder in both α and β forms, so that the trans-planar forms are described in terms of disordered modifications intermediate between limit disordered models (α′ and β′) and limit ordered models (α″ and β″). Melt crystallization procedures generally produce crystalline modifications of the α and β forms close to the limit ordered α″ and limit disordered β′ models, respectively [4]. Crystalline modifications close to the limit disordered α′ model are obtained by annealing amorphous SPS sample or samples in the crystalline γ form [4], whereas crystalline modifications close to the limit ordered β″ model are obtained by crystallization from solution [13]. The different conformations, trans-planar and helical, are associated with nearly equivalent minima in the conformational energy map, calculated for isolated chain models [15–17]. The interconversion energy barrier between helical and trans-planar was estimated to be about 2–3 kcal/mol [15]. A schematic representation of all the known conversions between the different polymorphic forms of SPS is shown in Figure 9.1.

Vapor

a

g

α′ Solvent

Rapid cooling

α″

Slow cooling Annealing T>180°C

Amorphous

Quenching

Solvent

Heating

e Melt Solvent

Heating

b

Vapor

Ultra-high pressure

Annealing T>140°C

Vapor Thermal aging

Slow cooling from high T

Solvent

Solvent

β′

d Heating T>150°C

β″

Precipitation

Solvent

Casting Solvent

Solution

Gel

Figure 9.1 Schematic representation of the interconversion conditions for the different polymorphic forms of SPS.


9.2.1

159

Crystallization from the Melt State

By melt crystallization, the α form can be obtained pure or mixed with the β form, depending on the crystallization conditions. Many studies [18–22] have focused on the following factors influencing the polymorphic behavior in samples crystallized on cooling from melt: • • • •

the the the the

crystallization temperature, maximum melt temperature, permanence time at the melting temperature, and cooling rate.

The results showed that the α crystal packing can be favored, or becomes an alternative route in SPS crystallization under three conditions: 1. fast cooling from the molten state, 2. melt crystallization at low temperatures (e.g., 230 °C or lower), and 3. cold crystallization from the quenched state. By comparison, melt crystallization at most accessible temperatures produces solidified SPS containing both α-type and β-type crystals of various fractions, although higher temperatures tend to favor larger fractions of β-type crystals [23]. The β-type crystals become the only discernible species if SPS is meltcrystallized at temperatures equal to or higher than 260 °C, suggesting that in conditions approaching equilibrium the β-crystals lamellae are the favored packing. By pressure annealing of the α form, the more densely packed β form is obtained [24], while the application of pressure on samples in the α form at temperatures below Tg results in a loss of the three-dimensional crystalline order [25]. Li et al. [20] found that the cooling rate from the melt should not be the intrinsic factor controlling the formation of the different phases. The crystallization temperature seems to have a more important influence. The crystallization rate of the α-form crystal is much faster than that of the β form, but the respective maxima are shifted in two different positions. Li et al. concluded that the SPS samples present a transition temperature (dependent on their molecular structures) [20], above which the crystallization rate of the β form is higher than that of the α form, and thus the α form can only be obtained below this temperature. Ho et al. [26] studied the possibility of α to β phase transformation in SPS during isothermal crystallization. The transition was found to occur in the later stage of crystallization as evidenced by the structural analysis of infrared spectroscopy experiments combined with enthalpic measurements. Studies on the crystallographic textures of the coexistent α″ and β′ single crystals indicate

160


that the growth of the β′ single crystals is on the surface of the α″ single crystals in terms of lamellar thickening process so as to form a dual-layer morphology. The lamellar crystals of both phases are grown as flat-on types and share a common c-axis in the crystal structures during the transition. The solid-state structure of SPS (Mw = 300,000 g/mol) after crystallization from the melt and the glassy states were examined by differential scanning calorimetry (DSC), density, and X-ray diffraction analysis [27]. The measurements confirmed the low density of both crystalline forms, which in the case of the α crystalline form was smaller and, in the case of the β crystalline form, was only slightly larger than the density of the glassy amorphous SPS. Bu et al. [28] studied the crystallization of the α- and β-form crystals of SPS and found that it is determined by the crystallization temperature. Being crystallized at various temperatures from the melt, SPS forms the β-form crystal at high temperatures, above approximately 230 °C, the α-form crystal at low temperatures, below approximately 170 °C, and a mixture of the α and β forms at intermediate temperatures. Cai and Han [29] also found that for melt crystallization, the contents of α-type and β-type crystals depend on the cooling rates. Polarized optical microscopy also indicated a difference in the final morphology of the sample, and the formation of β-type crystals was found to have larger absolute values of effective activation barrier than the formation of α-type crystals. Su et al. [30] also performed direct and nonintrusive observations of crystallization and melting behavior of α and β polymorphs in bulk SPS by means of temperature-programmed X-ray diffraction. Results indicated that the perfection of the less ordered α′ form into the better ordered α″ form within the α family occur in the vicinity of 270 °C. 9.2.2

Crystallization from the Glassy State

Depending on the cooling conditions from the melt, SPS can be obtained in the amorphous state and successively crystallized at temperatures higher than the glass transition temperature. In this sense, the processing histories of SPS can sensitively influence the microstructure and properties, because, as shown before, from the glassy state the α form is always obtained. The thermal crystallization of glassy films was studied using infrared, X-rays, and thermal analysis, in the range 120–220 °C [31]. It was found that the growth of the ordered structure occurred in two stages: a fast primary stage followed by a slower secondary process. The crystalline and the mesophase are formed in either process. The formation of the mesophase increases in the secondary stage of crystallization. To elucidate the nucleation mechanism during the crystallization process from the glassy state, Matsuba et al. [32] performed time-resolved Fourier transform infrared (FTIR) spectroscopy and depolarized light scattering measurements (DPLS) on glassy SPS kept at 120 °C. They found an induction


161

period during the first 30 min. The absence of crystallization in this induction period was also confirmed by X-rays. Therefore, Matsuba et al. investigated the FTIR spectra in the induction period, following the appearance of the conformational bands relative to the trans-planar form. After a decomposition procedure of the spectra in the region of 500–600 cm−1, they reported the dependence of the intensity of the 537 cm−1 band on the annealing time. This band, which includes trans-planar conformation, starts to increase in intensity just after rising to around the crystallization temperature, and continues to increase not only through the induction period but also when crystallization begins. Correspondingly, a decrease of the bands including gauche conformations was observed. Furthermore, the time-resolved DPLS measurements revealed that the orientation fluctuations due to parallel ordering of polymer chains begin to increase in the induction period. These results suggest that a certain minimum length of rigid segments is required for the parallel ordering of polymer chains. Wang et al. [33] studied the spherulitic growth rates and the microstructure of SPS cold-crystallized isothermally at various temperatures in the range 115–240 °C, by small-angle light scattering (SALS), optical microscopy, and transmission electron microscopy (TEM; Fig. 9.2).

117.5°C, Hv

117.5°C, Vv

117.5°C 20 μm

145°C, Hv

145°C, Vv

145°C

240°C, Hv

240°C, Vv

240°C

Figure 9.2 Light scattering patterns and phase contrast micrographs of SPS coldcrystallized at various isothermal crystallization. Reprinted from Wang et al. [33], with permission from Elsevier.

162


The derived activation energy for SPS chain mobility at the crystal-growing front is 5.4 kJ/mol, which is lower than that of isotactic polystyrene, which is 6.5 kJ/mol. Despite a wide range of Tc used, however, the sample crystallinity estimated by Fourier transform infrared spectroscopy remains unchanged. Interconnected domains with a width of approximately 1.8 μm are readily observable in all the crystallized samples under phase contrast microscopy, and the phase-separated structure is conserved within SPS spherulites whose diameters are increased with increasing Tc. Based on the above facts, the Wang et al. [33] concluded that there are two competitive transitions: liquid–solid crystallization and liquid-liquid demixing resulting from the spinodal decomposition (SD). At lower Tc, the unstable SD transition overwhelms the crystallization. Despite the low chain mobility, the coarsening process driven by the interfacial energies has reached a certain level before crystalline nucleation takes place. At higher Tc, on the other hand, cold crystallization becomes the dominant process due to the enhanced chain mobility, leading to the suppression of the ongoing SD coarsening process. At an intermediate Tc range, comparable competition of the phase separation and crystallization prohibits the development of ordered symmetry within spherulites. Handa et al. [34,35] showed that supercritical fluids or compressed gases produce a depression in the crystallization temperature of neat SPS. The authors also showed that when a cold crystallization of SPS amorphous sample is carried out in the presence of CO2 at 57 atmospheres the structure obtained is that of γ form. By strain-induced crystallization, the pure α form is obtained even at temperatures for which thermal crystallization from the glassy state is not observed. De Candia et al. [36] analyzed the drawing behavior of amorphous films of SPS at different temperatures. They found that strain-induced or thermalinduced crystallization was obtained, depending on the drawing temperature. At 110 °C thermal crystallization was not observed, whereas strain-induced crystallization occurred at high draw ratios, resulting in a substantial increase in the elastic modulus of the obtained samples. The effects of molecular orientation on the crystallization and polymorphic behavior of SPS and SPS/poly(2,6-dimethyl-1,4-phenylene oxide) (PPO) blends were studied with wide-angle X-ray diffraction (WAXD) and differential scanning calorimetry [37]. The oriented amorphous films of SPS and SPS/ PPO blends were crystallized under constraint at crystallization temperatures ranging from 140 to 240 °C. The degree of crystallinity was lower in the coldcrystallized oriented film than in the cold-crystallized isotropic film. It was inferred that the oriented mesophase was obtained in drawn films of SPS and that the crystallization of SPS was suppressed in that phase. The WAXD measurements showed that the crystal phase was more ordered in SPS/PPO blend than in pure SPS under the same annealing conditions. It was principally due to the decrease in the mesophase content. The crystal forms were found to be dependent on the crystallization temperature, blend composition, and


163

molecular orientation. Only the α′ crystalline form was obtained in SPS coldcrystallized, regardless of molecular orientation, whereas α′, α″, and β′ forms coexisted in the cold-crystallized SPS/PPO blends prepared at higher crystallization temperatures (200–240 °C). 9.2.3

Morphology Development in the Presence of Solvents

By solution crystallization or by sorption of suitable compounds (e.g., methylene chloride, toluene, chloroform) in amorphous SPS samples as well as in samples in the α form, clathrate crystalline phases, always containing s(2/1)2 helical chains, can be obtained [3,9,10,38,40–43]. The process of solvent-induced crystallization (SINC) of glassy SPS, leading to helical forms, was studied in different liquids [36,44–48]. The sorption of different liquids is a function of the solubility parameter and shows a maximum with respect to chloroform, indicating a maximum in the polymer-solvent interaction. Furthermore, it was found that the swollen samples were birefringent but appeared substantially amorphous on the basis of the X-ray diffractograms. Therefore, true crystallinity develops when the sample is dried, and it depends on the rate of evaporation of the solvent. In the case of toluene [36], the intensity of transmitted light in a polarizing microscope was followed during swelling, and it was found that there was development of order, shown by the increase of birefringence, and it was strongly diffusion controlled. The lack of sharp reflections in the WAXD of the swollen film and the simultaneous development of birefringent regions were explained by assuming either a low crystallinity or a development during the swelling of a liquidcrystalline-like order. The δ form consists of a polymer-solvent compound (designated sometimes as crystallo-solvates, intercalates, or chlathrates) and forms in a large variety of solvents, such as benzene, toluene, chloroform, bromoform, diethylbenzene, naphthalene, tetralin, and trans-decalin, which are either liquid or solid at room temperature. Surprisingly, no significant change in the experimental lattice parameters has been observed hitherto despite the formation of SPSsolvent compounds with solvent molecules of highly differing shapes and sizes. Solvent molar volumes typically range from 60 to 150 cm3/mol, and molecules with shapes as different as 1,2-dichloroethane and trans-decalin produce virtually the same diffraction pattern. The first observation of a different lattice, showing the existence of a second class of intercalates, was reported by Daniel et al. [49] for SPS/benzene. This structure could be described as a “swollen” form of the δ form. Similar results were later reported by Rastogi et al. [50–51] and more recently by Petraccone et al. [52]. These authors actually refer to the usual δ form as chlathrates and the “swollen” form as intercalates [53]. The crystallization of amorphous SPS films when induced by bulky solvents, whose molecules are too big to be enclosed as guest of SPS clathrate phases, generally leads to the formation of the γ phase. Moreover, the presence of

164


highly boiling solvents in the amorphous phase of these γ-form samples can induce a γ→β phase transition, as a consequence of thermal annealing procedures at atmospheric pressure [54]. As a consequence of suitable treatments [55–57], guest molecules can be removed and all clathrate phases can be transformed into the nanoporous ε phase [14]. This phase was thoroughly studied by Guerra et al. and is described in Chapter 10 of this book. SPS solutions cooled with bulky solvents, unsuitable as guests of SPS clathrate phases, show a paste-like gel morphology. These gels are characterized by a trans-planar chain conformation and poor mechanical properties. Strong elastic gels with s(2/1)2 helical polymer chains forming a three-dimensional network can be, instead, obtained with the same bulky solvents by the temporary presence of a volatile solvent which is instead a suitable guest for an SPS clathrate phase. After gel formation, the volatile solvent can be easily removed without breaking the gel structure, and the big size solvent molecules remain entrapped in the three-dimensional polymer network [58]. Gels of SPS with different solvents have been compared to clathrates. WAXD results using toluene (a good solvent for SPS) and decalin (a relatively poor solvent for SPS) show that the structure of the crystalline junctions of the gels is similar to that of the clathrate α phase. A difference can be found in the width of the (010) reflection, which is relative to the width of the (210) reflection, much broader for the gel than for the clathrate. This is caused by the difference in the mechanism involved in crystal formation in gels and clathrates. Experiments performed on quenched samples of SPS with the monomer benzyl methacrylate show that also for this gel the structure of the crystalline part is similar to that of the clathrate phase. This means that solvent is present in both the crystalline and the amorphous parts of the gel. By solidstate nuclear magnetic resonance (NMR) studies, a clear difference in the mobility of solvent molecules in the crystalline and amorphous parts of the gel has been observed [58]. Recently a new crystalline form, named ε, has also been discovered [39]. This new crystalline phase is also able to form co-crystals with long and highpolarity organic guest molecules. Moreover, it has been shown that in co-crystals obtained from the δ phase, the orientation of the guest molecular planes can be parallel to the polymer host chain-axes, rather than perpendicular, as generally observed for co-crystals obtained from the δ phase. 9.3 9.3.1

MORPHOLOGY OF THE ZIGZAG FORMS Crystal Structure of the α Form

X-ray diffraction has been widely used in the recognition and quantification of the different polymorphs in SPS. Guerra et al. [4,5] first gave the nomenclature of α and β crystalline forms and studied their formation in detail. Discriminating the zigzag forms, they

MORPHOLOGY OF THE ZIGZAG FORMS

165

Table 9.1 Summary of the Characteristic Peaks of the Different Melt-Crystallized Forms Characteristic Peaks (2θ)

Morphology α form β form α′ form α″ form β′ form β″ form γ form δ form (emptied) δ form (toluene 14% wt) δ form (o-DCB 14% wt) ε form Amorphous

6.7 °, 11.6 ° 6.1 °, 12.2 °, 18.6 ° 6.7 °, 11.6 °, 13.5 ° 10.3 °, 11.6 °, 15.6 ° 6.1 °, 12.2 °, 18.6 °, 20.3 ° 6.1 °, 11.8 °, 12.2 °, 15.8 °, 18.6 °, 20.3 ° 9.2 °, 10.3 °, 13.9 °, 15.9 °, 19.8 °, 28.0 ° 8.4 °, 10.5 °, 13.6 °, 17.0 °, 20.6 °, 23.4 °, 28.0 ° 8.0 °, 10.3 °, 17.4 °, 20.1 °, 23.4 °, 28.0 ° 7.9 °, 10.1 °, 16.8 °, 18.9 °, 20.0 °, 23.5 °, 25.0 °, 28.3 ° 6.9 °, 8.2 ° 10.6 °, 19.5 °

o-DCB: ortho-dichloro benzene.

found [13] that the α-form crystals exhibit characteristic peaks at 2θ = 6.7 ° and 11.6 °, while the β form at 2θ = 6.1 °, 12.2 °, and 18 °. The most important peak locations for the different crystalline form are summarized in Table 9.1. To quantify the fractions of the zigzag crystals, a procedure was suggested for evaluating, from the X-ray patterns, the relative amounts of the two crystalline forms α and β [4]. The 2θ region, 10–15 ° was considered and a baseline between the two intensity minimal located at 2θ = 10.8 ° and 2θ = 14.8 ° was drawn. The areas of the two peaks located at 2θ = 11.6 ° (α form) and 2θ = 12.2 ° (β form) was measured, and the percentage of the α form in the crystalline fraction was evaluated by the approximate relation (9.1): 1.8 × A (11.6°) A (12.2°) Pα = × 100 1.8 × A (11.6°) 1+ A (12.2°)

(9.1)

where 1.8 is the ratio between the intensities of the peaks at 11.6 ° and 12.2 ° of 2θ, for samples of equal thickness and crystallinity in the pure α and β forms, respectively. The α-form structure was first shown by Greis et al. [1] as made of clusters of three extended chains (triplets), with three triplets (nine chains) packed in a trigonal unit cell of parameters a = b = 26.26 Å and c = 5.04 Å. Greis et al. [1] also reported a one-triplet trigonal unit cell, obtained at low crystallization temperatures, later called α′. The α″ structure was progressively investigated and refined. The authors proposed a packing scheme of the triplets characterized by the fact that the azimuthal settings of two of the triplets are identical but differ by 180 ° from that of the third one, leading to a structure resulting from the coexistence of right- and left-handed triplets, as shown in Figure 9.3a.

166


0

0

3

0

3 0

3

0

3

0

3

3

0

0

0

3 3

0

(a)

2

4 1

5

1

4

5

2

4

2 5

1 0 3

3

0

(b)

0 3

Figure 9.3 Schematic representation of the crystal structure of the α″ form derived by: (a) Greis et al. [1] and (b) Corradini et al. Reprinted from [59], with permission from Elsevier.

Corradini and co-workers proposed a succession of several models of the α″ form, characterized by a progressive decrease of the high symmetry initially assumed and by a ∼30 ° rotation of the triplets as compared to the initial model. In the last models all the triplets are rotated by an additional angle of 7 °, as illustrated in Figure 9.3b [59]. In addition the structure keeps the 180 ° (or 60 °) rotation of the triplet relative to its two neighbors, thus creating the three triplets superstructure α″. In the successive model of De Rosa, the three triplets are also shifted by c/3 along the c-axis [60]. The one triplet α′ from structure was analyzed by Greis et al. and by De Rosa as a statistical assembly of the three triplets existing in the more stable α″ phase. Cartier et al. [7] re-evaluated the crystal structure of the α″ form, pointing out the frustrated character of all the previously proposed models. As matter


167

of a fact, a threefold symmetry of the clusters, packing of three clusters in a trigonal unit cell and, more importantly, different azimuthal settings of the clusters are features establishing frustration. On the basis of the type of frustration considered in the models, these authors suggested for the α″ structure a model that preserves many of its major features but in which the azimuthal settings of the triplets are not bound to be similar. They observed that SPS is original among frustrated polymer structures in that the building element is not a helix but rather a cluster of three extended chains with a threefold symmetry (as opposed to a threefold screw symmetry), as shown in Figure 9.4.

0

3

0

3 3

3

0

0 0

0

3

3 3 0

0

3

3

(a)

0

(b)

Figure 9.4 Schematic representation of the crystal structure of the α″ form derived by Cartier et al. Reprinted with permission from [7]. Copyright 1998 American Chemical Society.

168


Frustration rests on non-even interactions between the three triplets comprising the trigonal unit cell. The azimuthal settings and relative shifts of the triplets indicated that two triplets maximize their interactions, while the third one interacts less favorably with its neighbors. Application of the principles underlying frustration made it possible to derive a model that corresponds to a minimum in the packing energy and provides the best agreement with the available diffraction evidence, that is, diffraction patterns of single crystals with a resolution of up to 0.5 Å−1 [61]. 9.3.2

Crystal Structure of the β Form

The other crystalline form of SPS, in which the chains show a trans-planar conformation, is the β form [12,13]. The crystal structure was reported by De Rosa et al. [13] and was regarded as a regular alternation of two types of bimolecular layers (viz., A and B motifs). This regular structure is to be orthorhombic with a P212121 symmetry. Guerra et al. [4] found that the β form shows two types of extinction rules in the X-ray diffractograms depending on the preparation conditions. Intensity peaks corresponding to orthorhombic hk0 reflections with h + k = odd do not exist in the diffractogram of the meltcrystallized specimens. This crystalline form was named β′ modification. The other form in which hk0 reflections with h + k = odd exist was named the β″ modification. The structural difference between the β′ and β″ modifications was explained by De Rosa et al. [13] in terms of a statistical occurrence of a stacking fault producing a pseudo-centering in the ab-projection of the crystal lattice. The β′ modification may be very close to this limiting disordered structure, while the limiting ordered β″ modification corresponds to a regular alternation of motifs of A and B. Independently the existence of a kind of structural disorder that was designated as a stacking fault in the β-form single crystals isothermally grown from a dilute solution was reported [62–64]. It is a succession of motifs of the same type. The faulted part is related to a monoclinic cell with half the volume of the orthorhombic one. In this sense the β form with stacking faults results from the coexistence of monoclinic layers in the orthorhombic crystal, as shown in Figure 9.5. The probability of the presence of stacking faults in the β-form single crystals was derived by three different methods, and it showed a weak dependence on the crystallization temperature [10]. In Tosaka et al. [63] and Hamada et al. [64], the probability of the presence of stacking faults in the β-form single crystals of SPS, which were grown isothermally from a dilute solution at a temperature ranging from 150 to 210 °C, was estimated from the mean half-breadth of the streaked reflections. The probability was also estimated by counting the number of the faults recorded in the dark-field and the high-resolution electron microscopic images. The probability showed weak dependence on the crystallization temperature, with its maximum value at 165 °C. When the single crystals grown at 165 °C were annealed isothermally, even at 260 °C only a small decrease in the amount of

169


b a A1

A2

A1

A2

B1

B2

B1

B2

Figure 9.5 Packing model for β′ modification. Arrows indicate the domain boundary. Broken lines indicate the monoclinic unit cell. Reprinted with permission from [62]. Copyright 1999 American Chemical Society.

stacking faults was detected. The molecular mechanism and the difficulty in eliminating the stacking faults by annealing were discussed on the basis of the structural model of the fault. Tosaka et al. [63] found that, according to the analysis of electron diffraction pattern of the β′ modification, the probability values were greater than 0.5, indicating that the succession of the motifs of the same type comprising the monoclinic layer of the faulted structures is the dominant structure. The theory presented by Lotz et al. [62] was successively assessed by examining the structure of the β′ modification. The electron diffraction patterns from the single-crystal-like lamellae of the β′ modification showed a streaked feature supporting their prediction. The high-resolution images evidenced that the β′ modification is composed of monoclinic domains; the boundary between the domains is a twin plane. This feature was also detected in the dark-field images as irregularly spaced striations. The probability, p, to find the same type of bimolecular layer (viz., a motif) at the next position of a motif was estimated from the high-resolution and the dark-field images. The subsequent energetic analysis based on the experimental P values showed that the theory presented in the previous paper was reasonable. Napolitano et al. [65] investigated the adjacent re-entry folds of chains of SPS crystallized in the β form by molecular mechanics. Various models of chain fold along bilayers have been found. The results are in agreement with the literature experimental data indicating that the fold surface is

170


irregular. Both the conformational and the packing energy of folded chains have been minimized by various techniques using several sets of potential functions. 9.3.3

Lamellar and Spherulitic Morphology of the Zigzag Forms

Different morphologies can be obtained, depending on the crystallization conditions and on the crystalline forms. From the melt, SPS crystallizes according to a spherulitic morphology. Two different types of spherulites are observed: sheaf-like spherulites with a fibrosity of a few micrometers, and round spherulites, which are 50 μm in diameter. They all show positive birefringence and have the same isothermal radial growth rate [66,67]. Cimmino et al. [68] proved by optical microscopy that the growth dimension of SPS spherulites is very sensitive to crystallization temperature and time. Sun and Woo [69] showed the optical microscopy results for SPS meltcrystallized at different temperatures. At lower temperatures polygon-shaped spherulites with an average size of 20 μm are observed. As melt crystallization temperature increases, the spherulites became coarser and turn to a sheaf-like pattern. Figure 9.6 depicts the optical micrographs showing tiny spherulites of SPS melt-crystallized at (a) 230 and (b) 260 °C, which contained only the α crystal. Both optical microscopy results confirmed that the crystallization kinetics of the α-crystal SPS is of a heterogeneous nucleation, which remained so regardless of the melt crystallization temperature [78]. By contrast, Figure 9.7 shows the optical micrographs for the spherulites of SPS melt-crystallized at (a) 230 and (b) 260 °C, which contained only β crystal [78]. The morphology of the SPS samples obtained with the scanning electron microscopy technique show that inside the spherulites crystallized at low temperatures individual lamellae spread out from a nucleus center straightly (“flat-on” lamellae) [66]. The average size of these lamellae is about 350 nm in width and 3.5–10.4 μm in length. In SPS samples melt-crystallized at the highest temperatures the lamellae grow with the lamellar plane perpendicular to the radial direction (“edge-on” lamellae). These lamellae are thickened in comparison with the previous flat-on lamellae [66]. By heating the quenched amorphous material, lamellar structures can be obtained [70], as shown in Figure 9.8. The crystal thickness in the molecular direction was found to be about 200 Å;, whereas the width exceeds several micrometers. The lamellae emerge from a few nuclei and splay out, and, remarkably, single lamellae can be found belonging to several nuclei. The bulk samples are completely filled with lamellar stacks when crystallized at 240 and 250 °C, whereas only isolated lamellar stacks (sheaf-like spherulitic precursors) are frequently observed within the samples crystallized at 265 °C [71]. The interesting result is that the lamellar thicknesses obtained from SAXS closely agree with those measured from TEM micrographs.


171

(a)

(b)

10 μm

Figure 9.6 Optical micrograph showing highly-nucleated, tiny spherulites in SPS of α-crystal. Reprinted from Woo et al. [78], with permission from Elsevier.

Under appropriate thermal treatments SPS samples were prevalently crystallized into β′-crystal modification, as revealed by WAXD. Lamellar morphology and thickness of these samples, melt-crystallized at various temperatures, were studied by Wang et al. [71] using TEM and small-angle X-ray scattering (SAXS). In a successive study of Wang et al. [72], SPS samples melt-crystallized into neat α″ hexagonal and β′ orthorhombic modifications were thoroughly prepared at various temperatures for extensive morphological studies. The

172


(a)

(b)

10 μm

Figure 9.7 Optical micrograph showing highly nucleated, tiny spherulites in SPS of β crystal. Reprinted from Woo et al. [78], with permission from Elsevier.

lamellar morphologies of the as-prepared SPS samples were investigated with SAXS and TEM. Lamellar thicknesses were derived from SAXS and a good agreement was reached in comparison with TEM results. From exhaustive TEM observations on the RuO4-stained samples, long and parallel lamellae were readily observed in β′-form SPS. However, relatively irregular packing of lamellar stacks with short lateral dimensions was detected in the asprepared α″ form SPS, leading to the absence of spherulitic birefringence under polarized optical microscopy [72]. TEM micrographs of SPS singlecrystals growth in isothermal conditions are illustrated in Figure 9.9 [26].

MORPHOLOGY OF THE MESOMORPHIC PHASES

173

0.25 μm

Figure 9.8 Transmission electron micrograph of lamellae in an initially amorphous thin film crystallized by heating at 5 °C/min from room temperature to 155 °C. Reprinted from [70], with kind permission from Springer Science+Business Media.

9.4

MORPHOLOGY OF THE MESOMORPHIC PHASES

Mesomorphic phases, in which a conformational order is present, were recognized for either the zigzag or the helical conformations. Vittoria et al. [31,73] studied the permeability to dichloromethane vapors of films of SPS thermally crystallized from the amorphous phase in α form. Results showed that, for each degree of crystallinity, the fraction of impermeable phase is higher than the crystalline phase, at low vapor activity. Therefore, Vittoria and colleagues suggested that this was due to the presence of a mesomorphic phase, in addition to the crystalline and the amorphous. This mesophase, characterized by conformational order not sufficient to give rise to discrete X-ray reflections, is less permeable than the amorphous sample to the vapors at low activity and became permeable at higher activity (a > 0.6). By drawing amorphous films at 110 °C, the same authors induced noncrystalline, mesomorphic order, in the absence of the crystalline phase [74]. The infrared characterization of the drawn samples revealed the presence of the absorbance peak at 1222 cm−1, related to the skeletal vibration of chain sequences in zigzag planar conformations [75]. The presence of this conformational band in the infrared spectrum and the absence of crystalline spots in the X-ray pattern led to the conclusion that the oriented mesomorphic form had been obtained [74]. By slow annealing, both oriented and unoriented mesomorphic phases are gradually transformed, only into α-form crystals. On the basis of this evidence it was suggested that the local organization of triplets of trans-planar chains,

174


400 nm

(a)

(b)

(c)

Figure 9.9 TEM micrographs of α″, β′, and coexistent α″ and β′ single crystals SPS crystallized isothermally. Reprinted with permission from [26]. Copyright 1998 American Chemical Society.

THERMODYNAMIC AND KINETICS OF CRYSTALLIZATION

175

typical of the different modifications of the α form, would be largely present also in the disordered chain agglomerates of the mesomorphic form. The analysis of the structure of the mesomorphic form, through Fourier transform calculations on models constituted by large bundles of chains, provided further support that this form contained small and imperfect crystals of the α crystalline form [8]. Manfredi et al. [76] studied the structural changes in the δ form after consecutive annealing from 90 to 140 °C. The X-ray diffraction patterns of the emptied δ-form sample show the gradual disappearance of the peaks typical of the δ form up to the annealing temperature Ta = 110 °C. At this temperature only two broad peaks, centered at 2θ = 10 ° and 19.5 °, are present. The pattern remains unchanged for long annealing times and corresponds to a mesomorphic form. By annealing at higher temperatures, the typical peaks of the γ form gradually appear, and a well-developed pattern of the γ form is obtained for Ta = 130 °C, indicating that the recrystallization into the γ form occurred only well above the glass transition when the mobility of the chains was sufficiently high. The thermal transition of the δ form was also studied by de Candia et al. [77] by using different techniques such as thermal analysis, thermogravimetry X-ray diffraction, and infrared spectroscopy. Experimental evidence showed that the solvent included in the crystal lattice was partially released on heating the sample within the range of stability of the δ form. The solvent release depends on the time and temperature of annealing. Furthermore, the transition δ to γ occurs through an intermediate form characterized by conformational order, without crystalline order. This mesomorphic form was found to be impermeable to the vapor of dichloromethane at low activity, and it was possible to calculate its fraction. In the transition from δ to γ form, the value of crystallinity, as derived from the X-ray diffractograms, remains very similar (35% in the δ and 38% in the γ form), but there is a significant increase in the fraction of mesophase (from 20% in the δ form to 37% in the γ form).

9.5


Several authors have addressed crystallization measurements and kinetics of SPS [23–26,78–82]. The experiments were carried out in both isothermal and non-isothermal mode by means of differential calorimetry techniques (DSC) [24–26,78–82]. The effect of the processing parameters, in particular of the melt temperature, the cooling rate, the crystallization temperature and pressure, the annealing time and the crystallization time have been extensively studied [23,25,78–80]. The melt crystallization toward the α form is favored by fast cooling or by low isothermal temperatures. Crystallization at high temperatures (close to

176


the melting temperature) or under a slow cooling rate from the melt leads to the formation of the β form; otherwise always a mixture of the two phases (α and β) is obtained [81]. The analysis of these findings led to the conclusion that the β-crystal structure is thermodynamically more stable; however at low temperatures, the α phase is kinetically preferred. Ho et al. [26,82] studied the possibility of transformation from α phase to β phase during both isothermal tests and heating ramps. Under the thermal conditions analyzed, the transition was found to occur only after complete α crystallization [26,82]. Wesson [24], Sudduth et al. [83], and Chen [25] have analyzed the crystallization kinetics of SPS under isothermal and non-isothermal conditions using DSD techniques. All these authors found a good agreement between experimental data and the Hoffman-Lauritzen kinetic model. Chen et al. [25], in particular, observed a change in the kinetic regime at about Tc = 239 °C. They also estimated the values of growth rate parameters, such as lateral surface-free energy, fold surface-free energy and the average work of chain folding [25]. Lawrence and Shinozaki [84], for an SPS with a molecular weight of MW = 372,000 g/mol, determined the crystallization parameters by simultaneously fitting data of both melt and cold isothermal crystallization. The parameters identified were used to predict the crystallization behavior during non-isothermal experiments. The predictions were found to agree with the data of crystallization from the melt; vice versa, the predicted rates underestimated the data of crystallization from the amorphous solid. La Carubba et al. [85] proposed a two-phase model to describe the crystallization behavior of SPS. The samples were analyzed by macroscopic methods, such as density, WAXD, and microhardness (MH) measurements. The density was strictly related to the phase content, as confirmed by WAXD deconvolution [85]. Sorrentino et al. [81], for an SPS with a molecular weight of MW = 244,000 g/ mol, determined the kinetic parameters of each of the two phases (α and β) by controlling the memory effect. Afterward, the investigations were carried out in a very wide range of experimental conditions by using two different experimental techniques: (a) scanning calorimetry differential thermal analysis (DTA), exploring the high temperatures (close to the melting point), and low cooling rates ranges; (b) analysis of thin samples obtained by non-conventional quenching experiments, able to provide data of crystallization kinetics behavior at very low temperatures [86]. Experimental results were implemented in a multiphase kinetic model and discussed in terms of interactions between the two phases during the simultaneous growth [87]. The analysis of the melt crystallization data provided an Avrami index varying from about 2 to about 3.5. Lawrence et al. [84] also reported a value of n ranging between 2 and 3. The effect of pressure on crystallization behavior has been generally attributed to the effect of increasing melting point with pressure, which in turn is equivalent to amplifying the degree of supercooling.


177

However, as reported by Hohne [88,89], this behavior is not true for all the crystal phases. In particular, for the SPS, the α phase shows a decrease in the melting point with pressure, whereas that of the β phase shows an opposite trend. The opposite behavior of the melting temperature of the two crystal phases with pressure is a consequence of the fact that the density of the amorphous phase is smaller than the density of the β and larger than the density of the α phase. Furthermore, the glass transition temperature was found to considerably increase with pressure. Thus, a pressure increase produces a strong reduction in the crystallization range of the α phase, whereas the amplitude of the crystallization range of the β phase is almost unchanged (it is rigidly shifted versus higher temperature). 9.5.1

Thermodynamic and Kinetics of Crystallization

A connection between stability and kinetics data were established by Sorrentino et al. [79] (Fig. 9.10). A schematic representation of the thermodynamic stability lines of the two phases are reported in the left-hand side of Figure 9.10 as a function of the reciprocal lamellae thickness. At any value of the lamellae thickness, the stable phase is presented as a solid line, whereas the meta-stable phase is presented with a broken line. The intersection of the phase lines defines a triple point Q, where all three phases (the melt, the α, and the β crystals) can coexist as stable phases. From the viewpoint of kinetics, the smaller the critical nucleus, the faster the crystallization rate of crystalline phase. The crystallization rate of the α form is thus expected to be faster than that of the β form in the temperature range approaching the glass temperature and vice versa close

T Thermodynamics

T

Kinetics

Kb

Tma Tt

Ka

Standard experimental range

Tmb

Tg ct–1

Figure 9.10

Lamellae thickness –1

Log crystallization rate (1/s)

Comparison between kinetics and stability diagram of the SPS [79].

178


to the melting temperature. In the case of SPS, this conclusion has been further approved by the comparison of crystallization kinetic constants of the two phases. The maximum crystallization rate of the β form was found to be about 10 times smaller than that of the α form; however at high temperatures (close to melting points of the two phases), the relation inverts and the crystallization rate of the β form becomes considerably higher than that of the α form [23].

9.6

MELTING BEHAVIOR

Multiple melting peaks (two to four peaks) have been observed on heating scans of SPS samples that present a combination of α and β crystals. However, multiple melting peaks can also be observed in samples that are known to contain only a single type of unit cell [73]. Apparently, the multiple peaks phenomenon cannot be entirely attributed to multiple types of unit cells. Variation in the lamellar morphology may also be responsible for the complex thermal behavior. The number of melting peaks decreases with increasing scan rates. By contrast, only one broad and irregularly shaped peak for cold-crystallized SPS is observed [73]. Two main mechanisms have been commonly debated and proposed to explain the phenomenon of multiple melting endotherms: dual/multiple modifications and reorganization via melting/recrystallization/remelting [76]. In one case, each of the multiple melting peaks can be attributed to the various crystalline substructures, such as unit cells, lamellae, or spherulites, present in the polymer. In the alternative interpretation, multiple melting was attributed to the melting of thinner lamellae/crystals, recrystallization to thicker ones, and remelting of the thickened lamellae during DSC scanning. Woo et al. [90] suggest that a combination of these two mechanisms might be appropriate for providing a plausible explanation of the multiple melting behavior in SPS systems. Although there have been numerous studies, multiple melting peaks in relation to particular crystal forms in SPS have yet to be understood. Woo and Wu [91] found three melting peaks on heating SPS (Mw = 241,000 g/ mol) crystallized at a low temperature (250 °C). Furthermore, the disappeared endotherm was ascribed to the highest one (β′-form associated).

179

MELTING BEHAVIOR

Tmax=380 °C (0.5 min) and melt-crystallized at various temperatures for 120 min

P-II P-II α″ crystal only Pa,α P-II

Pa,α

P-IV P-IV P-II

Pa,α

230 (a)

Tc = 260°C

245 260 275 Temperature (°C)

250°C

240°C

P-I+P-III

Endothermic heat flow (offset scale)→

Endothermic heat flow (offset scale)→

Tmax=280 °C (1 min) and melt-crystallized at various temperatures for 120 min

P-I

β′ crystal only

Pa,β P-I

Tc = 260°C P-III 250°C

Pa,β

230°C

290

Pa,β P-III

Pa,β

230

240°C P-I

245 260 275 Temperature (°C)

230°C 290

(b)

Figure 9.11 DSC traces in SPS melt-crystallized at 230, 240, 250, and 260 °C after annealing at: (a) 280 °C and (b) 380 °C. Reprinted from Woo et al. [78], with permission from Elsevier.

Melt recrystallization of β crystals is evidenced by the presence of a small exothermic peak between the lowest peak (due to the melting of β crystals) and the intermediate melting peak (due to the α crystals) during heating under appropriate conditions [93]. Woo et al. [94] used a low molecular weight SPS (63,000 g/mol) in order to produce samples crystallized only in the β form. The samples obtained in this way show three different melting peaks. The lowest and the intermediate melting peaks are attributed to the preexisting thinner and less stable lamellae, whereas the highest melting peak is due to the lamellae formed only at high melt crystallization temperatures, or via reorganization upon melting of preexisting lamellae. Cold-crystallized SPS samples significantly differ from melt-crystallized SPS either in crystal forms or in melting behavior [95]. X-ray results indicated that only α-type crystal existed in cold-crystallized SPS, which yielded a single but broad-based melting endotherm, with the peak temperature remaining almost constant at 269 °C regardless of the temperature of cold crystallization (200 °C, 260 °C). Using DSC, Giannotti and Valvassori [96] studied the melting enthalpy of SPS crystals in the presence of diluents by a melting-temperature depression approach. A ΔHf value of 82.4 J/g was obtained, based on Flory’s theory of polymer solutions. In contrast, a relatively low value ΔHf = 53.2 J/g was

180


reported by Pasztor et al. [97] using extrapolation from the plot of heat capacity change at Tg versus the observed heat of fusion. Sorrentino et al. [81] found that the value of the melting enthalpy for each phase is essentially constant with Tc. In particular, values of 49.7 ± 2.5 and 56 ± 6 2.5 J/g were found for the α and the β phases, respectively. The discrepancy between the value reported by Giannotti [96] and that reported later [81,97] might be attributed to the polymorphic nature of SPS and to the solvent interactions. 9.6.1

Equilibrium Melting Temperature of α and β Crystals

While the characterization of the crystallization rates as a function of temperature is important in the evaluation of many processing characteristics, the measurement of the thermodynamic melting temperature is currently very important in the evaluation of the crystallization model. The presence of multiple melting peaks during the DSC heating scans of SPS samples strongly complicates the determination of this parameter. By birefringence measurements Cimmino et al. [95] determined the equilibrium melting temperature of SPS (Mw = 710,000 g/mol) to be 275 °C. In contrast, Arnauts et al. [98] found a value of 285.5 °C from DSC measurements (Mw = 79,000 g/mol). An even higher melting temperature of 291.5 °C (288 °C if considering zero heating rate) was proposed by Gvozdic and Meier [99–100] on SPS specimens with Mw = 360,000 g/mol. The authors proposed a detailed stepwise annealing process in which the temperature changes and the time of annealing at each temperature were closely controlled. Su et al. [30] performed observations of melting behavior of the α and β polymorphs by means of temperature-programmed X-ray diffraction. Results indicated that the highest sustainable temperature identifiable via WAXD using stepwise annealing at increasingly higher temperatures (Ta) for the perfected (with the initial crystallization temperature Tc = 245 °C, followed by annealing at stepwise at above 250 °C) the α″ phase may be at least 286 °C. In a similar manner, the highest sustainable temperature of the perfected (with Tc = 265 °C, followed by annealing at stepwise at Ta above 275 °C) the β″ phase may be at least 280 °C. It thus follows that equilibrium melting of the α and the β phases should occur at temperatures higher than 286 and 280 °C, respectively. Ho et al. [90] using different annealing temperatures were able to prepare samples crystallized either in neat α or β phase. For a resin with Mw = 67,000 g/ mol, the authors [90] were able to determine the thermodynamic melting temperature for both phases. They found an equilibrium melting temperatures equal to 272 and 278.6 °C for α and β crystals, respectively. Wang et al. [84] (Mw = 200,000 g/mol), using a press-molding procedure for preparing the specimens, reported the equilibrium melting temperatures equal to 281.3 and 291 °C for the α and β forms, respectively. In addition, they found that the

MELTING BEHAVIOR

181

Table 9.2 Summary of the Equilibrium Melting Temperatures Reported in Literature Temperature

Method

Values ( °C)

Reference

0 m

Birefringence measurements

275

95

0 m 0 m

DSC measurements

285.5

98

Stepwise annealing process

288

99–100

T T T

0 mα

0 mβ

X-ray diffraction annealing

286; 280

30

0 mα

0 mβ

T ;T

Linear extrapolation

272; 282

86

Tm0 α ; Tm0 β Tm0 α ; Tm0 β

Linear extrapolation Linear extrapolation

272; 278.6 281.3; 291

90 84

Tm0 α ; Tm0 β

T ;T

Nonlinear extrapolation

294; 320

84

0 mβ

T ;T


282; 295

86

Tm0 α ; Tm0 β


281; 288

85

0 mα

0 mα

0 mβ

Kinetic data

272; 283

86

0 mα

0 mβ

Gibbs–Thomson plot

282; 292

86

0 mα

0 mβ

Gibbs–Thomson plot

281; 292.7

78–115

T ;T

T ;T T ;T

assumptions on which the linear Hoffman-Weeks plot is based are not completely verified. A nonlinear extrapolation based on Marand’s method provided a value of 294 and 320 °C for the α and β forms, respectively [84]. Sorrentino et al. [86], using a protocol similar to that proposed by Ho et al. [84], determined the thermodynamic melting temperature for both α and β phases from the Hoffman-Weeks extrapolation. From such a linear extrapolation, the equilibrium melting temperature of the α form was determined as 272 °C, whereas the equilibrium melting temperature of the β form was determined as 282 °C (Table 9.2). Marand’s method was applied to the same experimental data. The resulting values for the equilibrium melting temperatures were 282 and 295 °C for the α and β forms, respectively. Woo et al. [85], adopting the nonlinear Hoffman-Weeks relation, found the equilibrium melting temperatures of α and β crystals to be 281 and 288 °C, respectively. It is interesting to note that for the β phase the experimental results reported by Sorrentino [86] are very close to the results obtained by Wang et al. [84] and to those reported by Woo et al. [101]. In spite of this, the results obtained for the equilibrium melting temperatures are quite different. These differences are mainly due to the different range of temperature investigated. However, all the literature results [84,85,86,101], and despite the extrapolation methods adopted, the thermodynamic equilibrium melting temperature of the β crystalline form in SPS was higher than that of the α form. On the basis of thermodynamic equilibrium, the higher value of the melting tempera-

182


ture in the β form may be attributed to the lower value of the entropy of fusion and/or to the higher value of heat of fusion. The orthorhombic chain packing of β-form crystals is more stable than the trigonal chain packing of the α-form crystals. However, the experimental melting temperature of the single-α form is higher than the melting temperature of the single-β form at low crystallization temperatures. This result seems to suggest that the SPS polymorphism exhibits a phenomenon of phase stability inversion with lamellar size [84,85]. Sorrentino et al. [86] determined the reference temperature for the crystallization kinetics (also called “zero growth rate temperature”) with the objective of providing a general discussion on problems occurring in their determination and of identifying the relationship between this temperature and the equilibrium melting temperature. It was found that the thermodynamic melting temperature resulting from DSC analysis provides a poor description when applied to the crystallization kinetic data. The reference temperature to be adopted in crystallization kinetic equations was found to be significantly lower. The discrepancy is probably related to the reliability of the kinetic model adopted to describe the experimental data [86]. 9.6.2

Memory Effects

The overall crystallization kinetics and morphology in semicrystalline polymer is characterized and mainly controlled by the presence of infusible heterogeneous nuclei as well as by the processing history. It is therefore important that the influences of impurities, additives, nucleating agents, and especially “crystalline memory effect” on crystallization behavior be thoroughly investigated. The latter refers to the cluster of molecules that retain their crystallographic arrangement of crystals as a result of insufficient or partial melting conditions. Memory of the α-form crystals in SPS is maintained, at least for a short time, in a wide temperature range above the melting temperature. De Candia et al. [81] suggested that it could be related to the existence of small agglomerates of chains in the mesomorphic form, which dissolve under very long annealing time. A comparison between the X-ray diffraction intensity of oriented samples of SPS in the mesomorphic form and the infrared spectra indicates that the mesomorphic form consists of small and imperfect crystals of the α crystalline form [8]. The effect of annealing temperature (Tann) on the polymorphism of SPS was first observed by Guerra et al. [4]. In samples annealed at high Tann (50 °C above Tm) only β crystals are formed, whereas in samples annealed at low Tann (close to Tm) only crystals with α form are obtained [4]. Chiu et al. [19,102] investigated the annealing temperature (Tann above Tm) dependency of isothermal and non-isothermal melt crystallization kinetics of SPS. Experimental results show that the percentage of the α phase increases with a decrease in either the crystallization or the annealing temperature. De Rosa et al. [21] analyzed the effect of several annealing temperatures (from 280 to 320 °C) and

STRUCTURE AND PROPERTIES OF THE CRYSTALLIZED SAMPLES

183

different annealing times (from 0.5 to 10 min) on the polymorphism of SPS. Their results show that the couple temperature-time determines the crystallization of the α and β phases: low annealing temperatures and short times allow the formation of the α phase, whereas higher temperatures and long times always induce a crystallinity in β form. An important opportunity related to a deep understanding of these memory effects is the possibility of obtaining SPS samples crystallized in a single crystalline phase [19,81,82,101,102]. Sorrentino et al. [81] determined the crystalline memory effect on the relative crystallinity content of each of the two phases after a given thermal treatment. This allowed the authors to define a kinetic of melting, which enabled a generalization of the effects of the couple time-temperature on the memory effect [81].

9.7 STRUCTURE AND PROPERTIES OF THE CRYSTALLIZED SAMPLES 9.7.1 Morphology of Injection Molded Samples The structural analysis of injection-molded SPS has recently attracted particular attention [103–109]. The final morphology distribution in the mold piece, characteristic of slowly crystallizing polymers, was found to be dependent on the cooling rate, injection speed, and packing pressure. All these parameters strongly affect the spectrum of relaxation times and the kinetics of crystallization. The effect is a development of a structure gradient in injection-molded parts [103]. Under common processing conditions, SPS forms a multilayer structure in the thickness direction. The development of the multilayer amorphous–semicrystalline–amorphous structure was explained by Hsiung et al. [107]. The formation of such a multilayer structure was a result of a complex interplay between the thermal and the stress history. Exemplar micrographs of injection molded SPS samples (about 2-mm thick) are shown in Figure 9.12 [106,109]. These samples show a distribution along the thickness direction of transparent, amorphous layers (white layers in the micrographs shown in Fig. 9.12), and opaque crystalline layers (black layers in the micrographs).

P=160 Bar

P=450 Bar

P=700 Bar

Figure 9.12 Micrographs of injection-molded samples in polarized light. The distinctive packing pressures are reported in the label.

184


As shown in Figure 9.12, an increase in the packing pressure produces a considerable enlargement of the amorphous core layer. Equivalently, a reduction in the molding temperature produces a thickening of all the amorphous layers, with particular effect on the skin layer. The situation is further influenced by the stress-induced crystallization. Lopez et al. [104] observed that the mold temperature produces significant differences in the macroscopic morphology and properties of the injected SPS samples. In particular, samples molded at high temperatures had higher resistance to the organic solvent than samples molded at low temperatures. The core of the molded samples analyzed by TEM appeared spherulitic [104]. However, the spherulites were not fully developed but appeared “sheaf structures” type with an elliptical profile. The intermediate region presented lamellar crystals oriented perpendicular to the flow direction. Evans et al. [105] found similar results through the depth of injectionmolded bars. In particular they found that the skin of samples molded with low mold temperatures was completely amorphous. The differences in the structure and morphology between the two groups are probably due to the techniques used for determining the crystallinity (WAXS and FTIR). Hsiung and Cakmak developed a structure-oriented model to simulate the crystalline structure developed in the injection molding of SPS [107]. Their model was elaborated by taking a Lagrangian approach and a three-dimensional mold geometry. The morphological structure of the SPS sample molded at a particular condition quantitatively matches the experimental observations. A cold compaction of SPS powders at a temperature well below its melting temperature (273 °C) was found possible [108]. In addition, the mechanical properties of the resulting material were comparable to those of the SPS submitted to compression molding after melting. Parallel experiments on poly(ethylene terephthalate) (PET) and linear low-density polyethylene (LLDPE) suggest that the behavior of SPS is a peculiarity of such polymers, which is likely connected to their polymorphic nature. 9.7.2 Relation between Morphology Structure, Processing, and Properties The temperature drastically affects the drawing behavior of SPS [36]. In the range 130–200 °C, the yield tension was found to decrease on increasing the temperature. After the yielding the drawing occurs at a constant τ in a strain range that depends on the drawing temperature [36]. Hot drawing was performed in the temperature range near the glass transition; it involved concurrent crystallization and orientation [110]. The longitudinal mechanical properties were correlated with the degree of orientation and crystallinity. Increasing crystallinity does not affect the mechanical properties to any large extent. In contrast, the orientation process produces a considerable increase in strength and modulus in the draw direction. In this latter case, crystalline unity cell was obtained with the axis preferentially aligned to the draw direc-

STRUCTURE AND PROPERTIES OF THE CRYSTALLIZED SAMPLES

185

tion. The anisotropy of the mechanical properties was measured using tensile stress and compared with a mathematical model. Failure and deformation mechanism have been studied and reported for SPS, and have been compared with atactic polystyrene [111,112]. The SPS material fails with a slow, controlled crack growth, while atactic polystyrene fails with a considerable crazing and some yield. Micrographs of fracture surface show that the spherulites are relatively stable structures, and it is the material at the interface of spherulites that draws and fails. After crack initiation, a layer structure appears near the boundary of the deformation zone. A mechanism of constrained crazing and void coalescence is proposed, with the spherulites themselves acting as stress concentrators for the initiation failure. The damage zone is then highly confined to the inter-spherulitic regions of the semicrystalline morphology. Nitta et al. [113] studied the crystallization mechanism of SPS using dynamic mechanical techniques. Temperature dependences of the storage E′ and loss modulus E″ were measured between −150 and 270 °C. All the samples showed a large decrease in the storage modulus at approximately 100 °C, which corresponds to the glass transition temperature. The magnitude of this change, however, varied from sample to sample. Samples slowly cooled from the melt showed a sigmoidal shape, whereas amorphous samples showed an increase in the modulus just after the Tg, due to crystallization of the meta-stable amorphous material. The loss modulus E″ presented a maximum in correspondence to the glass transition range, which was evidently related to the initial stage of the crystallization process. On samples previously drawn at different temperatures, de Candia et al. found similar behavior [36]. The analysis of the dynamic-mechanical behavior showed two effects: an appreciable increase in the elastic modulus on drawing and a relevant increase in the glass transition temperature. The modulus hardening can be considered as a further indication of the morphological and topological reorganization induced by drawing. The crystallization process of the glassy SPS film has been investigated by monitoring the dynamic loss tangent (tan δ) at a crystallization temperature above the glass transition temperature [113]. It was found that the SPS film isothermally heated during the induction period shows power law relaxation behavior or a critical point at which tan δ is independent of the frequency. For annealing times shorter than this critical time, tan δ decreases with increasing the frequency, as for a viscoelastic liquid. On the other hand, for annealing times longer than the critical time, tan δ increases with the frequency, as for a viscoelastic solid. To confirm the appearance of this critical state, the polarized infrared spectra and light scattering were measured during the straininduced crystallization point. The results suggest that three-dimensional network junctions, which act as a precursor of the crystal nuclei during straininduced crystallization, are formed at the critical state. Lawrence [77] found that samples slowly cooled from the melt are much stiffer than cold-crystallized samples with similar nominal amount of crystal-

186


linity. The authors showed that the reason for this difference is the constraint of the amorphous phase by the high-modulus lamellae, and the variation in phase contiguity. In slowly cooled samples the crystalline phase is mechanically contiguous and the inter-lamellar amorphous material is more highly constrained than in the quenched-annealed samples. Inoue et al. [114] performed dynamic viscoelasticity and dynamic birefringence measurements on a quenched amorphous sample of SPS. The data were analyzed with a modified stress optical role. The result showed that SPS had about two times larger stress-optical coefficient than atactic polystyrene, although the two polymers have approximately the same intrinsic birefringence. This suggests that the stereoregularity of SPS enhances the stiffness of the main chain, increasing the optical anisotropy of the segment. In other words, the Rouse segment size for SPS was estimated to be about two times larger than that for atactic polystyrene.

REFERENCES 1. Greis, O., Xu, Y., Asano, T., Petermann, J. Morphology and structure of syndiotactic polystyrene. Polymer, 30(4), 590–594 (1989). 2. Chatani, Y., Shimane, Y., Inoue, Y., Inagaki, T., Ishioka, T., Ijitsu, T., Yukinari, T. Structural study of syndiotactic polystyrene. 1. Polymorphism. Polymer, 33(3), 488-492 (1992). 3. Immirzi, A., de Candia, F., Iannelli, P., Zambelli, A., Vittoria, V. Solvent-induced polymorphism in syndiotactic polystyrene. Makromol. Chem. Rapid Commun., 9(11), 761–764 (1988). 4. Guerra, G., Vitagliano, V., De Rosa, C., Petraccone, V., Corradini, P. Polymorphism in melt crystallized syndiotactic polystyrene samples. Macromolecules, 23(5), 1539–1544 (1990). 5. De Rosa, C., Guerra, G., Petraccone, V., Corradini, P. Crystal structure of the α-form of syndiotactic polystyrene. Polymer Journal (Tokyo, Japan), 23(12), 1435–1442, (1991). 6. Chatani, Y., Shimane, Y., Ijitsu, T., Yukinari, T. Structural study on syndiotactic polystyrene: 3. Crystal structure of planar form I. Polymer, 34(8), 1625–1629 (1993). 7. Cartier, L., Okihara, T., Lotz, B. The α “superstructure” of syndiotactic polystyrene: A frustrated structure. Macromolecules, 31(10), 3303–3310 (1998). 8. Auriemma, F., Petraccone, V., Dal Poggetto, F., De Rosa, C., Guerra, G., Manfredi, C., Corradini, P. Mesomorphic form of syndiotactic polystyrene as composed of small imperfect crystals of the hexagonal (α) crystalline form. Macromolecules, 26(15), 3772–3777 (1993). 9. Chatani, Y., Shimane, Y., Inagaki, T., Ijitsu, T., Yukinari, T., Shikuma, H. Structural study on syndiotactic polystyrene: 2. Crystal structure of molecular compound with toluene. Polymer, 34(8), 1620–1624 (1993).

REFERENCES

187

10. Chatani, Y., Inagaki, T., Shimane, Y., Shikuma, H. Structural study on syndiotactic polystyrene: 4. Formation and crystal structure of molecular compound with iodine. Polymer, 34(23), 4841–4845 (1993). 11. Vittoria, V., De Candia, F., Iannelli, P., Immirzi, A. Solvent-induced crystallization of glassy syndiotactic polystyrene. Makromolekulare Chemie, Rapid Communications, 9(11), 765–769 (1988). 12. Vittoria, V., Ruvolo Filho, A., De Candia, F. Structural organization of syndiotactic polystyrene films crystallized in the β form. J. Macrom. Sci. Phys., 31(2), 133–148 (1992). 13. De Rosa, C., Rapacciuolo, M., Guerra, G., Petraccone, V., Corradini, P. On the crystal structure of the orthorhombic form of syndiotactic polystyrene. Polymer, 33(7), 1423–1428 (1992). 14. Tashiro, K., Ueno, Y., Yoshioka, A., Kobayashi, M. Molecular mechanism of solvent-induced crystallization of syndiotactic polystyrene glass. 1. Time-resolved measurements of infrared/Raman spectra and X-ray diffraction. Macromolecules, 34(2), 310–315 (2001). 15. Doherty, D. C., Hopfinger, A. J. Molecular modeling of polymers. 6. Intramolecular conformational analyses and molecular dynamics of syndiotactic polystyrene. Macromolecules, 22(5), 2472–2477 (1989). 16. Corradini, P., Napolitano, R., Pirozzi, B. Geometrical and energetical analysis of the chain conformations of syndiotactic polystyrene in the crystalline phase. Eur. Polym. J., 26(2), 157–161 (1990). 17. Napolitano, R., Pirozzi, B. Analysis of the modes of packing of the α and β forms of syndiotactic polystyrene. Macromolecules, 26(26), 7225–7228 (1993). 18. Haneda, T., Hashima, Y., Tanaka, K., Koyama, K. Crystallization behavior of syndiotactic polystyrene. Rep. Progr. Polym. Phys. Jpn., 32, 107–110 (1989). 19. Chiu, F., Shen, K., Tsai, S. H. Y., Chen, C. Pre-melting temperature effect on the isothermal melt crystallization and multiple melting behavior of syndiotactic polystyrene. Polym. Eng. Sci., 41(5), 881–889 (2001). 20. Li, Y., He, J., Qiang, W., Ho, X. Effect of crystallization temperature on the polymorphic behaviour of syndiotactic polystyrene. Polymer, 43, 2489–2494 (2002). 21. De Rosa, C., Ruiz de Ballesteros, O., Di Gennaro, M., Auriemma, F. Crystallization from the melt of α and β forms of syndiotactic polystyrene. Polymer, 44, 1861– 1870 (2003). 22. Sun, Z., Morgan, R. J., Lewis, D. N. Crystallization of syndiotactic polystyrene under pressure. Polymer, 33, 660–661 (1992). 23. Sorrentino, A., Tortora, M., Vittoria, V. New developments in syndiotactic polystyrene. In Recent Research Developments in Applied Polymer Science, Pandalai, S. G. (ed.), Research Signpost, Kerala, India, 2006, Chapter 4, pp 57–87. 24. Wesson, R. D. Melt crystallization kinetics of syndiotactic polystyrene. Polym. Eng. Sci., 34(14), 1157–1160 (1994). 25. Chen, Q., Yu, Y., Na, T., Zhang, H., Mo, Z. Isothermal and nonisothermal meltcrystallization kinetics of syndiotactic polystyrene. J. Appl. Polym. Sci., 83, 2528– 2538 (2002). 26. Ho, R., Lin, P., Hseih, P., Chung, T., Tsai, H. Isothermal crystallization-induced phase transition of syndiotactic polystyrene polymorphism. Macromolecules, 34, 6727–6736 (2001).

188


27. Hodge, K., Prodpran, T., Shenogina, N. B., Nazarenko, S. Solid-state structure of thermally crystallized syndiotactic polystyrene. J. Appl. Polym. Sci., 83(12), 2705– 2715 (2002). 28. Bu, W., Li, Y., He, J., Zeng, J. An interpretation of the formation of α- and β-form crystals in bulk syndiotactic polystyrene. Macromolecules, 32, 7224–7225 (1999) 29. Cai, J., Han, Y. Morphology, structure, and kinetic analysis of nonisothermal cold- and melt-crystallization of syndiotactic polystyrene. J. Appl. Polym. Sci., 103, 1311–1324 (2007). 30. Su, C. H., Chen, S. H., Su, A. C., Tsai, J. C. Crystallization and melting of α and β phases in bulk syndiotactic polystyrene as monitored in situ via high-temperature wide-angle X-ray diffraction. Journal of Polymer Research, 11(4), 293–298 (2005). 31. Vittoria, V., Ruvolo Filho, A., de Candia, F. Structure of syndiotactic polystyrene thermally crystallized from the glassy state. J. Macromol. Sci. Phys., B29(4), 411–428 (1990). 32. Matsuba, G., Kaji, K., Nishida, K., Kanaya, T., Imai, M. Conformational change and orientation fluctuations prior to the crystallization of syndiotactic polystyrene. Macromolecules, 32(26), 8932–8937 (1999). 33. Wang, C., Lin, C-C., Chu, C-P. Crystallization and morphological features of syndiotactic polystyrene induced from glassy state. Polymer, 46, 12595–12606 (2005). 34. Handa, Y. P., Zhang, Z., Wong, B. Effect of compressed CO2 on phase transitions and polymorphism in syndiotactic polystyrene. Macromolecules, 30, 8499–8504 (1997). 35. Zhang, Z., Handa, Y. P. CO2-assisted melting of semicrystalline polymers. Macromolecules, 30, 8505–8507 (1997). 36. de Candia, F., Romano, G., Russo, R., Vittoria, V. Drawing and dynamicmechanical behaviour of syndiotactic polystyrene. An introduction. Colloid Polym. Sci., 268(8), 720–725 (1990). 37. Dikshit, A. K., Kaito, A. Effects of the oriented mesophase on the cold crystallization of syndiotactic polystyrene and its blend with poly(2,6-dimethyl-1,4phenylene oxide). J. Polym. Sci. Part B: Polym. Phys., 41(14), 1665–1675 (2003). 38. Rizzo, P., Daniel, C., De Girolamo Del Mauro, A., Guerra, G. Nuova forma cristallina nanoporosa del polistirene sindiotattico e forme cristalline a complesso molecolare derivate. It. Pat. N. SA2006A22, 2006. 39. Rizzo, P., Daniel, C., De Girolamo Del Mauro, A., Guerra G. New host polymeric framework and related polar guest cocrystals. Chem. Mater., 19(16), 3864–3866 (2007). 40. Albunia, A. R., Musto, P., Guerra, G. FTIR spectra of pure helical crystalline phases of syndiotactic polystyrene. Polymer, 47(1), 234–242 (2006). 41. De Rosa, C., Rizzo, P., de Ballesteros, O. R., Petraccone, V., Guerra, G. Crystal structure of the clathrate delta form of syndiotactic polystyrene containing 1,2-dichloroethane. Polymer, 40, 2103–2107 (1999). 42. Yamamoto, Y., Kishi, M., Amatharani, D., Sivakumar, M., Tsujita, Y., Yoshimizu, H. Study on .delta.-form complex in syndiotactic polystyrene-organic molecules system III. Fine structures of complexes with xylene isomers by means of highresolution solid-state 13C NMR spectroscopy. Polymer, 35(6), 465–469 (2003).

REFERENCES

189

43. Tarallo, O., Petraccone, V. On the crystal structure of the clathrate forms of syndiotactic polystyrene containing carbon disulfide and iodine. Macromol. Chem. Phys., 205(10), 1351–1360 (2004). 44. Vittoria, V., Russo, R., de Candia, F. Solvent-induced crystallization of syndiotactic polystyrene in different liquids. Sorption and diffusion phenomena. Polymer, 32(18), 3371–3375 (1991). 45. Vittoria, V., Russo, R., de Candia, F. Solvent-induced crystallization of syndiotactic polystyrene in toluene. Makromol. Chem. Macromol. Symp., 39, 317–321 (1990). 46. Vittoria, V., Ruvolo Filho, A., de Candia, F. Vapour induced ordering phenomena in syndiotactic polystyrene amorphous films. Polym. Bull., 26(4), 445–449 (1991). 47. de Candia, F., Carotenuto, M., Guadagno, L., Vittoria, V. Noncrystalline order in swollen syndiotactic polystyrene. Macromolecules, 25(23), 6361–6364 (1992). 48. de Candia, F., Guadagno, L., Vittoria, V. Phase organization of solvent-crystallized syndiotactic polystyrene films. J. Macromol. Sci., B33(3–4), 347–356 (1994). 49. Daniel, C., De Luca, M. D., Brulet, A., Menelle, A., Guenet, J. M. Thermoreversible gelation of syndiotactic polystyrene in benzene. Polymer, 37(7), 1273–1280 (1996). 50. Rastogi, S., Goossens, J. G. P., Lemstra, P. J. An in situ SAXS/WAXS/Raman spectroscopy study on the phase behavior of syndiotactic polystyrene (sPS)/ solvent systems: Compound formation and solvent (dis)ordering. Macromolecules, 31(9), 2983–2998 (1998). 51. van Hooy-Corstjens, C. S. J., Magusin, P. C. M. M., Rastogi, S., Lemstra, P. J. A comparative study on gels and clathrates of syndiotactic polystyrene: Solvent mobility in polymer-solvent compounds. Macromolecules, 35(17), 6630–6637 (2002). 52. Petraccone, V., Tarallo, O., Venditto, V., Guerra, G. An intercalate molecular complex of syndiotactic polystyrene. Macromolecules, 38(16), 6965–6971 (2005). 53. Malik, S., Rochas, C., Guenet, J. M. Syndiotactic polystyrene intercalates from naphthalene derivatives. Macromolecules, 38(14), 6024–6030 (2005). 54. Rizzo, P., Albunia, A. R., Guerra, G. Polymorphism of syndiotactic polystyrene: Gamma phase crystallization induced by bulky non-guest solvents. Polymer, 46(23), 9549–9554 (2005). 55. Musto, P., Mensitieri, G., Cotugno, S., Guerra, G., Venditto, V. Probing by timeresolved FTIR spectroscopy mass transport, molecular interactions, and conformational ordering in the system chloroform-syndiotactic polystyrene. Macromolecules, 35(6), 2296–2304 (2002). 56. Manfredi, C., Del Nobile, M. A., Mensitieri, G., Guerra, G., Rapacciuolo, M. Vapor sorption in emptied clathrate samples of syndiotactic polystyrene. J. Polym. Sci. Polym. Phys., 35(1), 133–140 (1997). 57. Reverchon, E., Guerra, G., Venditto, V. Regeneration of nanoporous crystalline syndiotactic polystyrene by supercritical CO2. J. Appl. Polym. Sci., 74(8), 2077– 2082 (1999). 58. Daniel, C., Alfano, D., Guerra, G., Musto, P. Physical gelation of syndiotactic polystyrene in the presence of large molar volume solvents induced by volatile guests of clathrate phases. Macromolecules, 36(5), 1713–1716 (2003).

190


59. Corradini, P., De Rosa, C., Guerra, G., Napolitano, R., Petraccone, V., Pirozzi, B. Conformational and packing energy of the crystalline α modification of syndiotactic polystyrene. Eur. Polym. J., 30(10), 1173–1177 (1994). 60. De Rosa, C. Crystal structure of the trigonal modification (α form) of syndiotactic polystyrene. Macromolecules, 29(26), 8460–8465 (1996). 61. Sun, Z., Miller, R. L. Crystalline structure of α-form syndiotactic polystyrene. Polymer, 34(9), 1963–1965 (1993). 62. Tosaka, M., Tsuji, M., Kohjiya, S., Cartier, L., Lotz, B. Crystallization of syndiotactic polystyrene in β-form. 4. Crystal structure of melt-grown modification. Macromolecules, 32(15), 4905–4911 (1999). 63. Tosaka, M., Hamada, N., Tsuji, M., Kohjiya, S., Ogawa, T., Isoda, S., Kobayashi, T. Crystallization of syndiotactic polystyrene in β-form. 1. Positional identification of stacking faults in the solution-grown single crystals. Macromolecules, 30(14), 4132–4136 (1997). 64. Hamada, N., Tosaka, M., Tsuji, M., Kohjiya, S., Katayama, K. Crystallization of syndiotactic polystyrene in β-form. 2. Quantification of stacking faults in the solution-grown single crystals. Macromolecules, 30(22), 6888–6892 (1997). 65. Napolitano, R., Pirozzi, B. Models of adjacent re-entry folds in the beta form of syndiotactic polystyrene. Conformational and packing analysis by molecular mechanics. Macromol. Theor. Simul., 11(5), 472–480 (2002). 66. Sun, Y. S., Woo, E. M. Correlation between thermal behavior and crystalline morphology in β-form syndiotactic polystyrene. Macromol. Chem. Phys., 202(9), 1557–1568 (2001). 67. Koyama, K., Soeno, T., Haneda, T., Yonetake, K., Ishizuka, O. Melting and crystallization behavior of syndiotactic polystyrene. Polym. Prep. Jpn., 37, 1173– 1175 (1988). 68. Cimmino, S., Di Pace, E., Martuscelli, E., Silvestre, C. Syndiotactic polystyrenebased blends: Crystallization and phase structure. Polymer, 34, 2799–2803 (1993). 69. Sun, Y. S., Woo, E. M. Mechanisms of reorganization of lamellae in syndiotactic polystyrene. J. Polym. Sci. Polym. Phys., 38(24), 3210–3221 (2000). 70. Lawrence, S. S., Shinozaki, D. M. Micromorphology-dependent mechanical properties of syndiotactic polystyrene. J. Mater. Sci., 33(16), 4059–4068 (1998). 71. Wang, C., Cheng, Y-W., Hsu, Y-C., Lin, T-L. Lamellar morphology and equilibrium melting temperature of syndiotactic polystyrene in β′ crystalline form. J. Polym. Sci. Part B: Polym. Phys., 40, 1626–1636 (2002). 72. Wang, C., Liao, W-P., Cheng, Y-W. Strong diffuse scattering and lamellar morphologies of syndiotactic polystyrene: Polymorphic effects. J. Polym. Sci. Part B: Polym. Phys., 41, 2457–2469 (2003). 73. Vittoria, V., Ruvolo Filho, A., de Candia, F. Structural organization of syndiotactic polystyrene films crystallized in the α form. J. Macromol. Sci. Phys., B30(1–2), 155–170 (1991). 74. de Candia, F., Ruvolo Filho, A., Vittoria, V. Strain-induced mesomorphic order in glassy syndiotactic polystyrene. Makromol. Chem. Rapid Commun., 12(5), 295–299 (1991). 75. de Candia, F., Guadagno, L., Vittoria, V. Transport properties of mesomorphic syndiotactic polystyrene. J. Macromol. Sci. Phys., B34(3), 273–282 (1995).

REFERENCES

191

76. Manfredi, C., De Rosa, C., Guerra, G., Rapacciuolo, M., Auriemma, F., Corradini, P. Structural changes induced by thermal treatments on emptied and filled clathrates of syndiotactic polystyrene. Macromol. Chem. Phys., 196(9), 2795–2808 (1995). 77. de Candia, F., Guadagno, L., Vittoria, V. Annealing of the delta form of syndiotactic polystyrene. J. Macromol. Sci. Phys., B34(1–2), 95–103 (1995). 78. Woo, E. M., Sun, Y. S., Yang, C.-P. Polymorphism, thermal behavior, and crystal stability in syndiotactic polystyrene vs. its miscible blends. Prog. Polym. Sci., 26, 945–983 (2001). 79. Sorrentino, A., De Santis, F., Titomanlio, G. Polymer crystallization under high cooling rate and pressure: A step towards polymer processing conditions. In Lecture Notes in Physics: Progress in Understanding of Polymer Crystallization, Reiter, G., Strobl, G. (eds.), Springer, Berlin, 2007, vol. 714, Chapter 16, pp 329–344. 80. Sorrentino, A., Pantani, R., Picarella, D., Titomanlio, G. A novel apparatus for solidification of polymer samples under simultaneous high pressures and high cooling rates. Review of Scientific Instruments, 76(8), 0839011–0839015 (2005). 81. Sorrentino, A., Pantani, R., Titomanlio, G. Kinetics of melting and characterization of the thermodynamic and kinetic properties of syndiotactic polystyrene. J. Polym. Sci. Part B: Polym. Phys., 45, 196–207 (2007). 82. Ho, R-M., Lin, C-P., Tsai, H-Y., Woo, E-M. Metastability studies of syndiotactic polystyrene polymorphism. Macromolecules, 33, 6517–6526 (2000). 83. Sudduth, R. D., Yarala, P. K., Sheng, Q. A comparison of induction time and crystallization rate for syndiotactic polystyrene. Polym. Eng. Sci., 42(4), 694–706 (2002). 84. Lawrence, S. S., Shinozaki, D. M. Crystallization of syndiotactic polystyrene. Polym. Eng. Sci., 37(11), 1825–1832 (1997). 85. La Carrubba, V., Piccarolo, S., Brucato, V. Solidification of syndiotactic polystyrene by a continuous cooling transformation approach. J. Polym. Sci. Part B: Polym. Phys., 45, 2688–2699 (2007). 86. Sorrentino, A., Pantani, R., Titomanlio, G. Melting and zero growth rate temperatures of syndiotactic polystryrene. Colloid Polym. Sci., 286(8–9), 983–991 (2008). 87. Sorrentino, A., Pantani, R., Titomanlio, G. Two-phase crystallization kinetic model of syndiotactic polystyrene. Submitted. 88. Hohne, G. W. H. High pressure differential scanning calorimetry on polymers. Thermochim. Acta, 332(2), 115–123 (1999). 89. van Hooy-Corstjens, C. S. J., Hohne, G. W. H., Rastogi, S. Inverse melting in syndiotactic polystyrene. Macromolecules, 38(5), 1814–1821 (2005). 90. Woo, E. M., Wu, F. S. Polymorphic crystal forms and morphology of syndiotactic polystyrene in miscible states. J. Polym. Sci. Polym. Phys., 36(15), 2725–2735 (1998). 91. Woo, E. M., Wu, F. S. On the multiple melting behavior of polymorphic syndiotactic polystyrene and its behavior in a miscible state. Macromol. Chem. Phys., 199(9), 2041–2049 (1998). 92. Wang, C., Hsu, Y. C., Lo, C. F. Melting behavior and equilibrium melting temperature of syndiotactic polystyrene in α and β crystalline forms. Polymer, 42(20), 8447–8460 (2001).

192


93. Yuan, Z., Song, R., Shen, D. Study of multiple melting behaviour of syndiotactic polystyrene in β-crystalline form. Polym. Int., 49(11), 1377–1382 (2000). 94. Woo, E. M., Sun, Y. S., Lee M. L. Crystal forms in cold-crystallized syndiotactic polystyrene. Polym. Commun., 40(15), 4425–4429 (1999). 95. Cimmino, S., Di Pace, E., Martuscelli, E., Silvestre, C. Syndiotactic polystyrene: Crystallization and melting behavior. Polym. Commun., 32(6), 1080–1083 (1991). 96. Gianotti, G., Valvassori, A. Fusion enthalpy and entropy of syndiotactic polystyrene. Polymer, 31(3), 473–475 (1990). 97. Pasztor, A. J., Landes, B. G., Karjala, P. J. Thermal properties of syndiotactic polystyrene. Thermochim. Acta, 177, 187–195 (1991). 98. Arnauts, J., Berghmans, H. Equilibrium melting behaviour of syndiotactic polystyrene. Polym. Commun., 31(9), 343–345 (1990). 99. Gvozdic, N. V., Meier, D. J. On the melting temperature of syndiotactic polystyrene: 2 enhancement of the melting temperature of semicrystalline polymers by a novel annealing procedure. Polym. Commun., 32(16), 493–494 (1991). 100. Gvozdic, N. V., Meier, D. J. On the melting temperature of syndiotactic polystyrene. Polym. Commun., 32(6), 183–184 (1991). 101. Vittoria, V. Syndiotactic polystyrene. In Handbook of Termoplastics, Olabisi, O. (ed.), Dekker, New York, 1997, pp. 81–106. 102. Chiu, F-C., Peng, C-G., Fu, Q. Non-isothermal crystallization and multiple melting behavior of syndiotactic polystyrene—Pre-melting temperature effects. Polym. Eng. Sci., 40(11), 2397–2406 (2000). 103. Ulcer, Y., Cakmak, M., Miao, J., Hsiung, C. M. Structural gradients developed in injection-molded syndiotactic polystyrene (sPS). J. Appl. Polym. Sci., 60(5), 669– 691 (1996). 104. Lopez, L. C., Cieslinski, R. C., Putzig, C. L., Wesson, R. D. Morphological characterization of injection moulded syndiotactic polystyrene. Polymer, 36(12), 2331–2341 (1995). 105. Evans, A. M., Kellar, J. C., Knowles, J., Galiotis, C., Carriere, C. J., Andrews, E. H. The structure and morphology of syndiotactic polystyrene injection molded coupons. Polym. Eng. Sci., 37(1), 153–165 (1997). 106. Pantani, R., Sorrentino, A., Speranza, V., Titomanlio, G. Morphology distribution in injection moldings of syndiotactic polystyrene. PPS-Polymer Processing Society, 18th Annual Meeting, Taipei, Taiwan, November 4–8, 2002. 107. Hsiung, C. M., Cakmak, M. A Structure oriented model to simulate the shearinduced crystallization in injection molded polymers: A Lagrangian approach. Polymer, 37(20), 4555–4571 (1996). 108. Abbondanza, L., Garbassi, F., Gennaro, A., Giannotta, G., Marra, G., Po, R., Toscani, F. Molding of syndiotactic polystyrene under its melting temperature. J. Appl. Polym. Sci., 80(3), 377–383 (2001). 109. Pantani, R., Sorrentino, A., Speranza, V., Titomanlio, G. Injection moldings of syndiotactic polystyrene: Effect of processing conditions on morphology. IcheaP6, the Sixth Italian Conference on Chemical and Process Engineering, Pisa, Italy, June 8–11, 2003. 110. Yan, R. J., Ajji, A., Shinozaki, D. M. Mechanical properties of uniaxially oriented syndiotactic polystyrene. J. Mater. Sci., 34(10), 2335–2344 (1999).

REFERENCES

193

111. Jones, M. A., Carriere, C. J., Dineen, M. T., Balwinski, K. Failure and deformation studies of syndiotactic polystyrene. J. Appl. Polym. Sci., 64(4), 673–681 (1997). 112. Wu, S., Bubeck, R. A., Carriere, C. J. The effect of processing condition on the fracture behavior of syndiotactic polystyrene films. J. Appl. Polym. Sci., 62(9), 1483–1490 (1996). 113. Nitta, K., Nakatani, H., Yamada, H., Yamaguchi, M., Soga, K. Mechanical characterization of the crystallizing mechanism of syndiotactic polystyrene. Macromolecules, 31(10), 3317–3321 (1998). 114. Inoue, T., Ryu, D. S., Osaki, K., Takebe, T. Viscoelasticity and birefringence of syndiotactic polystyrene. I. Dynamic measurement in supercooled state. J. Polym. Sci. Polym. Phys., 37(4), 399–404 (1999). 115. Wang, C., Chen, C-C., Hung, C-H., Lin, K-S. Lamellar morphologies and crystal stability of syndiotactic polystyrene in α-crystalline form. Polymer, 45, 6681–6689 (2004).

CHAPTER 10

Preparation, Structure, Properties, and Applications of Co-Crystals and Nanoporous Crystalline Phases of Syndiotactic Polystyrene GAETANO GUERRA, ALEXANDRA ROMINA ALBUNIA, and CONCETTA D’ANIELLO Dipartimento di Chimica, University of Salerno, Fisciano (SA), Italy

10.1

INTRODUCTION

Systems composed of solid polymers and of low-molecular-mass molecules find several practical applications, including advanced applications [1–12]. In several cases, additives (often improperly referred to as guest molecules) are simply dispersed at the molecular level in polymeric amorphous phases, although frequently to reduce their diffusivity the active molecules are covalently attached to the polymer backbone, either by polymerizing suitable monomeric units or by grafting the active species onto preformed polymers [1–12]. The polymerization technique has been often limited by the difficulties in synthesizing and polymerizing highly functionalized monomers. The grafting technique has been limited by the often poor stability toward oxygen or water of the reactive polymeric substrates as well as by the need for several synthetic steps. In recent years, to reduce diffusivity of active molecules in the solid state and to prevent their self-aggregation, dendrimers have also been used [13–16]. A more simple alternative method to reduce diffusivity of active molecules in solid polymers and to prevent their self-aggregation consists of the formation of co-crystals with suitable polymer hosts. Polymeric co-crystalline phases are quite common for several regular and stereoregular polymers like syndiotactic polystyrene (SPS) [17–24], syndiotactic poly-p-methyl-styrene [25–31], syndiotactic poly-p-chloro-styrene [32], polyethyleneoxide [33–37], poly (muconic acid) [38,39], polyoxacyclobutane [40], or syndiotactic polymethylSyndiotactic Polystyrene, Edited by Jürgen Schellenberg Copyright © 2010 John Wiley & Sons, Inc.

194

CO-CRYSTALS

195

methacrylate [41]. Co-crystalline phases of syndiotactic polystyrene present the advantage to be formed by a cheap and robust polymer and are more flexible than those obtained from other polymers. In fact, different kinds of co-crystalline phases can be easily obtained with a large number of different guest molecules. Structure, preparation, properties, and applications of co-crystalline phases based on SPS are described in section 2 of this chapter. The removal of the low-molecular-mass guest molecules from polymer cocrystals generates host chain rearrangements, generally leading to crystalline phases that, as usual for polymers, exhibit a density higher than that of the corresponding amorphous phase. However, in the case of SPS [42–44] (and of SPS based copolymers) [32,45,46], by using suitable guest removal techniques [47,48], two different nanoporous crystalline phases (δ [49–52] and ε [53–56]), exhibiting a density (ρ = 0.98 g/cm3) definitely lower than that of the amorphous SPS (ρ = 1.05 g/cm3), can be obtained. In particular, the empty space is organized as isolated cavities and channels, for δ and ε crystalline phases, respectively. In this respect, it is worth noting that nanoporous crystalline structures can be achieved for a large variety of chemical compounds: inorganic (e.g., zeolites) [57,58], metal-organic [59–62] as well as organic [63–66]. These materials, often referred to as inorganic, metal-organic, and organic “frameworks” are relevant for molecular storage, recognition, and separation techniques. To our knowledge, SPS is presently the only polymer that is able to form “polymeric frameworks,” that is, semicrystalline polymeric materials presenting a nanoporous crystalline phase. Structure, preparation, properties, and applications of the nanoporous crystalline phases of syndiotactic polystyrene are described in section 3 of this chapter.

10.2

CO-CRYSTALS

As discussed in detail in the previous chapter of this book, SPS is a stereoregular polymer presenting a very complex polymorphic behavior including five crystalline phases (two of which being nanoporous) and is capable to form co-crystals with several low-molecular-mass guest molecules [17,19,21–24]. For most SPS co-crystals [17,19,21], isolated molecules are imprisoned as guests into cavities formed between layers of alternated enantiomorphous s (2/1)2 polymer helices, typical of the nanoporous δ phase. These SPS cocrystals have been defined as clathrate and are generally characterized by a guest/monomer-unit molar ratio 1/4. Due to the limited volume of the cavities, guests of these SPS clathrate phases have a molecular volume lower than 0.26 nm3 and exhibit orientation of their molecular plane nearly perpendicular to the axis of the polymer helices [17,19,21]. Recently, the occurrence of a second class of SPS co-crystals, defined as intercalate rather than clathrate, has been established [22–24]. Also, these co-

196

NANOPOROUS CRYSTALLINE PHASES OF SYNDIOTACTIC POLYSTYRENE

crystals are characterized by the same layers of alternated enantiomorphous s (2/1)2 polymer helices. However, in these co-crystals the guest molecules are not isolated into host cavities but contiguous inside layers intercalated with the polymer layers. Of course, these intercalate structures present a higher guest content with respect to the clathrate structures and the guest/monomer-unit molar ratio is generally 1/2 rather than 1/4. Presently known SPS intercalate phases present guest molecular volume in the range 0.15–0.36 nm3 [23,24]. More recently, the occurrence of a third class of SPS co-crystals has been established [53–56]. In this third class of SPS co-crystals, guest molecules are imprisoned into channels formed between enantiomorphous s (2/1)2 polymer helices, typical of the nanoporous ε phase [55]. As a consequence, longer guest molecules can be hosted by placing their molecular plane nearly parallel to the to the axis of the polymer helices [53,54]. The two classes of SPS co-crystals with guest molecules imprisoned in cavities and channels will be thereafter defined as δ clathrates and ε clathrates, respectively. 10.2.1

Crystalline Structures

Early characterization studies [42,67] on SPS had already recognized the occurrence of co-crystalline phase between SPS and low-molecular-mass molecules. However, the first crystalline structure of an SPS co-crystal (with toluene) has been described by Chatani et al. in 1993 [17]. Presently, the crystalline structure of several SPS co-crystals has been determined by X-ray diffraction, generally by measurements on axially oriented samples. In the case of well-ordered polymer co-crystals, this procedure allows, with a good accuracy level, determination of the guest location and orientation with respect to the axes of the crystalline phases. In some cases, relevant structural information has also been given by electron diffraction studies on solution-grown polymer single crystals [55]. The crystal structures of the three classes of SPS co-crystals present as a common feature the polymer conformation s (2/1)2 helices with a repetition period of nearly 0.78 nm. 10.2.1.1 δ Clathrates The structures of SPS co-crystals with toluene [17], iodine [18], 1,2-dichloroethane (DCE) [19], CS2 [20], and orthodichlorobenzene [21] have been completely characterized. The crystal structures of these SPS clathrate co-crystals present some common features: (a) a very efficient close packing of enantiomorphous helices in the ac plane; (b) a monoclinic P21/a symmetry; (c) the presence of isolated centrosymmetric guest locations, cooperatively generated by two enantiomorphous helices of two adjacent polymer layers (becoming, after guest removal, the cavity location of the nanoporous δ phase, see subsection 10.3.1.1); (d) the geometrical nature of the guest location tends to impose an orientation of the guest molecular plane nearly perpendicular to the helical axes. The above feature (c) has allowed these SPS co-crystals to be described

197

CO-CRYSTALS

a

guest

R

b L

(a)

L a

d010 = 1.34 nm

d010 = 1.05 nm

L

R

R

R

R R

guest

guest

L L

L

b

b L

(b)

a

guest

R (c)

Figure 10.1 Presentations of along-the-chain-projections of SPS co-crystalline phases: (a) δ clathrate phase with DCE; (b) intercalate phase with norbornadiene; (c) schematic ε clathrate phase with DCE.

as clathrates (from the latin word “clathratum,” i.e., imprisoned), since the polymer host forms a cage structure imprisoning the guest. As an example, the structure of the SPS/DCE co-crystalline phase is shown in Figure 10.1a. Due to the limited volume of the cavities, guests of these SPS clathrate phases have a molecular volume lower than 0.26 nm3 [17–21]. The above feature (a), that is, the efficient packing of the helices in the ac plane, corresponds to an a axis nearly equal for all clathrate co-crystals, independent of the chemical nature of the guest molecule (Fig. 10.2a). As the bulkiness of the guest increases, there is essentially only an increase of the spacing (d010 = bsin γ) between these layers [23]. This spacing, for all δ clathrates, remains in the range 1.05−1.2 nm, only slowly increasing with the guest molecular volume (see, e.g., fig. 10.4 of Reference 23). It is also worth noting that it has been found that generally each cavity hosts just one guest molecule, arranged approximately at its center, thus leading to a maximum molar ratio between polymer monomeric units and guest molecules equal to 4 [17–21]. The cases of I2 [18] and CS2 [20] clathrate forms, which can host up to two guest molecules per cavity, do not have to be considered exceptions since they present a molecular volume of about one-half of the estimated volume of the other guest molecules. 10.2.1.2 Intercalates Only recently, the occurrence of a second class of SPS co-crystals, defined as intercalate, has been established [22–24]. This new class of co-crystal presents two features being in common with the δ clathrates: (a) the same close packing of enantiomorphous helices in the ac plane (Fig. 10.2a); (b) a monoclinic P21/a symmetry. However, the guest molecules are not isolated into host cavities but contiguous inside layers intercalated with the monolayers of enantiomorphous polymer helices. Intercalate structures have been thoroughly described for the SPS cocrystals with bicyclo[2,2,1]-hepta-2,5-diene (norbornadiene, N) [22], 1,3,5trimethyl-benzene (TMB) [23] and 1,4-dimethyl-naphthalene (DMN) [23]. \As an example, the structure of the SPS/NB co-crystalline phase is shown in Figure 10.1b. Of course, this intercalate structure presents a higher guest

198


a/2 L

R

0.87 nm a // c//

c (c) (a)

a

a

c

a // c

(d)

a

(b) c (e)

a c//

Figure 10.2 Top (a) and lateral views (b) of the ac layer of s (2/1)2 parallel helices of SPS, that is, the high-density and low-energy structural feature that is common to the δ-nanoporous form and to the corresponding co-crystalline (both clathrate and intercalate) forms. The minimum interchain distance (0.87 nm) is achieved by alternating enantiomorphous helices (R and L stand for right-handed and left-handed, respectively). (c–e) Molecular models showing the three simplest orientations of the ac layers with respect to the film surface. The plane of the figure is assumed as parallel to the film plane. Arrows indicate the absence of axial orientation. (c) Both a and c axes are parallel to the film plane (a// c//); (d) a parallel and c perpendicular to the film plane (a// c⊥); (e) a perpendicular and c parallel to the film plane (a⊥ c//). These three uniplanar orientations can be achieved for γ, δ, and most co-crystalline phases of SPS. (See color insert.)

content with respect to the clathrate structures, in fact the guest/monomer-unit molar ratio is 1/2 rather than 1/4. The formation of intercalate structures have also been suggested for SPS co-crystals with 3-carene (3,7,7-trimethylbicyclo[4.1.0]hept-3-ene) [23], anthracene [23,68], and a styrene dimer (1,4-diphenyl-butane) [24]. The spacing (d010 = bsin γ) between the ac layers, for all the known SPS intercalates is larger than 1.3 nm and values as high as 1.75 nm have been observed. 10.2.1.3 ε Clathrates Very recently also, a third class of SPS co-crystals has been disclosed for which the guest molecules are imprisoned into the channel-shaped cavities typical of the recently discovered ε nanoporous phase

CO-CRYSTALS

199

[53–56] (which will be described in detail in subsection 10.3.1.2). Completely different guest orientations are generally imposed by the two nanoporous crystalline phases of SPS. For instance, for planar guest molecules, molecular planes generally tend to assume orientations roughly parallel (rather than perpendicular, as occurs for δ clathrates) to the crystalline chain axis [53–56]. We suggest defining this new class of SPS co-crystals as ε clathrates. A schematic presentation of the structure of a ε clathrate phase is shown in Figure 10.1c. The most relevant structural features of this new class of SPS co-crystals is that suitable guests are not only small molecules (both essentially apolar, like DCE[55] and highly polar, like 4-nitro-aniline [53]) but also long molecules like 4-(dimethyl-amino)-cinnamaldehyde, which exhibit a molecular volume too large to be enclosed as a guest into the δ phase. 10.2.2

Processing and Materials

Procedures suitable to prepare the different co-crystalline phases with different morphologies are reviewed in this section. Particular attention will be devoted to film processing, because SPS co-crystals depending on their preparation procedure can assume three different kinds of uniplanar orientation [69–75]. The control of three different uniplanar orientations for a same polymer is possibly an unprecedented phenomenon. Part of this section (10.2.2.4) is devoted to SPS physical gels presenting co-crystalline phases as knots of their physical networks [76–87]. 10.2.2.1 Solution Crystallization Procedures Most δ clathrate and intercalate co-crystals can be easily prepared by solution crystallization procedures (e.g., casting, spin-coating, gel desiccation [76–88]) or by solventinduced crystallization in amorphous SPS films [89–92]. All δ clathrate and intercalate co-crystals can be obtained by guest sorption into SPS samples presenting the δ phase [49–52] or by guest exchange in cocrystals [93–96]. In fact, for some guests, (mainly volatile, like carbon dioxide [CO2] [97], butadiene [97], or ethylene [98] but also bulkier guests like acetone, limonene, or carvone) the two latter are the only available procedures to obtain the corresponding SPS co-crystals. As for the ε clathrates, they can be only obtained by guest sorption in into SPS samples presenting the ε phase, which in turn can be obtained by chloroform sorption and desorption in γ form samples [53–56]. 10.2.2.2 Films and Three Different Uniplanar Orientations of the Cocrystalline Phases Film processing in the presence of suitable solvents can lead to the formation of co-crystalline films exhibiting three different kinds of uniplanar orientations of the co-crystalline phases [69–75,95]. The degree and the kind of uniplanar orientation depends on the selected technique (solution crystallization procedures or solvent-induced crystallization in amorphous

200


samples or solvent-induced recrystallizations of γ and α unoriented samples [74]) as well as on the chemical nature of the guest. It has been recently suggested that the structural feature determining these three different kinds of uniplanar orientations is the layer of close-packed alternated enantiomorphous helices [75] that characterizes the δ phase of SPS, as well as all related clathrate and intercalate co-crystalline phases with lowmolecular-mass guest molecules. In fact, thorough analyses of X-ray diffraction patterns of SPS films exhibiting different crystalline and co-crystalline phases, and related evaluations of degrees of orientation, have allowed the conclusion that the three observed uniplanar orientations correspond to the three simplest orientations of the high planar-density ac layers (i.e., of close-packed alternated enantiomorphous SPS helices, Fig. 10.2a,b) with respect to the film plane. In particular, it has been proposed that the three uniplanar orientations of SPS should be named a// c//, a// c⊥, and a⊥ c//, indicating crystalline phase orientations presenting the a and c axes parallel (//) or perpendicular (⊥) to the film plane (Fig. 10.2c–e) [75]. Information relative to quantitative evaluations of the uniplanar orientations [75] as well as to possible mechanisms leading to the three observed uniplanar orientations for SPS co-crystalline films, and their dependence on the crystallization technique and on the chemical nature of the guest, has been recently presented [71,75]. The three uniplanar orientations (without substantial loss of their degree of orientation) are maintained not only for the δ phase, as obtained by suitable guest-removal procedures, but also for the γ phase [67,69,99] as obtained by suitable thermal treatments [73–75]. The three uniplanar orientations can also be maintained [73–75] after guest-exchange procedures transforming the cocrystalline SPS phase with a given guest into a co-crystalline phase with a different guest [93–96]. Hence, the proposed nomenclature can be used unaltered for both γ and δ crystalline phases as well as for related co-crystalline phases of SPS. It is worth adding that suitable thermal treatments on SPS films presenting the three different uniplanar orientations of their helical crystalline phases can lead to films with planar orientations of their zigzag planar α [100–103] and β [104,105] crystalline phases [69,73]. In particular, the transformation of the helices into trans-planar chains occurs maintaining their preferential orientation (parallel or perpendicular) with respect to the film surface [73]. The availability of SPS films with three different kinds of uniplanar orientation not only allows establishment of fine structural features of SPS crystalline and co-crystalline phases (e.g., experimental evaluation of the orientation of transition-moment-vectors of host and guest vibrational modes, with respect to the host chain axes) [106,107] but it can also be relevant for practical purposes. For instance, it allows guest orientation control [108,109] for cocrystalline phases and guest diffusivity (and hence permeability) control [97,98,110,111] for the nanoporous phases. (See following subsection 10.3.3.3).

CO-CRYSTALS

201

10.2.2.3 Guest-Induced Recrystallizations The co-crystal formation by absorption of suitable guest molecules in the helical γ phase as well as in the trans-planar α phase was described in early papers on SPS [42,89,91]. It has also been found that the denser trans-planar β phase presents a higher stability and it is not transformed into any co-crystalline phase by guest absorption [89,91]. The formation of SPS co-crystalline phases by absorption of several guest molecules in the helical γ phase as well as in the trans-planar α phase, and related guest-induced orientation phenomena, has been thoroughly investigated recently mainly by X-ray diffraction measurements [74,75]. Guests, being able to induce co-crystal formation in amorphous SPS samples, are able to induce co-crystal formation from γ and α form films, by room temperature diffusion, only if their vapor pressure is higher than 20 torr and 60 torr, respectively. Guest-induced recrystallizations of unoriented γ form films generally lead to unoriented co-crystalline films, which in turn lead to essentially unoriented δ form films by guest removal [74,75,112]. By chloroform treatment of unoriented γ form films, a co-crystalline phase is obtained with orientation of chain axes prevailingly perpendicular to the film plane [74,75]. Moreover, the nanoporous ε form films obtained by chloroform removal [53–56], also present a preferential perpendicular orientation of chain axes. On the other hand, guest-induced recrystallizations of unoriented α form films generally lead to oriented co-crystalline and derived δ form films. The kind and degree of uniplanar orientation strongly depends on the guest molecule. In particular, recrystallizations of unoriented α form films by room temperature sorption of CS2, trichloroethylene (TCE) and CHCl3 lead to films with unoriented, a// c⊥ and a⊥ c// uniplanar orientations, respectively. The degrees of a// c⊥ and a⊥ c// uniplanar orientations, achieved by TCE and CHCl3 sorption in unoriented α form films, are similar to the maximum values until now obtained by other techniques, such as by solution casting [70] and by solvent-induced crystallization of amorphous films [73]. It is worth noting that nanoporous δ form films with a// c⊥ orientation, as derived by the TCE-induced recrystallization procedure, are much less brittle than those obtained by the previously known solution casting procedure. 10.2.2.4 Gels Gels consist of a three-dimensional network structure swollen by a liquid. For chemical gels, the cross-links that give rise to this network are covalent bonds while for physical gels the connectedness between polymer chains is achieved by intermolecular physical bonding forming junction zones (polymer-rich phase) that can be created and removed by cooling and heating, respectively [113]. It is well known that SPS easily forms physical gels with several organic compounds [76–88]. Particularly stable are gels where the polymer-rich phase is characterized by the helical s (2/1)2 conformation [88]. Comparative Fourier Transform InfraRed (FTIR) and X-ray diffraction characterizations of gels

202


and co-crystalline samples have allowed the clarification that the polymer-rich phase of helical SPS physical gels consists of co-crystalline phases [85]. In particular, it has been clearly established that in most cases the crystalline phases of the SPS gels are clathrate phases [85–88]. However, low 2θ peaks corresponding to d ≈ 1.5 nm have been observed for SPS gels in benzyl-methacrylate and in cyclohexyl-methacrylate [83,84], which could be easily rationalized by considering the possibility of formation of intercalate structures, analogous to those observed for semicrystalline films (see subsection 10.2.1.2). Particularly interesting is the case of SPS/benzene gels. In fact, the phase diagram study of the SPS/benzene system has suggested the possible occurrence of two molecular complexes with markedly different stoichiometries [81]. Moreover, by neutron diffraction of gels of deuterated SPS in deuterated benzene, it has been established that while for concentrated gels (57 wt%) the usual diffraction pattern of the SPS/benzene clathrate phase is present, for diluted gels (26 wt%) a different diffraction pattern, showing an intense low 2θ peak for d ≈ 1.5 nm, is observed [81]. These data can be rationalized by assuming that the two kinds of SPS/ benzene gels present as polymer-rich phase clathrate or intercalate co-crystals. It is worth adding that quenching procedures on SPS solutions, which use solvents whose molecules are bulkier than 0.16–0.17 nm3 and are unable to form co-crystals with SPS, give rise to the so-called gels of type II [80,88], which present in the polymer-rich phase trans-planar SPS chains and much lower elasticity. 10.2.3

Characterization Studies

Polymeric co-crystals are generally more complex than those with low-molecular-mass hosts, since the samples generally contain large amorphous fractions (e.g., typical degrees of crystallinity of SPS samples are lower than 50%). Hence, in general, guest molecules can be included both in co-crystalline and amorphous phases. Besides X-ray diffraction techniques, which for well-crystallized samples can allow a precise definition of the co-crystalline phases (including guest conformation, location, and three-dimensional orientation with respect to the crystalline axes), other techniques are highly informative relative to the structural organization of SPS co-crystalline samples. In this respect, it is worth adding that for all co-crystals, when the degree of occupancy of the crystalline positions available to the guest molecules is low, the X-ray diffraction method is not able to give reliable information relative to the guest. Particularly relevant are FTIR techniques, which can give information relative to the guest conformation, guest orientation with respect to the host polymeric crystalline framework, the possible occurrence of host-guest interactions, and most importantly, the partition of the guest molecules between the crystalline and amorphous phases [76,93,112,114–121]. Solid-state 2H-

CO-CRYSTALS

203

NMR techniques, mainly if supported by suitable line simulation procedures, can provide direct information on the dynamics of guest molecules in co-crystalline phases [122–126]. Additional information on host-guest interactions for SPS co-crystalline phases, presenting fluorescent guest molecules, can be obtained by fluorescence depolarization measurements [127–129]. These characterization studies have been supported by molecular simulations, which have given relevant contributions to the understanding of structure-property relationship of SPS co-crystals and nanoporous phases [51,110,130–134]. The main contributions of molecular simulation for SPS cocrystalline and nanoporous phases have been collected in a recent review [135]. In this review additional information from FTIR and, briefly, information from NMR and fluorescence depolarization studies are discussed. 10.2.3.1 FTIR FTIR spectroscopy, a technique with fast sampling rate, high sensitivity, and accurate quantitation, has proven to be capable of providing information at the molecular level and probing the local environment of both guest molecules and polymeric host of SPS co-crystalline phases. In fact, host-guest interactions not only often produce shifts of vibrational peaks, it can also change the conformational equilibria of the guest molecules. Particularly informative are linear dichroism measurements on uniaxially oriented samples as well as FTIR measurements on films presenting the three different uniplanar orientations. 10.2.3.1.1 Shifts of Vibrational Peaks As well known for other polymeric co-crystalline phases (e.g., the clathrate structures of poly [ethylene oxide] with para-disubstituted benzene) [136], guest peak shifting attributed to specific host-guest interactions has been observed for some SPS co-crystalline phases. In particular, a spectroscopic analysis of films presenting the SPS/chloroform δ clathrate phase [120] has shown that when chloroform is absorbed in the crystalline phase, a significant perturbation of its vibrational spectrum takes place. This effect mainly involves the 1219 cm−1 (δH-C-Cl) peak, which clearly shows a fine structure in the form of an unresolved component at a lower wavenumber (1210 cm−1), and is not present in the spectrum of the isolated molecule (vapor phase) and in the case of chloroform sorbed in the amorphous phase of SPS [120]. Hence, it is likely to be related to host-guest molecular interactions. Analogously, FTIR-ATR spectra of SPS/DCE gels show that during the progressive desorption of DCE, the reduction of the DCE trans conformer peak at 1232 cm−1 is accompanied by a splitting of the peak in two components at 1232 and 1235 cm−1 [87]. Moreover, it is observed that during DCE desorption the peak at 1232 cm−1 vanishes well before the peak at 1235 cm−1. This behavior can be explained by considering two types of trans DCE molecules in the gel: guest molecules included in the gel crystalline phase (characterized

204


by the peak at 1235 cm−1) and free molecules included in the polymer-poor phase (characterized by the peak at 1232 cm−1). At the initial stages of the desorption the amount of DCE included in the polymer-poor phase is much larger than the amount of DCE included in the clathrate phase, thus, the peak at 1232 cm−1 is largely prevailing. Then, as the DCE desorption from the polymer-poor phase proceeds at a higher rate than from the polymer-rich phase, there is observed a strong decrease of the 1232 cm−1 peak, which vanishes well before the 1235 cm−1 peak [87]. The resolution of the two peak components has allowed quantitative evaluations of the amount of chloroform in amorphous and crystalline phases [120] and of DCE in the polymer-poor and polymer-rich (crystalline) gel phases [87]. 10.2.3.1.2 Conformational Equilibria of the Guest Molecules FTIR studies relative to chlorinated hydrocarbons sorption into SPS films have shown that the trans conformation of DCE and 1,2-dichloropropane (DCP) is largely prevailing in the clathrate phase, while the trans and gauche conformations are nearly equally populated when both chlorinated compounds are sorbed in the amorphous phase [88,137–139]. Because essentially only their trans conformers are included into the clathrate phases (both δ- and ε clathrates), while both trans and gauche conformers are included in the amorphous phase, quantitative evaluations of vibrational peaks associated with these conformers allow the evaluation of the amounts of DCE and DCP confined as guest in the clathrate phase or simply absorbed in the amorphous phase [88,111,137–139]. 10.2.3.1.3 Polarized Spectra of Uniaxially Stretched Films: Linear Dichroism of Guest Peaks High linear dichroism of FTIR peaks of low-molecular-mass molecules can be achieved by enclosing them as guest of the crystalline nanoporous phases of uniaxially stretched SPS films [108,109,128,140]. As an example, the FTIR spectra in the wavenumber range 1650–780 cm−1, taken with polarization plane parallel and perpendicular to the draw direction (thin and thick lines, respectively) of uniaxially oriented SPS films exhibiting the SPS/nitrobenzene (NB) co-crystalline phase are reported in Figure 10.3a. The high dichroism observed for the peaks at 1440, 1276, and 934 cm−1, typical of the helical crystalline phase (labeled as h in Fig. 10.3a), clearly indicates the occurrence of a high degree of orientation of the host crystalline phase (fc,IR = 0.93). Moreover, the high dichroism of the guest peaks (labeled as g in Fig. 10.3a) clearly indicates the occurrence of an ordered orientation of the guest molecules in the crystalline host framework. It is worth adding that, since the guest molecules are highly oriented in the crystalline phase and essentially unoriented in the amorphous phase, the measured dichroic ratios (R = A///A⊥) strongly depend on the partition of the guest molecules between the two phases. As discussed in detail in the literature [138–140], after sorption into semicrystalline nanoporous samples, guest molecules can be partitioned almost evenly in amorphous and crystalline phases,

CO-CRYSTALS

2.0

205

g

Absorbance units

g

//

1.5

1.0

h g

0.5

g

h h 0.0

1400

(a)

1200 Wavenumber cm–1

1000

800

792 cm–1 0.5

S

α = 25°

0.0

1528 cm–1 α = 72° 1346, 851 cm–1 α = 75°

–0.5 0.0

(b)

0.2

0.4

0.6

0.8

1.0

fc,IR

Figure 10.3 (a) FTIR spectra in the wavenumber range 1650–780 cm−1 taken with polarization plane parallel (thin line) and perpendicular (thick lines) to the draw direction for a uniaxially oriented film exhibiting the SPS/nitrobenzene co-crystalline phase (h, host peak; g, guest peak). (b) Order parameter S of infrared peaks of NB guest molecules versus the axial orientation factor of the host polymer phase (fc, IR): (䊏) 792 cm−1 (out-of-plane); (䊊) 851 cm−1 (䉭) 1346 cm−1 (in-plane, parallel to the CN bond); (ⵧ) 1528 cm−1 (in-plane, perpendicular to the CN bond).

but the amorphous phase generally loses the guest molecules at a much faster rate than does the crystalline one and after substantial desorption, most of the residual low-molecular-mass molecules can be located in the crystalline phase. As a consequence, the order parameter (S = (R − 1)/(R + 2)) of the guest peaks tend to increase (in absolute value) with guest desorption, gradually reaching a plateau value [22,138–140]. On the basis of these plateau values, linear dichroism measurements can also allow evaluation of the partition of the guest molecules between the crystalline and the amorphous phase.

206


Polarized FTIR spectra of uniaxially stretched films including SPS clathrate phases, like those of Figure 10.3a, also allow an evaluation of the α angle between the polymer chain axis and the transition-moment-vector (for each guest vibrational peak), by applying the following equation [108]: fc, IR =

( R − 1) ( 2 cot 2 α + 2 ) ( R + 2 ) ( 2 cot 2 α − 1)

(10.1)

As an example, the order parameter S relative to some NB peaks versus the crystalline orientation factors (fc,IR) is plotted in Figure 10.3b. By applying Equation 10.1, the slope of these lines leads to the α angles between the crystallographic c axis and the guest transition-moment-vectors. By collecting data for films presenting largely different crystalline phase orientation (Fig. 10.3b), α angles can be evaluated with a good accuracy. The data plotted in Figure 10.3b indicate, for instance, that the angle between the guest CN bond and the host polymer chain axis is nearly 80 °. It is worth adding that the abovedescribed method can be applied if the concentration of the guest molecules in the amorphous phase is negligible or valuable by an independent method. These α angle evaluations, when associated with independent evaluation of the transition-moment-vector directions with respect to the molecular structure, allow getting information relative to the location of the guest molecule into the cavity of the SPS δ phase. Polarized FTIR spectra of uniaxially stretched films including SPS clathrate phases, like those of Figure 10.3, also allow a clear-cut discrimination between in-plane and out-of-plane guest transition moments [109]. In fact, in the case of planar guest molecules of SPS δ clathrates, it has been generally observed that they are oriented with their smallest cross-section nearly parallel to the SPS stretching direction [17–21,108,109,130–135]. As a consequence, in-plane and out-of-plane vibrational modes generally maximize their absorption intensities for light polarization nearly perpendicular (A⊥) and parallel (A//) to the stretching direction, respectively [109]. 10.2.3.1.4 Unpolarized Spectra of SPS Films Presenting Three Different Kinds of Uniplanar Orientation: Directions of Guest Transition-MomentVectors in the Host Unit-Cell A definition of the transition-moment-vector directions with respect to all the axes of the host unit cell can be achieved by comparing the relative intensities of different guest absorbance peaks, for unpolarized FTIR spectra of SPS co-crystalline films presenting the three different kinds of uniplanar orientation [141]. Due to the occurrence of a preferential orientation of guest molecules with respect to the host chain axes, the absorbance of the guest peaks is largely dependent on the different SPS orientations. As for δ clathrates, for instance, the peaks corresponding to out-of-plane vibrational modes, whose transitionmoment-vectors are nearly parallel to the chain axis, are of high intensity for the films where the c axis is in the film plane, that is, with a// c// and a⊥ c// uniplanar orientations, and of low intensity for the films with a// c⊥ orientation.

CO-CRYSTALS

207

In this case, more informative are the peaks corresponding to in-plane vibrational modes whose transition-moment-vectors are nearly perpendicular to the chain axis and hence of high intensity for films with a// c⊥ orientation and of low intensity for the films with a// c// and a⊥ c// orientations. In fact, some in-plane peaks are of higher intensity for films with a// c// orientation, hence their transition-moment-vectors are roughly parallel to the a axis. On the other hand, other in-plane peaks are of higher intensity for films with a⊥ c// orientation, hence their transition-moment-vectors are also roughly perpendicular to the a axis. Hence, by a quantitative analysis of the peak relative absorbances, it is in principle possible to completely define the directions of the guest transition moments with respect to the host unit cell. 10.2.3.2 Solid-State 2H-NMR By solid-state 2H-NMR measurements information about the reorientational dynamics of molecules in the microsecond time scale can be achieved. Some investigations have been focused on the guest molecules by preparing δ clathrate phases with deuterated molecules [124–126]. It has been demonstrated that the reorientation dynamics of molecules included in the clathrate phase is much more restricted with respect to the molecules absorbed in the amorphous phase [124–126]. As an example, the spectrum of benzene-d6 molecules included in the nanoporous crystalline SPS unoriented δ phase is reported in Figure 10.4a. The observed Pake pattern, showing a quadrupolar splitting of 67 kHz, clearly indicates that the guest motion is restricted to a rotation about its C6 symmetry axis [124–126]. Particularly informative can be 2H-NMR spectra of deuterated guest molecules included in uniaxially oriented δ form SPS films. In fact, highly anisotropic spectra of most volatile organic compounds can already be achieved at room temperature, which allow separation of components of solid-state 2 H-NMR molecular spectra. As an example, the 2H-NMR spectra of benzene-d6 included in the δ crystalline phase of a uniaxially oriented film for filmstretching direction parallel (B//, Fig. 10.4b) and perpendicular (B⊥, Fig. 10.4c) to the magnetic field, show large spectral differences. The main advantage of this confinement method is that the molecules are isolated guests in a crystalline host phase, rather than a solute in an amorphous phase, and as a consequence high guest orientation can be reached without altering the local molecular environment. Information about the orientation of the guest molecule into the host crystalline framework can be easily obtained from these spectra. For instance, the enhanced intensity of the outer components at 134 kHz in the // spectrum (Fig. 10.4b) and of the inner components at 67 kHz in the ⊥ spectrum (Fig. 10.4c) clearly indicates, in agreement with linear dichroism and X-ray diffraction measurements, that the benzene ring plane is roughly perpendicular with respect to the chain axis. 10.2.3.3 Fluorescence Depolarization This technique gives the opportunity to get information on the guest microenvironment [127–129]. In particular, it has been shown that naphthalene (NP) molecules absorbed into

208


67 kHz

λ=1 (a)

134 kHz

λ=3 (b)

//

67 kHz

λ=3 (c) –150

–100

–50

0

50

100

150

kHz 2

Figure 10.4 Solid-state H-NMR spectra of benzene-d6 molecules included in nanoporous SPS samples: unoriented (λ = 1, a) and oriented (λ = 3), collected for filmstretching directions parallel (//, b) and perpendicular (⊥, c) to the magnetic field. (Reproduced with permission from Albunia, A. R., Grassi, A., Milano, G., Rizzo, P., Venditto, V., Musto, P., Guerra, G. Oriented nanoporus host delta phases of syndiotactic polystyrene as a tool for spectroscopic investigation of guest molecules. Macromol. Symp., 234, 102–110 (2006) [141]).

amorphous phases, of atactic polystyrene or of SPS films, present the same fluorescence anisotropy (r ≈ 0.07) while definitely lower anisotropy (r ≈ 0.04) has been observed for NP molecules being guest of the δ clathrate phase of SPS [128]. According to diffusivity and NMR data, since NP guest molecules of the clathrate phase are less mobile than NP molecules adsorbed in the amorphous SPS phase, the observed reduction of fluorescence anisotropy could be tentatively attributed to a more efficient resonance energy transfer between guest molecules, possibly associated with their ordered positioning and orientation into the host polymeric crystallites. Very recently, SPS films with uniplanar-axial orientation of their cocrystalline phase with NP have been prepared and examined in detail [129] using Nishijima’s method [142,143], where angular distribution of NP polarized fluorescence intensities was measured at each setting film angle by the

CO-CRYSTALS

209

rotation of films around the excitation light beam. This polarized fluorescence method turns out to be quite effective for monitoring the orientation of fluorescent guest molecules in the films. In particular, in semicrystalline films presenting uniplanar-axial orientation of their SPS/NP clathrate phase, NP molecules were found to exhibit a high three-dimensional orientational order all over the films. The experimental data have also allowed determination of the orientation of the NP guest molecule with respect to the axes of the co-crystal unit cell, in satisfactory agreement with molecular modeling results [129]. 10.2.4

Properties and Applications

Systems composed of solid polymers and of low-molecular-mass active compounds find several practical and advanced applications. Recent studies have suggested that a simple method to reduce diffusivity of active molecules in the solid state and to prevent their self-aggregation consists of the formation of co-crystals with suitable polymer hosts. The formation of co-crystalline phases with SPS appears to be particularly efficient and versatile. In fact, studies of guest desorption kinetics [111,138] and of gas transport [97,144,145] on SPS films have shown that the guest solubility can be much higher in the crystalline phase (mainly for low-solute activities) while the solute diffusivity is generally much higher in the amorphous phase. This offers the opportunity to prepare and characterize samples including low-molecularmass molecules essentially only as guests of the host crystalline phase. As discussed above, X-ray diffraction and linear dichroism infrared studies have clearly shown that the guest molecules present well-defined average locations and orientations into co-crystalline phases. Moreover, solid-state 2 H-NMR studies have shown that the mobility of the solute molecules is heavily reduced when they are guests of the crystalline host phase, rather than simply absorbed in the amorphous phase [124–126]. As for possible applications of SPS co-crystalline films, particularly relevant is the possibility of achieving three different kinds of uniplanar orientation (see section 10.2.2.2), which allow control of the orientation of the guest molecules not only in the microscopic crystalline phase but also in macroscopic films. On these bases, films presenting SPS/active-guest co-crystals have been proposed as advanced materials, mainly for optical (chromophore [146], fluorescent [128,147], photo-reactive [148,149]) ferroelectric [54,150], and paramagnetic [151] materials. This constitutes an innovative approach in the area of functional polymeric materials that are instead characterized by a disordered distribution of active groups into amorphous phases, because generally prepared by simple dispersion of active molecules in amorphous phases or by chemical bonding of active groups to the polymer backbone (functionalization) or by copolymerization of standard monomers with comonomers containing suitable active groups.

210


In this review, only results relative to co-crystalline materials exhibiting fluorescent guests, photoreactive guests (possibly suitable for optical memories), and polar guest (possibly suitable for nonlinear optics and ferroelectricity) will be briefly reviewed. 10.2.4.1 Fluorescent Guests Transparent films exhibiting polymer/chromophore co-crystalline phases have been prepared by simple procedures on SPS, thus achieving low chromophore diffusivities [128,146,147]. Particularly interesting are the results reported for films exhibiting the cocrystalline phase between SPS and the fluorescent guest, TMB. X-ray diffraction measurements have proved the formation of two different kinds of polymer/chromophore host/guest co-crystals: intercalate, including layers of guest molecules intercalated with layers of polymer helices (corresponding to molar ratio guest/host-monomer-unit of 1/2) as well as clathrate, including isolated guest molecules into the cavities (corresponding to a molar ratio guest/host-monomer-unit of 1/4) [147]. UV absorption and emission studies on SPS films have shown that fluorescence phenomena are essentially additive when the chromophore is simply absorbed in the polymeric amorphous phase or isolated guest of the clathrate co-crystal. On the other hand, the fluorescence of the intercalate SPS/TMB co-crystal, when excited at its absorbance maximum, is red-shifted with respect to both host and guest emissions (Fig. 10.5). This phenomenon has been attributed to a fluorescence bleaching, which is related to the three-dimensional order of the intercalate SPS/TMB co-crystals [147]. The achieved fluorescence enhancement and red-shift could be relevant for optical and optoelectronic applications. Particularly relevant is the ability to

b

b c

a Intensity (a.u.)

a

Clathrate

240

260

d

a

280

b

Intercalate

300

320 λ (nm)

340

360

380

400

Figure 10.5 Emission spectra (excitation at 265 nm) of SPS films exhibiting clathrate (guest content 9 wt%) and intercalate (guest content 13 wt%) co-crystalline phases with 1,3,5-tri-methyl-benzene.

CO-CRYSTALS

211

hν (λ = 254 nm)

N Scheme 10.1

Δ

Q

Photoisomerization of norbornadiene (N) to quadricyclane (Q).

emit at longer wavelengths, which could bring the benefit of minimum losses due to reabsorption of the host phase. 10.2.4.2 Photoreactive Guests Photoisomerizations of organic dyes have been widely used as a means to record optical data. In particular, photoisomerization of norbornadiene and its derivatives (N, see Scheme 10.1) leading to quadricyclane and its derivatives (Q, see Scheme 10.1) has been deeply studied [152–155] and polymeric materials containing N derivatives (both as covalently bonded pendant groups or simply added to transparent amorphous phases) have been investigated for optical waveguides (utilizing photo-induced refractive index changes) and for data storage [156–160]. However, in all these optical materials there is a complete disorder in the spatial disposition of the photo-isomerizing molecules. It has been recently suggested that SPS films presenting co-crystalline phases with N and its derivatives could be suitable for data storage systems with molecular size marks [148,149]. In particular, it has been shown that in these films valence photoisomerization of N to Q can be easily achieved and that both reactant N and product Q present positional and orientational order with respect to the polymeric host crystallographic axes [148]. Infrared linear dichroism associated with gravimetric experiments on uniaxially stretched films has shown that it is possible to obtain samples with stable SPS/N intercalate co-crystalline phases and with negligible N content in the amorphous phases. Gravimetric and thermogravimetric measurements have shown that irradiation experiments, leading to N → Q photoisomerization, can be conducted without any significant loss of guest molecules in the co-crystalline phases. Moreover, N → Q photoisomerization reactions allow the preparation of patterns of micrometric size, by using suitable opaque masks, and an FTIR imaging technique has allowed characterization of the obtained N and Q patterns [149]. 10.2.4.3 Polar Guests After nearly 20 years from the discovery of the SPS co-crystals, although several dozens of SPS co-crystals have already been described, all the presented guests were apolar or poorly polar. In this respect, it is worth citing the relevant ability of the nanoporous δ and ε phases of SPS in separating molecules of different polarity (e.g., removing traces of organic pollutants from aqueous solutions, section 10.3.3.2). In 2007, it was shown that by sorption in nanoporous or in co-crystalline SPS samples of guests dissolved in suitable solvent-carriers, co-crystalline

212


Film plane

Film plane

c

c

Film thickness

(a)

Film thickness

(b)

Figure 10.6 Schematic presentation of the orientation of the co-crystal chain axes (c) and of guest molecules, as obtained by p-nitro-aniline absorption in ε form (a) and δ form (b) films, presenting an ideal a// c⊥ uniplanar orientation. (Reproduced with permission from Rizzo, P., Daniel, C., De Girolamo Del Mauro, A., Guerra, G. New host polymeric framework and related polar guest cocrystals. Chem. Mater., 19, 3864–3866 (2007) [54]).

phases can be easily obtained with molecules of very high polarity [140]. Solvents suitable as carriers are in general volatile guests of SPS co-crystals, like acetone or acetonitrile. These co-crystalline films present stable and threedimensionally ordered disposition of polar guests, being characterized by high first-order hyperpolarizability as for instance observed in trans-4-methoxy-βnitrostyrene (β = 17 × 10−30 esu) or in 4-(dimethyl-amino)-cinnamaldehyde (β = 30 × 10−30 esu), hence can be, in principle, used for nonlinear optical applications [140]. It is worth adding that the control of the orientation of the δ clathrate and ε clathrate phases allows, also at a macroscopic scale, the orientation control of the guest molecular dipoles. In this respect, particularly interesting can be the preferential orientation of the molecular dipoles perpendicular to the film surface. As schematically shown in Figure 10.6a, this guest dipole orientation can be achieved, by sorption of polar aromatic guests, like p-nitro-aniline, into SPS films exhibiting a a// c⊥ uniplanar orientation of the ε crystalline phase (i.e., presenting the polymer chains and the empty channels perpendicular to the film plane). The orientation of all guest dipoles nearly perpendicular to the film plane makes these materials suitable candidates for electrical poling processes, possibly leading not only to nonlinear optical but also to ferroelectric and piezoelectric properties [54].

10.3

NANOPOROUS CRYSTALLINE PHASES

The term polymeric nanoporous phase (and sometimes the term “polymeric framework”) is used to indicate crystalline polymorphic phases, whose density are lower than the density of the corresponding amorphous phase and are suitable to absorb suitable guest molecules at low activities (e.g., from diluted solutions).


10.3.1

213

Crystalline Structures

10.3.1.1 δ Phase The first nanoporous crystalline phase, the δ phase of SPS, was discovered and patented in 1994 [49]. The X-ray diffraction pattern of an unoriented sample exhibiting this crystalline phase is shown in Figure 10.7a and can be easily distinguished by the other crystalline (Fig. 10.7c,e) and co-crystalline phases (Fig. 10.7b,d) by an intense peak corresponding to the Bragg distance d = 1.05 nm (2θCuKα ≈ 8.4 °) and by low diffraction intensity in the 2θCuKα range 9 °–12 °. Few years later, the crystal structure of the δ form has been determined by the analysis of the X-ray fiber diffraction pattern and packing energy calculations. Chains in the helical s (2/1)2 conformation are packed in the monoclinic unit cell with axes a = 1.74 nm, b = 1.185 nm, c = 0.77 nm, and γ = 117 °, according to the space group P21/a (Fig. 10.8a,b).The structure is similar to the model proposed for the δ clathrate co-crystals: the b axis is shorter and the distance b sin γ between ac layers of macromolecules is shortened to 1.05 nm, as a consequence of the removal of the guest molecules.

0.96 nm

E

0.84 nm

Intensity (a.u.)

D /DCE

1.28 nm

C 1.08 nm

B /DCE A

5

1.05 nm

10

15

20

25

30

35

2 (°)

Figure 10.7 X-ray diffraction patterns (CuKα) of SPS semicrystalline powder samples presenting different helical crystalline and co-crystalline phases: (A) δ form; (B) δ clathrate with DCE; (C) ε form; (D) ε clathrate with DCE; (E) γ form. Bragg distances (d in nm) of relevant reflections are indicated.

214


L

R

b L

0 0.5

a

0

R

L

0

0.5

0.5

c

0

R

(a)

(b) R

R

0

0

0.5

0.5

b

(c)

b+a/2 L

L

1/4 0.5 f

0

0.5

0

0

0.5

0.5

0

0.5

R

1/4

0.5

0

f

L f

f

0

a

0

0

0.5

0.5

R

0.5

0.5

0

0

c

1/4

(d)

L (e)

b (f)

Figure 10.8 Top and lateral views of the crystalline structures of the two nanoporous crystalline phases of SPS. For the δ-(upper figures) and ε-(lower figures) phases, the porosity is distributed as cavities and channels, respectively. (See color insert.)

Shape and volume of empty space of the δ nanoporous phase of SPS have been evaluated by considering the space available to probe spheres of given radii into crystalline structures of different polymorphic forms of this polymer. This kind of analysis, for probe sphere radii higher than 0.13 nm, shows that the empty space into the nanoporous form corresponds to cavities (two per unit cell) centered on the center of symmetry of the crystal structure. For instance, the volume of these cavities, evaluated for probe spheres with a radius of 0.18 nm, is close to 0.115 nm3 [51]. This analysis has shown that the cavities of the δ phase are isolated and delimited mainly by eight parallel benzene rings belonging to two enantiomorphous adjacent helical chains (barycenters of four benzene rings at z and of four benzene rings at z + 1, Fig. 10.8a) [50,51]. In the δ form unit cell, the number of cavities is equal to the number of chains while the maximum number of guest molecules per cavity is a well-defined integer (generally one or two, depending on guest molecular volume) [17–24]. It is worth adding that the cavity is rather flat, that is, it presents its maximum dimension (nearly 0.8 nm) nearly perpendicular to the polymer chain axis (essentially along the a–b direction) while its minimum dimension (nearly 0.3 nm) essentially along the c axis [51]. This allows understanding of the


215

typical orientation of molecular planes of guests of δ clathrates, being nearly perpendicular to the polymer chain axes [109]. Figure 10.8 also compares a three-dimensional representation of the cavities of the δ form of SPS, being empty (b) or filled with the trans conformer of DCE (c). 10.3.1.2 ε Phase The nanoporous ε phase of SPS was discovered only in 2007 and its typical X-ray diffraction pattern for unoriented samples is shown in Figure 10.7c. This pattern presents, in the low 2θ range, two well-defined reflections for Bragg distances d = 1.28 nm and 1.08 nm (2θCuKα ≈ 6.9 ° and 8.2 ° in Fig. 10.7c), rather than the single peak at d = 1.05 nm typical of the δ phase [53–56]. The analysis of the reflections of patterns obtained for axially oriented samples indicates the occurrence of an orthorhombic unit cell with axes a = 1.61 nm, b = 2.18 nm, and c = 0.79 nm. The calculated density is 0.98 g/cm−3, with four chains of SPS in the s (2/1)2 helical conformation included in the unit cell, that is, very close to the density established for the δ phase [50] and definitely smaller than that one of the amorphous phase (1.05 g cm−3) [161]. The space group proposed is Pbcn, in accordance with the absence of hk0 reflections with h + k = 2n + 1, supported by the electron diffraction patterns [55]. The only symmetry element of the isolated s (2/1)2 helical chain preserved in the space group Pbcn is a binary axis perpendicular to the chain axis and the structural unit is reduced to two monomeric units [55]. A model of packing in the space group Pbcn corresponding to the minimum of the packing energy is shown in Figure 10.8d,e. The model of Figure 10.8d,e is characterized by channel-shaped cavities crossing the unit cells along the c axis and delimited, along b axis, by two enantiomorphic helical chains. These cavities, as those observed for the δ phase, are generated by a couple of s (2/1)2 enantiomorphous helices but they are rotated nearly 90 ° around their chain axis, with respect to the arrangement found for the δ form (cf. Fig. 10.8a,d). In these channels, planar guest molecules are expected to be hosted with their molecular planes roughly parallel to the polymer chain axis, as shown for instance for the trans conformer of DCE in Figure 10.8f. Moreover, the presence of channels easily allows rationalization of the formation of polymer co-crystals with guest molecules presenting a molecular axis much longer than the SPS chain axis periodicity, like 4-(dimethyl-amino)cinnamaldehyde, which are not able to form polymer co-crystals with the δ host phase [53,54]. 10.3.2

Processing and Materials

10.3.2.1 Preparation of the Nanoporous Crystalline Phases By suitable procedures of guest-removal from δ clathrate and intercalate SPS cocrystalline phases, the nanoporous crystalline δ phases can be easily obtained.

216


Particularly suitable are guest-extraction procedures by solvents that are temporary volatile guests of these nanoporous phases, like acetone, acetonitrile [43], or supercritical CO2 [47,48]. The same extraction procedures are suitable to obtain the nanoporous ε phase only when used on ε clathrate samples. The crystalline-phase orientation (axial, uniplanar, and uniplanar axial) that can be achieved for SPS co-crystalline phases can be maintained in the corresponding nanoporous crystalline phases [69–75,95]. 10.3.2.2 Aerogels Aerogels are a unique class of materials characterized by a highly porous network, being attractive for many applications such as thermal and acoustic insulation or catalysis. Many papers focusing on fabrication and characterization of aerogel materials have been published in the literature [162–165]. Although their preparation methods have been marked by several breakthroughs since the 1930s, most aerogels produced so far are characterized by chemically bonded three-dimensional networks. High porosity aerogels exhibiting physically bonded three-dimensional networks have been obtained by supercritical CO2 extraction of the solvent present in SPS physical gels [166,167] or by sublimation of the solvent [168,169]. The crystalline phase of the aerogels as well as their morphology depend on the crystalline structure of the junction zones of the starting gel, and by using suitable techniques can be well controlled [166,167]. In particular, the crystalline structure can be nanoporous (δ [166,167] or ε [56]) as well as dense (β [56,166,167] or γ [56]). When the polymer-rich phase of the gel is a co-crystal, the corresponding aerogel is generally formed by semicrystalline nanofibrils (fibril diameter range between 30 and 200 nm) (Fig. 10.9) [56,166–170].

Figure 10.9 Scanning electron micrography of an SPS aerogel with a porosity of 99%, as obtained by a toluene gel containing only 1 wt% of the polymer. The nanofibrils exhibit a crystallinity not far from 50% and the nanoporous δ crystalline phase.


217

Chloroform-vapor sorption measurements at low chloroform pressure have shown that δ form aerogels present the high sorption capacity characteristic of SPS δ form samples (due to the sorption of molecules as isolated guests of the host nanoporous crystalline phase) associated with the high sorption kinetics typical for areogels (due to the high porosity and hence high surface area) [166,167]. Thus, these new materials present fast sorption kinetics associated with a good handiness. Gravimetric and FTIR measurements show that the uptake of organic molecules from dilute aqueous solutions is negligible for SPS aerogels with dense β and γ crystalline phases, while is high for δ form aerogels (e.g., more than 5 wt% from a 1 ppm DCE solution) [170]. The volatile organic compounds (VOC) sorption is similar to those observed for other δ form samples and it is independent of aerogel porosity. These results indicate that the organic guest uptake from diluted aqueous solutions occurs essentially only in the crystalline nanocavities. Sorption and desorption experiments of DCE guest molecules have shown that the use of δ aerogels allows increase of the apparent guest diffusivity of several orders of magnitude (up to 7), with respect to δ form films [170]. 10.3.3

Characterization Studies

10.3.3.1 Thermal Transitions A recent study has compared the thermal behavior of δ form and ε form films, obtained by similar solvent sorption procedures starting from a same γ form film, by using several different techniques [44]. X-ray diffraction patterns of essentially unoriented δ form and ε form SPS films, annealed at different temperatures, show that both crystalline forms are transformed in the γ form. However, while for the δ form film this transformation involves an intermediate helical mesomorphic phase [45,52,161], for the ε form film a direct transformation toward the γ phase is observed [44]. Differential scanning calorimetry (DSC) scans also point out a clearly different behavior: the ε → γ transition produces poor enthalpic effects (only a very broad endothermic peak centered at 105 °C) while the δ → mesomorphic → γ transition presents a well-defined endothermic peak (at 102 °C, associated with formation of the mesomorphic form) followed by an exothermic peak (at 110 °C, associated with γ phase crystallization) [44]. Moreover, several features of the dynamic mechanical analysis (DMA) scans can be rationalized on the basis of the different routes of thermal transitions of ε and δ phases toward the γ phase, being direct and through a mesomorphic phase, respectively [44]. A general scheme of crystallization and interconversion routes between helical SPS crystalline forms is presented in Figure 10.10. 10.3.3.2 Guest Sorption The sorption behavior of low-molecular-mass compounds in semicrystalline polymeric materials is usually assumed to occur only in the amorphous domains; in fact, molecules are not likely to penetrate

218


γ Helical forms

ε

δ G G

G Helical co-crystals

Clathrates

Clathrates

Intercalates

Amorphous + guest

Figure 10.10 Schematic representation of the main interconversion conditions for the helical polymorphic crystalline forms of syndiotactic polystyrene. G stands for guest molecules of δ and ε phases.

and dissolve in the crystalline phase, which is usually denser than the amorphous one. As a consequence, in most cases, solubility in semicrystalline polymers has been assumed to be proportional to the volumetric fraction of the amorphous phase. There are few cases reported in the literature where the crystalline domain displays a nonnegligible solubility [171]. As for SPS, the two nanoporous crystalline phases (δ and ε) have a density (0.98 g/cm3) significantly lower than the amorphous phase of SPS (1.05 g/cm3), which is close to that of atactic polystyrene. Sorption studies from liquid and gas phases have shown that nanoporous phases of SPS are able to absorb suitable guest molecules, even when present at very low concentrations, and that guests sorption eventually leads to the formation of co-crystalline phases [172–174]. Chloroform sorption experiments at different temperatures have shown that if the relative pressure of chloroform is high enough, α, γ, δ, and ε phases are transformed into SPS/CHCl3 co-crystalline phases while the β phase remains unaltered [172]. Moreover, at low chloroform activities, SPS/CHCl3 co-crystalline phases are achieved only starting from the nanoporous δ- and ε-forms. Experimental analyses of chloroform sorption, performed using in situ FTIR spectroscopy, have also allowed a quantification of the amount sorbed into the amorphous and nanoporous crystalline phases [120]. Some sorption studies have been conducted on gaseous guest molecules and have shown that the gas sorption capacity of semicrystalline SPS samples presenting the nanoporous phases is generally much higher than for amorphous or for other semicrystalline SPS samples [144,145]. It has also been shown that several gas molecules can form true co-crystallline (clathrate) phases with SPS, by inclusion into its nanoporous host δ phase [97], thus leading to gas uptake and selectivity comparable with those of other nanopo-


219

rous materials. Moreover, the gas guest diffusivity is strongly reduced as a consequence of polymer/gas clathrate formation. These transport properties, being unique for polymeric materials, are associated with several practical advantages. In fact, the polymeric host framework can be obtained by a cheap commercial polymer and, due to its macromolecular nature, it is robust (chemical and thermal resistant), easy to process (can be easily shaped as film, fiber, membrane, foam, aerogel, etc.), and presents excellent mechanical properties. Hence, as discussed in detail in the section 10.3.4, SPS-based polymeric frameworks are suitable candidates as repeated use gas storage and controlled release materials. 10.3.3.3 Guest Diffusivity and Crystalline Phase Orientation Guest sorption studies from dilute aqueous solutions and from gas phases as well as desorption studies have been conducted for SPS films presenting the three different kinds of uniplanar orientation of the nanoporous δ phase (a// c//, a// c⊥, and a⊥ c//). These investigations have been affected mainly by FTIR measurements combined with gravimetric measurements. The reported sorption and desorption data not only confirm that at low guest activities the sorption occurs nearly only by the nanoporous crystalline phase but also show that the guest transport behavior is dependent on the kind of uniplanar orientation of the host crystalline phase. In particular, in agreement with predictions based on molecular simulations [110], the lowest diffusivity has been measured for films with a// c// uniplanar orientation while the highest diffusivity has been measured for films with a⊥ c// uniplanar orientation [97,98,111]. The possibility to change the guest diffusivity by controlling the crystalline phase orientation could be useful for several possible applications of films that exhibit nanoporous crystalline phases. For instance, for application as sensing elements of molecular sensors [175–179], high diffusivity films presenting the a⊥ c// uniplanar orientation would be in principle most suitable, because they should maximize the sensor response rates. On the other hand, for applications requiring a long-term stability of the co-crystals, like in the case of films including active guests (e.g., fluorescent, photoreactive) [146–151], low diffusivity films presenting the a// c// uniplanar orientation should be most suitable. 10.3.4

Applications

10.3.4.1 Molecular Separations Particularly relevant molecular separations are those implying the removal of VOC from water and air. To this purpose, the nanoporous crystalline phases of SPS are extremely suitable because they are able to include apolar molecules, as guest of their crystalline cavities, and to exclude the highly polar water molecules. In particular, several experiments have been conducted relative to the uptake from aqueous solutions of DCE. The choice of DCE was motivated

220


DCE equilibrium uptake g/100 g SPS

20 δ aerogel (P = 90%) δ powder γ aerogel (P = 90%)

15

Activated carbon

10

5

0

1

10 100 1000 DCE concentration (ppm)

10,000

Figure 10.11 Equilibrium DCE sorption at room temperature by δ aerogel (porosity [P] equal to 90%, filled circles), δ powder (empty circles), and γ aerogels (porosity [P] equal to 90%, triangles) as determined by FTIR measurements. For the sake of comparison DCE sorption from activated carbon is also reported (thin line; data from M. H Stanzel [182]). (Reproduced with permission from Daniel, C., Sannino, D., Guerra, G. Syndiotactic polystyrene aerogels: Adsorption in amorphous pores and adsorption in crystalline nanocavities. Chem. Mater., 20, 577–582 (2008) [170]).

by the additional information, which comes from its conformational equilibrium. In fact, as described in detail in subsection 10.2.3.1, because essentially only its trans conformer is included into the clathrate phase while both trans and gauche conformers are included in the amorphous phase, quantitative evaluations of vibrational peaks associated with these conformers allow the evaluation of the amounts of DCE confined as guest in the clathrate phase or simply absorbed in the amorphous phase. The choice of DCE was also motivated by its presence in contaminated aquifers and by its resistance to remediation techniques based on reactive barriers containing Fe0 [180,181]. DCE equilibrium uptakes from diluted aqueous solutions from γ and δ form samples, as obtained by FTIR measurements, are for instance compared in Figure 10.11. For the sake of comparison, equilibrium sorption capacity of DCE from activated carbon is also shown [182]. Particularly relevant are the results relative to the VOC uptake from the most diluted aqueous solution (1 ppm) showing that the sorption capacity is larger than 5 gDCE/100 gpolymer, that is, leading to a concentration increase of 50,000 times. Moreover, it is clearly apparent that the DCE uptake from the γ aerogel is always much lower than for the δ form samples and becomes negligible for dilute aqueous solutions. This clearly confirms that the DCE uptake occurs essentially only in the crystalline nanoporous phase. It is worth adding that ε form film and aerogels generally absorb lower amount of organic pollutants than the corresponding δ form samples. However,


221

they are particularly suitable to absorb long organic compounds that are not absorbed by the small cavities of the δ phase. 10.3.4.2 Gas Sorption The improvement of gas storage, recognition, and separation techniques represents a strategic industrial and environmental objective. Some techniques are based on gas absorption on high surface amorphous materials like activated carbons, cross-linked polymers, or carbon nanotubes. More selective techniques are based on inclusion of gas molecules into the cavities of crystalline materials leading to co-crystalline phases (generally clathrate phases). The available nanoporous crystalline frameworks can present a large variety of chemical structures. Mostly studied are inorganic frameworks (e.g., zeolites) [57,58] and, more recently, metal-organic frameworks [59–62], although the high molecular masses of the framework atoms can limit the gas weight uptake. In this respect, organic frameworks, that is, nanoporous organic crystals [63–66], often referred to as “organic zeolites,” could be more promising. Of particular economic relevance could be the removal of ethylene and CO2 from the environment storage of fruit and vegetables. Indeed, it is well known that the postharvest life and quality of many fruits, vegetables, and flowers are seriously shortened if they are exposed to trace amounts (also few ppb) of ethylene. For this purpose active packaging has been realized by adding zeolites, silica, or activated carbon to commercial polypropylene or polyethylene packaging films. The ethylene uptake of semicrystalline SPS films presenting its nanoporous host δ phase is much higher than for other polymeric materials, since it is related to the formation of a co-crystalline phase, where ethylene molecules are oriented nearly perpendicular to the crystalline polymer helices [98]. This ethylene uptake is much higher than for industrial polypropylene samples also containing large amounts (up to 25 wt%) of silica. As already observed for other guest molecules of SPS, the ethylene diffusivity in the host nanoporous crystalline phase is markedly reduced with respect to the diffusivity in the corresponding glassy amorphous phases. In addition, ethylene diffusivity can be further reduced by suitable selection of the uniplanar orientation ((a// c// or a// c⊥) of the host crystalline phase. In summary, the δ nanoporous crystalline phase of SPS presents high ethylene solubility and low ethylene diffusivity, which can also be controlled by the orientation of the crystalline phase, associated with negligible water uptake. These features make polymeric materials presenting the δ-nanoporous crystalline phase of SPS suitable for ethylene removal and storage and, due to its chemical and mechanical properties, it can be considered suitable candidates as (also repeated use) produce packaging. 10.3.4.3 Molecular Sensors Chemical sensors have an important and growing role in diverse fields including environmental (ground water and air pollution) monitoring, toxic chemical agent detection, and medical diagnosis.

222


In recent years, great efforts have been dedicated to realizing chemical sensors based on the integration of proper sensitive layers and suitable highsensitivity transducing mechanisms. The sensitive element should optimize specific interactions with a target analyte, should provide a fast and reversible diffusion of the penetrants and small recovery times, and should maintain the physical state as well as the geometry over several cycles of use, in order to avoid hysteresis effects, thus ensuring the reproducibility. Several studies have shown that SPS films, presenting the nanoporous crystalline phases, are suitable sensing elements for detection of organic pollutants, being effective with most VOCs (mainly chlorinated and aromatic), which are present in industrial wastes like benzene, toluene, chloroform, methylene chloride, tetrachloroethylene, and trichloroethylene. In particular, SPS films have been tested as sensing elements of resonant sensors for vapors of VOCs [175]. For instance, the response to chloroform vapor of quartz crystal microbalance (QCM) sensors, coated with films of semicrystalline δ form of SPS, has been analyzed and compared with analogous systems coated with films of amorphous atactic polystyrene. The sensitivity of sensors based on SPS films was found markedly higher than those based on atactic polystyrene, particularly for low chloroform pressures. Of course, the higher sensitivity of the semicrystalline SPS films is associated with the peculiar sorption mechanism; in fact, the organic compound, rather than being dissolved only into the amorphous phase, as it is generally the case for semicrystalline polymers, is mainly absorbed into the nanoporous crystalline phase, each molecule being confined into regularly spaced crystalline nanocavities, thus leading to a clathrate structure [175]. Optical transduction techniques are also very attractive in chemical sensing applications due to some unique characteristics such as immunity to electromagnetic interference, small size, lightweight, low cost, and the possibility of using them in a harsh environment. As for sensors based on SPS films, methods based on optical fiber technology are particularly relevant, because they allow detection of organic pollutants in water (even in deep water). Optical sensors based on the integration of SPS δ form films with optical fiber technology have demonstrated an in-water parts-per-million detection capability of chloroform and toluene [176–178]. More recently, a novel opto-chemical sensor employing long-period fiber gratings coated with SPS sensitive overlays has been also proposed [179]. 10.3.4.4 Sensors of Non-racemic Molecules Several methods for sensing chirality based on racemic host receptors interacting with target nonracemic guests have been proposed. For chromophore racemic receptor molecules, their non-covalent bonding to a non-racemic guest can provide induced circular dichroism (ICD) in the absorption region of the receptor. Particularly suitable are macromolecular receptors, which are able to form regular helices, hence in most cases are stereoregular [183–190]. In fact, racemic polymers can lead not only to detection but also to amplification of


223

chirality, since cooperative interactions with low-molecular-mass non-racemic compounds can generate prevalence of one polymer helical hand. These chirality transfer and amplification phenomena have been generally observed in solutions [183–190]. The occurrence of transfer and amplification of chiral information could be particularly relevant for solid polymer films, since they could have, in principle, applications in chiro-optical devices and data storage systems. Recently, it has been reported that non-racemic SPS films, exhibiting the nanoporous δ phase, are able to detect, amplify, and memorize the chirality of several volatile organic molecules [191,192]. ICD always presents a major Cotton band at 200 nm and a minor Cotton band of opposite sign at 223 nm, but its intensity is critically dependent on the film processing. In particular, maximum ICD intensities (in the presence of all the considered non-racemic molecules) have been observed for SPS films spincoated from chloroform solutions, for spin rates larger than 1600 rpm (Fig. 10.12). The ICD phenomena are instead negligible for spin-coating procedures with most solvents and always for spin rates lower than 100 rpm [192]. The observed ICD phenomena remain in the SPS films not only after complete guest removal but also after thermal procedures leading to transitions of the nanoporous δ phase toward the dense helical (γ) and trans-planar (α) crystalline phases (Fig. 10.12). The memory of the volatile non-racemic guest

240°C

trans planar)

R 25°C

(helical)

100

CD (mdeg)

160°C

(helical)

0

100

S 200

220

240

260

Wavelength (nm)

Figure 10.12 Room temperature CD spectra of SPS films spin-coated at 1600 rpm from 0.25 wt% chloroform solution onto quartz surface after exposure to vapors of R (thick lines) or S-carvone (dashed lines): after complete carvone removal by supercritical carbon dioxide (blue/middle [at about 198 nm] lines) followed by thermal treatments at 160 °C and 240 °C, leading to crystal-to-crystal transition between helical crystalline phases (δ → γ, green/inner lines) or from helical toward trans-planar crystalline phase (γ → α, red/outer lines).

224


molecules can be erased only by thermal treatments at temperatures higher than the SPS melting temperature (≈270 °C) or by long-term treatments with strong SPS solvents [192]. These results possibly indicate that the observed ICD phenomena (hence the chiral memory) are not associated with non-racemic molecular structures but with the formation of non-racemic supramolecular (possibly crystalline) structures [192]. The obtained SPS films are suitable not only for detection but also for memory of non-racemic molecules, hence suggest that they could possibly be used for chiro-optical memories.

10.4

CONCLUSIONS AND PERSPECTIVES

Several exciting new material based on co-crystalline and nanoporous crystalline phases of syndiotactic polystyrene have been achieved. In particular, several kinds of polymer co-crystalline phases have been prepared, belonging to three different classes: δ- and ε-clathrates and intercalates. Polymer cocrystals with active guest molecules show unusual physical properties, hence are promising for several kinds of advanced materials. Moreover, the unprecedented achievement of polymeric nanoporous crystalline phases (δ and ε) has given very interesting results in the fields of molecular separations, water/ air purification and sensorics. As for perspectives of applications of SPS-based co-crystalline phases, studies will be mainly devoted to films, also trying to exploit the unique availability of three different kinds of uniplanar orientations, which allow macroscopic control of the guest orientation. Particular attention will be devoted to possible applications of the recently discovered ε clathrates, mainly due the possibility of controlling the orientation of very long guest molecules, which could give relevant nonlinear optical properties. In recent years, advanced materials with special optical properties were mainly developed while for the future, by using co-crystalline phases with highly polar or paramagnetic guest molecules, the achievement of relevant materials with special electric and magnetic properties are expected. Relevant new materials could also be obtained by chemical reactions (e.g., polymerization) between guest molecules, for polymer co-crystals exhibiting guest-guest proximity (ε clathrates and intercalates). As for possible perspectives of the nanoporous crystalline phases, applications are expected in the field of controlled release of drugs and pesticides. As for nanoporous SPS films and aerogels, a relevant objective will also be the modification, with different kinds of functional groups, of the amorphous phase. The functionalization of the sole amorphous phase [193, 194] could bring several advantages, like increase of rates of guest sorption from the nanoporous crystalline phases.

REFERENCES

225

The field of co-crystalline and nanoporous polymer materials is expected to be widely expanded. In fact, until now, only materials based on syndiotactic polystyrene have been considered for possible applications. Relevant new cocrystalline and nanoporous materials are expected in the future for other stereoregular polymers. 10.5

ACKNOWLEDGMENTS

I thank Prof. V. Venditto, Dr. C. Daniel, Dr. G. Milano, and Dr. P. Rizzo of the University of Salerno; Prof. G. Mensitieri, Prof. V. Petraccone, and Dr. O. Tarallo of the University of Naples; Dr. P. Musto of the ICTP-CNR of Naples; and Dr. M. Giordano of the IMCB-CNR of Naples for their contributions to the reviewed research work as well as for the useful discussions. Financial support of the “Ministero dell’Istruzione, dell’Università e della Ricerca” (PRIN2007) and of “Regione Campania” (Legge 5) is gratefully acknowledged. REFERENCES 1. Okamoto, Y., Ueba, Y., Dzhnibekov, N. F., Banks, E. Rare earth metal containing polymers. 3. Characterization of ion-containing polymer structures using rare earth metal fluorescence probes. Macromolecules, 14, 17–22 (1981). 2. Burroughes, J. H., Bradley, D. D. C., Brown, A. R., Marks, R. N., Mackay, K., Friend, R. H., Burns, P. L., Holmes, A. B. Light-emitting diodes based on conjugated polymers. Nature, 347, 539–541 (1990). 3. Hosokawa, C., Kawasaki, N., Sakamoto, S., Kusumoto, T. Bright blue electroluminescence from hole transporting polycarbonate. Appl. Phys. Lett., 61, 2503– 2305 (1992). 4. Vestweber, H., Greiner, A., Lemmer, U., Mahrt, R. F., Richert, R., Heitz, W., Bässler, H. Progress towards processible materials for light-emitting devices using poly (p-phenylphenylenevinylene). Adv. Mater., 4, 661–662 (1992). 5. Robello, D. R., Perry, R. J., Urankar, E. J. Linear polymers for nonlinear optics. 3. Efficient grafting of chromophores to styrenic polymers via palladiumcatalyzed carbonylation and coupling reactions. Macromolecules, 26, 6940–6944 (1993). 6. Lüssem, G., Festag, R., Greiner, A., Schmidt, C., Unterlechner, C., Heitz, W., Wendorff, J. H., Hopmeier, M., Feldmann, J. Polarized photoluminescence of liquid crystalline polymers with isolated arylenevinylene segments in the main chain. Adv. Mater., 7, 923–925 (1995). 7. Cerullo, G., Stagira, S., Zavelani-Rossi, M., De Silvestri, S., Virgili, T., Lidzey, D. G., Bradley, D. D. C. Ultrafast forster transfer dynamics in tetraphenylporphyrin doped poly (9,9-dioctylfluorene). Chem. Phys. Lett., 335, 27–33 (2001). 8. Negres, R. A., Gong, X., Ostrowski, J. C., Bazan, G. C., Moses, D., Heeger, A. Origin of efficient light emission from a phosphorescent polymer/organometallic guest-host system. J. Phys. Rev., 68, 115209/1–115209/8 (2003).

226


9. Bignozzi, C. A., Ferri, V., Scoponi, M. Syntheses and characterization of luminescent polymers containing rhenium (I) pyridinyl-carbonyl complexes. Macromol. Chem. Phys., 204, 1851–1862 (2003). 10. Mastrorilli, P., Nobile, C. F., Grisorio, R., Rizzuti, A., Suranna, G. P., Acierno, D., Amendola, E., Iannelli, P. Polyacetylenes bearing chiral-substituted fluorene and terfluorene pendant groups: Synthesis and properties. Macromolecules, 37, 4488–4495 (2004). 11. Goudket, H., Nhung, T. H., Ea-Kim, B., Roger, G., Canva, M. Importance of dye host on absorption, propagation losses, and amplified spontaneous emission for dye-doped polymer thin films. Appl. Opt., 45, 7736–7741 (2006). 12. Pucci, A., Di Cuia, F., Signori, F., Ruggeri, G. Bis (benzoxazolyl)stilbene excimers as temperature and deformation sensors for biodegradable poly (1,4-butylene succinate) films. J. Mater. Chem., 17, 783–790 (2007). 13. Kawa, M., Frechet, J. M. Self-assembled lanthanide-cored dendrimer complexes: Enhancement of the luminescence properties of lanthanide ions through siteisolation and antenna effects. Chem. Mater., 10, 286–296 (1998). 14. Zhu, L., Tong, X., Li, M., Wang, E. Luminescence enhancement of Tb3+ ion in assemblies of amphiphilic linear-dendritic block copolymers: Antenna and microenvironment effects. J. Phys. Chem. B., 105, 2461–2464 (2001). 15. Chou, C. H., Shu, C. F. Synthesis and characterization of dendronized polyfluorenes. Macromolecules, 35, 9673–9677 (2002). 16. Furuta, P., Brooks, J., Thompson, M. E., Frechet, J. M. J. Simultaneous light emission from a mixture of dendrimer encapsulated chromophores: A model for single-layer multichromophoric organic light-emitting diodes. J. Am. Chem. Soc., 125, 13165–13172 (2003). 17. Chatani, Y., Shimane, Y., Inagaki, T., Ijitsu, T., Yukinari, T., Shikuma, H. Structural study on syndiotactic polystyrene: 2. Crystal structure of molecular compound with toluene. Polymer, 34, 1620–1624 (1993). 18. Chatani, Y., Inagaki, T., Shimane, Y., Shikuma, H. Structural study on syndiotactic polystyrene: 4. Formation and crystal structure of molecular compound with iodine. Polymer, 34, 4841–4845 (1993). 19. De Rosa, C., Rizzo, P., Ruiz de Ballesteros, O., Petraccone, V., Guerra, G. Crystal structure of the clathrate d form of syndiotactic polystyrene containing 1,2dichloroethane. Polymer, 40, 2103 (1999). 20. Tarallo, O., Petraccone, V. On the crystal structure of the clathrate forms of syndiotactic polystyrene containing carbon disulfide and iodine. Macromolecules, 205, 1351–1360 (2004). 21. Tarallo, O., Petraccone, V. Syndiotactic polystyrene containing orthodichlorobenzene: Crystal structure of the clathrate form. Macromol. Chem. Phys., 206, 672–679 (2005). 22. Petraccone, V., Tarallo, O., Venditto, V., Guerra, G. An intercalate molecular complex of syndiotactic polystyrene. Macromolecules, 38, 6965–6971 (2005). 23. Tarallo, O., Petraccone, V., Venditto, V., Guerra, G. Crystalline structures of intercalate molecular complexes of syndiotactic polystyrene with two fluorescent guests: 1,3,5-Trimethyl-benzene and 1,4-dimethyl-naphthalene. Polymer, 47, 2402–2410 (2006).

REFERENCES

227

24. Galdi, N., Albunia, A. R., Oliva, L., Guerra, G. Selective molecular-complex phase formation of syndiotactic polystyrene with a styrene dimer. Macromolecules, 39, 9171–9176 (2006). 25. Iuliano, M., Guerra, G., Petraccone, V., Corradini, P., Pellecchia, C. Polymorphism of syndiotactic poly (p-methylstyrene) unoriented samples. New Polym. Mater., 3, 133–144 (1992). 26. De Rosa, C., Petraccone, V., Guerra, G., Manfredi, C. Polymorphism of syndiotactic poly (p-methylstyrene): Oriented samples. Polymer, 37, 5247–5253 (1996). 27. Petraccone, V., La Camera, D., Pirozzi, B., Rizzo, P., De Rosa, C. Crystal structure of the clathrate form of syndiotactic poly (p-methylstyrene) containing tetrahydrofuran. Macromolecules, 31, 5830–5836 (1998). 28. Petraccone, V., La Camera, D., Caporaso, L., De Rosa, C. Crystal structure of the clathrate form of syndiotactic poly (p-methylstyrene) containing odichlorobenzene. Macromolecules, 33, 2610–2615 (2000). 29. La Camera, D., Petraccone, V., Artimagnella, S., Ruiz de Ballesteros, O. Crystal structure of the clathrate form of syndiotactic poly (p-methylstyrene) containing benzene. Macromolecules, 34, 7762–7766 (2001). 30. Petraccone, V., Esposito, G., Tarallo, O., Caporaso, L. A new clathrate class of syndiotactic poly (p-methylstyrene) with a different chain conformation. Macromolecules, 38, 5668–5674 (2005). 31. Tarallo, O., Esposito, G., Passarelli, U., Petraccone, V. A clathrate form of syndiotactic poly (p-methylstyrene) containing two different types of cavities. Macromolecules, 40, 5471–5478 (2007). 32. De Girolamo Del Mauro, A., Loffredo, F., Venditto, V., Longo, P., Guerra, G. Polymorphic behavior of syndiotactic poly (p-chlorostyrene) and styrene/pchlorostyrene cosyndiotactic random copolymers. Macromolecules, 36, 7577–7584 (2003). 33. Yokoyama, M., Ishihara, H., Iwamoto, R., Tadokoro, H. Structure of poly (ethylene oxide) complexes. III. Poly (ethylene oxide)-mercuric chloride complex. Type II. Macromolecules, 2, 184–192 (1969). 34. Point, J. J., Coutelier, C. Linear high polymers as host in intercalates. Introduction and example. J. Polym. Sci. Polym. Phys. Ed., 23, 231–239 (1985). 35. Delaite, E., Point, J. J., Damman, P., Dosière, M. Two allotropic forms for the poly (ethylene oxide)-resorcinol molecular complex. Macromolecules, 25, 4768– 4778 (1992). 36. Belfiore, L. A., Ueda, E. Solid-state carbon-13 NMR detection of molecular mixing in polymer blends that exhibit multiple eutectic phase transformations. Polymer, 33, 3833–3840 (1992). 37. Paternostre, L., Damman, P., Dosière, M. Polymorphism and crystal morphology of poly (ethylene oxide)-2-methyl resorcinol supramolecular complexes. Macromolecules, 32, 153–161 (1999). 38. Matsumoto, A., Odani, T., Sada, K., Miyata, M., Tashiro, K. Intercalation of alkylamines into an organic polymer crystal. Nature, 405, 328–330 (2000). 39. Oshita, S., Matsumoto, A. Fluorescence from aromatic compounds isolated in the solid state by double intercalation using layered polymer crystals as the host solid. Langmuir, 22, 1943–1945 (2006).

228


40. Kakida, H., Makino, D., Chatani, Y., Tadokoro, H. Structural studies of polyethers [-(CH2)mO-]n. VIII. Polyoxacyclobutane hydrate (modification I). Macromolecules, 3, 569–578 (1970). 41. Kusuyama, H., Yakase, M., Higashihata, Y., Tseng, H. T., Chatani, Y., Tadokoro, H. Structural change of syndiotactic poly (methyl methacrylate) (st-PMMA) on drawing, absorption and desorption of solvents. Polymer, 23, 1256–1258 (1982). 42. Guerra, G., Vitagliano, V. M., De Rosa, C., Petraccone, V., Corradini, P. Polymorphism in melt crystallized syndiotactic polystyrene samples. Macromolecules, 23, 1539–1544 (1990). 43. Rizzo, P., Albunia, A. R., Guerra, G. Polymorphism of syndiotactic polystyrene: G phase crystallization induced by bulky non-guest solvents. Polymer, 46, 9549– 9554 (2005). 44. Rizzo, P., D’Aniello, C., De Girolamo Del Mauro, A., Guerra, G. Thermal transitions of e crystalline phases of syndiotactic polystyrene. Macromolecules, 40, 9470–9474 (2007). 45. Manfredi, C., Guerra, G., De Rosa, C., Busico, V. P. C. Effects of p-methylstyrene comonomeric units on the polymorphic behavior of syndiotactic polystyrene. Macromolecules, 28, 6508–6515 (1995). 46. Loffredo, F., Pranzo, A., Venditto, V., Longo, P., Guerra, G. Clathrate phases of styrene/p-methylstyrene co-syndiotactic copolymers. Macromol. Chem. Phys., 204, 859–867 (2003). 47. Reverchon, E., Guerra, G., Venditto, V. Regeneration of nanoporous crystalline syndiotactic polystyrene by supercritical CO2. J. Appl. Polym. Sci., 74, 2077–2082 (1999). 48. Ma, W., Yu, J., He, J. Empty d crystal as an Intermediate form for the d to g transition of syndiotactic polystyrene in supercritical carbon dioxide. Macromolecules, 38, 4755–4760 (2005). 49. Guerra, G., Manfredi, C., Rapacciuolo, M., Corradini, P., Mensitieri, G., Del Nobile, M. A. Italy, IT1228915 (to CNR of Italy), 1994. 50. De Rosa, C., Guerra, G., Petraccone, V., Pirozzi, B. Crystal structure of the emptied clathrate form (d form) of syndiotactic polystyrene. Macromolecules, 30, 4147–4152 (1997). 51. Milano, G., Venditto, V., Guerra, G., Cavallo, L., Ciambelli, P., Sannino, D. Shape and volume of cavities in thermoplastic molecular sieves based on syndiotactic polystyrene. Chem. Mater., 13, 1506–1511 (2001). 52. Gowd, E. B., Shibayama, N., Tashiro, K. Structural changes in thermally induced phase transitions of uniaxially oriented de form of syndiotactic polystyrene investigated by temperature-dependent measurements of X-ray fiber diagrams and polarized infrared spectra. Macromolecules, 39, 8412–8418 (2006). 53. Rizzo, P., Daniel, C., De Girolamo Del Mauro, A., Guerra, G. Italy, SA2006A22, 2006. 54. Rizzo, P., Daniel, C., De Girolamo Del Mauro, A., Guerra, G. New host polymeric framework and related polar guest cocrystals. Chem. Mater., 19, 3864–3866 (2007). 55. Petraccone, V., Ruiz de Ballesteros, O., Tarallo, O., Rizzo, P., Guerra, G. Nanoporous polymer crystals with cavities and channels. Chem. Mater., 20, 3663– 3668 (2008).

REFERENCES

229

56. Daniel, C., Giudice, S., Guerra, G. Syndiotactic polystyrene aerogels with b, g and e crystallines phases. Chem. Mater., 21, 1028–1034 (2009). 57. Kuznicki, S. M., Bell, V. A., Nair, S., Hillhouse, H. W., Jacubinas, R. M., Braunbarth, C. M., Toby, B. H., Tsapatsis, M. A titanosilicate molecular sieve with adjustable pores for size-selective adsorption of molecules. Nature, 412, 720–724 (2001). 58. Zecchina, A., Bordiga, S., Vitillo, J. G., Ricchiardi, G., Lamberti, C., Spoto, G., Bjorgen, M., Lillerud, K. P. Liquid hydrogen in protonic chabazite. J. Am. Chem. Soc., 127, 6361–6366 (2005). 59. Eddaoudi, M., Li, H., Yaghi, O. M. Highly porous and stable metal-organic frameworks. Structure design and sorption properties. J. Am. Chem. Soc., 122, 1391–1397 (2000). 60. Pan, L., Adams, K. M., Hernandez, H. E., Wang, X., Zheng, C., Hattori, Y. K., Kaneko, K. Porous lanthanide-organic frameworks: synthesis, characterization, and unprecedented gas adsorption properties. J. Am. Chem. Soc., 125, 3062–3067 (2003). 61. Kitaura, R., Seki, K., Akiyama, G., Kitagawa, S. Porous coordination-polymer crystals with gated channels specific for supercritical gases. Angew. Chem. Int. Ed., 42, 428–431 (2003). 62. Millward, A. R., Yaghi, O. M. Metal-organic frameworks with exceptionally high capacity for storage of carbon dioxide at room temperature. J. Am. Chem. Soc., 127, 17998–17999 (2005). 63. Soldatov, D. V., Moudrakovski, I. L., Ripmeester, J. A. Organic zeolites: Dipeptides as microporous materials. Angew. Chem. Int. Ed., 43, 6308–6311 (2004). 64. Blau, W. J., Fleming, A. J. Materials science: Designer nanotubes by molecular self-assembly. Science, 304, 1457–1458 (2004). 65. Atwood, J. L., Barbour, L. J., Jerga, A. A new type of material for the recovery of hydrogen from gas mixtures. Angew. Chem. Int. Ed., 43, 2948–2950 (2004). 66. Sozzani, P., Bracco, S., Comotti, A., Ferretti, L., Simonutti, R. Methane and carbon dioxide storage in a porous van der Waals crystal. Angew. Chem. Int. Ed., 44, 1816–1820 (2005). 67. Immirzi, A., de Candia, F., Iannelli, P., Zambelli, A., Vittoria, V. Solvent-induced polymorphism in syndiotactic polystyrene. Makromol. Chem. Rapid Commun., 9, 761–764 (1988). 68. Tsujita, Y., Yoshimizu, H., Okamoto, S. Smart membrane: Preparation of molecular cavity and preferential sorption of small molecule. J. Mol. Struct., 739, 3–12 (2005). 69. Rizzo, P., Lamberti, M., Albunia, A. R., Ruiz de Ballesteros, O., Guerra, G. Crystalline orientation in syndiotactic polystyrene cast films. Macromolecules, 35, 5854–5860 (2002). 70. Rizzo, P., Costabile, A., Guerra, G. Perpendicular orientation of host polymer chains in clathrate thick films. Macromolecules, 37, 3071–3076 (2004). 71. Daniel, C., Avallone, A., Rizzo, P., Guerra, G. Control of crystal size and orientation in polymer films by host-guest interactions. Macromolecules, 39, 4820–4823 (2006).

230


72. Rizzo, P., Albunia, A. R., Milano, G., Venditto, V., Guerra, G., Mensitieri, G., Di Maio, L. Crystalline orientation and molecular transport properties in nanoporous syndiotactic polystyrene films. Macromol. Symp., 185, 65–75 (2002). 73. Rizzo, P., Spatola, A., De Girolamo Del Mauro, A., Guerra, G. Polymeric films with three different uniplanar crystalline phase orientations. Macromolecules, 38, 10089–10094 (2005). 74. Albunia, A. R., Annunziata, L., Guerra, G. Guest-induced syndiotactic polystyrene cocrystal formation from g and a phases. Macromolecules, 41, 2683–2688 (2008). 75. Albunia, A. R., Rizzo, P., Tarallo, O., Petraccone, V., Guerra, G. Layers of closepacked alternated enantiomorphous helices and the three different uniplanar orientations of syndiotactic polystyrene. Macromolecules, 41, 8632–8642 (2008). 76. Kobayashi, M., Nakaoki, T., Ishihara, N. Molecular conformation in glasses and gels of syndiotactic and isotactic polystyrenes. Macromolecules, 23, 78–83 (1990). 77. Deberdt, F., Berghmans, H. Phase behavior of syndiotactic polystyrene-o-xylene. Polymer, 35, 1688–1693 (1994). 78. Daniel, C., Dammer, C., Guenet, J. M. On the definition of thermoreversible gels: The case of syndiotactic polystyrene. Polymer, 35, 4243–4246 (1994). 79. Kobayashi, M., Yoshioka, T., Kozasa, T., Tashiro, K., Suzuki, J.-I., Funahashi, S., Izumi, Y. Structure of physical gels formed in syndiotactic polystyrene/solvent systems studied by small-angle neutron scattering. Macromolecules, 27, 1349–1354 (1994). 80. Roels, T., Deberdt, F., Berghmans, H. Thermoreversible gelation in syndiotactic polystyrene solvent systems. Progr. Colloid Polym. Sci., 102, 82–85 (1996). 81. Daniel, C., Deluca, M. D., Guenet, J. M., Brulet, A., Menelle, A. Thermoreversible gelation of syndiotactic polystyrene in benzene. Polymer, 7, 1273–1280 (1996). 82. Daniel, C., Menelle, A., Brulet, A., Guenet, J.-M. Thermoreversible gelation of syndiotactic polystyrene in toluene and chloroform. Polymer, 38, 4193–4199 (1997). 83. Rastogi, S., Goossens, J. G. P., Lemstra, P. J. An in situ SAXS/WAXS/Raman spectroscopy study on the phase behavior of syndiotactic polystyrene (sPS)/ solvent systems: Compound formation and solvent (dis)ordering. Macromolecules, 31, 2983–2998 (1998). 84. Van Hooy-Corstjens, C. S. J., Magusin, P. C. M., Rastogi, S., Lemstra, P. J. Comparative study on gels and clathrates of syndiotactic polystyrene: Solvent mobility in polymer-solvent compounds. Macromolecules, 35, 6630–6637 (2002). 85. Daniel, C., Musto, P., Guerra, G. Clathrate phase in syndiotactic polystyrene gels. Macromolecules, 35, 2243–2251 (2002). 86. Daniel, C., Avallone, A., Guerra, G. Syndiotactic polystyrene physical gels: Guest influence on structural order in molecular complex domains and gel transparency. Macromolecules, 39, 7578–7582 (2006). 87. Daniel, C., Alfano, D., Guerra, G., Musto, P. Evaluation of the amount and composition of the polymer-rich and polymer-poor phases of syndiotactic polystyrene gels with binary solvent Mixture. Macromolecules, 36, 5742–5750 (2003). 88. Daniel, C., Alfano, D., Guerra, G., Musto, P. Physical gelation of syndiotactic polystyrene in the presence of large molar volume solvents induced by volatile guests of clathrate phases. Macromolecules, 36, 1713–1716 (2003).

REFERENCES

231

89. Vittoria, V., Russo, R., de Candia, F. Effect of thermal treatments on the structure of syndiotactic polystyrene. J. Macromol. Sci. Phys., B28, 419–431 (1989). 90. Vittoria, V., Russo, R., de Candia, F. Solvent-induced crystallization of syndiotactic polystyrene in different liquids—Sorption and diffusion phenomena. Polymer, 32, 3371–3375 (1991). 91. Rapacciuolo, M., De Rosa, C., Guerra, G., Mensitieri, G., Apicella, A., Del Nobile, M. A. Different solvent stability of the crystalline polymorphic forms of syndiotactic polystyrene. J. Mater. Sci. Lett., 10, 1084–1087 (1991). 92. Gupper, A., Chan, K. L. A., Kazarian, S. G. FT-IR imaging of solvent-induced crystallization in polymers. Macromolecules, 37, 6498–6503 (2004). 93. Yoshioka, A., Tashiro, K. Polymer-solvent interactions in crystalline d form of syndiotactic polystyrene viewed from the solvent-exchange process in the d form and the solvent evaporation phenomenon in the thermally induced d-g phase transition. Macromolecules, 36, 3593–3600 (2003). 94. Uda, Y., Kaneko, F., Kawaguchi, T. Selective guest uptake from solvent mixtures in the clathrate phase of syndiotactic polystyrene. Macromol. Rapid Commun., 25, 1900–1904 (2004). 95. Rizzo, P., Della Guardia, S., Guerra, G. Perpendicular chain axis orientation in s-PS films: Achievement by guest-induced clathrate formation and maintenance after transitions toward helical and trans-planar polymorphic forms. Macromolecules, 37, 8043–8049 (2004). 96. Uda, Y., Kaneko, F., Kawaguchi, T. Guest exchange with n-alkanes and hostguest interactions in the clathrate phase of syndiotactic plystyrene. Macromolecules, 38, 3320–3326 (2005). 97. Annunziata, L., Albunia, A. R., Venditto, V., Mensitieri, G., Guerra, G. Polymer/ gas clathrates for gas storage and controlled release. Macromolecules, 39(26), 9166–9170 (2006). 98. Albunia, A. R., Minucci, T., Guerra, G. Ethylene removal by sorption from polymeric crystalline frameworks. J. Mater. Chem., 18, 1046–1050 (2008). 99. Tamai, Y., Fukuda, M. Thermally induced phase transition of crystalline syndiotactic polystyrene studied by molecular dynamics simulation. Macromol. Rapid Commun., 23, 891–895 (2002). 100. De Rosa, C., Guerra, G., Petraccone, V., Corradini, P. Crystal structure of the a-form of syndiotactic polystyrene. Polymer J., 23, 1435–1442 (1991). 101. De Rosa, C. Crystal structure of the trigonal modification (a form) of syndiotactic polystyrene. Macromolecules, 29, 8460–8465 (1996). 102. Cartier, L., Okihara, T., Lotz, B. The “a” superstructure of syndiotactic polystyrene: A frustrated structure. Macromolecules, 31, 3303–3310 (1998). 103. Torres, F. J., Civalleri, B., Pisani, C., Musto, P., Albunia, A. R., Guerra, G. Normal vibrational analysis of a trans-planar syndiotactic polystyrene chain. J. Phys. Chem. B., 111, 6327–6335 (2007). 104. De Rosa, C., Rapacciuolo, M., Guerra, G., Petraccone, V., Corradini, P. On the crystal structure of the orthorhombic form of syndiotactic polystyrene. Polymer, 33, 1423–1428 (1992). 105. Chatani, Y., Shimane, Y., Ijitsu, T., Yukinari, T. Structural study on syndiotactic polystyrene: 3. Crystal structure of planar form I. Polymer, 34, 1625–1629 (1993).

232


106. Albunia, A. R., Rizzo, P., Guerra, G., Torres, F. J., Civalleri, B., Zicovich-Wilson, C. M. Uniplanar orientations as a tool to assign vibrational modes of polymer chain. Macromolecules, 40, 3895–3897 (2007). 107. Torres, F. J., Civalleri, B., Meyer, A., Musto, P., Albunia, A. R., Rizzo, P., Guerra, G.B. J. Chem. Phys. B, 113, 5059–5071 (2009). 108. Albunia, A. R., Di Masi, S. B, Rizzo, P., Milano, G., Musto, P., Guerra, G. Chlorinated guest orientation and mobility in clathrate structures formed with syndiotactic polystyrene. Macromolecules, 36, 8695–8703 (2003). 109. Albunia, A. R., Milano, G., Venditto, V., Guerra, G. Clear-cut experimental method to discriminate between in-plane and out-of-plane molecular transition moments. J. Am. Chem. Soc., 127, 13114–13115 (2005). 110. Milano, G., Guerra, G., Müller-Plathe, F. Anisotropic diffusion of small penetrants in the d crystalline phase of syndiotactic polystyrene: A molecular dynamics simulation study. Chem. Mater., 14, 2977–2982 (2002). 111. Venditto, V., De Girolamo Del Mauro, A., Mensitieri, G., Milano, G., Musto, P., Rizzo, P., Guerra, G. Anisotropic guest diffusion in the d crystalline host phase of syndiotactic polystyrene: Transport kinetics in films with three different uniplanar orientations of the host phase. Chem. Mater., 18, 2205–2210 (2006). 112. Albunia, A. R., Musto, P., Guerra, G. FTIR spectra of pure helical crystalline phases of syndiotactic polystyrene. Polymer, 47, 234–242 (2006). 113. Keller, A. Aspects of polymer gels. Farad. Discus., 101, 1–49 (1995). 114. Reynolds, N. M., Savage, J. D., Hsu, S. L. A spectroscopic study of syndiotactic polystyrene. Macromolecules, 22, 2867–2869 (1989). 115. Ruvolo Filho, A., Vittoria, V. Correlation between crystallinity and some infrared bands in the spectra of syndiotactic polystyrene. Makromol. Chem. Rapid. Commun., 11, 199 (1990). 116. Guerra, G., Musto, P., Karasz, F. E., MacKnight, W. J. Fourier-transform infrared spectroscopy of the polymorphic forms of syndiotactic polystyrene. Makromol. Chem., 191, 2111–2119 (1990). 117. Musto, P., Tavone, S., Guerra, G., De Rosa, C. Evaluation by Fourier transform infrared spectroscopy of the different crystalline forms in syndiotactic polystyrene samples. J. Polym. Sci. Polym. Phys., 35, 1055–1066 (1997). 118. Tashiro, K., Ueno, Y., Yoshioka, A., Kobayashi, M. Molecular mechanism of solvent-induced crystallization of syndiotactic polystyrene glass. 1. Time-resolved measurements of infrared/Raman spectra and X-ray diffraction. Macromolecules, 34, 310–315 (2001). 119. Gowd, E. B., Nair, S. S., Ramesh, C. Crystalline transitions of the clathrate (d) form of syndiotactic polystyrene during heating: Studies using high-temperature X-ray diffraction. Macromolecules, 35, 8509–8514 (2002). 120. Musto, P., Mensitieri, G., Cotugno, S., Guerra, G., Venditto, V. Probing by time-resolved FTIR spectroscopy mass transport, molecular interactions, and conformational ordering in the system chloroform-syndiotactic polystyrene. Macromolecules, 35, 2296–2304 (2002). 121. Musto, P., Rizzo, P., Guerra, G. Host-guest interactions and crystalline structure evolution in clathrate phases formed by syndiotactic polystyrene and 1,2dichloroethane: A two-dimensional FTIR spectroscopy investigation. Macromolecules, 38, 6079–6089 (2005).

REFERENCES

233

122. Capitani, D., Segre, A. L., Grassi, A., Sykora, S. Pulsed proton NMR relaxation in crystalline syndiotactic polystyrene. Macromolecules, 24(2), 623–624 (1991). 123. Capitani, D., De Rosa, C., Ferrando, A., Grassi, A., Segre, A. L. Polymorphism in syndiotactic polystyrene: A proton NMR relaxation study. Macromolecules, 25(15), 3874–3880 (1992). 124. Trezza, E., Grassi, A. Dynamics of aromatic molecules clathrated in crystalline syndiotactic polystyrene: A solid state 2H NMR investigation of the host/guest complexes. Macromol. Rapid Commun., 23, 260–263 (2002). 125. Albunia, A. R., Graf, R., Guerra, G., Spiess, H. W. 2H NMR study of aromatic guest dynamics in clathrate phases of syndiotactic polystyrene. Macromol. Chem. Phys., 206, 715–724 (2005). 126. Albunia, A. R., Graf, R., Grassi, A., Guerra, G., Spiess, H. W. Macromolecules, 42, 4929–4931 (2009). 127. Itagaki, H., Mochizuki, J. Size and distribution of free volume in thermoreversible gels of syndiotactic polystyrene. Macromolecules, 38, 9625–9630 (2005). 128. Venditto, V., Milano, G., de Girolamo Del Mauro, A., Guerra, G., Mochizuki, J., Itagaki, H. Orientation and microenvironment of naphthalene guest in the host nanoporous phase of syndiotactic polystyrene. Macromolecules, 38, 3696–3702 (2005). 129. Itagaki, H., Sago, T., Uematsu, M., Yoshioka, G., Correa, A., Venditto, V., Guerra, G. Guest orientation in uniplanar-axial polymer host films and in cocrystal unit-cell, determined by angular distributions of polarized guest fluorescence. Macromolecules, 41, 9156–9164 (2008). 130. Milano, G., Guerra, G., Cavallo, L. A preliminary study of host-guest interactions in polymeric clathrates. An ab initio study of the model complexes benzene/X2 (X = F, Cl, Br, I). Eur. J. Inorg. Chem., 1, 1513–1517 (1998). 131. Milano, G., Guerra, G., Cavallo, L. Ab initio and molecular mechanics study of conformational selectivity of chlorinated compounds adsorbed in the clathrate phase of syndiotactic polystyrene. The role of electrostatic host-guest interactions. Macromol. Theory Simul., 10, 349–354 (2001). 132. Tamai, Y., Fukuda, M. Effect of encaged aromatic guests on the shape and connectivity of molecular cavity in crystalline polystyrene evaluated by molecular simulations. J. Chem. Phys., 121, 12085–12093 (2004). 133. Tamai, Y., Fukuda, M. Sorption of organic solvents on the surface of crystalline syndiotactic polystyrene studied by molecular dynamics simulation. J. Molec. Struct., 739, 27–32 (2005). 134. Tamai, Y., Tsujita, Y., Fukuda, M. Reorientational relaxation of aromatic molecules in the molecular cavity of crystalline syndiotactic polystyrene studied by molecular dynamics simulation. J. Mol. Struct., 739, 33–40 (2005). 135. Milano, G., Guerra, G. Understanding at molecular level of nanoporous and cocrystalline materials based on Syndiotactic Polystyrene. Progr. Mater. Sci., 54, 68–88 (2009). 136. Point, J. J., Jasse, B., Dosiere, M. Intercalates with linear polymer hosts. Part 2. Fourier transform infrared study of poly (ethylene oxide)-p-disubstituted benzene intercalates. J. Phys. Chem., 90, 3273–3277 (1986).

234


137. Guerra, G., Manfredi, C., Musto, P., Tavone, S. Guest conformation and diffusion into amorphous and emptied clathrate phases of syndiotactic polystyrene. Macromolecules, 31, 1329–1334 (1998). 138. Musto, P., Manzari, M., Guerra, G. Isothermal guest desorption from crystalline and amorphous phases of syndiotactic polystyrene. Macromolecules, 32, 2770– 2776 (1999). 139. Musto, P., Manzari, M., Guerra, G. Spectroscopic investigation of host-guest interactions into clathrate phases of syndiotactic polystyrene containing chlorinated compounds. Macromolecules, 33, 143–149 (2000). 140. Daniel, C., Galdi, N., Montefusco, T., Guerra, G. Syndiotactic polystyrene clathrates with polar guest molecules. Chem. Mater., 19, 3302–3308 (2007). 141. Albunia, A. R., Grassi, A., Milano, G., Rizzo, P., Venditto, V., Musto, P., Guerra, G. Oriented nanoporous host delta phases of syndiotactic polystyrene as a tool for spectroscopic investigation of guest molecules. Macromol. Symp., 234, 102– 110 (2006). 142. Nishijima, Y., Onogi, Y., Asai, T. Fluorescence method for studying molecular orientation in polymer solids. J. Polym. Sci., Polym. Symp., 15, 237–248 (1966). 143. Nishijima, Y. Fluorescence methods in polymer science. J. Polym. Sci., Polym. Symp., 31, 353–373 (1970). 144. Larobina, D., Sanguigno, L., Venditto, V., Guerra, G., Mensitieri, G. Gas sorption and transport in syndiotactic polystyrene with nanoporous crystalline phase. Polymer, 45, 429–436 (2004). 145. Mensitieri, G., Larobina, D., Guerra, G., Venditto, V., Fermeglia, M., Pricl, S. Chloroform sorption in nanoporous crystalline and amorphous phases of syndiotactic polystyrene. J. Polym. Sci. Part B: Polym. Phys., 46, 8–15 (2008). 146. Uda, Y., Kaneko, F., Tanigaki, N., Kawaguchi, T. The first example of a polymercrystal-organic-dye composite material: The clathrate phase of syndiotactic polystyrene with azulene. Adv. Mater., 17, 1846–1850 (2005). 147. De Girolamo Del Mauro, A., Carotenuto, M., Venditto, V., Petraccone, V., Scoponi, M., Guerra, G. Fluorescence of syndiotactic polystyrene/ trimethylbenzene clathrate and intercalate co-crystals. Chem. Mater., 19, 6041– 6046 (2007). 148. Stegmaier, P., De Girolamo Del Mauro, A., Venditto, V., Guerra, G. Optical recording materials based on photoisomerization of guest molecules of a polymeric crystalline host phase. Adv. Mater., 17, 1166–1168 (2005). 149. D’Aniello, C., Musto, P., Venditto, V., Guerra, G. Photoisomerization patterns based on molecular complex phases of syndiotactic polystyrene. J. Mater. Chem., 17, 531–535 (2007). 150. Biasucci, M., Bobba, F., Cucolo, A., Daniel, C., D’Aniello, C., Guerra, G., in preparation. 151. Kaneko, F., Kajiwara, A., Uda, Y., Tanigaki, N. Molecular-complex formation of syndiotactic polystyrene with stable radical molecules. Macromol. Rapid Commun., 27, 1643–1647 (2006). 152. Dauben, W. G., Cargill, R. L. Photochemical transformations. VIII. Isomerization of bicyclo[2.2.1]hepta-2,5-diene to quadricyclo[2.2.1.02.6.03,5]-heptane (quadricyclene). Tetrahedron, 5, 197–201 (1961).

REFERENCES

235

153. King, R. B., Sweet, M. Polymer-anchored cobalt tetraarylporphyrin catalysts for the conversion of quadricyclane to norbornadiene. J. Org. Chem., 44, 385–391 (1979). 154. Cuppoletti, A., Dinnocenzo, J. P., Goodman, J. L., Gould, I. R. Bond-coupled electron transfer reactions: Photoisomerization of norbornadiene to quadricyclane. J. Phys. Chem. A, 103, 11253–11256 (1999). 155. Dubonosov, A. D., Bren, V. A., Minkin, V. I. CRC Handbook of Organic Photochemistry and Photobiology, 2nd edn., CRC Press, Boca Raton, 2004. 156. Maruyama, K., Tamiaki, H., Kawabata, S. Development of a solar energy storage process. Photoisomerization of a norbornadiene derivative to a quadricyclane derivative in an aqueous alkaline solution. J. Org. Chem., 50, 4742–4749 (1985). 157. Maafi, M., Aaron, J.-J., Lion, C. Effect of the formation of inclusion complexes with b-cyclodextrin on the photoisomerization and fluorescence properties of aromatic norbornadiene derivatives. J. Inclus. Phenom., 30, 227–241 (1998). 158. Kato, Y., Muta, S., Takahashi, K., Horie, T., Nagai, T. Large photoinduced refractive index change of polymer films containing and bearing norbornadiene groups and its application to submicron-scale refractive-index patterning. Polymer J., 33, 868–873 (2001). 159. Nagai, T., Shimada, M., Ono, Y., Nishikubo, T. Synthesis of new photoresponsive polymers containing trifluoromethyl-substituted norbornadiene moieties. Macromolecules, 36, 1786–1792 (2003). 160. Robello, D., Farid, S. Y., Dinnocenzo, J. P., Gillmore, J. G., U.S. Patent Appl., 00381461A1, 2004. 161. Manfredi, C., De Rosa, C., Guerra, G., Rapacciuolo, M., Auriemma, F., Corradini, P. Structural changes induced by thermal treatments on emptied and filled clathrates of syndiotactic polystyrene. Macromol. Chem. Phys., 196, 2795–2808 (1995). 162. Kistler, S. S. Coherent expanded aerogels and jellies. Nature, 127, 741 (1931). 163. Gesser, H. D., Goswami, P. C. Aerogels and related porous materials. Chem. Rev., 89, 765–788 (1989). 164. Fricke, J., Tillotson, T. Aerogels: Production, characterization, and applications. Thin Solid Films, 297, 212–223 (1997). 165. Reichenauer, G. Kirk-Othmer Encyclopedia of Chemical Technology. John Wiley and Sons, 2008. 166. Daniel, C., Alfano, D., Venditto, V., Cardea, S., Reverchon, E., Larobina, D., Mensitieri, G., Guerra, G. Aerogels with a microporous crystalline host phase. Adv. Mater., 17, 1515–1518 (2005). 167. Guerra, G., Mensitieri, G., Venditto, V., Reverchon, E., Daniel, C. Italy, WO 2005012402 (to University of Salerno), 2005. 168. Malik, S., Rochas, C., Guenet, J. M. Syndiotactic polystyrene/naphthalene intercalates: Preparing thermo reversible fibrillar gels from a solid solvent. Macromolecules, 38, 4888–4893 (2005). 169. Malik, S., Roizard, D., Guenet, J. M. Multiporous material from fibrillar syndiotactic polystyrene intercalates. Macromolecules, 39, 5957–5959 (2006). 170. Daniel, C., Sannino, D., Guerra, G. Syndiotactic polystyrene aerogels: Adsorption in amorphous pores and absorption in crystalline nanocavities. Chem. Mater., 20, 577–582 (2008).

236


171. Puleo, A. C., Paul, D. R., Wong, P. K. Gas sorption and transport in semicrystalline poly (4-methyl-1-pentene). Polymer, 30, 1357–1366 (1989). 172. Manfredi, C., Del Nobile, M. A., Mensitieri, G., Guerra, G., Rapacciuolo, M. Vapor sorption in emptied clathrate samples of syndiotactic polystyrene. J. Polym. Sci. Part B: Polym. Phys., 35, 133–140 (1997). 173. Yamamoto, Y., Kishi, M., Amutharani, D., Sivakumar, M., Tsujita, Y., Yoshimizu, H. Study on d-form complex in syndiotactic polystyrene-organic molecules system III. Fine structures of complexes with xylene isomers by means of high-resolution solid-state 13C NMR spectroscopy. Polymer J., 35, 465–469 (2003). 174. Saitoh, A., Amutharani, D., Yamamoto, Y., Tsujita, Y., Yoshimizu, H., Okamoto, S. Structure and properties of the mesophase of syndiotactic polystyrene IV. Release of guest molecules from d form of syndiotactic polystyrene by time resolved FT-IR and WAXD measurement. Polymer J., 35, 868–871 (2003). 175. Mensitieri, G., Venditto, V., Guerra, G. Polymeric sensing films absorbing organic guests into a nanoporous host crystalline phase. Sens. Actuat. B, 92, 255–261 (2003). 176. Giordano, M., Russo, M., Cusano, A., Cutolo, A., Mensitieri, G., Nicolais, L. Optical sensor based on ultrathin films of d-form syndiotactic polystyrene for fast and high resolution detection of chloroform. Appl. Phys. Lett., 85, 5349–5351 (2004). 177. Giordano, M., Russo, M., Cusano, A., Mensitieri, G., Guerra, G. Syndiotactic polystyrene thin film as sensitive layer for an optoelectronic chemical sensing device. Sens. Actuat. B, 109, 177–184 (2005). 178. Cusano, A., Pilla, P., Contessa, L., Iadiccio, A., Campopiano, S., Cutolo, A., Giordano, M., Guerra, G. High-sensitivity optical chemosensor based on coated long-period gratings for sub-ppm chemical detection in water. Appl. Phys. Lett., 87, 234105/1–234105/3 (2005). 179. Cusano, A., Pilla, P., Contessa, L., Campopiano, S., Cutolo, A., Giordano, M., Guerra, G. Coated long-period fiber gratings as high-sensitivity optochemical sensors. J. Lightwave Technol., 24, 1776–1786 (2006). 180. Di Molfetta, A., Sethi, R. Clamshell excavation of a permeable reactive barrier. Environ. Geol., 50, 361–369 (2006). 181. Jeen, S. W., Jambor, J. L., Blowes, D. W., Gillham, R. W. Precipitates on granular iron in solutions containing calcium carbonate with trichloroethene and hexavalent chromium. Environ. Sci. Techn., 41, 1989–1994 (2007). 182. Stanzel, M. H. Remove organics by activated adsorption. Chem. Eng. Progr., 89, 36–43 (1993). 183. Green, M. M., Kathri, C., Peterson, N. C. A macromolecular conformational change driven by a minute chiral solvation energy. J. Am. Chem. Soc., 115, 4941–4942 (1993). 184. Rowan, A. E., Nolte, R. J. M. Helical molecular programming. Angew. Chem., 37, 63–68 (1998). 185. Yashima, E., Maeda, K., Okamoto, Y. Memory of macromolecular helicity assisted by interaction with achiral small molecules. Nature, 399, 449–451 (1999). 186. Green, M. M., Park, J. W., Sato, T., Teramoto, A., Lifson, S., Selinger, R. L. B. The macromolecular route to chiral amplification. Angew. Chem., 38, 3139–3154 (1999).

REFERENCES

237

187. Nakano, T., Okamoto, Y. Synthetic helical polymers: Conformation and function. Chem. Rev., 101, 4013–4038 (2001). 188. Borovkov, V. V., Lintuluoto, J. M., Inoue, Y. Supramolecular chirogenesis in zinc porphyrins: Mechanism, role of guest structure, and application for the absolute configuration determination. J. Am. Chem. Soc., 123, 2979–2989 (2001). 189. Nakashima, H., Koe, J. R., Torimitsu, K., Fujiki, M. Transfer and amplification of chiral molecular information to polysilylene aggregates. J. Am. Chem. Soc., 123, 4847–4848 (2001). 190. Yashima, E., Maeda, K., Nishimura, T. Detection and amplification of chirality by helical polymers. Chem. Eur. J., 10, 42–51 (2004). 191. Buono, A. M., Immediata, I., Rizzo, P., Guerra, G. Detection and memory of nonracemic molecules by a racemic host polymer film. J. Am. Chem. Soc., 129, 10992–10993 (2007). 192. Guadagno, L., Raimondo, M., Silvestre, C., Immediata, I., Rizzo, P., Guerra, G. Processing, thermal stability and morphology of chiral sensing syndiotactic polystyrene films. J. Mater. Chem., 18, 567–572 (2008). 193. Napoli, M., Daniel, C., Immediata, I., Longo, P., Guerra, G. Nanoporous crystalline and cross-linked polymeric materials. Macromolecules, 42, 5566–5571 (2009). 194. Borriello, A., Agoretti, P., Ambrosio, L., Fasano, G., Pellegrino, M., Venditto, V., Guerra, G. Syndiotactic polystyrene films with sulfonated amorphous phase and nanoporous crystalline phase. Chem. Mater., 21, 3191–3196 (2009).

CHAPTER 11

Crystallization Thermodynamics and Kinetics of Syndiotactic Polystyrene TOMOAKI TAKEBE1 and KOMEI YAMASAKI2 1 2

Chemicals Development Center, Idemitsu Kosan Co., Ltd., Chiba, Japan Polymer Development Center, Idemitsu Kosan Co., Ltd., Chiba, Japan

11.1 INTRODUCTION Syndiotactic polystyrene (SPS) has a high stereoregularity, that is, syndiotactic structure, which leads to the specific aspects of a high melting point (Tm = 270 °C) and a fast crystallization rate. Isotactic polystyrene (IPS) is also well known as a crystalline polystyrene with high stereoregularity, that is, isotactic structure [1,2]. The crystallization of IPS is extremely slow [3]; thus, it has been of no commercial use. On the contrary, SPS has a crystallization rate high enough to be utilized for various purposes. Thus, there is a special interest in the fact that the crystallization rate of SPS is much faster than that of IPS. When SPS is crystallized from the bulk, a large number of spherulites generate spontaneously and grow with a rate dependent on the crystallization temperature. Usually the spherulites are actually spherical in shape only during the initial stage of the crystallization. During the later stage of crystallization, the spherulites impinge on their neighbors. In many cases, the spherulites are nucleated simultaneously, so that the shape of the spherulites is polygonal and the truncation of the spherical contour arises from impinging on one another. Finally, the spherulites form structures that pervade the entire mass of the material. Figure 11.1 shows the photograph of SPS spherulites observed by scanning electron microscopy. The volume filling spherulites can be clearly seen. It is important to establish the equilibrium melting point Tm0 as accurately as possible because the crystallization parameters derived from the kinetic data are rather sensitive to Tm0 . Thus, Tm0 has carefully been established from the dependence of the melting point on the fold thickness. Syndiotactic Polystyrene, Edited by Jürgen Schellenberg Copyright © 2010 John Wiley & Sons, Inc.

238

THEORETICAL BACKGROUND

239

10 μm

Figure 11.1 microscopy.

Photograph of SPS spherulites observed by scanning electron

The characteristics of the sample used are Mw = 1.9 × 105 g/mol, Mw/Mn = 2.5, and [rrrr]>95 mol%. The light scattering apparatus for the measurements was especially constructed in the laboratory.

11.2

THEORETICAL BACKGROUND

The radial growth rates of polymeric spherulites are expressed as a function of undercooling from their equilibrium melting point according to the wellknown equation of Hoffman and Lauritzen [4,5], Kg U* ⎡ ⎤ ⎡ ⎤ G = G0 exp ⎢ − exp ⎢ − ⎥⎦ ⎥ ( ( ) ) R T − T T Δ T f ⎣ ⎣ c c ∞ ⎦

(11.1)

where G0 is a pre-exponential factor that includes all terms that are taken as effectively independent of temperature. The first exponential term contains the contribution of the diffusional process to the growth rate and it is closely analogous to the segmental jump

240

CRYSTALLIZATION THERMODYNAMICS AND KINETICS

rate encountered in the viscoelastic analyses. U* is the activation energy for chain motion, T∞ is the hypothetical temperature below which all viscous flow ceases, and Tc is the crystallization temperature. The second term in Equation 11.1 is a strong function of the crystallization temperature Tc, and of the undercooling ΔT, where ΔT ( = Tm0 − Tc ) is measured from the equilibrium melting point Tm0 of the samples. The factor f is a correction term given by f =

2Tc T + Tc 0 m

(11.2)

This is close to unity at a high temperature limit, but becomes of importance at large undercooling ΔT where the heat of fusion significantly depends on temperature. The term Kg is an important factor that is related with the thermodynamic driving force of nucleation, and given by Kg =

nbσσ eTmo kB ΔHf

(11.3)

where n is a constant that equals 4 for Regimes I and III, and 2 for Regime II, b is the thickness of the monolayer, σ is the lateral surface free energy and σe is the fold surface free energy, and ΔHf is the enthalpy of fusion. In Regime I the rate of nucleation is so small that one surface nucleus causes the completion of the entire substrate by spreading. In Regime II the nuclei spreading rate along the crystal growth front is comparable to the nuclei formation rate. In Regime III the nuclei formation rate is substantially faster than the spreading rate.

11.3 EQUILIBRIUM MELTING POINT OF SPS In order to determine the equilibrium melting point, the lamellar structure of SPS has been characterized by small-angle X-ray scattering (SAXS) [6]. SAXS data were collected using a one-dimensional position-sensitive proportional counter. Figure 11.2 shows a plot of the scattering intensity versus the scattering vector q, which is given by (4π/λ)sin(θ/2), where λ is the wavelength of X-ray and θ is the scattering angle. The scattering maximum shifts to lower q as the crystallization temperature increases, which indicates the increase in the long period of the crystalline lamellae. A quantitative analysis of the lamellar structure can be made by using the correlation function analysis [7,8]. Performing the Fourier transformation on the observed SAXS data in Figure 11.2, the one-dimensional electron density correlation function γ(r) can be calculated [7], γ (r ) =

∞

1 1 cos qz ⋅ 4 πq2 ⋅ I ( q) dq 3 ∫ re ( 2 π ) 0

(11.4)

EQUILIBRIUM MELTING POINT OF SPS

241

10,000 260°C 8000

Intensity (a.u.)

250°C

6000 246°C 240°C 4000 236°C

(Crystallization temperature)

2000

0 0.0

0.2

0.4 q /nm–1

0.6

0.8

Figure 11.2 SAXS profiles at various crystallization temperatures Tc.

where z is a unit vector along the lamellar normal direction, and re is the classical electron radius, which is 2.81 × 10−15 m. The curves shown in Figure 11.3 are typical correlation functions. These γ(r) curves were analyzed using a pseudo-two-phase structure model, which has a broad distribution of long period and lamellar thickness and a diffuse phase boundary at the interface of the crystal and the amorphous regions as shown by Figure 11.4. The long period L is determined from the position of the first maximum in γ(r). The number-averaged thickness of the amorphous layer is given by the intersection of the straight line to the baseline of γ(r) [7,8]. The characteristic interfacial thickness E can be estimated by analyzing the asymptotic behavior of the SAXS profile at large q (Porod’s law). An empirical equation based on the model, assumed to be the phase boundary expressed by the Gaussian curve, is given by [9] I ( q) ~ q−4 exp ( −σ b2 q2 )

(11.5a)

ln [ I ( q) q4 ] = const − σ b2 q2

(11.5b)

or

242


Tc = 240°C

Baseline

Figure 11.3 One-dimensional electron density correlation function γ(r) calculated by performing the Fourier transformation on the SAXS profile at Tc = 240 °C. r (r) L

E

rc

ra r

Figure 11.4 Electron density profile of pseudo-two-phase structure model. ρc,ρa: electron density of crystalline and amorphous layers; , : number-averaged thickness of crystalline and amorphous layers; L: long period; E: interfacial thickness.

where σb is a parameter related to the thickness of the interfacial layer. Figure 11.5 shows the plot to evaluate the value of σb from which the value of E was determined by [10] E = 2πσ b

(11.6)

243


Tc = 240°C f = 0.52

ln(q4I(q))

–5

–10

0

2

Figure 11.5

4 q2×105 (nm–2)

6

8

Plot of ln[q4I(q)] against q2.

TABLE 11.1 Strucural Parameters Obtained from the Analyses of Correlation Functions Tc (°C) 240 244 250 254 265

φ

Tm (°C)

L (nm)

(nm)

E (nm)

lc (nm)

0.52 0.53 0.53 0.55 0.64

259 263 266 270 275

195 206 200 213 220

122 134 137 149 173

12 17 12 12 13

110 117 125 137 160

The results of the analyses of the lamellar structure of SPS are summarized in Table 11.1. Also the degree of crystallinity ϕ and the melting point Tm obtained by DSC measurements are listed in Table 11.1. It should be noted that the larger the fold thickness, the higher the melting point. If the lateral dimension of a crystal is much larger than the fold thickness, the melting point of the crystal of a finite thickness is given by ⎛ 2σ e ⎞ Tm = Tm0 ⎜ 1 − ⎝ ΔHf lc ⎟⎠

(11.7)

where σe denotes the fold surface free energy and ΔHf the heat of fusion. According to Equation 11.7, Tm0 is yielded when lc is infinite, that is, lc−1 is going to zero. Figure 11.6 shows the plot of Tm against lc−1, and from the intercept Tm0 is determined as 583K, which is higher than the Tm0 of 558.5K obtained from the Hoffman-Weeks plot [11].

244


600 Tm = Tm0[1 – 2se/ΔHf lc] Tm0 = 583K

Tm (K)

580

560

540

0

0.005 1/lc (Å–1)

Figure 11.6

11.3.1

0.01

Plot of Tm against lc−1.

Evaluation of Spherulitic Growth Rate G

The growth and the structure of spherulites may also be studied by small-angle light scattering. The sample is placed between polarizers, the laser beam is passed through, and the resultant scattered beam is photographed. When the incident radiation is vertical in polarization but the analyzer is horizontal (crossed nicols), an Hv pattern is obtained as shown by Figure 11.7. This scattering pattern has been theoretically calculated by Stein and Rhodes, and is given by [12] I Hv = KVS2 (α r − α t ) sin 2 2μ S (U ) U 6

(11.8)

sin x dx x

(11.9)

2

S(U ) = 4 sin U − U cosU − 3∫

U

0

U=

4 πR θ sin λ 2

(11.10)

where VS is the volume of spherulite with the radius R, given by 4πR3/3, (αr - αt) is the optical anisotropy of the spherulite, θ and μ are the scattering angle and the azimuthal angle, respectively, and K is a constant. The maximum scattering angle θm (or Um) that occurs in the radial direction is related to R, the radius of the spherulite, given by


245

Figure 11.7 Typical Hv scattering pattern from SPS film crystallized at Tc = 180 °C. The angle mark of 2.5 ° indicates the scattering angle in air.

Um =

4 πR θ sin m = 4.09 λ 2

(11.11)

The time-resolved light scattering experiment was carried out to evaluate the spherulitic growth rate as a function of Tc. When Tc is above the Tc,max ∼ 200 °C, where SPS crystallizes with a maximum rate, the crystallization was induced by the rapid temperature drop from 300 °C to Tc (T-drop). On the other hand, when Tc is below Tc,max, the amorphous samples, which were obtained by quenching the samples from 300 to 0 °C, were heated up to Tc to be crystallized (T-jump). Figure 11.8 shows the typical time evolution of light scattering profiles I(q,t) which were measured under crossed nicols during the isothermal crystallization at Tc = 263 °C, where q is the scattering vector defined as (4π/λ)sin(θ/2). The following trend was observed: the scattering maximum appears after the incubation time for nucleation ti, the maximum intensity Im increases, and the magnitude of the scattering vector qm, which gives rise to the maximum intensity shifts toward smaller values with lapse of time, and qm(t) and Im(t) cease to change at time t > t*, where t* is the time for spherulitic growth to be terminated due to impinging of spherulites. This trend reflects the increase in the spherulitic radius R, defined as [12]


Intensity

246

Figure 11.8 Time-evolution of Hv scattering profiles I(q,t) during the isothermal crystallization at Tc = 180 °C. Each profile was obtained at a certain time (in seconds) after the incubation time for nucleation ti.

R = 4.09 qm

(11.12)

Figure 11.9 shows the change of R and Im with time at T = 180 °C. It should be noted that the time dependences of R(t) and Im(t) are written by R (t ) ~ t α

(11.13)

I m (t ) ~ t β

(11.14)

and

From Figure 11.9, α = 1.0 and β = 6.2 were obtained, and the exponents satisfy the relation β/α = 6, which indicates that the scattering intensity agrees with the prediction by Stein and Rhodes, that is, IHv ∼ R6 [12]. Figure 11.10 shows the spherulitic growth behavior in T-jump (a) and T-drop (b). It was observed that the spherulites grow linearly with time for all Tc. Calculating the growth rates G from the slope of the lines, it was found that G is increasing when Tc is lower than 200 °C; however, G is decreasing when Tc is higher than 200 °C. This suggests that G is a convex function of Tc with the maximum near 200 °C (Fig. 11.11).


247

6.2

1.0

Figure 11.9 Change in spherulitic radius R and maximum scattering intensity Im with time during isothermal crystallization at Tc = 180 °C. t* indicates the time at which the spherulitic growth is terminated. t - ti means the effective time after subtracting ti, which represents the time lag due to the incubation for nucleation and due to the stabilization of oven temperature, from the real time. 30 197°C

25

3

168°C

20 239°C 249°C

2 158°C

R (μm)

R (μm)

178°C

255°C

15 10

1 258°C

5 0 0 (a)

10

20

30 40 t – ti (s)

50

0

60 (b)

0

50

100 t – ti (s)

150

200

Figure 11.10 Change in spherulitic radius with time during isothermal crystallization at various Tc, (a) at lower than 200 °C by T-jump, and (b) at higher than 200 °C by T-drop.

248


Figure 11.11 Spherulitic growth rate curves for SPS obtained in this work and IPS reported by Suzuki et al. [3].

11.4 ANALYSES OF SPHERULITIC GROWTH RATE G The quantitative analysis of the growth rate was made on the basis of the Hoffman-Lauritzen theory expressed by Equation 11.1. The universal values of U* = 1500 cal/mol and T∞ = Tg − 30K were used in these calculations [4]. It is often most convenient to rearrange Equation 11.1 as log G +

Kg U* = log G0 − 2.3R (Tc − T∞ ) 2.3Tc ( ΔT ) f

(11.15)

Plotting the left-hand term of Equation 11.15 against 1/[Tc(ΔT)f] as shown in Figure 11.12, one can obtain the Kg value from the slope and the pre-exponential factor G0 from the intercept. It appears obvious to fit the experimental data with two straight lines, and supports a distinct regime behavior; that is, the regime transition Regime I to II is readily observed in the vicinity of 243–248 °C. The values of the kinetic parameters for SPS determined from the slopes and intercepts of the Regimes I and II are provided in Table 11.2. The value of K gI K gII experimentally obtained is 3.0, which is significantly larger than the theoretical value of K gI K gII = 2 [4,5]. Using Tm0 = 583 K, ΔHf = 5.1 × 107 J/m3 [6,12] and b = 7.2 × 10−10 m [14–16], the product of two crystal surface free energies σσe is calculated to be 52.5 × 10−6 J2/m4. The lateral surface free energies may be estimated from [4,5] σ = 0.1ΔHf ( ab)

12

(11.16)

COMPARISON BETWEEN SPS AND IPS

249

Figure 11.12 Kinetic analysis of the growth data using U* = 1500 cal/mol and T = 343K. Tx(=245.3 °C) indicates the crossover temperature between Regimes I and II.

TABLE 11.2 Fitting Parameters in the Different Regimes

Regime I Regime II

Range of Temperature

G0 (nm/s)

Kg (K2)

243–248 150–160

126 × 10−3 0.12 × 10−3

3.3 × 105 1.2 × 105

where a is the molecular width. Thus, σ is obtained as 4.1 × 10−3 J/m2, and hence the value of the fold surface free energy σe was determined to be 12.8 × 10−3 J/ m2 when σσe was divided by σ. The work of chain folding is obtained directly from the fold surface free energy as [4,5] Q = 2σ e ab

(11.17)

The calculated work of chain folding Q for SPS is 2.4 kcal/mol.

11.5 COMPARISON BETWEEN SPS AND IPS The thermodynamic parameters determined from the spherulitic growth rate in Regime II are listed in Table 11.3 for SPS and IPS [3]. The Q value for SPS is about one-third as small as that derived for IPS. The work of chain folding is closely correlated with molecular structure, and

250


TABLE 11.3 Thermodynamic Parameters Characterizing the Spherulitic Growth for SPS and IPS SPS Mw (g/mol) Tm0 (K) ΔHf (erg/cm3) a (cm) b (cm) G0 (cm/s) σ (erg/cm2) σe (erg/cm2) Q (cal/mol)

1.9 × 105 583 5.1 × 108 8.8 × 10−8 7.2 × 10−8 0.12 4.1 12.8 2.4

IPS 2.2 × 106 515.2 9.11 × 108 12.8 × 10−8 5.5 × 10−8 2.1 × 10−2 7.64 34.8 7.1

the most important contribution to its relative magnitude is thought to be the inherent stiffness of the chain itself. It has been known that polymers with flexible chains such as polyethers have Q values of about 3–4 kcal/mol, intermediate ones (e.g., polyethylene) have Q values close to 5 kcal/mol, and stiffer ones (e.g., IPS) have Q values in the range of 7–9 kcal/mol. Therefore, the work of chain folding has been found to be the most important parameter governing the temperature dependence of the growth rate. From an estimate above, it appears that the work required to form a chain folding for SPS is roughly one-third of that for IPS, and hence chain folding is less of a hindrance to the crystallization for SPS than IPS. This relatively large difference in Q is apparently associated with the planar zig-zag conformation in the SPS molecular chain [17], in contrast to the 3/1 spiral conformation in IPS [18].

REFERENCES 1. Natta, G., Piero, P., Corradini, P., Danusso, F., Mantica, E., Mazzanti, G., Moraglio, G. Crystalline high polymers of a-olefins. J. Am. Chem. Soc., 77, 1708–1710 (1955). 2. Hay, J. N. Crystallization kinetics of high polymers-isotactic polystyrene. J. Polym. Sci. Part A, 3, 433–437 (1965). 3. Suzuki, T., Kovacs, A. J. Temperature dependence of spherulitic growth rate of isotactic polystyrene. Critical comparison with the kinetic theory of surface nucleation. Polymer J., 1, 82–100 (1970). 4. Lauritzen, J. I., Hoffman, J. D. Extension of theory of growth of chain-folded polymer crystals to large undercoolings. J. Appl. Phys., 44, 4340–4352 (1973). 5. Hoffman, J. D., Miller, R. L. Kinetics of crystallization from the melt and chain folding in polyethylene fractions revisited: Theory and experiment. Polymer, 38, 3151–3212 (1997). 6. Takebe, T., Uchida, T., Yamasaki, K. Polym. Prep., 42, 4309 (1993).

REFERENCES

251

7. Strobl, G. R., Schneider, M. Direct evaluation of the electron density correlation function of partially crystalline polymers. J. Polym. Sci., Polym. Phys., 18, 1343– 1359 (1980). 8. Balta-Calleja, F. J., Vonk, C. G. X-ray Scattering of Synthetic Polymers, Elsevier Science Publishing, Amsterdam, 1989, Chap. 7. 9. Porod, G. The x-ray small-angle scattering of close-packed colloid systems. I. Koll. Z., 124, 83–114 (1951). 10. Hashimoto, T., Shibayama, M., Kawai, H. Domain-boundary structure of styreneisoprene block copolymer films cast from solution. 4. Molecular-weight dependence of lamellar microdomains. Macromolecules, 13, 1237–1247 (1980). 11. Arnauts, J., Berghmans, H. Equilibrium melting behavior of syndiotactic polystyrene. Polymer Commun., 31, 343–345 (1990). 12. Stein, R. S., Rhodes, M. B. Photographic light scattering by polyethylene films. J. Appl. Phys., 31, 1873–1884 (1960). 13. Pasztor, A. J., Landes, B. D., Karjala, P. J. Thermal properties of syndiotactic polystyrene. Thermo Chimica. Acta, 177, 187–195 (1991). 14. Greis, O., Xu, Y., Asano, T., Petermann, J. Morphology and structure of syndiotactic polystyrene. Polymer, 30, 590–594 (1989). 15. Chatani, Y., Shimane, Y., Ijitsu, T., Yukinari, T. Structural study on syndiotactic polystyrene: 3. Crystal structure of planar form I. Polymer, 34, 1625–1629 (1993). 16. DeRosa, C., Guerra, G., Petraccone, V., Corradini, P. Crystal structure of the αform of syndiotactic polystyrene. Polymer J., 23, 1435–1442 (1991). 17. Kobatashi, M., Nakaoki, T., Ishihara N. Polymorphic structures and molecular vibrations of syndiotactic polystyrene. Macromolecules, 22, 4377–4382 (1989). 18. Natta, G., Danusso, F., Moraglio, G. Volumetric determination of the degree of crystallinity of isotactic polystyrenes. Macromol. Chem., 28, 166–172 (1958).

PART IV

COMMERCIAL PROCESSES FOR MANUFACTURING OF SYNDIOTACTIC POLYSTYRENE

CHAPTER 12

Processes for the Production of Syndiotactic Polystyrene MASAO AIDA,1 DAVID HABERMANN,2 HANS-JOACHIM LEDER,3 and JÜRGEN SCHELLENBERG3 1

Process Development Center, Idemitsu Kosan Co., Ltd., Chiba, Japan Information Systems Management, The Dow Chemical Company, MI, USA 3 R&D Dow Central Germany, Dow Olefinverbund GmbH, Schkopau, Germany 2

12.1

INTRODUCTION

Many types of commercial styrene polymerization processes are applied. However, the process for the production of syndiotactic polystyrene (SPS) is completely different from those for atactic polystyrene polymerizations. Catalysts are sensitive to the impurities in styrene monomer and SPS is insoluble in aromatic solvents. The SPS process is divided into eight sections. They are monomer purification section, catalyst section, polymerization section, styrene removal from SPS, deactivation section, pelletizing section, blending section, and shipping section. Each section will be explained from the patent information.

12.2

MONOMER PURIFICATION SECTION

Like other polymerization plants, fresh monomer is pretreated to reduce any catalyst poisons. The SPS catalyst is very sensitive like other transition metal catalysts used in coordination polymerizations [1]. Oxygen, water, and phenylacetylene are well-known catalyst poisons. Fresh styrene is prepared through an oxygen stripping column (V-110), an alumina column (V-120), and a hydrogenation column (V-130) to remove oxygen, water, and phenylacetylene. Figure 12.1 shows the flow diagram of the monomer purification section.


255

256

PROCESSES FOR THE PRODUCTION OF SYNDIOTACTIC POLYSTYRENE

To scrubber

V-140 To scrubber

V-110

V-130

F-140

Styrene V-150

V-120

V-100 Inert gas

Fresh styrene

To V-300 To Q-300

Hydrogen

Figure 12.1 Flow diagram of the monomer purification section. V-100: styrene storage tank; V-110: oxygen stripping column; V-120: alumina column; V-130: hydrogenation column; V-140: surge vessel; F-140: filter; V-150: styrene feed tank.

12.3 CATALYST SECTION Styrene is polymerized using a catalyst of a titanium compound and a cocatalyst. The catalyst components are dissolved in an organic solvent and are fed to the reactor. These components can be mixed before the addition to the reactor (Figure 12.2).

12.4 POLYMERIZATION SECTION The SPS polymerization is very unique. Liquid styrene monomer is polymerized with homogeneous catalysts and produces SPS powder in the reactor. This is in contrast to Ziegler–Natta catalysts, which have solid components and powders that can easily be obtained by the growth of the solid components in bulk polyethylene and polypropylene plants. Under limited operational conditions and with a special equipment, SPS powder can be obtained. The SPS polymerization systems are very difficult to design. There are three types of SPS reactor processes from the patent information.

POLYMERIZATION SECTION

Titanium compound

M

Cocatalyst

M

V-200 Organic solvent

MAO

257

V-210

V-220

Catalysts/organic solvent

To V-300 To Q-300

Figure 12.2 Flow diagram of the catalyst section. V-200: titanium compound vessel; V-210: methylaluminoxane (MAO) vessel; V-220: cocatalyst vessel.

The first is a continuous stirred tank reactor process [2]. The SPS polymerization is carried out with a continuous stirred tank reactor with a paddletype agitator [3–5], a helix-type agitator [2,4], an anchor-type agitator [5], or a pipe blade [5]. The powder level is indicated by a driving radiant ray device [6]. High power consumption and high SPS conversion are needed to avoid adhesions. In this process, the polymerization heat (160 kcal/kg) is removed by the jacket coolant. In a larger plant reactor, a heat transfer area proportional to the reactor volume is needed. The second process is a continuous fluidized bed reactor. In an Idemitsu Petrochemical patent [7], a helical ribbon mixer is adopted. High power consumption and high SPS conversion are important to avoid adhesions. In this process, the polymerization heat is removed by the evaporation heat of styrene and by the jacket coolant. The third process is a continuous self-cleaning reactor process [5,8–10]. In the self-cleaning reactor, low adhesion powders can be easily obtained under low SPS conversions. The heat transfer coefficient in the self-cleaning reactor is relatively high. Multiple reactors are necessary to achieve efficiency in the production scale. 12.4.1

Continuous Stirred Tank Reactor Process

Fresh styrene and catalyst components are fed to the jacketed stirred tank reactor with a multi-paddle agitator (V-300). In the reactor, styrene is polymerized to SPS in the presence of catalysts. The SPS powder is discharged from the bottom of the stirred tank reactor. The polymerization heat is removed only by the jacket coolant.

258


From V-200, V-210, V-220 M


V-300

To V-400 SPS powder

Figure 12.3 Flow diagram of a continuous stirred tank reactor process. V-300: stirred tank reactor.

Figure 12.3 shows the flow diagram of the continuous stirred tank reactor process. 12.4.2 Continuous Fluidized Bed Reactor Process Fresh styrene and catalyst components are fed to the fluidized bed reactor (V-300). In the reactor, styrene is polymerized to SPS in the presence of catalysts. The SPS powder is fluidized in the jacketed vessel with a helical ribbon mixer. The polymerization conditions are 70 °C and 225 mmHg under nitrogen flow. Unreacted styrene is taken out from the reactor by a nitrogen flow. Polymerization heat is removed by the evaporation heat of styrene and the jacket coolant. SPS powder is discharged from the bottom of the fluidized bed reactor. The styrene vapor from the top of the reactor is cooled with cooling water (E-310) and chilled with brine (E-320) in heat exchangers. Most of the styrene vapor is condensed in heat exchangers. The styrene condensate is recycled to the reactor. Figure 12.4 shows the flow diagram of the continuous fluidized bed reactor process. 12.4.3

Continuous Self-Cleaning Reactor Process

Styrene is polymerized in two reactors. The first one is a self-cleaning reactor (Q-300) and the second one is a stirred tank reactor with a helical ribbon mixer (V-300) [11]. The catalyst components are fed to the reactors. Fresh styrene is fed only to the self-cleaning reactor. In the reactors, styrene is polymerized to SPS in

POLYMERIZATION SECTION

259

To scrubber Nitrogen

E-310 CW

E-320 BR

From V-200, V-210, V-220 C-320

M


V-300

V-310

V-320 To V-400 SPS powder

Figure 12.4 Flow diagram of a continuous fluidized bed reactor process. V-300: fluidized bed reactor; E-310: 1. Heat exchanger; V-310: 1. Surge tank; E-320: 2. Heat exchanger; V-320: 2. Surge tank; C-320: vacuum pump.

From V-150 Fresh styrene

From V-200, V-210, V-220

M

Catalysts/organic solvent M

V-300 Q-300

SPS powder

To V-400

Figure 12.5 Flow diagram of a continuous self-cleaning reactor process. Q-300: selfcleaning reactor; V-300: stirred tank reactor.

the presence of catalysts. The SPS powder is discharged from the bottom of the stirred tank reactor. Polymerization conditions are 70 °C and 760 mmHg. Polymerization heat (160 kcal/kg) is removed by the jacket coolant. Figure 12.5 shows the flow diagram of a continuous self-cleaning reactor process.

260


12.5 STYRENE STRIPPING SECTION The SPS powder is dried in the dryer. Under high temperatures, SPS powders melt and the adhesion problem occurs in the dryer [12]. Under low temperatures, the drying rate is low. Therefore, drying temperature and pressure are important. The SPS powder is discharged from the bottom of the reactor (V-300) having a styrene monomer content of 20% and is fed to the dryer (Q-400). Subsequently, the SPS powder is discharged from the bottom of the dryer. The styrene vapor from the top of the dryer is cooled with cooling water (E410) and is chilled with brine (E-420) in heat exchangers. Most of the styrene vapor is condensed in heat exchangers. The styrene condensate is recycled to the styrene feed tank (V-150). Drying conditions are 150 °C and 30 mmHg [10]. Figure 12.6 shows the flow diagram of the styrene drying section. 12.6 DEACTIVATING SECTION After the drying section, the catalyst components still show chemical reactivity. Under air, the physical properties of SPS deteriorate and the color changes from white to yellow. This section describes how the residual catalysts are deactivated [13,14]. The dried SPS powder is fed to the deactivation vessel (V-520) and is mixed with a methanolic solution of sodium hydroxide. After the deactivation, the From V-300 SPS powder To scrubber

V-400 E-410 CW

E-420 BR

C-420 M

Q-400 Styrene V-410

V-420

To V-100 To V-521

SPS powder

Figure 12.6 Flow diagram of the styrene drying section. V-400: feed hopper; Q-410: dryer; E-410: 1. Heat exchanger; V-410: 1. Surge tank; E-420: 2. Heat exchanger; V-420: 2. Surge tank; C-420: vacuum pump.

261

DEACTIVATING SECTION

SPS cake is separated from the methanolic solution slurry by a first centrifuge (Q-520). Subsequently, this SPS cake is fed to the washing vessel (V-530) and is mixed with methanol. After the washing process, the SPS cake is separated from the methanolic slurry by a second centrifuge (Q-530) and is fed to the dryer (Q-540). The SPS powder is discharged from the bottom of the dryer. The separated liquids are purified in the methanol column (V-560) and are recycled to the methanol tank (V-500) for reuse. Figures 12.7–12.9 show the flow diagrams of the deactivating section. From Q-400 SPS powder

To V-550 Methanol

NaOH M

V-511

V-521 Q-520

M

Methanol

M

M

M

V-510

V-520

V-530

Q-530

To Q-540 V-500

SPS cake

Figure 12.7 Flow diagram of the deactivating section (1). V-500: methanol tank; V-510: NaOH–methanol tank; V-511: NaOH feed hopper; V-520: deactivation vessel; V-521: SPS feed hopper; Q-520: 1. Centrifuge; V-530: purification vessel; Q-530: 2. Centrifuge; Q-540: methanol dryer; V-550: methanol surge tank.

Air

From Q-530 SPS cake

To scrubber

M

Q-540 To Q-600, V-700A,B,C

Figure 12.8 Flow diagram of the deactivating section (2). Q-540: methanol dryer.

262


E-561 CW

V-561

Methanol

From Q-520, Q-530

To V-500

E-560 STM

Methanol To waste tank V-550

V-560

Figure 12.9 Flow diagram of the deactivating section (3). V-550: methanol surge tank; V-560: methanol column; V-561: methanol receiver; E-560: V-560 reboiler; E-561: V-560 cooler.

12.7 PELLETIZING SECTION The SPS powder shows a particle size distribution and contains also some small particles. This powder may be caught in the wind and may cause environmental problems. Many customers dislike the powdery form, so that the SPS polymer is transformed into pellets. In the extruder, some additives are added to the polymer to improve selected polymer properties as known from other polymers. The dried SPS powder is sent to the powder hopper (V-600A,B). SPS powder from the powder hopper and additional additives are fed through the extruder (Q-600) to the pelletizer (Q-610). Volatile residuals in the SPS polymer are vented and removed from the extruder [15]. Figure 12.10 shows the flow diagram of the pelletizing section.

12.8 BLENDING SECTION In all polymer plants, the polymer pellets show a slight fluctuation in properties. In this section, the SPS pellets are blended and transferred to homogeneous lots. The SPS pellets are sent to the pellet blend silo (V-700A,B) and are blended. After blending, the SPS pellets are sent to the pellet silo. Figure 12.11 shows the flow diagram of the blending section.

SHIPPING SECTION

263

From Q-540 SPS powder

V-600A,B

To V-700 M

SPS pellet Q-600

Z-600

Q-610

Additives

Figure 12.10 Flow diagram of the pelletizing section. V-600A,B: powder hopper; Q-600: extruder; Z-600: strand bath; Q-610: pelletizer.

SPS pellet To V-800A,B From Q-610 SPS pellet

Air V700A

Figure 12.11

V700B

Flow diagram of the blending section. V700A,B: blend silo.

12.9 SHIPPING SECTION The SPS pellets are sent to the pellet silo (V-800A,B) and are shipped by bulk. Some pellets are sent to the packing silo (V-810) and are packed (Z-810). SPS products in containers are sent to the warehouse. Figure 12.12 shows the flow diagram of the shipping section. Some more details on useful analytical methods to control the SPS production process are given in Reference 16.

264


SPS pellet

From V-700

V-810 V800A,B

Air

To warehouse Bulk shipping SPS product

SPS product Z-810

Figure 12.12 Flow diagram of the shipping section. V800A,B: pellet silo; V-810: packing silo; Z-810: packer.

REFERENCES 1. Schellenberg, J. Effect of impurities on the syndiospecific coordination polymerization of styrene. Macromol. Mater. Eng., 290, 833–842 (2005). 2. Shirota, D. JP 63-030049 (to Idemitsu Petrochemical), 1988. 3. Ishikawa K. JP 63-226357 (to Idemitsu Petrochemical), 1988. 4. Ishikawa K. JP 63-226358 (to Idemitsu Petrochemical), 1988. 5. Imabayashi, H. JP 63-226359 (to Idemitsu Petrochemical), 1988. 6. Hirose K. Japan Kokai 03-087303 (to Idemitsu Petrochemical), Process for Producing Thermal Plasticity Polymers, 1991. 7. Yamamoto, K. European Patent 379,128 (to Idemitsu Petrochemical), Process for Producing Styrene-Based Polymers, 1990. 8. Shirota, D. European Patent 328,975 (to Idemitsu Petrochemical), Process for Producing Styrene-Based Polymer and Apparatus for Producing Said Polymers, 1989. 9. Yamamoto, K. Japan Kokai 04-216835 (to Idemitsu Petrochemical), Process for Producing Styrene-Based Polymers, 1992. 10. Imabayashi, H. European Patent 535,582 (to Idemitsu Petrochemical), Process for Producing Styrenic Polymer, 1993. 11. Yamamoto, K. Japan Kokai 01-009929 (to Idemitsu Petrochemical), Process for Producing Styrene-Based Polymers, 1989.

REFERENCES

265

12. Schellenberg, J. Melting properties of syndiotactic polystyrenes and effect of hydrogen on molecular weight distribution. Macromol. Mater. Eng., 290, 675–680 (2005). 13. Kuramoto M. Japan Kokai 01-193877 (to Idemitsu Petrochemical), Process for Purification of Styrene-Based Polymers, 1989. 14. Teshima H. Japan Kokai 03-253570 (to Idemitsu Petrochemical), Process for Purification of Styrene-Based Polymers, 1991. 15. Yamamoto, K. Japan Kokai 01-199331 (to Idemitsu Petrochemical), Process for Removal of Volatile Materials from Styrene-Based Polymers, 1989. 16. Schellenberg, J., Leder, H.-J. Syndiotactic polystyrene: Process and applications. Adv. Polym. Technol., 25, 141–151 (2006).

PART V

PROPERTIES, PROCESSING, AND APPLICATIONS OF SYNDIOTACTIC POLYSTYRENE

CHAPTER 13

Properties of Syndiotactic Polystyrene TOMOAKI TAKEBE,1 KOMEI YAMASAKI,2 KEISUKE FUNAKI,3 and MICHAEL MALANGA4 1

Chemicals Development Center, Idemitsu Kosan Co., Ltd., Chiba, Japan Polymer Development Center, Idemitsu Kosan Co., Ltd., Chiba, Japan 3 Polymer Development Center, Idemitsu Kosan Co., Ltd., Chiba, Japan 4 R&D Advanced Technologies, The Dow Chemical Company, Auburn Hills, MI, USA 2

13.1

INTRODUCTION

The precise synthesis by homogeneous catalysts created a new polymer, syndiotactic polystyrene (SPS), in 1985. SPS is categorized as an engineering plastic based on its performance. Since the discovery of SPS, many studies have been done to characterize and understand the nature of SPS. In this chapter, the properties of SPS are presented in terms of rheological, mechanical, and electrical properties.

13.2

RHEOLOGICAL PROPERTIES OF SPS

This section treats the flow behavior of SPS and particularly focuses on the temperature and molecular weight dependencies. The subject of this section is limited to the flow behavior of SPS in the molten state above its melting point. The effects of the tacticity on the rheological behavior in the amorphous state are discussed by comparing SPS with atactic polystyrene (APS) and isotactic polystyrene (IPS). A cone and plate rheometer was used in order to measure the complex melt viscosity η*(ω) as a function of frequency ω [1]. Figure 13.1 illustrates the master curve of η*(ω, T) data measured at various temperatures T for SPS using the time–temperature superposition principle [1]. The reference temperature Tr is 290 °C. All the data are converted by shifting the curves to overlap the original 290 °C curve.


269

270

PROPERTIES OF SYNDIOTACTIC POLYSTYRENE

Temp. = 290°C

η∗/aTaM (Pa·s)

105

5

Mw = 2.89 × 10 Mw/Mn = 2.59 T = 290°C 280 300 310

T = 290°C 5

Mw = 3.42 × 10 Mw/Mn = 2.47 5

6.03 × 10 5 7.17 × 10

2.53 2.39

ωaTaM (1/s)

Figure 13.1 Master curve of the reduced melt viscosity η*/aTaM as a function of the reduced frequency ωaTaM.

0.30 0.20 0.10

In aT

0.00 –0.10 –0.20 –0.30 –0.40 –0.50 –0.60 1.700

Figure 13.2

1.750 1/T × 103/K–1

1.800

Plot of ln aT against 1/T.

The shift factor aT, shown in Figure 13.2, is found to obey the Arrhenius equation ln aT =

Ea ⎛ 1 1 ⎞ ⎜ − ⎟, R ⎝ T Tr ⎠

(13.1)

RHEOLOGICAL PROPERTIES OF SPS

271

and the activation energy of flow could be obtained as Ea = 21 kcal/mol. The experimental value obtained for SPS is almost equivalent to Ea = 25 kcal/mol for APS [2] and Ea = 20 kcal/mol for IPS [3]. Analogous to the time–temperature superposition principle, a master curve for various SPS polymers of different molecular weights has been tested. The dependence of the molecular weight on the melt viscosity can be conveniently represented by the ratio of any specific relaxation time τp for a polymer with a molecular weight M to its value for one with an arbitrary molecular weight Mr: 3.4

⎛M⎞ = aM = ⎜ . r ⎝ Mr ⎟⎠

[ τ p ]M [ τ p ]M

(13.2)

The corresponding reduced plot of the complex viscosity η*(ω, M) for various molecular weights is shown in Figure 13.1. It should be noted that the flow curves are successfully superposed on a single master curve. The molecular weight dependence of the zero shear viscosity η0 for polymers having the weight averaged molecular weight Mw above a critical value to cause entanglement is empirically given by [1] η0 = KMw3.4,

(13.3)

where K is a constant for the molecular weight. Figure 13.3 shows the molecular weight dependence of η0 experimentally obtained for SPS. The 105

η0 (Pa·s)

104

103

102

101 105

Mw

106

Figure 13.3 Plot of the zero shear viscosity η0 against the weight average molecular weight Mw. The solid line represents η0 = 4.2 × 10−16 Mw3.4 calculated by the best fit to the experimental results.

272


dependence of the viscosity on Mw3.4 was confirmed, and the numerical relationship η0 = 4.2 × 10 −16 Mw3.4

(13.4)

was obtained. Comparing Equation 13.2 with η0 = 2.8 × 10−16 Mw3.4 for IPS [4] and with η0 = 3.3 × 10−16 Mw3.4 for APS [4], it is found that the molecular weight dependences can be expressed as similar equations for all polystyrenes. Notice that the values of the front factor in the molecular weight dependences are different among SPS, IPS, and APS, and the relation SPS > APS > IPS holds in the front factor. The theory of the tube model by Doi and Edwards gives [5] 2 ⎛b ⎞ 3 η0 ∝ ⎜ M, ⎝ Me ⎟⎠

(13.5)

where b is Kuhn’s statistical segment length and Me is the molecular weight between entanglement points. The front factor contains the parameter (b2/Me), which is associated with the stiffness of the polymer chain. Kobayashi et al. have measured the characteristic ratio C∞ for SPS, IPS, and APS defined by /Nb2 (: mean square end-to-end distance) using the small-angle neutron scattering technique and have shown that C∞SPS > C∞APS > C∞IPS, which indicates that the SPS chain is the stiffest with regard to the other polystyrenes, APS and IPS [6]. Therefore, the zero shear viscosity is remarkably affected by differences in the tacticity, and the effect of the tacticity on the melt viscosity is found to be included in the front factor of the molecular weight dependence of the zero shear viscosity in terms of chain stiffness. 13.3

BASIC PHYSICAL MECHANICAL PROPERTIES OF SPS

SPS is very different from conventional amorphous polystyrene (APS) because it is a crystalline polymer. The high stereoregularity, which is precisely controlled by the catalyst, allows SPS to crystallize, resulting in a high melting point of 270 °C and a fast crystallization rate, which enables SPS to be used in the market as a newcomer of an engineering plastic. 13.3.1

Thermal Properties of SPS

A typical melting pattern of SPS measured by a differential scanning calorimeter (DSC) is shown in Figure 13.4. Upon heating of SPS of an almost amorphous state, the glass transition temperature (Tg) appears at about 100 °C, then the cold crystallization takes place at around 140–150 °C, and finally it melts at about 270 °C (Fig. 13.4a). At the cooling from the molten state, SPS crystallizes at about 230–240 °C, showing a sharp exothermic peak

Endotherm


273

First heating

Exotherm

Heat flow

Cooling

Second heating

60

80

100

120

140

160

180

200

220

240

260

280

Temperature (°C)

Figure 13.4 DSC thermogram of SPS measured at a heating rate of 20 °C/min and a cooling rate of −20 °C/min. (a) First heating of the almost amorphous sample. (b) Cooling of the sample melted at 300 °C. (c) Second heating of the crystallized sample. TABLE 13.1 Thermodynamic Parameters of SPS and IPS [9]

SPS IPS[9]

Tm0 (K)

ΔHm0 (kJ/mole)

ΔSm0 (kJ/mole/K)

583[7] 513

5.53[8] 9.00

9.49 17.54

(Fig. 13.4b). When well-crystallized SPS is heated in DSC, Tg appears at about 100 °C followed by the melting point at about 270 °C (Fig. 13.4c). The equilibrium melting point (Tm0) of SPS was determined to be 583K (310 °C) by extrapolating the measured Tm and the lamellar thickness measured by small-angle X-ray scattering [7]. The heat of fusion, ΔH f0, of SPS was determined to be 5.53 kJ/mole (53.2 J/g) [8]. The entropy change at melting, ΔSm0, is determined by dividing ΔH f0 by Tm0. Table 13.1 summarizes thermodynamic parameters of SPS compared to those of IPS [9]. These parameters are important to characterize the crystallinity and the crystallization rate of SPS. The degree of polymerization of SPS can be controlled by choosing the polymerization conditions without changing the syndiotacticity. The Tg and Tm of SPS with various molecular weights are listed in Table 13.2, whereas Tg and Tm are plotted against the number average molecular weight, Mn, in Figure 13.5. As Mn becomes larger, Tg increases and levels off at about 100 °C, and Tm also becomes higher and reaches about 270 °C.

274


TABLE 13.2 Tg and Tm of SPS of Various Molecular Weights (g/mol) Mw 3,400 7,100 20,200 41,700 80,000 150,000 250,000 410,000 625,000 1,126,000 3,045,000

Mn

Mw/Mn

Tg (°C)

Tm (°C)

1,400 2,800 8,300 12,800 26,900 52,600 96,900 158,000 245,000 467,000 1,378,000

2.43 2.54 2.43 3.26 2.97 2.85 2.58 2.59 2.55 2.41 2.21

71.5 81.2 85.9 93.4 96.5 96.7 96.2 99.2 99.1 97.1 98.4

224 244 251 253 255 266 267 267 265 267 268

120 110

Tm (°C)

Tg (°C)

100 90 80 70 60 50 1000

(a)

10,000

100,000 1,000,000 10,000,000

Mn

300 290 280 270 260 250 240 230 220 210 200 1000

(b)

10,000

100,000 1,000,000 10,000,000

Mn

Figure 13.5 Plot of Tg and Tm of SPS as a function of Mn (g/mol). (a) Tg, (b) Tm.

13.3.2

Mechanical Properties of SPS

It is important to know the effect of the degree of crystallinity for SPS since it is a crystalline polymer and the degree of crystallinity, Xc, has an influence on the thermal mechanical properties of SPS. The dynamic elastic moduli (E′ and E″) are plotted against the temperature in Figure 13.6 for SPS specimens with different Xc. At high Xc, E′ drops at Tg, then it shows a gradual decline near the melting point, while at low Xc, E′ shows a steep drop at Tg. Thus, the degree of crystallinity is very critical to the thermal mechanical properties of SPS. The crystallization rate of SPS is much faster than that of IPS [10]; however, a fast cooling below Tg does not allow SPS to fully crystallize, resulting in “halfway” crystallization. When SPS is injection molded, the degree of crystallinity and the morphology of SPS is greatly affected by the mold temperature [11]. When neat SPS is molded at a high mold temperature, a well-developed lamellar structure is observed in the specimen and Xc is as high as about 50% across the whole cross section of the molded specimen, from the surface to


275

104

E′

E′, E″ (MPa)

103

102

E″

101

100

10–1 0

50

100

150

200

250

300

Temperature (°C)

Figure 13.6 Temperature dependence of the dynamic elastic moduli (E′, E″) of SPS measured at a frequency of 10 Hz. Xc: +, 58%; 䊊, 46%: 䊏, 10%.

the core part of the molded specimen. On the contrary, when the mold temperature is lower than the Tg of SPS, the specimen with a lower Xc and a poorly developed lamellar structure in the surface is obtained because the low mold temperature prevents the crystallization of SPS in the surface. But in the center of the specimen, the crystallinity is high enough because the slower cooling allows SPS to crystallize. Thus, a high enough mold temperature is necessary in order to attain a high degree of crystallinity for SPS. The mechanical properties of SPS also depend on the molecular weight as well as on the crystallinity. Figure 13.7 shows the crystallinity of injectionmolded SPS specimens for various average molecular weights, Mw. SPS was melted at 290 °C and was injection molded into test pieces at various mold temperatures. The Xc of the specimens became higher as the mold temperature was increased and leveled off at about 50%. On the other hand, a higher Xc was obtained for SPS with lower molecular weight at a fixed-mold temperature. As Mw becomes higher, the flexural strength increases for all mold temperatures (Fig. 13.8). Furthermore, the flexural strength gradually decreases with an increase of the mold temperature, which suggests that an increase of the crystallinity makes SPS more brittle. The deflection temperature under load (DTUL) is plotted in Figure 13.9 against the mold temperature. The DTUL at a low load (0.45 MPa) became

276


Figure 13.7 Crystallinity of injection-molded SPS specimens of various Mw (g/mol). Symbols represent Mw: 䉬, 155,000; ⵧ, 256,000; 䉱, 300,000; 䊊, 385,000.

Figure 13.8 Flexural strength of SPS for various Mw (g/mol). Symbols represent Mw: 䉬, 155,000; ⵧ, 256,000; 䉱, 300,000; 䊊, 385,000.

higher with an increase of the mold temperature, while the DTUL at a high load (1.82 MPa) remained unchanged. Raising the crystallinity of the SPS increases the heat resistance under small stress; however, the heat resistance under high stress was governed by the glass transition temperature regardless of the crystallinity. The mechanical properties of SPS are affected by the ambient temperature. Figure 13.10 shows the flexural strength and the flexural modulus of


277

Flexural strength (MPa)

Flexural modulus (MPa)

Figure 13.9 Deflection temperature under load (DTUL) of SPS (Mw = 256,000 g/mol). Applied stress: 䉬, 0.45 MPa; 䊏, 1.82 MPa.

Temperature (°C)

Temperature (°C)

Figure 13.10 Flexural strength and flexural modulus of SPS injection molded at the mold temperature of 150 °C. (a) Flexural strength. (b) Flexural modulus.

SPS, which was injection molded at a mold temperature of 150 °C. As the ambient temperature goes up to the Tg of SPS, both flexural strength and flexural modulus drop steeply. A similar behavior was also observed in the E′ change at Tg in Figure 13.6. This change in the mechanical properties below and above Tg should be taken into account in the practical use of SPS.

278


TABLE 13.3 Mechanical Properties of Neat Polymers SPS Specific gravity Melting point Glass transition temperature Flexural strength Flexural modulus Notched Izod impact DTUL (1.82 MPa) Vicat softening temperature Dielectric constant (23 °C, 1 MHz)

3

GPPS

PBT

PA6

PA66

PPS

kg/m

1040

1040

1310

1140

1140

1340

°C °C

270 100

— 100

224 30

224 45

260 70

285 92

MPa

75

65

80

100

110

95

MPa

3000

2900

2400

2600

2800

3800

kJ/m2

1.2

1.7

4.4

4.4

5.4

2.5

°C

96

89

60

64

80

138

°C

254

104

215

215

250

270

2.6

2.6

3.2

3.5

3.4

3.1

GPPS: general purpose polystyrene; PBT: poly(butylene terephthalate); PA6: polyamide 6; PA66: polyamide 6, 6; PPS: poly(phenylene sulfide).

In Table 13.3, the mechanical properties of plastics are listed. SPS has a high melting point almost comparable to those of other plastics. An outstanding feature of SPS is that SPS has the lowest specific gravity and dielectric constant. The disadvantage of SPS is its brittleness; the impact resistance is low compared to those of other engineering plastics. SPS shows a lower impact strength than APS. The study of the failure and deformation behavior of SPS revealed that the breakage of SPS occurs with a slow and controlled crack growth at a much lower energy level than APS. During the deformation, many craze bands appear in the APS, while no visual evidence of crazing was observed in SPS before the break [12]. The critical stress intensity factor, K1c, and the fracture energy, G1c, of SPS are smaller than those of APS. These results show that SPS is more brittle compared with APS. In order to improve the mechanical properties and to raise the heat resistance, reinforcement by glass fibers or carbon fibers is generally employed. The flexural strength and the DTUL of SPS reinforced by glass fiber (GFSPS) are shown in Figures 13.11 and 13.12, respectively. By reinforcing with glass fiber, strength and DTUL were significantly raised compared with those of neat SPS.


279

Figure 13.11 Flexural strength of GFSPS (glass fiber content 30 wt%) of various Mw (g/mol), molded at a mold temperature of 150 °C. Symbols represent Mw: 䉬, 80,000; 䊏, 163,000; 䊉, 305,000.

Figure 13.12 DTUL at 18.2 MPa of GFSPS (glass fiber content 30 wt%) of various Mw (g/mol), molded at a mold temperature of 150 °C. Symbols represent Mw: 䊏, 163,000; 䊉, 305,000.

Flexual modulus (MPa)


Flexual strength (MPa)

280

Testing temperature (°C)


Izod impact strength (with notch) (kJ/m2)

Figure 13.13 Effects of glass fiber content on flexural properties (strength and modulus) of GFSPS molded at a mold temperature of 150 °C. Symbols represent glass fiber content in weight percent: 䉬, 10; 䊏, 20; 䉱, 30; 䊉, 40.


Figure 13.14 Effects of glass fiber content on lzod impact strength. Test bars were molded at a mold temperature of 150 °C. Symbols represent glass fiber content in weight percent: 䉬, 10; 䊏, 20; 䉱, 30; 䊉, 40.

The effects of the glass fiber loading on flexural properties (strength and modulus) and the Izod impact strength are plotted in Figures 13.13 and 13.14, respectively. Flexural strength and modulus increased with the increase of the glass fiber content. The decrease in strength and modulus at Tg became smaller than those of neat SPS. The Izod impact strength of SPS was improved by glass fiber reinforcement. Since glass fiber reinforcement improves the impact resistance of SPS, GFSPS becomes competitive with other engineering plastics with the excellent balances of mechanical and electrical properties, heat resistance, and dimensional stability (Table 13.4).

ORIENTATION OF SPS AND PROPERTIES OF ORIENTED SPS

281

TABLE 13.4 Comparison of Mechanical Properties of 30 wt% Glass FiberReinforced Polymers

Specific gravity Water absorption Mold shrinkage (MD) Tensile strength Tensile elongation Flexural strength Flexural modulus Notched Izod impact DTUL (1.82 MPa) DTUL (0.45 MPa) CLTE (MD) Dielectric constant Dissipation factor Breakdown voltage

kg/m3 % % MPa % MPa MPa kJ/m2 °C °C ×105/°C (1 MHz) (1 MHz) kV/mm

GFSPS

GF PBT

GF PET

GF PA66

GF PPSa

1,270 0.05 0.35 118 1.8 185 9,000 9 251 269 2.5 2.9 260 >260 2.2 3.9 0.001 16

a

40% GF. CLTE: coefficient of linear thermal expansion.

13.4 ORIENTATION OF SPS AND PROPERTIES OF ORIENTED SPS Stretching is employed to improve the mechanical strength and the transparency of polymers. Uniaxial stretching and biaxial stretching are industrially utilized to obtain films with good mechanical properties. Giving orientation to SPS improves the brittle nature of SPS. In this section, the properties of oriented SPS are presented. 13.4.1 Properties of Uniaxially Oriented SPS Generally, the molecular chains are oriented parallel to the direction of the applied stress when molecules are exposed to a stress field. APS has a negative refractive index and a negative birefringence due the benzene rings protruding from the main chain; the polarizability perpendicular to the main chain is much larger than those parallel to the main chain. SPS, isomeric but different in stereoregularity to APS, also shows a negative refractive index and a negative birefringence. The difference between SPS and APS is that SPS crystallizes under the stress generated by the flow, while APS does not. Figure 13.15 shows the birefringence behavior of SPS against the applied stress during the melt extrusion. From the slope, the stress-optical coefficient is calculated to be −9.6 × 10−9 Pa−1. The study of the dynamic viscoelasticity and the dynamic birefringence measurements demonstrated that a similar value (−9.5 × 10−9 Pa−1) is obtained for the stress-optical coefficient of SPS, and this value is almost twice as much of that of APS (−4.7 × 10−9 Pa−1) obtained by the same measurements [13].

282


Figure 13.15 Birefringence of SPS against applied stress. The stress-optical coefficient is calculated to be −9.6 × 10−9 Pa−1 from the slope.

An SPS web with an almost amorphous state can be obtained by extrusion and by quenching of the extruded web below Tg. An orientation of the amorphous web can be achieved by heating the web above Tg. The birefringence of a uniaxially oriented SPS film is shown in Figure 13.16. Stretching was done at 120 °C in a table tenter, and the obtained film was transparent. The birefringence of a stretched film gradually increases as the stretch ratio λ is increased and steeply increased where λ was larger than 3, then it leveled off. Pictures of the Hv pattern of the small-angle light scattering (SALS) for the uniaxially oriented films are shown in Figure 13.17. At a stretch ratio of λ = 2, the scattering pattern of a slightly deformed spherulite-like structure is observed, whereas the deformation became larger at λ = 3, and a much larger deformation is observed at λ = 3.5 and 4.0. The birefringence change and the SALS patterns suggest that the orientation-induced crystallization takes place during stretching, and the orientation becomes larger where the stretch ratio λ exceeds 3. The mechanical strength of uniaxially stretched films is listed in Table 13.5. A remarkable improvement in strength and modulus is attained by a uniaxial stretching of SPS. 13.4.2

Properties of Biaxially Oriented SPS (BoSPS)

Biaxial orientation is widely utilized because excellent balances of properties, such as high tensile strength, dimensional stability, transparency, and electrical properties, are provided. The amorphous web of SPS can be biaxially oriented

ORIENTATION OF SPS AND PROPERTIES OF ORIENTED SPS

283

Figure 13.16 Birefringence of a uniaxially oriented SPS film. Stretching temperature: 120 °C.

P, SD A

λ = 2.0

λ = 3.0

λ = 3.5

λ = 4.0

Figure 13.17 Hv patterns of small-angle light scattering of uniaxially stretched films. Polarizer (P) is set parallel to the stretch direction (SD).

284


TABLE 13.5 Tensile Strength and Modulus of Uniaxially Stretched Films Draw Ratio 1.0 3.5 4.5

Tensile Strength (MPa)

Tensile Modulus (GPa)

56 145 105

2.8 5.8 6.5

Figure 13.18 Comparison of the dimensional stability of a biaxially oriented SPS film and a PET film under cyclic humidity change (film thickness: 175 μm).

above Tg (100 °C), simultaneously or sequentially. In simultaneous stretching, a stretching to the machine direction (MD) and to the transversal direction (TD) is done at once. For SPS, a preferable stretching temperature is between 100 and 130 °C for the simultaneous stretching. In sequential biaxial stretching, a uniaxial stretching is done in the first step followed by a transversal stretching in the second step. The stretching of the second step requires a higher stretching temperature because the crystal formed in the first stretching step resists to the deformation in the second step. In the second stretching step, rotation, breakage, or deformation of crystallites take place, followed by the reordering of the structure. The properties of a BoSPS film are summarized and compared with biaxially oriented films of other engineering plastics in Table 13.6. BoSPS has comparable mechanical properties with other engineering plastic films. Outstanding features of BoSPS are the dimensional stability against humidity, electrical properties, and transparency. The dimensional stability of BoSPS against a cyclic change of relative humidity is shown in Figure 13.18. BoSPS is very stable against humidity

285

0.5–0.8 2.6 0.0003 300 90–92 2

ppm/%RH

— — kV/mm % %

3.3 0.002 300 88 3

12.0

1400 260 0.4 25 4.0 120 230 1–6 17

PET

3.0 0.002 260 85 —

1.5

1350 280 0.05 22 4.0 55 250 1.5 20

PPS

3.3 0.002 325 59 4

18.0

1430 — 1.3 28 3.7 80 300< 0.2 30

PI

2.3 0.0001 200 — —

—

2150 315 0.01 4 0.5 460 300< 2.1 15

PTFE

2.9 0.005 340 82 14

11.0

1360 275 0.3 27 6.2 85 250 2.0 13

PEN

PET: poly(ethylene terephthalate); PI: polyimide; PTFE: polytetrafluoroethylene; PEN: poly(ethylene naphthalate); RH: relative humidity.

1040 272 0.04 15 4.0 40 250 1–5 10–20

SPS

kg/m °C % MPa GPa % °C % ppm/°C

3

Properties of Biaxially Oriented Films

Density Melting point Water absorption Tensile strength Tensile modulus Elongation at break Soldering resistance Thermal shrinkage at 200 °C Coefficient of linear thermal expansion Humidity coefficient of linear expansion Dielectric constant at 1 kHz, 25 °C tan δ at 1 kHz, 25 °C Breakdown voltage (AC) Transmittance Haze

TABLE 13.6

3.4 0.02 — 90 —

100.0

1200 230 2.2 25 2.0 100 — — —

PA6

286


change in comparison to the biaxially oriented polyethylene terephthalate film. With these excellent properties, BoSPS can be used in many applications where high accuracy is required.

13.5 OTHER IMPORTANT PROPERTIES OF SPS SPS has inherited many properties from APS such as low specific gravity, excellent electrical properties, hydrolytic resistance, and good moldability. Moreover, it has acquired heat resistance and chemical resistance by the ability to crystallize. 13.5.1 Electrical Properties of SPS Most engineering plastics are synthesized by condensation reactions of compounds involving heteroatoms. SPS, however, is synthesized by a chain polymerization reaction. SPS only consists of carbon atoms and of hydrogen atoms, which provides SPS with excellent electrical properties. Figure 13.19 shows dielectric constants and loss tangents (tan δ) of various engineering plastics. SPS as well as GFSPS have a low dielectric constant and a low tan δ compared to other engineering plastics. The frequency dependence of the dielectric constant and the loss tangent of BoSPS is shown in Figure 13.20. In BoSPS, the low dielectric constant and the small loss tangent are unchanged over the wide range of frequency. Moreover, the low dielectric constant and the small loss tangent are maintained over a wide range of ambient temperatures (Fig. 13.21).

Figure 13.19 Map of the dielectric properties of plastics (bulk samples). IMSPS, impact-modified SPS; IRGFSPS, ignition-resistant GFSPS. PC: polycarbonate; PPE: poly(phenylene ether); PE: polyethylene; PAr: polyarylate; PFS: polysulfone.

OTHER IMPORTANT PROPERTIES OF SPS

287

Figure 13.20 Frequency dependence of dielectric constant and loss tangent of biaxially oriented films. PP: polypropylene.

Figure 13.21 Dielectric constant and loss tangent of biaxially oriented films against temperature.

In electrical and electronic applications, tracking resistance is sometimes required. SPS has a good tracking resistance and a high breakdown voltage (Fig. 13.22). Thus, SPS has excellent electrical properties in bulk and in film. 13.5.2

Chemical Resistance of SPS

APS is resistant to acids and alkalis but is not resistant to organic solvents. The crystallization of SPS provides resistance to organic solvents in addition to the resistance to acids and alkalis. Table 13.7 summarizes the chemical

288


Figure 13.22 Tracking resistance and breakdown voltage of bulk samples of glass fiber-reinforced engineering plastics. LCP: liquid crystal polymer; UL: Underwriters Laboratories.

TABLE 13.7 Chemical Resistance of Glass Fiber-Reinforced SPS, PA66, and PBTaHCl/H Chemicals

Acid Alkali Salt Alcohol Organic solvents

Automotive chemicals

Temperature (°C) Blow-bye watera NaOH aq (10%) CaCl2aq (10%) Methanol Ethyl acetate Acetone Methyl ethyl ketone Toluene Gasolineb Gas oil Engine oil Diesel oil Gear oil Brake oil Silicone grease Antifreeze Window washing

100 80 80 60 70 50 70 80 80 80 150 150 150 80 150 120 80

SPS

PA66

PBT

GF30%

GF30%

GF30%

ଛ

× 䉭 × 䉭

䊊 ଛ ଛ

䉭䉭䉭䉭䉭 ଛ ଛ ଛ ଛ ଛ ଛ

䊊 ଛ

ଛ

䉭 ଛ ଛ ଛ ଛ ଛ ଛ ଛ ଛ ଛ

× 䉭

After immersion for 30 days: ଛ, excellent; 䊊, good; 䉭, swell; ×, erosion, dissolution. a HCl/H 2SO4/HNO3 mixed acid (pH = 3). b Premium type.

× × ଛ

䊊䊊䊊䊊䊊 ଛ ଛ

䊊䊊䉭 ଛ ଛ

× 䊊

REFERENCES

289

resistance of GFSPS, glass fiber-reinforced polyamide 6, 6 (GFPA66), and glass fiber-reinforced poly(butylene terephthalate) (GFPBT). GFSPS has good chemical resistance except for some organic solvents. The precise control of the stereoregularity gave birth to a new engineering plastic on the basis of styrene monomer as a very common chemical; therefore, SPS is very cost competitive. SPS will be widely used as an engineering plastic in the market based on its unique properties, especially its electrical properties.

REFERENCES 1. Ferry, J. D. Viscoelastic Properties of Polymers, John Wiley & Sons, New York, 1980. 2. Nielsen, L. E. Polymer Rheology, Mercel Dekker, New York, 1977. 3. Suzuki, T., Kovacs, A. J. Temperature dependence of spherulitic growth rate of isotactic polystyrene. Critical comparison with the kinetic theory of surface nucleation. Polymer J., 1, 82–100 (1970). 4. Brandrup, J., Immergut, E. H., Grulke, E. A., eds. Polymer Handbook, 3rd edn., Wiley, New York, 1989. 5. Doi, M., Edwards, S. F. The Theory of Polymer Dynamics, Oxford Science Publications, Oxford, 1986. 6. Kobayashi, M., Hanafusa, S., Yoshioka, T., Koizumi, S. Polym. Prepr. Jpn., 44, 728 (1995). 7. Takebe, T., Uchida, T., Yamasaki, K. Polym. Prepr., 42, 4309 (1993). 8. Pasztor, A. J., Landes, B. D., Karjala, P. J. Thermal properties of syndiotactic polystyrene. Thermochimica Acta, 177, 187–195 (1991). 9. Boyer, R. F. The relation of transition temperatures to chemical structure in high polymers. Rubber Chem. Technol., 63, 1303–1421 (1963). 10. Takebe, T., Funaki, K., Yamasaki, K. 4th SPSJ International Polymer Conference, 175 (1992). 11. Lopez, L. C., Cieslinski, R. C., Putzig, C. L., Wesson, R. D. Morphological characterization of injection molded syndiotactic polystyrene. Polymer, 36, 2331–2341 (1995). 12. Jones, M. A., Carriere, C. J., Dineen, M. T., Balwinski, K. M. Failure and deformation studies of syndiotactic polystyrene. J. Appl. Polym. Sci., 64, 673–681 (1997). 13. Inoue, T., Ryu, D.-S., Osaki, K., Takabe, T. Viscoelasticity and birefringence of syndiotactic polystyrene. I. Dynamic measurement in supercooled state. J. Polym. Sci. Part B, 37, 399–404 (1999).

CHAPTER 14

Melt Processing of Syndiotactic Polystyrene DAVID BANK,1 KEVIN NICHOLS,2 HAROLD FOWLER,3 JASON REESE,1 and GERRY BILLOVITS4 1

Dow Automotive R&D, The Dow Chemical Company, Midland, MI, USA R&D Dow Building Solutions, The Dow Chemical Company, Midland, MI, USA 3 Ventures & Business Development, The Dow Chemical Company, Midland, MI, USA 4 Core R&D New Products, The Dow Chemical Company, Midland, MI, USA 2

14.1

INTRODUCTION

Because of its semicrystalline nature, syndiotactic polystyrene (SPS) exhibits performance attributes that are significantly different from amorphous styrenic materials. These properties include a high melting point, excellent hydrocarbon resistance, a high degree of dimensional stability, and excellent electrical performance. This combination of properties opens a wide array of potential applications. These applications are arrived at through a combination of formulation science and melt processing to enable maximum benefit from the crystalline structure of the material [1]. SPS is differentiated from conventional styrenic polymers on the basis of molecular structure. Figure 14.1 compares the molecular structure of atactic polystyrene with its isotactic and syndiotactic analogs. The most important distinction here is that atactic or general purpose polystyrene is produced with random stereochemistry resulting in nonspecific placement of the cyclic aromatic portion of the molecule in space. In contrast, isotactic polystyrene and SPS are produced with stereospecific catalysis techniques that result in highly ordered molecular structures. This high degree of molecular order gives rise to the ability of SPS and isotactic polystyrene to crystallize from the melt forming discrete crystalline domains, and hence a semicrystalline microstructure. The kinetics of crystallization for syndiotactic material are several orders of magnitude faster than that of isotactic. For these reasons the syndiotactic configuration is preferred because it provides rapid crystallization resulting in ease of processing and fabrication of finished goods. Syndiotactic Polystyrene, Edited by Jürgen Schellenberg Copyright © 2010 John Wiley & Sons, Inc.

290

INTRODUCTION

H3C

291

CH3 n

Atactic polystyrene

H3C

CH3 n

Isotactic polystyrene

H3C

CH3 n

Syndiotactic polystyrene

Figure 14.1 Comparison of the chemical structures of atactic polystyrene, isotactic polystyrene, and syndiotactic polystyrene.

As SPS cools from the melt, molecules organize and form distinct regions of crystallinity. Analysis of the microstructure of the SPS by microscopy clearly indicates that the resultant crystallinity manifests itself as individual crystalline lamellae or tightly packed spherulitic structures. Figure 14.2 provides a transmission electron photomicrograph showing differences in crystalline morphology that resulted via injection molding of an unmodified SPS sample. In this case the mold temperature was maintained at a low level to induce a thermal gradient and the corresponding change in crystalline morphology as a function of depth in the molded sample. Studies have shown that crystalline content and morphology can be controlled by alteration of the SPS molecular weight, addition of nucleating agents, or control of temperature during processing [2]. The level of crystallinity achieved following melt processing of SPS can be determined using differential scanning calorimetry. The calculated heat of fusion of 100% crystalline SPS is 53.2 J/g. Integration of the melting peak and

292

MELT PROCESSING OF SYNDIOTACTIC POLYSTYRENE

Skin

0.2 μm

Intermediate

1 μm

Core

2 μm

Figure 14.2 Transmission electron micrographs taken from various locations of an injection-molded component. Crystalline morphology differences between the skin, the intermediate region, and the core of the sample are compared.

correction for the heat of fusion indicated crystallinity level from 15% to 60% depending on processing conditions; however, typical injection-molded or extruded components generally have a level of crystallinity in the range of 50%. The rate at which crystallinity develops in SPS upon cooling from the melt can be controlled through tailoring of the SPS molecular weight distribution or addition of nucleating agents. In injection molding, the temperature of the mold surface has also been shown to have a major effect on crystallization. Each of these factors can be utilized to vary the rate and ultimate level of crystallinity development in molded articles. Melt processing conditions are generally selected to ensure that maximum crystallinity is attained in fabricated parts. Analysis of the melting behavior of SPS indicates a melting point of 270 °C. It is interesting to note that this represents one of the highest melting temperatures for a thermoplastic system that is produced from a single-monomer feedstock. Further, SPS has a glass transition temperature similar to that of atactic polystyrene measured at 100 °C. A unique attribute of SPS is the observation that the crystalline and amorphous densities are equal. Careful measurement indicated a density of 1.04 g/ cm3 irrespective of crystallinity level. This aspect of SPS performance is important in that varying degree of crystallinity in a product that is melt processed and cooled in a non-isotropic condition will yield a gradient of crystallinity as a function of depth in the component. In resins exhibiting a difference in crystalline and amorphous densities such a change would result in warpage or distortion in molded components. In the case of SPS compounds this

INTRODUCTION

293

10,000 300°C 310°C Viscosity, poise

320°C

1000 10

100

1000

Shear rate, 1/s

Figure 14.3 [3].

Melt rheology data for high molecular weight syndiotactic polystyrene

unique feature helps to yield exceptional dimensional stability in molded components. Melt rheology forms an important part of the discussion regarding processability of any thermoplastic resin. Here practical information useful for understanding the flow behavior of SPS as a molten polymer is summarized. The viscosity for high molecular weight SPS homopolymer as a function of apparent shear rate is shown in Figure 14.3 for typical melt processing temperatures [3]. As typical for most thermoplastics, the viscosity approaches a limiting, high value at low shear rates and decreases with increasing shear rate, indicating shear thinning behavior. As the processing temperature increases, the entire viscosity curve shifts to lower values. From a fundamental viewpoint, the dynamic mechanical behavior of amorphous SPS below its glass transition temperature is virtually identical to that of atactic polystyrene. In addition to the expected glass transition, SPS also exhibits a beta relaxation, virtually identical to that of atactic polystyrene in terms of temperature and activation energy. No relaxations are observed between the glass transition temperature and the melt temperature of SPS. The rubbery plateau modulus for SPS is somewhat higher than the atactic polystyrene value, which leads to the molecular weight between entanglements calculated to be 13,000 g/mol, compared with 18,100 g/mol for atactic polystyrene. The higher plateau modulus for SPS results in melt strengths that are higher for SPS than for atactic polystyrene at a given temperature. However, the high temperature required to process SPS versus atactic poly-

294


SPS

100,000

LCP Viscosity (poise)

PCT PBT

10,000

1000

100 10

100

1000 10,000 Shear rate (1/s)

100,000

Figure 14.4 Capillary rheology of 30% glass fiber-reinforced SPS versus 30% glass fiber-reinforced liquid crystalline polymer (LCP), poly(1,4 cyclohexamethylene terephthalate) (PCT), and polybutylene terephthalate (PBT). All resins were tested 50 °C above their melting point.

styrene decreases the viscosity of SPS significantly relative to atactic polystyrene and offsets that effect. From a practical standpoint, the dynamic mechanical behavior of SPS reveals that SPS softens appreciably at its glass transition temperature, thus to maintain mechanical strength up to the melting point, the polymer needs to be reinforced. Figure 14.4 illustrates that SPS, specifically SPS compounds containing glass fiber, is generally more shear thinning than formulated resins based on other high-performance engineering polymers. This implies that if a compound based on SPS can be processed at a high shear rate, it will exhibit lower apparent viscosity than other materials. Such an observation leads to the ability to fill more complex tooling during injection molding, or to improve throughput in extrusion applications. 14.2 COMPOUNDING 14.2.1

Introduction

Melt mixing is a common method of improving the properties or appearance of a thermoplastic by incorporating a polymer with a combination of other polymer(s), filler(s), reinforcement(s), functional additive(s), or colorant(s). This process is referred to as compounding. Since SPS in its base form has inferior strength and ductility for most target applications, commercial products of SPS are mostly compounds. In this section, the equipment and processing conditions typically used to create SPS compounds are summarized. Fibrous reinforcement (glass, carbon, etc.) has been successfully applied to SPS to produce a range of products with high heat resistance, good dimensional stability, excellent electrical performance, and resistance to hydrocarbon-based fluids. Glass fiber is added using typical compounding extruders to

COMPOUNDING

295

produce pelletized feedstocks for injection molding. Reinforced grades generally contain 20% to 40% glass fiber and in some cases a small amount of impact modifier. 14.2.2

Compounding Equipment

14.2.2.1 Feeders SPS compounds can be categorized into five main categories: (a) glass fiber-reinforced neat SPS, (b) ignition-resistant glass fiberreinforced SPS, (c) mineral-filled SPS, (d) impact-modified neat SPS, and (e) glass fiber-reinforced SPS/polyamide blends. The glass fiber-reinforced compounds often include some impact modifiers to enhance ductility. These compounds require the ability to feed the ingredients described in Table 14.1 to the compounding extruder. In many cases, rework material is also fed to the compounding extruder. The pellet ingredients in Table 14.1 are typically fed to the compounding extruder using a loss-in-weight or volumetric single-screw feeder or belt feeder. Depending on the number of ingredients versus the number of feeders that are available, it may be necessary to pre-blend some of the ingredients in a medium-intensity blender and then introduce the blend to the feeder. Normally, these pre-blends do not include high amounts of mineral filler, the glass fibers, or the ignition-resistance additives. The major powder ingredients are usually fed to the compounding extruder through loss-in-weight twin-screw extruder-type feeders. The refill and feed hoppers for these feeders are generally agitated. When low amounts of powder are to be added, it is sometimes the practice to blend these powders with a ground SPS-bulking agent and to add this blend to the twin-screw powder feeder.

TABLE 14.1 Typical Form of Ingredients for Compounding Syndiotactic Polystyrene Formulations Ingredient Syndiotactic polystyrene Polyamide Impact modifier Color concentrates Color pigments Glass fiber Fillers Antioxidants, UV stabilizers Glass compatibilizer Nucleators Ignition resistance additives Mineral oil

Form Pellets Pellets Crumb or pellet Pellets Powder Fiber bundles Powder or fibrous powder Powder Pellets or powder Powder Pellets or powder Liquid

296


Liquid ingredients are normally added with a gear pump having a proportional-differential flow meter. Fibrous additives are generally introduced to the compounding extruder with loss-in-weight single- or double-screw feeders that feed into the side of the compounding extruder at a point the resin is already melted. The feed system must minimize glass movement to avoid fiber attrition. Vibratory trays should not be used for fibrous additives. 14.2.2.2 Extruder The role of the compounding extruder is to melt the plastic resin and knead, smear, fold, stretch, wipe, compress, and shear mix it with all the other additives. Ideally, the shear and mixing forces are uniform throughout the compounding process. The additives should be dispersed and distributed equally throughout the mixture to form a uniform mix coming from the compounding process. To accomplish this, a batch mixer, a continuous mixer, a single-stage or two-stage single-screw extruder or a twin-screw extruder can be used. The design and operation of these mixers and extruders for compounding are described in detail elsewhere [4]. To date, a co-rotating intermeshing twin-screw extruder has been used the vast majority of time to compound SPS formulations; this is what will be described in the remainder of this section. 14.2.2.3 Pelletizers The extrudate from the compounding extruder is pelletized. These pellets are then used to feed customer polymer processing equipment to make final products. Pelletization can be accomplished through chopping strands or underwater die face cutting the extrudate [4]. The technical considerations for strand-cut pelletizers are summarized in the next section. 14.2.3

Compounding Process Conditions

As mentioned previously, this section will concentrate on compounding SPS with a co-rotating intermeshing twin-screw extruder. Many of the concepts discussed here are applicable and easily transferable to any style of compounding extruder that might be used to compound SPS formulations. 14.2.3.1 Feeder Operation It is common to sequence the addition of SPS ingredients to the compounding extruder. The pellets, crumb, oil, and powders are added at the feed funnel of the extruder. Glass fiber and fibrous fillers are fed through the side of the extruder barrel at the point the resin is melted and well mixed with the other ingredients that were introduced at the feed funnel. This is done to avoid excessive attrition of the fibrous additives. The amount of material that can be added to the extruder is dependent on the torque capabilities of the extruder. It is likely that when powders are added to the compounding extruder, dust control will be an issue, and a dry vent system will be necessary. There is a tendency for the extruder to overload or lock up if the ratio of powder to resin

COMPOUNDING

297

is too high at the feed funnel. If the concentration of powder in the formulation is large, it may be required to introduce some powder at the feed funnel and some powder downstream in the extruder after the resin is melted. 14.2.3.2 Screw Design The design of the compounding extruder screw is critical to making SPS compounds with the desired properties. The screw imparts the dispersive and distributive mixing that yields the morphology resulting in the final properties of the compound [4]. In the feed section of the screw, there are typically long pitch screw elements. The melting section of the screw is made up of kneading blocks to disperse and distribute powder and impact modifier additives. Many times there is a reverse screw element or reverse kneading blocks at the end of the melting zone to hold material up in the melting zone. Next, there is a section of long pitch screw elements where fibrous material (if applicable) is fed into the side of the extruder. This is followed by tooth mixers and/or forwarding kneading blocks that are designed to disperse and distribute the fibrous material with as little breakage as possible. Often there is a short reverse screw element at the end of this section to slightly hold up the material. Finally, in the vent area there are standard long-pitch screw elements and in the pumping section there are medium-pitch screw elements. 14.2.3.3 Vacuum Having appropriate vacuum on the compounding extruder is essential to having SPS compound pellets with good appearance and to having good strand stability during the pelletization step. The vacuum removes volatiles, mostly air and water, from the melt. A very high level of vacuum is required to remove styrene residuals. The amount of vacuum usually needed is (–) 0.5–0.75 bar. A signal that the amount of vacuum is appropriate is if the cut surface of the pellet appears dense and not foamy. In some cases, the ignition resistant additives may cause the pellets to appear foamy regardless of the level of vacuum. Too much vacuum can cause small pieces of melt, oils, and/or wax to be removed from the melt and condensed in the vacuum line. Proper venting for any open sections of the barrel and at the die is recommended. 14.2.3.4 Compounding Extruder Barrel Screw and Temperature Settings When compounding with the co-rotating intermeshing twin-screw extruder, the screw speed is 200–400 rpm. The screw speed is adjusted to affect melt temperature or obtain the desired morphology, and hence properties. For SPS formulations, the barrel and die temperatures are set at 290–340 °C (recall that the melting point of SPS is 270 °C). The target melt temperatures are generally 300–340 °C, usually less than 330 °C to avoid degradation. In the vent area, the barrel is set at 315–325 °C to reduce/stop vent flow. For ignitionresistant formulations, it is customary to have a die set point of 275–290 °C to stop degradation at the wall.

298


14.2.3.5 Pelletizer Operation The pellet is the product that is delivered to the customer, so it is imperative to have an excellent pelletization process for SPS compounds. In this section, the conditions to obtain good pellets from chopped strands are discussed. The first consideration is avoiding degradation along the wall of the strand die, especially with ignition-resistant formulations. This degradation causes discoloration of pellets. A second consideration is making sure that the flow is uniform through each strand hole on the die. This is particularly important for highly reinforced formulations. The extruded strands are cooled in a water bath or preferably with a cooling belt to a temperature that they can be chopped in the pelletizer. If the strand is too hot entering the pelletizer the pellets will smear and deform. If the strand is too cold entering the pelletizer, shards will break off the strand during the cutting process due to the brittle nature of the material. A strand temperature near 100 °C is usually suitable for palletizing SPS formulations. To avoid excessive fines, care should be taken to keep the cutter knife sharp and maintain rotor-bed knife gap on the pelletizer. 14.2.3.6 Purging Considerations Dependent of the formulation, it is possible that SPS will degrade if left in the barrel of the compounding extruder for more than 2 to 3 min at compounding temperatures. Therefore, it is important to purge the barrel of the extruder during start-up, before shutdown, and quickly after any unplanned shutdown (if possible). High-density polyethylene has been found to be an excellent purge material. Note that SPS will freeze off at approximately 288 °C, so it is important to have all SPS purged from the extruder before barrel and die temperatures are decreased below this value.

14.3 14.3.1

INJECTION MOLDING Introduction

The introduction of Injection Molding occurred in 1872 when J. W. Hyatt melted and shaped a mixture of nitrocellulose and camphor with his “packing machine” (United States Patent 133229) [5]. The next 135 years showed advances in materials and equipment for injection molding, but the basic concept has remained: A plastic material is softened by heating, formed under pressure, and solidified through cooling [6]. The injection-molding process has historically been a cost-effective manufacturing solution to produce highvolume complex plastic designs. SPS is well suited for injection molding with excellent rheology profiles and the ability to produce thinner than average wall thickness designs. Successful manufacturing of injection molding requires a combination of excellent part design, mold (tool) design and construction, machine design and operation, as well as resin selection. This section will address how SPS relates to each of the key factors to successful injection molding.

INJECTION MOLDING

14.3.2

299

General Product Design

Plastic product design with thermoplastics can be very complex. Following a few fundamental principles of product design will minimize problems during the injection molding cycle. The key principles of part design that affect the injection molding cycle are nominal wall thickness, avoiding sharp corners, generous use of radii, and use of draft angles. 14.3.2.1 Nominal Wall Thickness In general, a uniform nominal wall thickness should be maintained throughout the part as much as possible. Necessary variations should be controlled through transition zones shown in Figure 14.5, and abrupt wall thickness changes should be avoided. Parts with wall thicknesses greater than 4 mm should be avoided, as these designs are subject to voids and sinks as shown in Figure 14.6 during the injection molding cycle. 14.3.2.2 Avoiding Sharp Corners Avoid part designs with sharp corners. Sharp corners act as notches, which concentrate stress and reduce the part’s impact strength. A corner radius, as shown in Figure 14.7, will increase the

Appropriate

Inappropriate

Figure 14.5 Wall thickness transition.

Sink Void

Sink

Figure 14.6 Thick wall section causing voids.

300


T Minimum radius

T 4

Inside radius + T Recommended

Not recommended

Figure 14.7

Corner radius.

Release draft

Figure 14.8

Draft angles.

strength of the corner, and improve mold filling during the injection molding cycle. 14.3.2.3 Generous Use of Radii The generous use of radii will maximize part strength, improve polymer flow balance during the injection phase of the injection molding cycle, and promote uniform shrinkage during the cooling phase of the injection molding cycle. Radii should be considered at the corners and edges of parts, at the intersection of ribs and bosses with the supporting nominal wall, and at all internal corners. Radii should be designed to target 50% of the nominal wall thickness, with a minimum of 25% and a maximum of 75%. A radius that is too large will only add wall thickness and could result in voids and sink. 14.3.2.4 Use of Draft Angles To ease the demolding (ejection) of parts made from SPS, the largest possible draft angle shown in Figure 14.8 should be used. Draft angles as low as 0.1 ° have been successfully used on parts with shallow depths of draw; however, a draft angle of 1.0 ° or more is preferred.

INJECTION MOLDING

301

Spiral flow length of various material 120

Filled length (cm)

100 80

PBT 30% GF340°C PPS 40% GF340°C

60

40% GF SPS300°C

40

40% GF IR SPS-300°C

20 0

0

20

40

60 80 100 120 Injection speed (mm/s)

140

160

180

Figure 14.9 Spiral flow length of various materials. PBT: polybutylene terephthalate; GF: glassfiber reinforced; PPS: polyphenylene sulfide; IR: ignition resistant.

14.3.3

Thin Wall Product Designs

Due to SPS’s unique low viscosity (an example is shown in Fig. 14.9), it is often injection-molded into parts with thin wall sections (100 : 1) are more ductile with lower SDR fibers being more brittle. Subsequent redrawing of

FIBER SPINNING

317

10,000

Specific modulus (gpd)

Carbon PBO 1000

Vectran LCP

PE Kevlar

SPS PAN PET 100

Acetate cellulose Wool

PPS Cotton

E glass

S glass

PVA Acrylic Nylon PP

10 1

10

100

Tenacity (gpd)

Figure 14.21 Comparison of the tenacity and specific modulus of various fiberforming materials with SPS [22]. PAN: polyacrylonitrile; PP: polypropylene; PE: polyethylene; PET: polyethylene terephthalate; PBO: polybenzoxazole; PVA: polyvinylacetate.

these fibers can increase the tenacity of the fibers but with reductions in their elongation. Redraw ratios of about 2.2 times at 180 °C provide a good balance of strength and elongation of fibers with an SDR of about 200 : 1 [22]. The relationship between the modulus and the tenacity of the SPS fibers that can be achieved with varying processing conditions compared with typical modulus-tenacity performance of other fibers can be seen in Figure 14.21. Similarly, the relationship of SPS fiber tenacity to its elongation can be seen in Figure 14.22 as compared with other fibers. As documented in a recent patent [23], blends of SPS with other resins, especially polyamides, can also find utility in fiber applications. While these blends are typically immiscible, when appropriately compatibilized, this technology offers the opportunity to differentiate the appearance and feel of fiber-based products as compared with standard polyamide fibers. Figure 14.23 shows a TEM image of a cross-section of a polyamide 6/SPS fiber, taken in the fiber direction, where the SPS has been stained dark to provide contrast [24]. The dispersed SPS phase takes the form of submicron sized domains, some of which are highly extended in the fiber direction. The presence of this secondary phase within the polyamide matrix comprising the fiber leads to roughness on the surface, which affects the look and feel of the fiber. The fiber-to-fiber frictional behavior is altered and the appearance is delustered and less shiny. These attributes make these blend fibers more similar to natural fibers, such as cotton, making them desirable for applications such as carpet fiber and apparel.

318


300

Specific modulus (gpd)

SPS

200 Acetate cellulose PAN

PET

100

PVA Acrylic Nylon Wool PPS

0

0

2

Cotton 4

PP

6 8 Tenacity (gpd)

10

12

Figure 14.22 Comparison of the tenacity and elongation of various fiber-forming materials with SPS [22].

1 μm

Figure 14.23 Transmission electron micrographs image of a cross-section, taken in the fiber direction, from a polyamide 6/SPS blend fiber. The SPS phase is stained dark in this image [24].

REFERENCES

319

REFERENCES 1. Bank, D., Wessel, T., Kolb, J. J. Syndiotactic polystyrene: A new polymer for high performance automotive applications. Proceedings from the SEE International Congress and Exposition, Detroit, MI, March 1993. 2. Cieslinski, R. C., Dineen, M. T., Wood, C. J. Dow polymer compendium. Dow Internal Chemical Report, ML-AL-94-301287, July 1994. 3. Powers, J. R., Spalding, M. A., Hughes, K. R, Meyette, G. Extrusion of syndiotactic polystyrene. Dow Internal Report, November 1999. 4. Giles, H. F., Wagner, J. R., Mount, E. M. Extrusion: The Definitive Processing Guide and Handbook, William Andrew Publishing, Norwich, NY, 2005. 5. Rubin, I. I. Injection Molding—Theory and Practice, Wiley, New York, 1972. 6. Berins, M. L. Plastics Engineering Handbook, 5th edn., Chapman and Hall, New York, 1991. 7. Dow Plastics, Form#301-03014-999SMG. QUESTRA Crystalline Polymers Design Guide, The Dow Chemical Company, Midland, MI, 1999. 8. Menges, G., Mohren, P. How to Make Injection Molds, 2nd edn., Hanser, New York, 1993. 9. Rees, H. Mold Engineering, Hanser, New York, 1995. 10. Powers, J. R., Spalding, M. A., Wessel, T. E. Extrusion of syndiotactic polystyrene. Proceedings from the Society of Plastics Engineers 55th Annual Technical Conference, Toronto, vol. 1, pp. 34–42 (1997). 11. Colby, P. N., Colby, P. T., Colella, M., et al. Plasticating Essentials, Spirex Corporation, Youngstown, 2006. 12. Hyun K. S., Spalding, M. A. Use of process data obtained from a data acquisition system for optimizing and debugging extrusion processes. Adv. Polym. Tech., 15, 29 (1996). 13. Wessel, T. Initial extruder screw design for syndiotactic polystyrene extrusion. Dow Internal Report, September 1994. 14. Kohan, M. Nylon Plastic Handbook, Hanser/Gardner, Cincinnati, 1995. 15. Dow Plastics Literature. QUESTRA EA/WA Crystalline Polymers: Suggested Injection Molding Conditions for Glass-Filled Resins, The Dow Chemical Company, Midland, MI, 2001. 16. Dow Plastics Literature. QUESTRA N WA Crystalline Polymer Blends: Suggested Injection Molding Conditions for Glass-Filled Resins, The Dow Chemical Company, Midland, MI, 2001. 17. Huang, Y., Nonnemacher, G., Wessel, T. A competitive analysis of syndiotactic polystyrene film with major performance films. Dow Internal Report, April 1997. 18. Giles, H. F., Wagner, J. R., Eldridge, E. M. Extrusion; The Definitive Processing Guide and Handbook, William Andrew Publishing, Norwich, NY, 2005. 19. Hensen, F., ed. Plastics Extrusion Technology, Hanser/Gardner, Cincinnati, 1997. 20. Wu, S., Bubeck, R., Carriere, C. The effect of processing conditions on the fracture behavior of syndiotactic polystyrene films. J. Appl. Polym. Sci., 62(9), 1483–1490 (1996).

320


21. Hensen, F., ed. Plastics Extrusion Technology, Hanser/Gardner, Cincinnati, 1997. 22. Turek, D. A. Summary of the physical properties of continuous melt-spun fibers formed from syndiotactic polystyrene and syndiotactic copolymers of styrene and paramethylstyrene. Dow Internal Report, September 1994. 23. Menning, B. A., Guenard, R. D., Henton, D. E., Pressly, T. G., Sen, A., Warakomski, J. M. Thermoplastic compositions for the preparation of fibers and films. PCT WO 2002/055768 A1 (to The Dow Chemical Company), 2002. 24. Pressly, T. G., Barnes, C. G., Henton, D. E. Selective etching of nylon in nylon/SPS fibers for determination of SPS morphology. Dow Internal Report, May 2002.

CHAPTER 15

Applications of Syndiotactic Polystyrene TOM FIOLA,1 AKIHIKO OKADA,2 MASAMI MIHARA,3 and KEVIN NICHOLS4 1

Xarec and SPS Products, Idemitsu Chemicals USA Corporation, Southfield, MI, USA Engineering Plastics Department, Idemitsu Kosan Co., Ltd., Tokyo, Japan 3 Idemitsu Kosan Co., Ltd., Chiba, Japan 4 R&D Dow Building Solutions, The Dow Chemical Company, Midland, MI, USA 2

15.1

INTRODUCTION

Since the commercial introduction of polystyrene in the 1930s, styrenic-based thermoplastic products have become widely adopted for a variety of consumer and industrial applications due to styrenic’s favorable balance of performance versus cost combined with an ease of processing. General-purpose polystyrene (GPPS) is usually processed by injection molding and is a clear, glossy, and relatively brittle polymer. Molded parts are typically rigid and insensitive to moisture, possess excellent electrical properties and good dimensional stability, and are resistant to acids and alkalis [1]. The principal performance limitations of GPPS are brittleness; limited heat resistance; and poor chemical resistance to oils, aromatic and chlorinated hydrocarbons, esters, ketones, and aliphatic hydrocarbons. Filled grades of GPPS are available but offer limited improvements in mechanical properties and sacrifice transparency and clarity. The blending, alloying, and copolymerization of polystyrene is frequently employed to improve toughness, offering some improvement in chemical resistance but little to improve heat resistance [2]. Improvements in the performance of homopolymer GPPS through polymerization with traditional Ziegler–Natta catalysts are primarily limited to flow and glass transition temperatures. Polymerization into an isotactic backbone configuration results in a semicrystalline structure that improves both heat and chemical resistance; however, isotactic polystyrene (IPS) is slow to crystallize and therefore of little practical use with modern injection molding production techniques [3]. Syndiotactic Polystyrene, Edited by Jürgen Schellenberg Copyright © 2010 John Wiley & Sons, Inc.

321

322

APPLICATIONS OF SYNDIOTACTIC POLYSTYRENE

With the invention and introduction of metallocene catalysts for polystyrene in the early 1980s, by the Idemitsu Kosan Co., cost-efficient performance enhancements via polymerization of polystyrene were realized. Metallocene catalysts employ single active sites, permitting control of the stereoregularity of the polymer chain, making syndiotactic polystyrene (SPS) configurations a practical reality. SPS, in combination with suitable nucleating agents, boasts good crystallization rates that are compatible with injection molding processes. In contrast to atactic GPPS, the SPS configuration results in a dramatic improvement in the heat and chemical resistance of polystyrene due to the semicrystalline structure of SPS. With a melting point of 270 °C and improvements in the chemical resistance to oils, aliphatic hydrocarbons, esters, and ketones, SPS extends the performance envelope of polystyrene and offers designers another option to capture the inherent benefits of the polystyrene structure. 15.2

THE PERFORMANCE CAPABILITIES OF SPS

SPS is a semicrystalline polymer with a glass transition temperature (Tg) of 100 °C and a crystalline melting point (Tm) of 270 °C. SPS readily accepts fillers, and by compounding with short glass fibers, the heat deflection temperature can be increased to close to the crystalline melting point. A comparison of the Tg and Tm of SPS with other engineering thermoplastic materials is shown in Figure 15.1. The typical properties of glass-filled SPS formulations versus other thermoplastics are shown in Table 15.1. SPS is not inherently flame-resistant and requires modification with flameretardant additives to achieve Underwriters Laboratories (UL) 94 V0 or 5VA performance. Typical formulations, modified for UL 94 V0 flame test performance or unmodified, have the capability to withstand brief temperature spikes of up to 260 °C, making SPS an attractive alternative for the electronic/ electrical connector industry, which utilizes temperatures in this range (lead350 Tm Tg

300

°C

250 200 150 100 50 0 PPA

PPS

SPS

PA66

PET

PBT

Figure 15.1 Glass transition (Tg) and melting temperatures (Tm) of SPS and other thermoplastics.


323

TABLE 15.1 Mechanical Properties of SPS and Other Glass-filled Thermoplastics

Density, g/cm3 Water absorption, % 24 h Tensile strength, (MPa) Elongation (%) Flexural strength (MPa) Notched Izod (Kj/m2) DTUL, 1.82 MPa, °C CLTE (MD), ×10−5/°C UL 94 UL RTI, °C, electrical Mold shrinkage (%)

SPS 30% GF

SPS 30% GF

PPS 40% GF

PPA 33% GF

PBT 30% GF

LCP 30% GF

1.42

1.25

1.67

1.68

1.58

1.65

0.09

0.01

0.02

0.2

0.06

0.05

120

115

147

199

117

145

2.0

2.0

1.5

1.7

2.0

2.2

185

175

206

224

186

220

10

10

9

8.5

7.0

14

238

240

260

277

191

235

1.5

2.1

2.2

4.5

4.5

0.3

V0 125

HB 110

V0 220

V0 140

V0 130

V0 130

0.2

0.2

0.25

0.3

0.4

0.1

Source: Idemitsu Kosan, Dow Chemical, BASF, Solvay, DuPont. Properties: dry as molded. CLTE (MD), coefficient of linear thermal expansion; DTUL, distortion temperature under load.

free reflow) to manufacture printed circuit boards in infrared (IR) reflow soldering operations. SPS competes against a number of other high-temperature polymers for applications of this type and represents a viable lower cost alternative (Table 15.2). The long-term, continuous-use temperature resistance ranges from about 120 to 125 °C, as measured by the UL 746B relative thermal index (RTI) test, which places its performance closer to that of the “engineering” polymers such as polybutylene terephthalate (PBT), liquid crystal polymers (LCP), and polyethylene terephthalate (PET) than to that of “performance” polymers such as polyphenylene sulfide (PPS), polyether sulfone (PES), and polyamide imide (PAI). Combined with SPS’s excellent short-term resistance to temperature “spikes,” SPS has proven to be an excellent cost-savings alternative to perfor-

324


TABLE 15.2 Dimensional Stability of SPS and Other Thermoplastics at Peak Reflow Temperatures Material

Glass Content (%)

SPS PPS PPA PET PCT LCP

30 40 30 30 30 30

Peak On-board Temperature (°C) During Reflow Soldering 250–260

260–270

270–280

‡ ‡ ‡ + ‡ ‡

+ ‡ ‡ — + ‡

— — ‡ — — ‡

‡ ‡ ‡ — ‡ ‡

— — ‡ — — ‡

— — ‡ — — +

280–290 — — — — — +

— — — — — —

Source: Dow Chemical Co. ‡ Stable; + Slight change; — Deformation. PCT: polycyclohexylene dimethylene terephthalate.

Displacement (mm)

15

10

5

0 30% GF SPS

30% GF PBT

30% GF SPS/PA66

1.58 mm HDT bar, 200°C, 20 min

Figure 15.2 Creep resistance of GF SPS, GF PBT, and GF SPS/PA66 blend. GF, glass fiber; HDT: heat distortion temperature.

mance materials such as polyphthalamide (PPA) in automotive forwardlighting sockets, where the proximity to halogen light bulbs can result in localized, on-socket temperatures of 200 °C. Automotive head lamp lens assemblies have also benefited from the minimal outgassing of low molecular weight substituents during bulb operation—a result of SPS’s relatively clean addition polymerization process. As such, the use of SPS reduces the potential for lens fogging. The “nonpolar” nature of the chemical structure of the SPS polymer is the source of many of the most important performance attributes, including its electrical properties as well as moisture and chemical resistance, resulting in a tradeoff in which SPS’s creep resistance is diminished due to the lack of the hydrogen bonding that would otherwise occur with a polar chemical structure. Blending SPS with Nylon 66 improves creep resistance significantly (Fig. 15.2), with some tradeoff in moisture resistance.


325

Similarly, adhesion to common bonding substrates is problematic without the contribution of a polar chemical structure. SPS benefits from the use of secondary processes such as corona or UV treatment to promote bonding. Conversely, SPS’s low energy surface is ideal for applications that depend on a lack of adhesion. In-home garbage compacters benefit from the fact that adhesion to the SPS surface is limited. SPS demonstrates good resistance to a wide range of chemicals, including acids, bases, and most organic solvents. Exceptions are aromatic and chlorinated hydrocarbons. The data in Figures 15.3 and 15.4 show the retention of tensile strength after a 30-day immersion in various fluids, and illustrates the broad range of stability afforded by the SPS chemical structure.

Figure 15.3 Retention of tensile strength (%) after 30-day immersion at room temperature. GF, glass fiber.

Figure 15.4 Retention of tensile strength (%) after 30-day immersion at room temperature. GF, glass fiber .

326


Figure 15.5 glass fiber.

24-h water absorption (50% relative humidity, room temperature). GF,

Figure 15.6 Tensile strength retention after exposure to water at 93 °C (29 psi).

The hydrolytic stability of SPS is excellent. Short-term water absorption of nonflame retardant grades is only 0.01% (Fig. 15.5), and even extended exposure to moisture at elevated temperatures and pressure leaves SPS relatively intact in comparison with other engineering thermoplastics (see Figure 15.6). The dimensional stability of molded components benefits from several unique aspects of SPS. First, the amorphous and crystalline phases have equivalent density, and this tends to minimize the crystalline gradients that are typical of semicrystalline, thermoplastic materials. A comparison of the amorphous phase density with the crystalline phase density of SPS and other semicrystalline thermoplastics is shown in Table 15.3. Materials with large density differences are prone to warpage due to the volume rearrangement that occurs during the cooling phase of the injection molding process. This unique characteristic of SPS helps minimize warpage as components cool after manufacturing.

327


TABLE 15.3 Crystalline versus Amorphous Densities of SPS versus Other Thermoplastics Thermoplastic SPS PPS PBT PA66 PA46

Figure 15.7

Crystalline Density g/cm3

Amorphous Density g/cm3

Difference %

1.05 1.43 1.41 1.23 1.28

1.05 1.32 1.28 1.08 1.08

0 7.7 9.2 12.2 15.6

Capillary melt viscosity of SPS versus other thermoplastics.

Second, the low melt viscosity of SPS formulations helps to reduce moldedin stress from the injection molding process. The data in Figure 15.7 show typical melt viscosities of SPS in comparison with other semicrystalline thermoplastics. The low melt viscosity of SPS results in low hydraulic pressures during injection molding, which helps to minimize molded-in stress. Low viscosity also contributes to SPS’s ability to fill complex parts with thin walls and its relative lack of flash during injection molding. The results in Table 15.4 show the practical benefit of SPS’s dimensional stability in a molded part. The amount of warpage or “skew” of a molded SPS and an SPS/Nylon (polyamide [PA]) part is compared with that of a PBT part. Both SPS compounds exhibit less warpage than the PBT compound. The data in Figure 15.8 show that the bulk density of SPS is low in comparison with other thermoplastics.

328


TABLE 15.4 Skew of a Tray Wall Injection Molded from GF PBT, GF SPS/PA, and GF SPS Material

Skewa

GF GF GF GF

0.016 0.007 0.006 0.005

PBT (30%) SPS/PA (30%) SPS/PA (20%) SPS (30%)

a

Defined as the deflection/length of an injection molded wall. A tray configuration (8.5″ × 11″) is molded, and the length between two edges are measured and compared with the vertical distance from a reference point midpoint on an attached wall.

Figure 15.8

Density reduction of part weight with SPS. GF, glass fiber.

The final component weight can be minimized, and less energy is required to melt-process SPS. This specific gravity advantage results in a 10% to 20% greater part yield per pound than other glass–fiber-reinforced resins. The electrical properties of SPS are excellent. Volume and surface resistivity are high, and, combined with a low dielectric constant and high breakdown strength, SPS provides durable electrical insulation. The dissipation factor (or power loss) is nearly zero, providing an advantage to cell phone antennas and microwave oven components since SPS minimizes interference in broadcast frequencies and does not undergo excessive heat rise upon exposure to microwaves. The comparative tracking resistance (CTI), an electrical performance characteristic important for the connectors used in the new generation of hybrid electric automobiles, is high (600 V, PLC 0), minimizing the potential for short circuits in power connectors. The electrical properties of SPS are stable over a wide range of temperature and humidity due to SPS’s nonhygroscopic nature. The electrical properties of SPS are compared and contrasted in Figures 15.9 and 15.10.

CONNECTORS FOR AUTOMOTIVE AND ELECTRONIC APPLICATIONS

329

0.1 1 MHz, 25°C

Dissipation factor

Epoxy PA66 (GF 30%)

0.01 PES SPS (GF 30%)

0.001

PBT (GF 30%) PPS (GF 40%)

PE PTFE

0.0001 2

3

4

5

Dielectric constant

Figure 15.9 Dielectric constant versus dissipation factor. GF, glass fiber; PE, polyethylene; PTFE, polytetra fluoro ethylene.

Figure 15.10

Breakdown voltage versus UL tracking resistance. GF, glass fiber.

15.3 CONNECTORS FOR AUTOMOTIVE AND ELECTRONIC APPLICATIONS The growth in SPS usage in automotive applications has benefited from several long-term trends, including an increase in underhood operating temperatures (more components, less space, and restricted air flow), the miniaturization of components to accommodate increased electrical functionality (global positioning system devices, DVD players, etc.), more compact designs, and the need for improvements in warranty performance, especially as it relates to durability in humid environments. As it has become necessary to design more automotive features into the same or smaller package areas, automotive designers have responded by reducing the size of the wiring used for signal transmissions in harness connectors, which allows an increase in electrical connections per unit area. This

330


trend toward 0.64-mm diameter wire terminals results in the need for thinner wall sections in the connectors that are used to insulate between terminal connections. The low viscosity and excellent dimensional stability of SPS helps to ensure the capability to manufacture these precision connectors. The hydrolytic and heat stability of SPS makes it well suited for compliance with United States Council for Automotive Research (USCAR) guidelines for the exposure of automotive connectors to heat and humidity. Typical wire harness connector designs incorporate small internal features, called “terminal locking fingers,” which help to maintain the position of metal terminals inside the housing. SPS and SPS/PA66 blends help connector designers meet the stringent temperature and humidity requirements (USCAR 4) and provide additional benefits to injection-molded connectors because of their low viscosity and excellent dimensional stability. Similarly, SPS is well positioned to benefit from the trend toward electric hybrid vehicles. The superior electrical tracking resistance of SPS (Fig. 15.10) is well suited for the connectors, terminal and fuse holders, etc., that are needed for this evolving market. With a CTI of 600 V, SPS offers good protection against terminal-to-terminal short circuits and other benefits to the electronic architecture needed to integrate 600-V electrical power systems into vehicles. The fact that SPS typically offers weight reduction, in comparison with other materials, is an added benefit for all automotive applications as it helps to improve fuel economy. In addition to wire harness connectors, printed circuit board-attached connectors, for automotive or electronic applications, also benefit from many of SPS’s performance characteristics, including low viscosity, which helps to fill thin walls and heat resistance to maintain dimensional stability through leadfree reflow soldering operations (see Table 15.2). The trend toward miniaturization in automotive electronic control modules has increased the need for compact surface mount connector technology, as opposed to bulkier throughhole and wave-soldered alternatives. Both board-attached and wire harness connectors benefit from SPS’s excellent thermal diffusivity, defined as the ratio of thermal conductivity to volumetric heat capacity. As a result, cycle time models show that injection-molded parts cool quickly in the mold, helping to minimize cycle time and improve dimensional stability of their complex, molded features such as the terminal snap fit features on harness connectors. Tool designers typically cannot place cooling lines in close proximity to these features and the cooling characteristics of SPS are essential to minimizing the cycle time during manufacturing.

15.4

ELECTRONIC COMPONENTS: PLATED AND NON-PLATED

The low electric power loss inherent in the SPS structure, combined with the capability to easily fill thin walled, complex tools, has resulted in significant growth of SPS into electronic components such as embedded antennas and various interconnect devices that rely on consistent electrical performance.

INDUSTRIAL AND APPLIANCE COMPONENTS

331

Quite often, two-shot injection molding techniques are used to mold glassfilled SPS blends that have been modified for electroless plating. In the production of a two-shot antenna, deposition of a nickel coating onto an SPS formulation modified for plating is carried out in secondary operations. Antenna performance is enhanced by the fact that the non-platable SPS substrate rejects deposition of the nickel coating since the low-energy, nonpolar surface of SPS does not readily adhere to nickel. The adhesion of SPS to another SPS substrate is excellent and an advantage for two-shot component designs.

15.5


Industrial applications of SPS take advantage of the heat, dimensional stability, and chemical resistance of SPS. For example, liquid-filled transformer components require long-term exposure to the mineral oils used for coolants as well as heat resistance to handle the elevated temperatures that occur during peak demand periods. Because of the exceptional hydrolytic stability of SPS, it is utilized in many components associated with water handling, including pump housings, impellers, and water heaters. The chlorine resistance of SPS also makes it suitable for components exposed to domestic water supplies, even at elevated temperatures. Good resistance to steam, pressurized hot water, ethylene oxide, and even gamma radiation makes SPS a good choice for applications that require sterilization. Favorable biocompatibility ratings (as measured by ISO 10993 testing) emphasize the capability to utilize SPS compounds for non-fluid/tissue contact medical devices. Heating, ventilation, and air conditioning (HVAC) applications rely on SPS’s hydrolytic and heat resistance for moisture drip pans that are close to heating elements, and chemical resistance to corrosive flue gases generated by combustion. Excellent dimensional stability ensures consistent manufacturing within the tight tolerances required for precision components. Electric meter bases capitalize on SPS’s excellent electrical insulation and ease of processing, which in turn permits cost reductions through parts consolidation. Various appliance components take advantage of SPS’s hydrolytic, heat, and dimensional stability, including coffee makers, garbage compactors, humidifiers, and washer/dryer components. Portable arc welding machines and plasma cutters utilize the heat and dimensional stability of SPS for the molded fixtures that hold steel heat sinks in place. Their performance relies on compressed air (plasma cutter) or inert gases (welding), combined with a large amount of energy, therefore generating a great deal of heat to manage within the framework of the unit. SPS has the necessary heat and dimensional stability to maintain the critical tolerances for assembly and performance of the units.

332


The majority of SPS resin is sold as fiberglass-reinforced compounds for injection molding; however there are end uses for other forms of SPS. The use of SPS in fibers has been explored: first, as an additive for Nylon 6 in carpet fiber applications, where improvements in tactile quality and appearance have been observed; and second, as a low-cost replacement in fiber applications that require heat and chemical resistance, such as hot stack gas filter bags. The rapid crystallization rate of homopolymer SPS reduces the practical benefit of cast film production. However, copolymerization of the SPS structures helps to delay the onset of crystallization during cast film production until a secondary biaxial orientation operation. SPS crystallites improve the dimensional stability and chemical resistance of films and have also been investigated for unique optical properties. Table 15.5 summarizes the common applications for the SPS material [4–8]. TABLE 15.5 Applications for SPS Electrical Electric meter base Coil bobbin Transformer components Sensors Lighting support Battery/fuel cell components

Electronic Antenna Cable interconnects Board-attached (SMT) connectors Fiber optic connectors

Business Machine and Appliance Printer heat guards Printer toner supports Coffee maker components Timer housings Washer and dryer components Microwave oven components Nonstick refuse container/compactor Fan blades Plasma cutter heat sinks Baking tins

Pump and Water Handling Housings Impellers Water heater components Valve housings Oil well pump components Test tube racks (steam sterilized)

Automotive Harness connectors Board connectors Electronic control modules Fuse holders Transmission connectors/modules Oil and transmission fluid handles Encapsulated valves Power distribution centers Halogen light sockets High-voltage connectors

HVAC Humidifier components Air conditioner drip pan Collector boxes Vent conduits


333

Photographs of selected common applications of SPS in the automotive, electrical, business machine, appliance, electronic, and HVAC areas are shown in Figures 15.11–15.18.

Figure 15.11 Automotive power distribution center.

Figure 15.12

Coil bobbins.

334


Figure 15.13

Microwave oven spindle support.

Figure 15.14

Printer fuser shield.


Figure 15.15 Transformer tap changer lock nut.

Figure 15.16

Cell phone antenna.

335

336


Figure 15.17

Electrical circuit meter base.

Figure 15.18 Air conditioner drip pan.

REFERENCES

337

REFERENCES 1. Saechtling, H. International Plastics Handbook, Hanser Press, Wien, 1983. 2. Hampel, C., Hawley, G. Encyclopedia of Chemistry, 3rd edn., Van Nostrand Reinhold Co., New York, 1973. 3. Schwartz, S., Goodman, S. Plastics Materials & Processes, Van Nostrand Reinhold Co., New York, 1982. 4. Yamasaki, K., Tomotsu, N., Malanga, M. Characterization, properties and applications of syndiotactic polystyrene. In Modern Styrenic Polymers: Polystyrenes and Styrenic Copolymers, Scheirs, J., Priddy, D. B. (eds.), John Wiley & Sons, New York, 2003, pp. 389–409. 5. Schellenberg, J., Leder, H.-J. Syndiotactic polystyrene: Process and applications. Adv. Polym. Technol., 25, 141–151 (2006). 6. Bank, D. H., Wessel, T. E., Kolb, J. J. Syndiotactic polystyrene: A new polymer for high performance automotive applications. SAE Technical Paper Series 1993, number 930088. 7. Koch-Reuß, U. Syndiotactic polystyrene (SPS) properties and applications. Kunststoffe, 88, 1139–1144 (1998). 8. Hermanson, N. J., Wessel, T. E. Syndiotactic polystyrene: A new polymer for highperformance medical applications. Medical Plastics and Biomaterials 1998, http:// www.devicelink.com/mpb/archive/98/07/001.html (accessed September 25, 2009).

CHAPTER 16

Blends of Syndiotactic Polystyrene with Polyamide KEVIN NICHOLS,1 AKIHIKO OKADA,2 and HIROKI FUKUI3 1

R&D Dow Building Solutions, The Dow Chemical Company, Midland, MI, USA Engineering Plastics Department, Idemitsu Kosan Co., Ltd., Tokyo, Japan 3 PP Automotive Materials Division, Prime Polymer Co., Ltd., Shizuoka, Japan 2

16.1

INTRODUCTION

Blends of two immiscible polymers are created to yield a material with properties that could not be obtained otherwise. Each component of the blend overcomes the property deficiencies of the other component of the blend. In the case of syndiotactic polystyrene (SPS)/polyamide (PA; nylon) blends, the blends have improved strength, ductility, and creep versus SPS formulations, and the blends have improved dimensional stability and flow versus nylon compounds. Other attributes of the SPS/nylon blends are low specific gravity (lower weight parts), high thermal diffusivity (low cycle time), excellent electrical properties, good chemical resistance, and excellent United States Council for Automotive Research (USCAR) electric wiring components test performance. In this chapter, the composition, properties, and applications for SPS/ nylon blends will be reviewed.

16.2

COMPOSITION OF SPS/NYLON BLENDS

In this section, the polyamides (PAs) used in SPS/nylon blends are briefly explained. This is followed by a summary of SPS/nylon blend formulations; the compositions of SPS/nylon blends that are claimed in patents are listed; and finally SPS/nylon blends detailed in journal articles are described.


338

PROPERTIES OF SPS/NYLON BLENDS

16.2.1

339

Polyamides Used in SPS/Nylon Blends

The vast majority of SPS/nylon blends to date have been formulated using nylon 6 and nylon 6,6. nylon 6 and nylon 6,6 have essentially the same properties; the difference is in the way these nylons are synthesized. Nylon 6 is made by ring opening polymerization of the monomer caprolactam, while nylon 6,6 is made from two monomers: adipoyl chloride and hexamethylene diamine. The synthesis, properties, and applications of nylons are described in [1]. 16.2.2

SPS/Nylon Blend Formulations

To date, the practical formulations of SPS/nylon blends have consisted of SPS dispersed in a nylon matrix. There are 10, 20, and 30 weight percent glass fiber-reinforced SPS/nylon blend commercial products. In some cases, the SPS/nylon compositions include olefinic impact modifiers that are compatibilized with the matrix nylon via a maleic anhydride-functionalized olefin. The concept is that the maleic anhdydride portion of the molecule reacts with the nylon, and the olefin part of the molecule is miscible with the olefin impact modifier. Since nylon is the matrix phase, the glass fiber-reinforced SPS/nylon blends contain glass fibers that are surface treated to be compatible with nylon. 16.2.3

SPS/Nylon Blend Composition Patents

Table 16.1 contains a list of compositions for SPS/nylon blends that have been claimed in patents. No attempt was made to identify patents that might include SPS/nylon blends in the teachings. 16.2.4 SPS/Nylon Blend Compositions Described in Technical Journals Table 16.2 contains a list of compositions for SPS/nylon blends that have been described in technical journals.

16.3


The possibility of a synergistic combination of properties for SPS/nylon blends has been recognized, and, therefore, the structural and physical properties [46] as well as the morphology and mechanical properties [47] of PA 6,6 (nylon 6,6)/SPS blends have been the subject of two research articles. Below is a summary of the mechanical properties, rheological properties, moisture response, dimensional stability, USCAR test results, and solvent resistance of SPS/nylon blends that make these blends attractive for some applications, such

340

BLENDS OF SYNDIOTACTIC POLYSTYRENE WITH POLYAMIDE

TABLE 16.1 SPS/Nylon Blend Compositions Claimed in Patents Patent SPS/Nylon Composition Claim Basic SPS/nylon blend patent SPS/nylon blends impact modified with maleic anhydridefunctionalized poly (styrene-b-(ethylene-co-butylene)-b-styrene) MA-SBS, or MA-SEBS SPS/nylon blends impact-modified with oil extended impact modifier and containing domain-forming agent Brominated SPS/nylon blends A composition having improved adherence with an addition-curable material SPS/nylon blend compositions for making fiber and films Compatibilized SPS/nylon blend SPS/nylon blends to make a nanocomposite A resin composition that can be SPS/nylon that contains a graft copolymer A composite magnetic material A thermally developable photosensitive layered material SPS/nylon blends impact-modified with a rubber having a specific particle size Compatibilized SPS/nylon blends with a copper stabilizer SPS/nylon blend compatibilized with acid-modified polyphenylene ether Impact-modified SPS/nylon blend that is compatibilized with a compatibilizer that can react with nylon and is compatible with SPS A process for making an SPS/nylon blend containing block copolymer A vibration damping engineering plastic A flame-retardant SPS/nylon blend SPS/nylon with mineral fibers to create a low dielectric loss tangent resin SPS/nylon blends with functionalized reinforcement SPS/nylon blend with styrenic block copolymer SPS/nylon blend with impact modifier and optionally filler A process for making a nanocomposite masterbatch Impact-modified SPS/nylon blend with reinforcement and mineral filler for improved gloss SPS/nylon/clay nanocomposite composition nylon/SPS with maleic anhydride polystyrene compatibilizer SPS/nylon blend containing two different nylons and a compatibilizer SPS/nylon blends formulated for fibers and films

Reference 2 3

4 5 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

as under-the-hood automotive connectors. Most of the data below are for SPS/ nylon blends that contain a ratio of 70:30 nylon 6,6/SPS in the polymer matrix. 16.3.1

Mechanical Properties of SPS/Nylon Blends

The room temperature dry-as-molded (DAM) tensile stress-strain curves for 30% glass fiber-reinforced SPS/nylon blends versus 30% glass fiber-reinforced

341


TABLE 16.2 SPS/Nylon Blend Compositions Described in Technical Journals SPS/Nylon Blend Described in Journal Articles An Sb-free ignition-resistant nylon/syndiotactic polystyrene (SPS) blend with improved mechanical properties Preparation of a new grafted syndiotactic polystyrene (g-SPS), to be used as a compatibilizer for SPS/polyamide (PA) 66 blends Preparation and properties of nanocomposite composed of SPS/ PA 66/g-SPS/nano-Al2O3 particles The effect of adding sulfonated syndiotactic polystyrene ionomer (SPS-Zn), to immiscible blends to SPS/PA6 Preparation and use in SPS/nylon blends of maleic anhydride-grafted syndiotactic polystyrene (SPS-g-MA) Analysis of sulfonated syndiotactic polystyrene (SPS-H) as a compatibilizer for blends of SPS/PA6 The use of maleated copolymer of paramethylstyrene and styrene and PA as a glass coupling system for SPS compositions The toughening behavior of semicrystalline polymers was investigated using SPS/ PA6 blends compatibilized with maleic anhydridefunctionalized poly (styrene-b-(ethylene-co-butylene)-b-styrene) SEBS-MA triblock copolymer Comparison of phthalic anhydride terminated polystyrene (PS-An) and styrene-maleic anhydride copolymer (SMA) as compatibilizers for PA 66/polystyrene (PS) blends A review covering the characteristics of SPS, the latest studies on modifications of SPS by blending with various polymers, for example, PPO, elastomers, PA, and polyesters, and applications of SPS The compatibilization of SPS/PA6 blends with maleic anhydridegrafted syndiotactic polystyrene (SPS-g-MA) as a reactive compatibilizer Use of styrene/glycidyl methacrylate (SG) copolymers as compatibilizers for blends of PA6 with SPS Studies of sulfonated syndiotactic polystyrene (SSPS-H) as compatibilizer in the blend of SPS/PA6 The preparation and characterization of novel syndiotactic polystyrene/ polyamide-6/sulfonated syndiotactic polystyrene/dodecylaminemodified montmorillonite (SPS/PA6/SSPS/MTA) nanocomposite by a melt-intercalation-blend process Compositions consisting of SPS, a PA resin, glass fibers, and a fluorescent mineral filler (for enhancing UV stability, e.g., CaO/SiO2)

Reference 31 32 33 34 35 36 37 38

39

40

41

42 43 44

45

SPS and 30% glass fiber-reinforced polybutylene terephthalate (PBT) are illustrated in Figure 16.1. The data demonstrate that 30% glass fiberreinforced SPS/nylon has 47% higher tensile strength than 30% glass fiberreinforced SPS and PBT, and 22% higher tensile elongation than 30% glass fiber-reinforced SPS. The improvement in elongation has been important for designing functional latches for automotive connectors.

342


30,000 30% GF SPS/PA

Stress (psi)

25,000 20,000

30% GF SPS 30% GF PBT 1

15,000

30% GF PBT 2

10,000 5,000 0 0.0

1.0

2.0

3.0

4.0

% Strain

Figure 16.1 Room temperature dry-as-molded tensile stress–strain curves for 30% glass fiber-reinforced SPS/nylon blends versus 30% glass fiber-reinforced SPS and 30% glass fiber-reinforced PBT.

30,000

30% GF GFSPS/PA SPS/PA 30% 30%GF GFPBT PBT 30%

DAM

Conditioned at 40°C, 50% r.h.

Stress (psi)

25,000 20,000 15,000

Conditioned Conditioned DAM

10,000 5,000 0 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 % Strain

Figure 16.2 Dry-as-molded (DAM) versus moisture-conditioned room temperature tensile properties for 30% glass fiber-reinforced SPS/nylon blends and 30% glass fiberreinforced PBT. (Moisture-conditioned samples were conditioned at 40°C and 50% r.h. for 1 week.).

It is well known that nylon can absorb a few percent of water depending on the environment, and this absorbed water plasticizes the nylon, thus reducing strength and modulus and increasing ductility [48]. Figure 16.2 demonstrates that this phenomenon also holds true for SPS/nylon blends since, as mentioned, these blends, to date, have nylon as the continuous phase. Figure 16.2 shows that after exposure to conditions that have been shown to yield the highest moisture absorption that could be expected in use, 30% glass fiber reinforced. SPS/nylon blends still have 18% higher strength than 30% glass


343

1000

Viscosity (Pa·s)

30% GF SPS/PA 30% GF PBT 30% GF IM SPS 30% GF Nylon

100

10 10

100

1000 10,000 Shear rate (1/s)

100,000

Figure 16.3 Apparent shear viscosity versus apparent shear rate for 30% glass fiberreinforced SPS/nylon, PBT, impact-modified (IM) SPS, and nylon at mid-range processing temperatures.

fiber-reinforced SPS or 30% glass fiber-reinforced PBT. The effect of moisture is to increase the elongation to break. This improved ductility is even more beneficial for the robustness of connector latches. 16.3.2

Rheology of SPS/Nylon Blends

Viscosity versus shear rate curves for 30% glass fiber-reinforced PBT, impact modified SPS, nylon, and SPS/nylon at mid-range processing temperatures are depicted in Figure 16.3. The data show that the glass fiber-reinforced SPS/ nylon has a much lower viscosity than glass fiber-reinforced PBT and nylon at the high shear rates (>5000/s) seen in typical injecting molding processes. Note that the glass fiber-reinforced SPS/nylon adopts some of the lower viscosity characteristics of nylon at low shear rates and adopts the shear thinning nature of glass fiber-reinforced SPS at high shear rates. This allows the glass fiber-reinforced SPS/nylon to flow into thinner walls than the glass fiberreinforced PBT or nylon. Also, there is less molded in stress with the glass fiber-reinforced SPS/nylon because of the lower viscosity. 16.3.3 Moisture Absorption and Moisture Growth of SPS/Nylon Blends It is well known that nylon-based materials absorb water and that this affects properties and results in moisture growth. In this section, the amount of moisture absorbed and resultant moisture growth for glass fiber-reinforced SPS/ nylon blends is compared with glass fiber-reinforced nylon. The moisture growth and moisture absorption measurements were conducted on 5″ × 5″ × 1/8″ plaques. Before exposure to a given environmental

344


condition, the plaques were taken directly from the injection molding machine and placed in moisture barrier bags; this is defined as DAM or conditioning time equal to zero condition. The moisture conditioning entailed placing five plaques in a humidity oven set at 90 °C, 95% relative humidity (r.h.). This was thought to represent the most severe humidity conditions a part might experience. Moisture growth and absorption measurements were made at 0, 3, 10, 15, 23, 31, 39, 51, 60, 66, 80, and 87 days. After each exposure time given above, the dimensions of each plaque were measured in four places to determine dimensional stability, and the plaque was weighed to determine the amount of moisture absorbed. Samples from the water bath and the 95% humidity oven were allowed to cool to room temperature for approximately one half hour before measurement. The data for percent moisture absorption versus time of the 30% glass fiber-reinforced SPS/nylon grades versus 30% glass fiber-reinforced nylon after exposure to 90 °C, 95% r.h., are shown in Figure 16.4, and the data for moisture growth of these materials are plotted in Figure 16.5.

% Moisture absorption

30% Glass fiber reinforced SPS/nylon blend 4 3.5 3 2.5 2 1.5 1 0.5 0 0

20

40

60

80

100

60

80

100

Time (h)

% Moisture absorption

30% Glass fiber reinforced nylon 6,6 4 3.5 3 2.5 2 1.5 1 0.5 0 0

20

40 Time (h)

Figure 16.4 Moisture absorption of 30% glass fiber-reinforced SPS/nylon blends and 30% glass fiber-reinforced nylon exposed to 90°C and 95% r.h.


345

30% Glass fiber reinforced SPS/nylon blend With flow middle of part With flow edge of part

% Change in dimensions

Across flow middle of part

0.7

Across flow end of flow

0.6 0.5 0.4 0.3 0.2 0.1 0 0

20

40

60

80

100

Time (h)

30% Glass fiber reinforced nylon 6,6 With flow middle of part With flow edge of part

% Change in dimensions

Across flow middle of part

0.7

Across flow end of flow

0.6 0.5 0.4 0.3 0.2 0.1 0 0

20

40

60

80

100

Time (h)

Figure 16.5 Moisture growth of 30% glass fiber-reinforced SPS/nylon blends and 30% glass fiber-reinforced nylon exposed to 90°C and 95% r.h.

All materials reached moisture equilibrium in less than 3 h in the 90 °C, 95% r.h. environment. The 90 °C, 95% r.h conditioning data show that the 30% glass fiber-reinforced nylon absorbed 1.5% more moisture than the 30% glass fiber-reinforced SPS/nylon at equilibrium. The moisture growth in the 90 °C, 95% r.h. conditioning for 30% glass fiber-reinforced SPS/nylon was about half of that of 30% glass fiber-reinforced nylon. These data illustrate the positive effect of adding SPS to nylon on moisture absorption and moisture growth properties. Of course, the more SPS that is added to the blend, the greater the effect on these properties.

346


16.3.4

Dimensional Stability of SPS/Nylon Blends

Glass fiber-reinforced SPS/nylon blends have shown a synergistic combination of two aspects of dimensional stability for some applications. The first is outof-the-mold dimensional stability, or the ability to hold the dimensions of the mold during injection molding, and the second is creep resistance, or resistance to permanent deflection under load at elevated temperatures. These are illustrated briefly below. To compare the out-of-the-mold dimensional stability of glass fiberreinforced SPS/nylon with other glass fiber-reinforced resins, the skew of a wall of a divided tray, as shown in Figure 16.6, was measured right after injection molding. Figure 16.6 demonstrates that glass fiber-reinforced SPS/nylon blends are comparable to the excellent out-of-the-mold dimensional stability of glass fiber-reinforced SPS, and superior to glass fiber-reinforced PBT. The glass fiber SPS/nylon blends also were much better than glass fiber-reinforced nylon 6,6 in this characteristic. To measure creep resistance in an accelerated manner, the permanent deflection of test bars exposed to 154 psi stress at 200 °C for 20 min was measured for glass fiber-reinforced SPS/nylon versus glass fiber-reinforced SPS and PBT. Figure 16.7 illustrates that the glass fiber-reinforced SPS/nylon blends had better creep resistance than 30% glass fiber-reinforced PBT and much better creep resistance than glass fiber-reinforced SPS. The glass fiberreinforced SPS/nylon blends were comparable to glass fiber-reinforced nylon 6,6 in creep resistance at the stated conditions.

0.018 0.016

Edge 1 Edge 2 Deflection

0.014

Skew

0.012

• Skew = deflection/Length • Length = distance between Edge 1 and Edge 2 • Deflection = vertical distance from reference

0.010 0.008 0.006 0.004 0.002 0 30% GF PBT

20% GF SPS/PA 30% GF SPS/PA1 30% GF SPS/PA2 30% GF SPS

Figure 16.6 Skew of a wall of a divided tray, right after injection molding with glass fiber-reinforced SPS/nylon blends versus glass fiber-reinforced SPS and PBT.

Average net deflection (in.)


347

0.5 0.45 0.4 0.35 0.3 0.25 0.2 0.15 0.1 0.05 0

30% GF IM SPS

30% GF PBT

30% GF SPS/PA1

30% GF SPS/PA2

Figure 16.7 Permanent deflection of test bars exposed to 154 psi stress at 200°C for 20 min for glass fiber-reinforced SPS/nylon versus glass fiber-reinforced SPS and PBT.

TABLE 16.3 Definition of Temperature Class for USCARPF-1 Standard Class I II III IV

16.3.5

Ambient Temperature (°C) Range

Maximum Temperature (°C) (Ambient + Rise)

−40 ⇔ + 85 −40 ⇔ + 100 −40 ⇔ + 125 −40 ⇔ + 155

+105 +120 +145 +175

USCAR Performance of SPS/Nylon Blends

The USCAR Electric Wiring Components Applications Partnership developed a PF-1 Specification whose objective is to ensure that the components of the electrical connection systems will maintain functionality for at least 100,000 h or 150,000 driving miles, whichever comes first. There are three tests in the PF-1 standard that could degrade plastics and hence affect the performance of the electrical interconnect system components. Since a major application for SPS/nylon blends is in automotive underthe-hood connectors, it is important that these blends perform well in these tests. The first relevant test is to expose the electrical interconnect system components to the top temperature in the ambient temperature range for a given USCAR class, as defined in Table 16.3, for 1008 h. In a second test, the electrical interconnect system components are exposed to 40 cycles of extreme temperatures (−40°C to maximum temperature for temperature class) with r.h. levels ranging from 10% to 95%.

348


TABLE 16.4 USCAR Class III and IV Thermal Stability Test Results Product

30% GF SPS/ PA 10% GF SPS/ PS

Initial Properties

Class III Thermal Stability Test (125°C for 45 days)

Class IV Thermal Stability Test (155°C for 45 days)

Tensile Strength (psi)

% Elongation


% Elongation


% Elongation

26,600

2.8

25,600

2.3

21,800

1.6

11,400

3.2

12,200

2.9

6,000

1.0

GF SPS/PA, glass fiber syndiotactic polystyrene/polyamide.

TABLE 16.5 USCAR Class III and IV Hydrolysis Test Results Product

30% GF SPS/ PA 10% GF SPS/ PS

Initial Properties Tensile Strength (psi)

% Elongation

26,600

2.8

11,400

3.2

Class III Hydrolysis Test Tensile Strength (psi)

Class IV Hydrolysis Test

% Elongation


% Elongation

25,700

2.7

22,500

3.1

11,500

3.5

10,300

3.7

GF SPS/PA, glass fiber syndiotactic polystyrene/polyamide.

Table 16.4 shows that glass fiber-reinforced SPS/nylon blends start to lose strength and elongation in the Class IV thermal stability test. In practice, the connectors exposed to this testing have maintained their functionality with no degradation in performance. Table 16.5 illustrates the results for the temperature/humidity cycling of the second test described above for glass fiber-reinforced SPS/PA. The SPS/PA performs exceptionally well in this test. The plasticization effect of moisture on then nylon phase of the blends is evident in the results. Glass fiber-

APPLICATIONS OF SPS/NYLON BLENDS

349

TABLE 16.6 Automotive Fluid Resistance (5-min immersion) Brake Fluid Oil Gasoline Engine coolant Automatic Transmission Fluid Windshield Washer Fluid Power Steering Fluid Diesel Fuel M85 Methanol Fuel

SAE RM66-04 ASTM IRM-902 ASTM Reference Fuel C ASTM Service Fluid 104 SAE J311

at 50 °C at 100 °C at 25 °C at 100 °C at 100 °C

— ASTM IRM-903 ASTM Reference Fuel F ASTM Reference Fuel K

at 25 °C at 100 °C at 25 °C at 25 °C

reinforced nylon 6,6 and PBT show large drops in properties when exposed to Class III and Class IV temperature/humidity cycling. A third potentially degradative USCAR PF-1 test is 5-min immersion of the electrical interconnect system component in various automotive fluids at different temperatures as shown in Table 16.6. The SPS/PA blends perform very well in these tests. 16.3.6

Environmental Stress Crack Resistance of SPS/Nylon Blends

The environmental stress crack resistance (ESCR) of glass fiber-reinforced SPS/nylon blends was characterized by exposing the surface of test bars with 0% and 1% strains applied to drops of solvent and then observing the damage to the surface. In addition, 7-day and 30-day chemical immersion resistance were evaluated. The results of this testing for glass fiber-reinforced SPS/nylon blends are shown in Table 16.7.

16.4


The combination of properties for SPS/nylon blends described above has led to substantial amounts of these blends being used in under-the-hood connectors in the automotive industry. These blends have also received much interest in the carpet fiber industry. There are a great number of patents where SPS/ nylon blends are a candidate material for the application being patented. In some patents, SPS/nylon is the required material. These applications are described in more detail below. 16.4.1

SPS/Nylon Blend Under-the-hood Automotive Connectors

SPS/nylon blends have found wide acceptance in automotive under-the-hood connectors for all of the technical reasons described above. In addition, the SPS/nylon blends have the required electrical insulation properties for the

350


TABLE 16.7 Summary of Chemical Resistance of Glass Fiber-reinforced SPS/PA Blends Solvent Acetic acid 5% Acetic acid Acetone 10% Ammonium hydroxide Benzene Bleach Ethyl acetate Ethyl alcohol Ethylene dichloride 37% Hydrochloric acid 13% Hydrochloric acid 3% Hydrogen peroxide Methyl ethyl ketone 1% Soap 10% Sodium chloride 60% Sodium hydroxide 10% Sodium hydroxide Toluene

30% GF SPS/PA

20% GF SPS/PA

10% GF SPS/PA

Excellent Excellent Excellent Good



Excellent Good Excellent Good Good Poor Poor Fair Excellent Good Excellent Good Good Excellent



Excellent = Minimal to no change in properties after exposure. Good = Minor changes in properties after exposure. Fair = Moderate changes in properties after exposure; not recommended for long-term use. Poor = Significant changes in properties after exposure; not recommended for use. GF SPS/PA, glass fiber syndiotactic polystyrene/polyamide.

connector application. Examples of these under-the-hood connectors are given in Figure 16.8. 16.4.2

SPS/Nylon Blend Carpet Fibers

SPS/nylon blends have been used to form bicontinuous, self-bulking fibers resulting from melt spinning joined melt streams, one of which comprises SPS, and the other a melt spinnable polymer other than SPS (e.g., nylon 66) [49]. More typically, SPS has been used as an additive in nylon 6 to form carpet fibers that match the morphology and hence the luster and softness of natural fibers such as wool. This technology is described briefly below. It was determined that the size and shape of the dispersed phase (SPS) in the nylon 6/SPS fiber are critical to obtaining surface roughness. Three factors affect this morphology: compatibilization of the blend, viscosity ratio of the nylon 6 and SPS phases, and concentration of SPS in the blend. Figure 16.9 depicts the effect of viscosity ratio on the morphology of the SPS/nylon blends.


351

Figure 16.8 Photograph of selected automotive under-the-hood connectors.

50 micron

50 micron

-

50 micron

-

-

Figure 16.9 Effect of viscosity ratio on particle size of SPS in the fiber. The photographs are scanning electron microscopy pictures of the SPS phase that remains after dissolving away the nylon 6. These have been collected on filters so the spatial arrangement is not significant. MFR, melt flow rate; RV, relative viscosity.

352


Extruder

Spin beam/spinneret

Die swell

Melt stretch 20× draw, 300°C to 100°C

Quench cabinet

300°C

Air at 10°C

First roll 25ºC

First draw roll 75ºC

Second draw roll (hot draw) 200ºC

Texturizer

3× Draw

Figure 16.10

210ºC

Fiber spinning overview.

Final morphology develops after hot draw

25 μm

Extruder Pellet

First roll First draw roll

Quench cabinet

Quench

Second draw roll (hot draw) BCF

Figure 16.11 SPS phase morphology development during the spinning of nylon 6/SPS fibers. The photographs are scanning electron microscopy pictures of the SPS phase that remains after dissolving away the nylon 6. These have been collected on filters so the spatial arrangement is not significant. BCF, bulk continuous fiber.


Quench

353

Complete surface roughness develops after hot draw Second (hot) draw roll Quench cabinet

Extruder

First roll

First draw roll

BCF

Figure 16.12 Fiber surface morphology development during the spinning of nylon 6/ SPS fibers. BCF, bulk continuous fiber.

Figures 16.10–16.12 illustrate the morphology and surface roughness development during the spinning process. Figure 16.10 shows a schematic of the fiber spinning process, whereas Figures 16.11 and 16.12 demonstrate the SPS phase morphology and fiber surface morphology development in the process. These figures demonstrate that a significant portion of the surface roughness developed after the hot draw. The conclusion of the nylon 6/SPS fiber work was that, to obtain correct particle sizing, the viscosity ratio (SPS/PA6) at low shear rates ( TS,M1911 > TS,M1913. 18.4.2 Evaluation of Domain Size and Interfacial Thickness HISPS used in this experiment comprises SPS (Mw = 2.0 × 105 g/mol, 2.8 × 105 g/mol, 4.5 × 105 g/mol), PPO, elastomer, and compatibilizer. Elastomers are EPR (Mw = 3.0 × 105 g/mol, Mw/Mn = 2.2) and ethylene-propylene-diene rubber (EPDM) (Mooney viscosity at 100°C = 88), and compatibilizers MA-gSEBS (Tuftec M1913), SEBS (H-1081, 60% styrene), poly(styrene)-blockethylenepropylene-block-styrene (SEPS) (Septon 2105, 50% styrene), and

COMPATIBILIZER EFFECTS

389

poly(styrene-block-ethylenebutylene-block-crystalline olefin) (SEBC) (HSB575, 45% styrene). The blend sample was compression molded into a thin film with 25-μm thickness at 300°C mold temperature and then was quickly quenched in ice water. The film specimen prepared in this procedure is amorphous. The light scattering (LS) technique was employed to evaluate domain size and interfacial thickness. The scattered intensity was corrected for transmission and was converted to the intensity per unit solid angle [14]. Subtracting the scattering of glass cell from total scattered intensity, the true scattering can be obtained from a sample I(q), where q is a magnitude of scattering vector defined as q = (4nπ/λ0)sin(θ/2), where λ0 is the wavelength of the incident beam in the vacuum, n is the refractive index of the sample, and θ is the scattering angle. The scattered intensity distribution at a small q is given by the Guinier equation [15] ⎛ Rq ⎞ I (q ) = I ( 0 ) exp ⎜ − ⎟, ⎝ 3 ⎠ 2

(18.16)

where R is the domain size and I(0) is the scattered intensity at q = 0. Figure 18.15 shows the plot of ln I(q) against q2 for SPS/EPR (80/20) and SPS/EPR/MA-g-SEBS (80/16/4) blends. R can be obtained from the initial slope of the straight line. In Figure 18.15, R of SPS/EPR/MA-g-SEBS is found to be smaller than that of SPS/EPR, which indicates that MA-g-SEBS acts as a “compatibilizer.” MA-g-EB chains are adsorbed on the surface of EPR and 12

11 SPS/EPR/MASEBS(80/16/4) R g = 850 nm

Ln I(q)

10

9

8

7

SPS/EPR (80/20) R g = 1108 nm

6 0

5

10 15 q2 (nm–2) × 106

20

25

Figure 18.15 Plot of ln [I(q)] versus q2 for SPS/ethylene-propylene rubber (EPR) and SPS/EPR/MA-g-SEBS.

390

COMPATIBILIZERS FOR IMPACT-MODIFIED SYNDIOTACTIC POLYSTYRENE

–14.0 SPS/EPR/MA-g-SEBS –14.5

–15.0 SPS/EPR –15.5

–16.0 10

Figure 18.16

20 q2 (nm–2)

30

40 × 10–6

Plot of ln [q4I(q)] versus q2.

styrene chains of MA-g-SEBS dissolved into the SPS matrix. This phase structure causes the reduction of interfacial tension, and hence the domain size becomes smaller (compatibilizer effect). Analyzing the LS curve at the Porod region where I(q) obeys the power law of I ∼ q−4, the interfacial thickness can be evaluated. If the interface is assumed by the sigmoidal curve defined as Gaussian with standard deviation σ, the scattering at the Porod region can be expressed as [16] I (q ) = κ η2 Σq4 exp ( −σ 2 q2 )

(18.17)

1 q2 I (q ) dq πφ (1 − φ ) ∫

(18.18)

and κ η2 =

where is the mean square of fluctuations, Σ is the surface area per unit volume, ϕ is the volume fraction of the elastomer, and κ is a constant. Plotting ln q4I(q) against q2, the slope and intercept of the straight line give σ and Σ, respectively, as shown by Figure 18.16. The characteristic interfacial thickness t is calculated using t = (2π)1/2σ [1]. Results are summarized in Table 18.6. The interfacial thickness and the surface area are found to increase with the addition of MA-g-SEBS. These results can be interpreted as the compatibilizer effect. However, this system does not indicate enough impact strength because the domain size is too large. Thus, it is necessary to make the domains much smaller by adjusting the primary structure of the polyolefin and the compatibilizer. Table 18.7 shows the domain size obtained from small-angle light scattering (SALS) experiments. From the table, it is obvious that R is decreasing by the addition of compatibilizers (compatibilizer effect). It is found that the Izod impact strength tends to increase with decreasing R. However, the impact

391

COMPATIBILIZER EFFECTS

TABLE 18.6 Structural Parameters Obtained from Light Scattering Experiments

SPS/EPR SPS/EPR/MA-g-SEBS

TABLE 18.7 Elastomers EPR

SEBS

Rg (nm)

t (nm)

Σ (nm−1)

1108 850

134 211

1.65 × 10−7 2.83 × 10−7

Domain Size R and Izod Impact Strength of HISPS Compatibilizer

R (nm)

Izod (Jm−2)

None SEBS SEBC SEPS None MA-g-SEBS

1055 742 600 788 548 470

2.4 4.1 5.6 4.0 22 36

strength of the EPR system is much smaller than that of the SEBS system. This is because the domain size of the EPR system is too large in comparison with that of the SEBS system. Thus, the domain size should be decreased in order to get higher-impact strength equivalent to the SEBS system. The domain size can be controlled by the molecular weight of the domain or the matrix. The dependence of the domain size on molecular weight is given by Wu’s equation, which was established for incompatible blends [17]: R∝

σ I γ ⎛ ηd ⎞ ⎜ ⎟ ηm ⎝ ηm ⎠

0.84

,

(18.19)

where ηm and ηd are the viscosity of the matrix and the domain, respectively, and this is the case of ηm < ηd; σI is the interfacial tension and γ is the shear rate. According to Equation 18.19, the increase of ηm gives rise to the decrease of R. The domain size R obtained from LS and from the Izod impact strength for HISPS are summarized in Table 18.8. By increasing the viscosity of SPS matrix ηm, the domain size R is decreased with the increase of ηm. It is obvious from the table that, by increasing the Mw of SPS, R is decreased and the Izod impact strength is increased. Contrary to the expectation, the impact strengths are much smaller than those of the SEBS system even though R of the 446k series are equivalent to that of the SEBS system. This can be explained by two factors: (a) The adhesion between the domain and the matrix in the EPR and EPDM systems is weaker than in the SEBS system, and (b) the impact strength of HISPS is reflected by the mechanical properties of the elastomer itself. Although there is no information about the former at present, the differences in the mechanical properties among the elastomers are clearly observed. Figure 18.17 shows the flow behaviors of the elastomers, SEBS, EPR, and EPDM.

392

COMPATIBILIZERS FOR IMPACT-MODIFIED SYNDIOTACTIC POLYSTYRENE

TABLE 18.8 Domain Size R and Izod Impact Strength of EPR and EPDM Systems as a Function of Mw of SPS Elastomer

EPR

Compatibilizer

Mw of SPS (g/mol)

SEBS

2.78 × 105

3.51 × 105

4.46 × 105

699 3.6 898 3.6 769 4.2

615 5.7 657 5.5 614 5.2

534 7.8 610 7.6 701 12.2

470 7.6 464 12.7 469 15.5

R (nm) Izod (Jm−2) R (nm) Izod (Jm−2) R (nm) Izod (Jm−2)

SEPS EPDM

1.95 × 105

SEBC

4

T = 130°°C SEBS

Viscosity (Pa·s)

2

10 48 EPDM

6 4

EPR

2

10 38 6 4 2

10

2

1

2

3 4 5 6 78

10

2

3 4 5 6 78

100

2

3 4 5 6 78

1000

Shear rate (1/s)

Figure 18.17 Flow behaviors measured by the capillary rheometer for SEBS, EPR, and EPDM at 300°C.

It is found that SEBS is highly viscous and exhibits strong non-Newtonian flow. The flow behavior of SEBS is caused by the microphase-separated structure in which cylindrical styrene domains act as physical cross-linking points. It is supposed that the existence of the ordered structure in the domain, as seen in SEBS, should be a key factor to improve the impact strength. Based on this speculation, the idea of organizing some ordered structures inside the EPR domains might be interesting, for example, a two-phase structure, which is formed by the phase separation between EPR and PP.

REFERENCES

393

REFERENCES 1. Hashimoto, T., Shibayama, M., Kawai, H. Domain-boundary structure of styreneisoprene block copolymer films cast from solution. 4. Molecular-weight dependence of lamellar microdomains. Macromolecules, 13, 1237–1247 (1980). 2. Hashimoto, T., Fujimura, M., Kawai, H. Domain-boundary structure of styrene-isoprene block copolymer films cast from solutions. 5. Molecular-weight dependence of spherical microdomains. Macromolecules, 13, 1660–1669 (1980). 3. Strobl, G. R., Schneider, M. Direct evaluation of the electron density correlation function of partially crystalline polymers. J. Polym. Sci. Polym. Phys., 18, 1343– 1359 (1980). 4. Balta-Calleja, F. J., Vonk, C. G. The small-angle x-ray scattering of polymers. In X-ray Scattering of Synthetic Polymers, Jenkins, A. D. (ed.), Elsevier Science Publishing, Amsterdam, 1989, pp. 241–306. 5. Keith, H. D., Padden F. J. Jr., Spherulitic crystallization from the melt. I. Fractionation and impuritiy segregation and their influence on crystalline morphology. J. Appl. Phys., 35, 1270–1285 (1964). 6. Sakurai, S., Hasegawa, H., Hashimoto, T., Hargis, I. G., Aggarwal, S. L., Han, C. C. Microstructure and isotopic labeling effects on the miscibility of polybutadiene blends studied by the small-angle neutron scattering technique. Macromolecules, 23, 451–459 (1990). 7. Shibayama, M., Yang, H., Stein, R. S., Han, C. C. Study of miscibility and critical phenomena of deuterated polystyrene and hydrogenated poly(vinyl methyl ether) by small-angle neutron scattering. Macromolecules, 18, 2179–2187 (1985). 8. Mori, K., Hasegawa, H., Hashimoto, T. Effects of galbanic acid on thermal and thermo-oxidative stabilities of LLDPE. Polymer J., 17, 799–806 (1985). 9. Hashimoto, T., Kowsaka, K., Shibayama, M., Kawai, H. Time-resolved small-angle X-ray scattering studies on the kinetics of the order-disorder transition of block polymers. 2. Concentration and temperature dependence. Macromolecules, 19, 754–762 (1986). 10. Mori, K., Tanaka, H., Hashimoto, T. Scattering functions for disordered twocomponent polymer systems including block polymers. Macromolecules, 20, 381– 391 (1987). 11. It is described in Chapter 17, section 17.4 of this book. 12. Leibler, L. Theory of microphase separation in block copolymers. Macromolecules, 13, 1602–1617 (1980) 13. de Gennes, P.-G. Scaling Concepts in Polymer Physics, Cornell University Press, New York, 1979. 14. Stein, R. S., Keane, J. J. The scattering of light from thin polymer films. I. Experimental procedure. J. Polym. Sci., 17, 21–44 (1953). 15. Guinier, A., Fournet, G. Small-Angle Scattering of X-ray, Wiley, New York, 1955. 16. Porod, G. The X-ray small-angle scattering of close-packed colloid systems. I. Koll. Z., 124, 83–114 (1951). 17. Wu, S. Formation of dispersed phase in incompatible polymer blends: Interfacial and rheological effects. Polym. Eng. Sci., 27, 335–343 (1987).

PART VI

POLYMERS BASED ON SYNDIOTACTIC POLYSTYRENES

CHAPTER 19

Functionalization and Block/Graft Reactions of Syndiotactic Polystyrene Using Borane Comonomers and Chain Transfer Agents T. C. MIKE CHUNG Department of Materials Science and Engineering, The Pennsylvania State University, University Park, PA, USA

19.1

INTRODUCTION

One of the most interesting advances in metallocene technology is the preparation of syndiotactic polystyrene (SPS) [1], which is a novel engineering plastic made from an inexpensive commodity monomer. SPS exhibits a high melting point (∼270 °C) and a high crystallization rate that is several orders of magnitude [2] faster than that in isotactic polystyrene. Despite its unique properties, SPS polymers also have several drawbacks that pose concerns in many commercial applications. Due to a high melting point, the required melt processing temperature at >300 °C (near the decomposition temperature) causes a major problem in polymer processability. A modified SPS polymer with a slightly reduced melting point (to about 250 °C) would be a very desirable material. In addition, SPS polymers have low surface energy, similar to traditional atactic polystyrene prepared by free radical and anionic processes. The absence of polar groups in hydrophobic polystyrene restricts their end uses, especially where adhesion to substrates (metals, ceramics, glass, etc.) and compatibility with polar polymers are desired. There are few reports discussing the modification of SPS polymer, including sulfonation [3] and bromination [4] of SPS, hydroxylated SPS [5] prepared via poly(styrene-co-4tert-butyldimethylsilyloxystyrene) precursor, and amino-SPS homopolymers [6] prepared by direct polymerization of functional styrene monomers containing a silane-protected amino group.


397

398

CHAIN TRANSFER AGENTS

Several years ago, Chung [7] developed a unique strategy in the functionalization of polyolefins, including polyethylene (PE), polypropylene (PP), and ethylene/propylene (EP) copolymers. The chemistry involves boranecontaining monomers [8] or chain transfer agents [9] during direct metallocene-mediated α-olefin polymerization. Due to the excellent stability of borane moieties to transition metal catalysts and the good solubility of trialkylborane in hydrocarbon media, the borane reagents behave like hydrocarbons and can be incorporated in the side chains or in the chain end of polyolefin without sacrificing catalyst activity or homogeneity of polyolefin molecular structures with narrow molecular weight and composition distributions. In addition, the incorporated borane groups in the polyolefin copolymers are very versatile, which can be effectively interconverted to various functional groups under mild reaction conditions. They are also auto-oxidized by oxygen to form “control” polymeric radicals for graft/block reactions [10] to prepare polyolefin graft/block copolymers, such as PP-g-poly(methyl methacrylate) (PMMA), PP-b-PMMA, PE-g-PMMA, and PE-b-PMMA. These borane comonomer and chain transfer agent approaches were also extended to SPS to prepare side-chain [11] and chain-end [12] functionalized SPS polymers, as well as SPS graft and block copolymers. 19.2

FUNCTIONALIZATION OF SPS VIA BORANE COMONOMERS

Scheme 19.1 illustrates the chemical route to prepare the side-chain functionalized SPS polymers and SPS graft copolymers. The chemistry is centered on an intermediate SPS copolymer (II) containing some pending borane groups, which is prepared by half-sandwich metallocene-mediated styrene copolymerization with 4-[B-(n-butylene)-9-BBN]styrene (B-styrene) comonomer (I). The pending borane groups in the copolymer (II) are very versatile and can be quantitatively interconverted to the desirable functional groups, such as hydroxy and anhydride, under mild reaction conditions. In addition, the pending borane groups are also transformed to stable polymeric radicals for control radical graft-from polymerization to prepare SPS graft copolymers, such as SPS-g-PMMA containing polar polymer side chains. 19.2.1

Copolymerization of Styrene and B-styrene

Several conventional half-sandwich metallocene catalysts, including Cp*Ti(OMe)3/methylaluminoxane (MAO), Cp*TiCl3/MAO, and CpTiCl3/ MAO complexes [1] that exhibit high syndiospecificity and good catalyst activity in styrene polymerization, were applied in the copolymerization of styrene and B-styrene. The resulting copolymers, poly(styrene-co-4-[B-(n-butylene)9-BBN]styrene) (II), were completely soluble in tetrachloroethane at elevated temperature (110 °C). Some copolymers with high B-styrene comonomer contents (>8.4 mol %) are soluble in toluene or tetrahydrofuran (THF) at ambient temperature. 11B nuclear magnetic resonance (NMR) spectra of all copoly-

399


x CH2 =CH

+

y CH2 =CH B

(I)

=

B

(CH2 )4 B

Cp*Ti(OMe)3 /MAO

(CH2 -CH)x (CH 2 -CH)y (CH2 -CH)x (CH 2 -CH)y

H2 O2 /NaOH

(IV)

(CH2 )4 O H

(II) (CH2 )4 B

(CH2 -CH)x (CH 2 -CH)y

O2 O

O O

(V)

(CH2 -CH)x (CH 2 -CH)y

(CH2 )4 O O

O O

(III)

MMA (CH2 )4

(CH2 -CH)x (CH 2 -CH) y

O * * O B

(VI)

(CH2 )4 O PMMA

Scheme 19.1

mers show a single chemical shift at 87 ppm (vs. BF3.OEt2) corresponding to the pendant alky-9-BBN moieties. Due to the oxygen sensitivity of the borane moiety, the borane-containing SPS copolymers were usually oxidized by NaOH/H2O2 to form the corresponding stable hydroxy-containing SPS copolymers (SPS-OH) for analysis. As will be discussed later, all of the transformation reactions were quantitative even in heterogeneous conditions. Table 19.1 summarizes the results of styrene/B-styrene copolymerization reactions under various reaction conditions. In general, both monomers show similar reactivity. The para-substituted borane group exhibits almost no effect on the monomer incorporation, which also indicates the unique advantages of trialkylborane moiety with excellent stability to the transition metal catalysts and good solubility in hydrocarbon

400

Cp*Ti(OMe)3 Cp*Ti(OMe)3 Cp*Ti(OMe)3 Cp*Ti(OMe)3 Cp*Ti(OMe)3 Cp*Ti(OMe3) Cp*Ti(OMe)3 Cp*Ti(OMe)3 Cp*Ti(OMe)3 CpTiCl3 Cp*TiCl3

Catalyst

0 0.63 1.76 2.59 5.38 7.16 12.2 24.3 100 6.10 6.10

B-Sin Feed (mol %)

1.73 1.73 1.73 1.73 1.59 2.14 2.04 1.13 0 1.41 1.41

[S] (mol/l)

0 0.011 0.031 0.046 0.091 0.165 0.284 0.363 0.730 0.091 0.091

[B-S] (mol/l)

3.48 4.44 4.66 5.13 5.36 6.73 6.04 2.98 1.91 3.52 2.98

Yield(g)

67.0 85.1 88.3 96.2 106.3 97.3 86.7 66.7 87.3 78.2 66.2

Catalyst Activityb

0 1.0 2.1 3.0 6.1 8.4 16.7 32.2 100.0 6.7 5.4

B-S Content in Copolymer (mol %) 20.3 21.5 20.7 15.5 14.4 12.7 9.8 5.3 — — —

Mw × 104g/ mol

7.0 8.0 8.3 6.7 6.3 5.5 5.2 2.8 — — —

Mn × 104g/ mol

2.9 2.7 2.5 2.3 2.3 2.3 2.0 2.0 — — —

PDI Mw/Mn

Copolymerization of Styrene (S) and 4-[B-(n-butylene)-9-BBN]styrene (B-S) with Syndiospecific Titanocene/MAO

Polymerization conditions: [cat] = 100 μmol/l; Al/Ti = 2000; temperature = 50 °C; time = 3 h; solvent = 40 ml toluene. kg of polymer/mol Ti·h·mol/l.

b

a

A-1 A-2 A-3 A-4 A-5 A-6 A-7 A-8 A-9 B-1 C-1

Run Number

TABLE 19.1 Catalystsa


401

0.8

(1-F)/f

0.3

0 –0.2

0.4

0.8

1.2

1.6

F/f 2

–0.7

–1.2

Figure 19.1 Fineman and Ross plots for copolymerization of styrene/B-styrene and the least square best-fit line.

reaction media. The copolymer composition is basically controlled by the comonomer feed ratio, and we have prepared a whole composition range of styrene/B-styrene copolymers. In general, the catalyst efficiency in the copolymerization reaction increases somewhat compared to that of homopolymerization. The better diffusibility of monomers in the copolymer structures (due to lower crystallinity) may help provide better polymerization conditions. The best way to investigate a copolymerization is to measure the reactivity ratio of the comonomers. A series of experiments were carried out by varying the monomer feed ratio and by comparing the resulting polymer compositions at low conversion (

b c d e f g

75.0

125.0 175.0 225.0 275.0 Temperature (°C)

Figure 19.5 DSC curves of (a) SPS homopolymer and several SPS–OH copolymers containing (b) 1.0, (c) 2.1, (d) 6.1, (e) 8.4, (f) 16.7, and (g) 32.2 mol % of OH groups.

groups at 145.3 shows that these polymers are highly syndiotactic. The same conclusion can be reached by considering two aliphatic carbons in the polymer backbone. Two sharp resonances at 44.5 and 41.0 ppm correspond to methine and methylene in a highly stereoregular environment. In addition, a new resonance at 63.2 ppm corresponds to the methylene carbon atom in the –CH2–OH group. Overall, the sharp resonances in the copolymers resemble those of SPS and are very different from several reported atactic polystyrene derivatives containing chloro and methoxy groups [15] prepared with other catalyst systems. Figure 19.5 compares the DSC curves of an SPS homopolymer (A-1) and several SPS–OH copolymers (runs A-2, A-3, A-5, A-6, A-7, and A-8 in Table 19.2). As expected, both the melting point (Tm) and crystallinity (χc) of the copolymer are affected by the side chains. The higher the density of the side chain, the lower the Tm and χc. The detailed results of the melting point (Tm) and crystallinity (χc) of the copolymers are summarized in Table 19.2. The crystallinity (χc) of the copolymer was calculated from fusion enthalpy according to the following equation: χc = Hf/Hf × 100, where Hf and Hf (53 J/g) [16] are the fusion enthalpies of the copolymer and the SPS homopolymer, respectively. Sample A-3, with an average of one side chain per 50 styrene units, still possesses >40% crystallinity with a sharp melting peak at Tm = 250 °C. In addition to its functionality, this SPS–OH copolymer may also benefit from a lower processing temperature. The melting peak in the copolymer completely disappears at >8.4 mol % of B-styrene concentration. Only a single Tg was observed

406

CHAIN TRANSFER AGENTS

TABLE 19.2 Summary of the Thermal Propertiesa of Hydyoxylated SPS Copolymers Entry A-1 A-2 A-3 A-4 A-5 A-6 A-7 A-8

OH– Content (mol %)

Tg (°C)

TM (°C)

TcB (°C)

ΔH (J/g)

0 1.0 2.1 3.0 6.1 8.4 16.7 32.2

101.5 97.6 94.1 93.1 87.8 87.9 81.3 79.8

272.5 263.0 250.1 247.2 223.9 ND ND ND

233.1 226.9 208.7 197.7 ND ND ND ND

30.1 26.8 21.9 20.2 16.6 — — —

Xcc (%) 56.8 50.5 41.3 38.1 31.3 — — —

a

Based on DSC measurement with a heating and cooling rate of 20 °C/min. Crystallization temperature. c Crystallinity. ND, not detected. b

throughout the whole composition range, and the Tg systematically decreases with increasing side-chain density. All of the experimental results indicate that the copolymer microstructure is homogeneous, and that the copolymer morphology becomes completely amorphous with more than 8.4 mol % of B-styrene units. As illustrated in Scheme 19.1, the pending borane groups were selectively oxidized by oxygen to form polymeric radicals (III) that are associated with “stable” borinate radicals [10]. The polymeric radicals react in situ with maleic anhydride (MA) to produce MA-grafted SPS (SPS-g-MA) (V) with a single MA unit in each side chain. Figure 19.6 shows the infrared (IR) spectrum of an SPS-g-MA copolymer containing 8 mol % of MA units, and the insert compares the absorption region of the νC=O vibration modes between SPS and three SPS-g-MA copolymers with 1, 3, and 8 mol % MA contents, respectively. After the MA graft reaction, several new absorption peaks were observed at 1860 and 1780 cm−1 corresponding to two νC=O vibrational stretching modes [17] in succinic anhydride. The concentration of incorporated MA units was determined by the IR carbonyl group absorption intensity and sample thickness. Overall, this route provides SPS with very useful anhydride functional groups with high concentration. 19.2.3

SPS Graft Copolymers

The pending stable polymeric radicals associated with dormant borinate radicals in the SPS copolymer (III) are very useful in the preparation of SPS graft copolymers. The alkoxyl radical (C-O*) is active in initiating polymerization of functional monomers. On the other hand, the borinate radical (B-O*) is too stable to initiate polymerization due to the back donating of electrons to


407

5.00 1780 cm–1

4.00 Absorbance

d)

Absorbance

3.00

c)

b) a)

2.00

1880 1860 1840 1820 1800 1780 1760 1740

Wave number (cm–1)

1.00

0.00

3000

2500

2000 Wave number

1500

1000

500

(cm–1)

Figure 19.6 IR spectrum of an SPS-g-MA copolymer containing 8 mol % of MA units; the inset compares the absorption region of νC=O vibration modes between (a) SPS and three SPS-g-MA copolymers with (b) 1, (c) 3, and (d) 8 mol % MA content, respectively.

the empty p-orbital of boron. However, this “dormant” borinate radical may form a reversible bond with the radical at the growing chain end to prolong the lifetime of the propagating radical [10]. In the presence of methyl methacrylate (MMA) monomers, the free radical graft polymerization results in SPS-g-PMMA (VI) graft copolymers. The molecular weight of the incorporated PMMA side chain is proportional to the MMA monomers used, and the graft density is governed by the incorporated borane content. After the graft reaction, the product, isolated by filtration and washed with boiling MeOH, was then extracted with acetone in a Soxhlet apparatus for 24 h to remove the PMMA homopolymer (

Syndiotactic Polystyrene: Synthesis, Characterization, Processing, and Applications

Palladacycles: Synthesis, Characterization and Applications

Polymer Brushes: Synthesis, Characterization, Applications

Hybrid Materials: Synthesis, Characterization, and Applications

Graphene – Synthesis, Characterization, Properties and Applications

Hybrid Nanomaterials: Synthesis, Characterization, and Applications

Advanced Inorganic Fluorides: Synthesis, Characterization and Applications

Metal Nanoparticles: Synthesis Characterization & Applications

Synthesis, Characterization and Thermal Behaviour

Synthesis and Characterization of Glycosides

Colloidial Polymers Synthesis and Characterization

Polypropylene Handbook: Polymerization, Characterization, Properties, Processing, Applications

Polyimides and Other High Temperature Polymers: Synthesis, Characterization and Applications

Carbon Nanotubes and Related Structures: Synthesis, Characterization, Functionalization, and Applications

Characterization of Silicon Processing (Materials Characterization)

Handbook of Biodegradable Polymers: Synthesis, Characterization and Applications

Semiconductor Nanostructures for Optoelectronic Devices: Processing, Characterization and Applications

Peptide Synthesis and Applications

Heterophase Network Polymers - Synthesis Characterization and Properties

Cournot Oligopoly: Characterization and Applications

Cournot Oligopoly: Characterization and Applications

Peptide Synthesis and Applications

Peptide Synthesis and Applications

Polymer Synthesis and Characterization: A Laboratory Manual

Starches: Characterization, Properties, and Applications

Starches: Characterization, Properties, and Applications

Starches: Characterization, Properties, and Applications

Energetic Materials: Processing and Characterization of Particles

Oligonucleotide Synthesis Methods and Applications

Nanomaterials: synthesis, properties, and applications

Polyynes: Synthesis, Properties, and Applications

Syndiotactic Polystyrene: Synthesis, Characterization, Processing, and Applications

Palladacycles: Synthesis, Characterization and Applications

Polymer Brushes: Synthesis, Characterization, Applications

Hybrid Materials: Synthesis, Characterization, and Applications

Graphene – Synthesis, Characterization, Properties and Applications

Hybrid Nanomaterials: Synthesis, Characterization, and Applications

Advanced Inorganic Fluorides: Synthesis, Characterization and Applications

Metal Nanoparticles: Synthesis Characterization & Applications

Synthesis, Characterization and Thermal Behaviour

Synthesis and Characterization of Glycosides

Colloidial Polymers Synthesis and Characterization

Polypropylene Handbook: Polymerization, Characterization, Properties, Processing, Applications

Polyimides and Other High Temperature Polymers: Synthesis, Characterization and Applications

Carbon Nanotubes and Related Structures: Synthesis, Characterization, Functionalization, and Applications

Characterization of Silicon Processing (Materials Characterization)

Handbook of Biodegradable Polymers: Synthesis, Characterization and Applications

Semiconductor Nanostructures for Optoelectronic Devices: Processing, Characterization and Applications

Peptide Synthesis and Applications

Heterophase Network Polymers - Synthesis Characterization and Properties

Cournot Oligopoly: Characterization and Applications

Cournot Oligopoly: Characterization and Applications

Peptide Synthesis and Applications

Peptide Synthesis and Applications

Polymer Synthesis and Characterization: A Laboratory Manual

Starches: Characterization, Properties, and Applications

Starches: Characterization, Properties, and Applications

Starches: Characterization, Properties, and Applications

Energetic Materials: Processing and Characterization of Particles

Oligonucleotide Synthesis Methods and Applications

Nanomaterials: synthesis, properties, and applications

Polyynes: Synthesis, Properties, and Applications

Recommend Documents