Functionalization of Polyole®ns
This Page Intentionally Left Blank
Functionalization of Polyole®ns
T.C. (Mike) Chu...
70 downloads
895 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
Functionalization of Polyole®ns
This Page Intentionally Left Blank
Functionalization of Polyole®ns
T.C. (Mike) Chung
The Pennsylvania State University, University Park, PA 16802, USA
An Elsevier Science Imprint San Diego San Francisco New York Boston London Sydney Tokyo
This book is printed on acid-free paper. Copyright # 2002 by ACADEMIC PRESS All Rights Reserved. No part of this publication may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, recording, or any information storage and retrieval system, without permission in writing from the publisher. Academic Press An Elsevier Science Imprint Harcourt Place, 32 Jamestown Road, London NW1 7BY, UK http://www.academicpress.com Academic Press An Elsevier Science Imprint 525 B Street, Suite 1900, San Diego, California 92101-4495, USA http://www.academicpress.com ISBN 0-12-174651-8 Library of Congress Catalog Number: 2001093793 A catalogue record for this book is available from the British Library
Typeset by Newgen Imaging Systems (P) Ltd, Chennai, India Printed and bound in Great Britain by MPG Books Ltd, Bodmin, Cornwall 02 03 04 05 06 07 MP 9 8 7 6 5 4 3 2 1
Contents
Preface About the Author 1 Polyole®n Development and Prospects 1 2 3 4
Introduction Historical Aspects: Catalyst-driven Process Ziegler±Natta Catalysts and Polymerization Future Prospects: Material-driven Process
2 General Approaches in Functionalization of Polyole®ns 1 2 3 4 5
Introduction Direct Polymerization Post-polymerization Reactive Polyole®ns Polyole®n Block and Graft Copolymers
xi xiii 1 1 2 4 6 9 9 11 13 14 16
Direct Polymerization Approach
19
3 Direct Polymerization of B- and Si-containing Monomers
21
1 2 3 4 5
Introduction Borane-containing Monomers Stability of Borane Moiety to Transition Metal Catalyst Homopolymerization of Borane Monomers Copolymerization of Borane Monomers and -Ole®ns 5.1 Copolymerization of Ethylene and Borane Monomer 5.2 Copolymerization of Propylene, 1-Butene, and Borane Monomer 5.3 Copolymerization of 1-Octene and Borane Monomer
21 23 24 25 27 28 30 33
vi
Contents
6 Polymerization of Si-containing monomers 7 Summary 4 Direct Polymerization of O-, N-, and Cl-containing Monomers 1 2 3 4 5 6 7 8 9
Introduction Protection of Functional Groups Amino-group-containing Monomers Synthesis of Syndiotactic Polystyrene Containing Amino Groups Polymerization of OH-containing Monomers Polymerization of COOR-containing Monomers Polymerization of Halogen-containing Monomers Late Transition Metal Catalysis Summary
36 38 39 39 40 41 44 54 56 57 58 61
Post-polymerization Approach
63
5 Functionalization by Post-polymerization Processes
65
1 Introduction 2 Chemical Modi®cation via Free Radical Reaction Mechanism 3 Bulk Functionalization by Reactive Process 3.1 Maleic-anhydride-grafted PP Polymers (PP-MA and PP-SMA) 3.2 Maleic-anhydride-modi®ed PE (PE-MA) 3.3 Maleic-anhydride-modi®ed EP (EP-MA) 4 Surface Functionalization 4.1 Chemical Process 4.2 Radiation and Photo Processes 4.3 Corona and Plasma Processes 5 Summary
65 66 68 69 72 73 74 75 75 76 77
Reactive Copolymer Approach
79
6 Functionalization via Reactive Polyole®ns Containing Borane Groups
81
1 Introduction 2 Borane-containing Polyole®ns
81 83
Contents
3 Hydroxylation Reactions 3.1 Poly(hexen-6-ol) and Poly(octen-8-ol) 3.2 Polyethylene Containing OH Groups (PE-OH) 3.3 Polypropylene Containing OH Groups (PP-OH) 3.4 Poly(1-octene) Containing OH Groups (PO-OH) 3.5 EPDM Containing OH Groups (EP-OH) 4 Selective Auto-oxidation Reaction of Alkyl-9-BBN 5 Maleic-anhydride-modi®ed Polyole®ns 6 Applications 6.1 Adhesion Studies of PP/Al and PP/Glass Laminates 6.2 Hydrophilic PP Membranes 6.3 Immobilized Catalysts on Polyole®ns 7 Summary 7 Functionalization via Reactive Polyole®ns Containing p-Methylstyrene Groups 1 Introduction 2 Poly(ethylene-co-p-Methylstyrene) Copolymers 2.1 Copolymerization of Ethylene and p-Methylstyrene (p-MS) 2.2 p-Methylstyrene Reactivity Ratio 3 Poly(propylene-co-p-Methylstyrene) Copolymers 3.1 Metallocene Copolymerization 3.2 Ziegler±Natta Copolymerization 4 Polyole®n Elastomers Containing p-Methylstyrene Groups 4.1 Poly(Ethylene-ter-Propylene-ter-p-Methylstyrene) (EP-p-MS) 4.2 Poly(Ethylene-ter-1-Octene-ter-p-Methylstyrene) (EO-p-MS) 5 Functionalization of p-Methylstyrene Containing Polymers 5.1 Lithiation and Subsequent Transformation Reactions 5.2 Free Radical Maleic Anhydride Grafting Reaction 5.3 Halogenation Reactions 5.4 Oxidation Reaction 6 Summary 8 Functionalization via Reactive Polyole®ns Containing Unsaturated Groups 1 Introduction 2 Diene Monomers 3 Co- and Terpolymerization of -Ole®ns and Dienes 3.1 1,4-Hexadiene 3.2 5-Vinyl-2-norbornene
vii
83 84 85 86 89 89 91 93 95 96 98 102 103 105 105 107 107 110 111 111 113 114 115 116 119 119 125 128 132 133 135 135 136 137 137 139
viii
Contents
3.3 p-(3-Butenyl)styrene 3.4 Divinylbenzene 4 Functionalization of Unsaturated Polyole®ns 4.1 Hydroboration Reaction 4.2 Crosslinking Reactions 4.3 Maleation Reaction 4.4 Epoxidation Reaction 5 Summary 9 Synthesis of Polyole®ns with a Terminal Functional Group 1 Introduction 2 Living Polymerization Approach 3 Chain Transfer Reaction Approach 3.1 Zinc-alkyl-terminated Polyole®ns 3.2 Silyl-terminated Polyole®ns 3.3 Borane-terminated Polyole®ns 3.4 Styrene-derivative-terminated Polyole®ns 4 Chemical Modi®cation of Chain-end Unsaturated Polymer 4.1 Synthesis of Chain-end Unsaturated Polypropylene (u-PP) 4.2 Hydroboration and Oxidation 5 Summary
140 140 145 145 149 150 153 153 155 155 155 157 157 158 158 166 171 172 172 175
Polyole®n Block and Graft Copolymers
177
10 Synthesis of Functional Polyole®n Diblock Copolymers
179
1 Introduction 2 Living Ziegler±Natta (Metallocene) Polymerization 3 Transformation from Metallocene to Living Free Radical Polymerization 3.1 Polyethylene Diblock Copolymers 3.2 Polypropylene Diblock Copolymers 3.3 s-PS Diblock Copolymers 4 Transformation from Ziegler±Natta to Living Anionic Polymerization 5 Transformation from Living Anionic to Ziegler±Natta Polymerization 6 Coupling Reaction 7 Summary 11 Synthesis of Functional Polyole®n Graft Copolymers 1 Introduction 2 Living Radical Graft-from Reaction via Borane Reactive Sites
179 180 181 182 185 186 189 193 194 195 197 197 198
Contents
2.1 PP Graft Copolymers 2.2 EP Graft Copolymers 3 Living Anionic Graft-from Reaction via p-Methylstyrene Reactive Sites 3.1 PE-g-PS and PE-g-PMS Graft Copolymers 3.2 Synthesis of PE-g-PMMA and PE-g-PAN Graft Copolymers 3.3 PP-g-PB, PP-g-PS and PP-g-PMS Graft Copolymers 3.4 PP-g-PMMA and PP-g-PAN Graft Copolymers 4 Ring-Opening Graft-from Reaction 5 Graft Reactions via Divinylbenzene Sites 5.1 Anionic Reactions 5.2 Metallocene Reactions 6 Compatibilization of Polyole®n Blends 6.1 PP/PMMA Blends 6.2 PP/PC Blends 6.3 PE/PS Blends 7. Summary 12 New Maleic-anhydride-modi®ed and Long-chain-branched Polyole®ns 1 2 3 4 5 6
Introduction Comparison of Free Radical Grafting Processes New Maleic-anhydride-terminated Polyole®ns Applications in PP/Polyamide Reactive Blends Applications in Long-chain-branched Polyole®ns Summary
ix
199 201 203 204 208 209 213 213 216 217 219 220 220 222 222 225 227 227 228 230 235 240 245
References
247
Index
257
This Page Intentionally Left Blank
Preface Since the discovery of low-pressure high-density polyethylene and isotactic polypropylene about a half century ago, functionalization of polyole®ns has been a scienti®cally challenging and industrially important area. The constant interest, despite the lack of effective functionalization chemistry, is obviously due to the strong desire to improve polyole®ns' poor interactive properties. The hydrophobicity and low surface energy of polyole®ns has limited their applications, especially in the areas of coating, blends, and composites, in which adhesion, comparability, and paintability are paramount. In the recent decade of ®erce worldwide competition of single site transition metal coordination catalysis, across-board (from early to late transition metals) catalyst systems have been developed with powerful catalytic activities and well-controlled polymerization mechanisms. With designed ligands, the new catalyst systems can be tuned to produce polyole®ns with predictable microstructure (tacticity, molecular weight, etc.) and copolymer compositions. Many research activities were also geared toward the preparation of functional polyole®n copolymers. The objective has been the direct copolymerization of -ole®n with functional (polar) comonomers. Some successes have been achieved. My interest in functional polyole®ns dates back to 1984, when I had just completed my postdoctoral study and started my industrial career at Exxon, Corporate Research. With my research background in conducting polymers and familiarity in electrochemistry and electric properties of polymers, it was somewhat unusual to completely change my research area to polymer synthesis and to choose polyole®ns (commodity) as my major research subject. Maybe I was in the right place and at the right time to bene®t from some early exciting development of metallocene catalysis by Exxon Chemical Co. The potential of metallocene catalyst in the copolymerization reactions intrigued me, with the initial thought of searching for a facile ``reactive'' comonomer that could be effectively incorporated into polyole®ns. Then, the incorporated ``reactive'' groups can be selectively interconverted to polar functional groups, or preferably be transformed to initiators for graft-from polymerization of functional monomers. In other words, the chemistry leads to the preparation of a polyole®n graft polymer having a polyole®n backbone and several functional (polar) polymer side chains. My remote connection to Ziegler±Natta catalyst (from the synthesis of polyacetylene and its derivatives) was very helpful in
xii
Preface
formulating the idea of using organoborane as the reactive group. After the decadelong efforts of my postdoctoral and graduate students in my research group at Penn State University pursuing this idea of the ``reactive'' comonomer approach, which is coincidental with the continuous evolution of metallocene catalysis, we have developed three useful reactive comonomer systems: organoborane, p-methylstyrene, and divinylbenzene. In addition, the chemistry has been applied to ``reactive'' chain transfer agents to incorporate a reactive group at the polymer chain end and consequently to prepare polyole®n functional diblock copolymers. To my knowledge, there is no book speci®cally dedicated to the functionalization of polyole®n. Some book chapters describe individual chemical and physical modi®cations of speci®c polymers and their properties. With high interest in this research area and many new exciting experimental results available, it seems the right time to systematically summarize the experimental results into several basic chemical approaches and to discuss the scope and limitations (advantages and disadvantages) of each method. It will also be useful for some researchers to know the functional polyole®n materials available today. However, the major goals of this book are to provide a reference for researchers, and especially to inspire new researchers and stimulate their ideas in this important area. This book starts with the historical development of and future prospects for polyole®ns in Chapter 1. The functionalization chemistry, classi®ed into four general approaches, is introduced in Chapter 2. In the following chapters, I discuss each chemical approach with some experimental results to explain the chemistry. The ®rst approach of direct copolymerization of -ole®n and functional (polar) monomers is discussed in Chapters 3 and 4. The second functionalization approach via post-polymerization, employing the chemical reactions to the existing commercial polyole®n, is discussed in Chapter 5. The third ``reactive'' polyole®n approach is centered on the incorporation of ``reactive'' groups in polyole®n and interconversion of reactive groups to functional (polar) groups. Three reactive comonomers, namely organoborane, p-methylstyrene and dienes, are presented in Chapters 6, 7, and 8, respectively. Chapter 9 focuses on the results of incorporating a reactive group at the polymer chain end. The fourth approach involves the grafting reaction of functional polymer chains to polyole®ns, in other words, the preparation of functional polyole®n block and graft copolymers, presented in Chapters 10 and 11, respectively. Chapter 12 deals with the newly developed methods in the preparation of desirable maleic-anhydride-modi®ed polyole®ns and long-chain branched polyole®ns. This book by no means covers all the experimental results. Instead, it is intended to show the available functionalization approaches with the discussion of their scope and limitations, and some signi®cant experimental results (especially, some results from our laboratory that I know best). T.C. Mike Chung Professor of Polymer Science Pennsylvania State University University Park, PA 16802, USA February 2001
About the Author T.C. Mike Chung received his B.S. in Chemistry from Chung Yuan University (Taiwan) in 1976. He came to the United States for his graduate study in the Department of Chemistry, University of Pennsylvania, in 1979. After ®nishing his Ph.D work on conducting polymers (with Professor Alan G. MacDiarmid) in 1982, he spent two years as a research scientist at the Institute for Polymers and Organic Solids (with Professor Alan J. Heeger), University of California, Santa Barbara. Professors MacDiarmid, Heeger, and Shirakawa shared the 2000 Nobel Prize in Chemistry. Between 1984 and 1989, he was on the research staff in Corporate Research, Exxon Company. In 1989 he joined the faculty of the Pennsylvania State University as an associate professor and later became professor of polymer science in the Department of Materials Science and Engineering in 1993. His research interests are in synthetic polymer chemistry, with a current emphasis on synthesis of borane-containing polymers, functionalization and graft/block reactions of polyole®ns, polyole®n blends and composites, ferroelectric polymers, and high thermal stable boron/carbon ®ber. He is the author of more than 120 professional publications and holds 38 US patents.
This Page Intentionally Left Blank
1 Polyole®n Development and Prospects
1
INTRODUCTION
Polyole®ns ± including polyethylene (PE), polypropylene (PP), poly(1-butene), poly(1-octene), poly(4-methyl-1-pentene), ethylene±propylene elastomer (EPR), and ethylene±propylene±diene rubber (EPDM) ± are the most widely used commercial polymers, with over 53 billion pounds (24 million tons)1 US annual production (about 40% worldwide) in 1999, or close to 60% of the total polymer produced. By controlling crystallinity and molecular weight, polyole®ns with a wide range of thermal and mechanical properties have been produced for wide range of commercial applications.2 The products are presented in various forms, such as ®lms, sheets, ®bers, and even wax and viscous liquid. They are greatly in¯uencing our day-to-day life, and range from common items like bread bags, garbage bags, milk jugs, bottles, containers, hoses, outdoor±indoor carpets, tires, bumpers, and trims, to bullet-proof jackets. It is very interesting to note that the simplest molecular structure of linear polyethylene (HDPE) with ultra-high molecular weight (such as the commercial spectrum ®ber used in bullet-proof jackets) has near 100% crystallinity and melting temperature (Tm) ~140 C. Its mechanical strength is greater than any known existing materials,3 including steel, Kevlar, carbon ®ber, etc. On the other hand, the ethylene±propylene copolymers (EPR), containing about 40± 45 mol% of propylene, are completely amorphous elastomers with glass transition temperature (Tg) as low as 65 C.4 Polyole®ns are the most preferred choice among commercial polymers because of their excellent combination of chemical and physical properties along with low cost, superior processibility, and good recyclability. The low cost of ole®n monomers and extremely high polymer/catalyst turnovers position polyole®ns as the least expensive high-performance polymers. Most polyole®ns are routinely fabricated by all available methods ± including extrusion, injection molding, compression molding, injection blow molding, and calendering ± into ®lm, sheet, pipe, and ®ber products with various shapes and sizes. PE and PP are also known to be the most recyclable commercial polymers.5
2
Functionalization of Polyolefins
With its great commercial success, it is very interesting to look back at the evolution of this technology and to refresh the most important discoveries and breakthroughs during the past half century. It is even more important to think of possible future developments. What are the shortfalls of polyole®n materials that prevent their even wider usage in many areas currently occupied by other polymers that are much more expensive and less environmentally friendly materials? What are the needed technologies? How much has been done to address the technological challenges? 2
HISTORICAL ASPECTS: CATALYST-DRIVEN PROCESS
Polyole®n development dates back to the early 1930s with the production of lowdensity polyethylene (LDPE) by free radical initiators.6 The polymerization required elevated temperature and high-pressure reaction conditions to produce LDPE, and the polymer contains both long and short branches. As expected, the branched structure affects the polymer chain packing, and LDPE exhibits relatively low density, low crystallinity, and low melting point. Due to low glass transition temperature (Tg ~ 120 C) of polyethylene, the thermoplastic properties and application temperature are critically linked to its crystallinity and melting temperature. In the early 1950s, the historic discovery7 of Professor Karl Ziegler at the Max Planck Institute in Germany by using early transition metal catalyst opened up the new age in the preparation of polyole®ns, as well as the new frontier for the utility of transition metal coordination chemistry in catalytic chemical reactions. In 1953, the ®rst actual Ziegler polymerization, as we know it, used zirconium acetylacetonate and AlEt3 to form high-molecular-weight linear polyethylene. The process was soon improved by the use of TiCl4 and AlEt3. The transition metal coordination polymerization can take place at ambient temperature and low-pressure conditions. The polyethylene (HDPE) produced has a linear molecular structure with very low concentration of branches, high density, high crystallinity, and high melting point.8 In 1954, the second major breakthrough9 took place. Professor Giulio Natta, a consultant with Montecatini Co. under an agreement with Ziegler, realized that the Ziegler catalyst is capable of stereoregularly polymerizing -ole®ns, such as 1-propene, into a polymer with stereoregular repeating units. The most important stereospeci®c polymer structures, isotactic and syndiotactic, are shown in Fig. 1.1. Isotactic polypropylene (i-PP) leads to helical chains, which can ef®ciently pack together in crystals. The catalyst-controlled isotactic structure is responsible for the excellent thermal and mechanical properties of stereoregular polyole®ns. Ziegler and Natta were jointly awarded the Nobel Prize for Chemistry in 1963 for their roles in the start of the ``Golden Age'' of polymer science. Immediately after the initial discovery of Ziegler and Natta, many larger research groups were set up in the USA, UK, Japan, Italy, and Germany, then across the
Polyolefin Development and Prospects R
R
R
R
R
R
R
3
R Isotactic
R
R R
R
R
R R
R
Syndiotactic
Figure 1.1
world in the industrial and academic laboratories, to study Ziegler±Natta catalysis and the resulting polymers. The tremendous research efforts, with many thousands of papers and patents, were generating the discovery of catalysts with superior activity and stereospeci®city, as well as leading to economically viable production processes and product developments. The enormous combined success in both science and technology catapulted the industry forward by expanding new products and making possible the very inexpensive raw materials. In the late 1970s, Kaminsky and Sinn10,11 discovered a new class of Ziegler± Natta catalyst, based on metallocene/methylaluminoxane (MAO), to prepare polyole®n with high molecular weight and even higher activity than the commercially used Ziegler±Natta catalysts. This homogeneous catalyst system offers tremendous advantages in understanding the polymerization mechanism and allows the design of catalysts to prepare new polymers (especially copolymers) with well-de®ned molecular structures, i.e. narrow molecular weight and composition distributions. In the early 1980s, Brintzinger and coauthors12,13 reported the synthesis of ansatitanocene and -zirconocene with well-de®ned active centers. In 1984, Ewen14 synthesized both isotactic and atactic polypropylene (PP) by using a rac/meso mixture of ansa-ethylene(bisindenyl)titanium dichloride [Et(Ind)2TiCl2] with MAO as a catalyst, and correlated catalyst symmetry with polymer microstructure and reaction mechanism. In 1985, Kaminsky, Brintzinger, and coauthors15 reported highly isotactic PP by using chiral ansa-zirconocene/MAO catalyst. Since then, worldwide industrial and academic research in metallocene catalysis for ole®n polymerization has taken off. It has been growing so fast that after 10 years, this advanced technology has been brought to the commercialization stage. Polyole®n technology and industry have been changing the landscape by broadening the monomer pool and introducing new high-performance products and new applications. Several excellent recent review papers have been published to cover the whole story.16±19 The single-site catalyst technology was further expanded to late transition metals. In the past, the late transition metal system was expected to be less reactive due to the less available empty d-orbitals and also prone to side reactions to prevent the formation of high polymers.20±22 In the 1990s, Brookhart et al. reported the success of achieving high polymers with exceptionally high catalyst activity by some welldesigned Ni, Pd, Co, Fe catalysts,23±25 containing bulky diimine ligand (nonmetallocene type). A highly branched polymer structure was formed with the degree
4
Functionalization of Polyolefins
of branching controlled by reaction conditions. Because of the less oxophilic nature of late transition metal, the catalyst system is also more tolerant to functional groups. This catalyst, astonishing us with its power and scope of transition metal coordination polymerization, truly elevates polyole®n catalysis to another level and opens the door for new products and applications. Overall, the past half-century of development of polyole®n technology has been centered on the discovery of new catalyst systems, including all three components ± ligands, cocatalyst, and transition metal catalyst itself. The successful catalyst systems, having the unique combination of these three components, were also accompanied with new production processes, such as gas phase, suspension, and solution, for industrial-scale production. In general, the current advanced technology allows us to tailor-design catalysts and polyole®n polymers.
3
ZIEGLER±NATTA CATALYSTS AND POLYMERIZATION
It is very interesting to brie¯y overview the key components and steps in the coordination polymerization. In the broadest de®nition, Ziegler±Natta catalysts are a ``mixture of a metal alkyl of group I to III metal and a transition metal of group IV to VIII.''26 Both homogeneous and heterogeneous catalysts have been extensively investigated and shown useful in -ole®n polymerization. In the past, the heterogeneous systems offered better stereospeci®c addition for the preparation of isotactic structure. The soluble systems are generally used for polyethylene and EP copolymers where tacticity is irrelevant. However, the new metallocene-based homogeneous catalysts clearly demonstrate the stereospeci®c insertion largely dependent on the structure symmetry at the active site. In traditional heterogeneous Ziegler±Natta catalysts, the base metal component, most often an aluminum alkyl, serves as an alkylating agent for the transition metal salt. The most common one, TiCl3, is a crystal of alternating layers of Ti3 and Cl ions. At the edges, and in cracks or defects, the alkylation takes place on titanium atoms with an unsaturated coordination sphere. The exact polymerization mechanism had been controversial for a long time. One major stumbling block was that the ill-de®ned active sites26 coexisted and were simultaneously involved in the propagation step. The most reasonable theory states that the polymer growth takes place at a transition metal±carbon bond. Cossee27,28 was the ®rst to propose a mechanism in which the polymer chain grows by successive ole®n coordination and insertion. Cossee's mechanism was later proven by deuterium labeling experiments done by Grubbs.29 Figure 1.2 illustrates the polymerization mechanism for isotactic polymers. The reaction cycle starts with the coordination of -electrons in -ole®n to the vacant d-orbital of the octahedral Ti metal (I) to form metal±ole®n complex (II), before inserting into the Ti±C bond (III). The insertion regenerates the vacant d-orbital (IV) with a con®guration opposite that of the original (I). In isotactic
Polyolefin Development and Prospects
C X
X
5
CHR
M CH2
X X (I) C X X
M
C
Isospecific
CHR
polymerization
X
CH2
X CHR
M CH2
X X
X X
(II) CHR
C
(IV)
X X
M
X X
1,2-insertion
CH2 (III)
Figure 1.2
polymerization, with very limited opening at the active site, only one prochiral face of the monomer (with less steric hindrance primary carbon) is able to approach the Ti±C bond. The steric repulsion between the active site (ligand) and the bulky substitute (R) in the monomer regulates -ole®n inserting into the Ti±C bond in a 1,2-fashion. In addition, the steric hindrance around the active site (IV) also prevents immediate -ole®n coordination/insertion. Instead, the following propagation only continues after the polymer chain in the active site (IV) migrates back to its original site (I) with the regeneration of the original con®guration of the vacant dorbital. In summary, the combination of stereoregular coordination/insertion and facile polymer chain migration in each reaction cycle results in the isospeci®c addition reaction of -ole®n and assures the formation of isotactic polymer. The metallocene catalyst, having cyclopentadienyl (Cp) ligands, is a new class of homogeneous Ziegler±Natta catalyst with the unique characteristics of well-de®ned molecular structure and good solubility in hydrocarbon media. There are many experimental evidences indicating the cationic nature30 of the active species. In a CpnMX2/MAO system (M: Ti, Zr, etc., X: halide, n 1 or 2), the active species is CpnM(CH3), the cation is stabilized by a non-coordinating [MAO X] anion.31,32 In CpnMR2/borate system,33,34 the active species is CpnM R[B(C6F5)4] or CpnMR[B(C6F5)3R] . The weakly coordinated anion provides high catalyst activity of metallocene cation in the polymerization (coordination and insertion) of -ole®ns. In general, the single-site metallocene catalyst offers tremendous details of the transition metal coordination polymerization reaction. The relationship
6
Functionalization of Polyolefins
between the stereostructure (symmetry and chirality) of a catalytic site and the tacticity of the resulting polymer provides a strong foundation for designing reaction conditions and predicting the products. Based on stereostructure, metallocenes can be divided into ®ve categories. The achiral C2v-symmetric catalysts, such as Cp2MCl2 and [Me2Si(Cp)2]MCl2 and meso-Et(Ind)2ZrCl2, produce atactic PP.35 Chiral C2-symmetric metallocenes, such as rac-Et(Ind)2ZrCl2, produces isotactic PP.15 Cs-symmetric metallocenes, such as [Me2C(Cp)(Flu)]MCl2, produce syndiotactic PP.36,37 C1-symmetric metallocenes, such as [Me2C(Flu)(3-RCp)]ZrCl2, produce hemiisotactic PP,38,39 thermoplastic elastomeric PP (TPE-PP) containing both isotactic and atactic PP segments,40,41 or isotactic PP.42,43 Oscillating metallocene catalysts, such as (2-Ph-Ind)2ZrCl2/MAO, produce TPE-PP.44 Furthermore, in each catalyst system, by adjusting the position and bulkiness of the substituents on the Cp-rings, one can further control the catalyst activity, stereoregularity, and molecular weight of the polymers. There is some evidence that in the major termination reactions in most transition metal (especially metallocene) polymerization of -ole®ns are -H elimination45±47 and/or -H chain transfer with monomers.48,49 -CH3 elimination has also been observed in some catalysts for propylene polymerization.50 -H transfer reaction can be effectively suppressed by adjusting the substituents in the Cp-rings, in which case very high molecular weight PP has been achieved.47,51 On the other hand, these well-de®ned termination reactions could be utilized to produce PP with chain-end unsaturation, which is reactive for further modi®cation.52±55
4
FUTURE PROSPECTS: MATERIAL-DRIVEN PROCESS
The new catalysis technology has achieved most of its major objectives in the current industrial production of polyole®ns, with the catalyst activity reaching the kinetic limitation, and the polymer having well-designed molecular structure (i.e. composition, molecular weight, molecular weight distribution, tacticity). On the other hand, the world market for polyole®ns has steadily increased with an even higher increase in worldwide production capacity. Competition exists not only between the established producers in industrial countries, but also between them and newcomers in oil production and the rapidly developing Paci®c Rim countries. To maintain a competitive edge, the traditional approaches of increasing catalyst ef®ciency, improving production processes, and lowering operating costs will not be suf®cient for achieving long-term objectives. A successful polyole®n producer will have the technological advantages to provide new and better products that have the desirable properties for non-traditional polyole®n applications. The current strategy for increasing the applications of polyole®ns has been to transform them from commodity to semispecialty or specialty materials through new catalyst technology. Polyole®ns would be the choice of material (cost-effective, easily processable, and recyclable) for applications if the properties meet the
Polyolefin Development and Prospects
7
requirements. In fact, some new metallocene polyole®n polymers are gradually changing the landscape of polyole®n usage. For example, the new m-LLDPE, having a well-designed molecular structure with narrow molecular weight and composition distributions, is a potential replacement for the ¯exible PVC in ®lms, tubing, etc., and the ethylene/styrene copolymers are challenging the rigid PVC for ¯oor coverings. In addition, new syndiotactic polystyrene (s-PS) with a high melting temperature and polycyclole®ns with a high glass transition temperature are certainly the contenders of engineering plastic and high-value materials in microelectronic applications. The most effective areas for broadening polyole®n applications may be in polymer blends and composites, which represent the economic venues for expanding new products and enhancing product performances. Traditionally, these are dif®cult areas for polyole®n, which is known to be inadequate for applications where adhesion, compatibility, wettability, printability, or reactivity (i.e. functional chemical groups) are required. Instead, they are favored in applications requiring low cost, chemical and thermal stability, easy processing, and excellent mechanical properties. For example, polyole®n is most commonly used for ®lms and molded articles where a single polyole®n is required. The major stumbling block has been the poor adhesion and incompatibility between polyole®ns and other materials, such as pigments, paints, glass ®bers, metals, carbon black, and most polymers. Due to lack of chemical functionality and semicrystalline morphology (in PE, PP, etc.), polyole®n naturally exhibits low surface energy. If the aforementioned problems could be overcome, it would phenomenally expand the available market for polyole®n applications. Most of these dif®culties should be resolved by introducing polar functionality or by grafting suitable polar monomers to polyole®n, which could enhance interactions between polyole®n and other materials in polymer blends and composites. In fact, since the commercialization of HDPE and PP in the 1950s the chemical modi®cation of polyole®ns has been an area of intense interest in both academic and industrial communities. Unfortunately, the overall progress has been mixed, leading to only a few commercially available technologies and products. But recent advances in catalyst technology have also brought new light to this dif®cult area with many promising results. This book intends to capture some of the most important recent advances in the ®eld, with the emphasis on functionalization approaches and an understanding of their scope and limitations. Chapter 2 will outline the general functionalization approaches. Chapters 3 and 4 discuss the direct copolymerization reactions of -ole®ns and functional monomers. Chapter 5 summarizes the current commercial method of a post-polymerization approach involving chemical modi®cations of the preformed polyole®n. Recently, a new powerful approach of ``reactive'' polyole®ns has been emerging in the functionalization of polyole®ns, based on the combination of metallocene catalysis and several reactive comonomers. Chapters 6, 7, and 8 summarize the results of reactive polyole®n containing borane, p-methylstyrene, and dienes, respectively. The reactive polymer approach has been extended to
8
Functionalization of Polyolefins
prepared polymers with very desirable molecular structures, including polymers with a terminal functional group (Chapter 9), block copolymers (Chapter 10), and graft copolymers (Chapter 11). The last chapter (Chapter 12) will discuss new reactive polyole®n chemistry in the preparation of the most commercially important maleic anhydride polymers.
2 General Approaches in Functionalization of Polyole®ns
1
INTRODUCTION
As early as the late 1950s, Natta already realized the importance of introducing functional groups into polyole®ns. He and his co-workers reported some of the early work56,57 in the polymerization of heteroatom-containing monomers via the transition metal coordination process. Many research efforts,58±65 especially from industrial research laboratories, have been pursuing the technology with enthusiasm and skepticism. The need for improving the compatibility and adhesion of polyole®n in many application areas keep interest strong in the industry. However, the signi®cant efforts have met with very limited results, which dampens overall con®dence in developing industry-acceptable functionalization technology. Currently, only very few functionalization processes and products are available, and most commercial functionalized polyole®n products have ill-de®ned molecular structures. This chapter intends to provide a global view of the functionalization approaches and discuss their scope and limitations. Then, the following chapters will show detailed experimental results, including some exciting ones that are emerging very quickly and in my view may provide the solution to this long-standing scienti®c and technological problem. Theoretically, there are three possible approaches to the functionalization of polyole®ns, as illustrated in Scheme 2.1. They include (a) direct copolymerization of -ole®n with functional monomer, (b) chemical modi®cation of the preformed polymer, and (c) a reactive copolymer approach, incorporating reactive comonomers that can be selectively and effectively interconverted to functional groups. The ®rst two approaches are more obvious and naturally have enjoyed the most attention. In the past, they were referred to as direct and post-polymerization processes, respectively. The direct process could be an ideal one (involving an only one-step reaction) if the copolymerization reaction with functional monomers would be as effective and straightforward as the corresponding homopolymerization reaction. Unfortunately, some fundamental chemical dif®culties (which will be discussed in Section 2.2) have prevented serious consideration of the direct process for
10
Functionalization of Polyolefins
(a)
(b)
(c)
α-Olefin + Functional monomer
Preformed polyolefin
α-Olefin + Reactive comonomer
Copolymerization
Chemical modification
Copolymerization Interconversion
Functional polyolefin
Reactive polyolefin “intermediate”
Scheme 2.1
commercial application. So far, most of the commercial functionalization processes are based on post-polymerization. Chemical modi®cation of the preformed polyole®n homopolymers has been usually carried out in situ during the fabrication process to reduce the production cost, as well as to relieve the signi®cant concern (in many cases) of reducing processibility of polyole®n after functionalization reaction. However, the combination of the inert nature of polyole®n (requiring highly energetic reaction conditions) and a very short reaction time (during the processing) causes great dif®culties in controlling polymer composition and structure. As will be discussed in Section 2.3, the most frequently used free radical functionalization reactions are usually accompanied with many undesirable side reactions and byproducts. Overall, the current commercial process is far away from the ideal one, and the products are barely satisfactory for industrial needs. The third approach is a relatively new one (mostly developed in our laboratory). The basic idea is to circumvent the chemical dif®culties in both direct and postpolymerization processes by designing a reactive copolymer ``intermediate'' that can be effectively synthesized and subsequently interconverted to functional polymer. This approach has bene®ted greatly from metallocene technology, especially due to its superior capability in the copolymerization reactions. Several new reactive comonomers have been effectively incorporated into polyole®ns with narrow molecular weight and composition distributions, similar to those of commercial polyole®n copolymers, such as m-LLDPE and poly(ethylene-co-styrene). As will be discussed in Section 2.4, this approach has opened up the opportunity to prepare a broad range of new functional polyole®ns with compositions and structures that would be very dif®cult to prepare by other methods. It is very interesting to note that the reactive polyole®n approach has been broadened to prepare a polymer containing only a reactive group at the polymer chain end. The well-de®ned polymerization mechanism of metallocene catalysis leads to a precise control of the chain transfer reaction. With the design of a chain transfer agent containing a reactive group, the in situ chain transfer reaction
General Approaches in Functionalization of Polyolefins
11
produces a polymer having a terminal reactive group. In turn, this reactive end group opens up a lot of possibilities to produce new polyole®n products, including diblock copolymers and long-chain branching polymers (discussed in Section 2.5). The following sections will discuss the scope and limitations of each process. The chemical reasoning may provide a general perspective of each functionalization approach.
2
DIRECT POLYMERIZATION
As discussed in Chapter 1, the Ziegler±Natta and metallocene catalysts, using early transition metals, are the most important method in the preparation of polyole®ns. It is obvious that the most direct route to prepare functional polymers is the copolymerization of -ole®n with functional monomer containing a desired functional group. Unfortunately, this direct process is normally very dif®cult, because of catalyst poisoning and other side reactions.66,67 The Lewis acid components (Ti, Zr, Hf, V, and Al) of the catalyst will tend to complex with nonbonded electron pairs on N, O, and X (halides) of functional monomers in preference to complexation with the -electrons of the double bonds. The net result is the deactivation of the active polymerization sites by formation of stable complexes between catalysts and functional groups, thus inhibiting polymerization, as illustrated in Scheme 2.2.
CH2CH ⱍ (CH2)n ⱍ OH
COOH NH2
• Using non-oxophilic catalyst
Ti, Zr, Hf, V Catalyst
CH2CH ⱍ (CH2)n ⱍ OH Cat.
Logical approaches to prevent catalyst deactivation: • Blocking functional group
Catalyst deactivation Scheme 2.2
In addition to the catalyst poison, the solubility of functional groups in polymerization solution is a serious concern. Normally, only pure hydrocarbon solvents ± such as butane, hexane, toluene, or the monomer itself ± are acceptable in the Ziegler±Natta polymerization reaction, due to the catalyst sensitivity to the heteroatoms. The nonpolar solvents cause serious concerns in the polymerization
12
Functionalization of Polyolefins
involving polar monomers. The insoluble functional groups in the propagating polymer chain tend to associate with each other to form phase-separated domains or cause high solution viscosity. Both conditions limit monomer diffusion and reduce catalyst ef®ciency, polymer yield, and molecular weights. So far, most research activities related to these problems have been focused on the prevention of catalyst poisoning. The approach can be classi®ed into the three general routes shown below: (i) using functional monomers containing neutral or acidic heteroatoms; (ii) protecting sensitive functional groups from catalyst poisoning; (iii) employing catalysts that are less oxophilic and more stable to heteroatoms. The ®rst route is mostly limited to neutral silane57±60 and acidic borane68±70 groups. In fact, both are the most logical candidates, considering stability, solubility, and availability. Several silane- and borane-containing comonomers were studied in the homo-, co- and ter-polymerization with -ole®ns, such as ethylene, propylene, 1-butene, and 1-octene (the detailed experimental results will be discussed in Chapter 3). In general, the borane groups show excellent stability and solubility in transition metal polymerization reactions, and borane-containing monomers behave like a high -ole®n, such as 1-octene, in the copolymerization reactions. In turn, borane groups in polyole®n are very versatile, and can be used as the reactive site to prepare many new polyole®n structures (which will be discussed in Chapters 6, 10, 11, and 12). On the other hand, the silane-containing monomers show relatively low reactivity. In many reactions, low-yield and low-molecular-weight products were observed, which may indicate that some weak interaction still existed between silane and the active site. In addition, silane functional groups are not polar enough to greatly affect the polyole®n properties, and are also too stable for further chemical modi®cation. The most desirable functional groups are the polar ones, such as ±OH, ±COOH, ±NH2, and halides, which can provide strong interactions with most materials and substrates having polar surfaces. However, they also deactivate the catalysts, especially the early transition metals. The protection chemistry (second) route was usually adopted, with the strategy of steric71±76 or electronic77±83 protection of the polar functional group from contact with the catalyst. The detailed results will be discussed in Chapter 4. In general, both methods have their own concerns and limitations. In electronic protection, the formed acid±base complex in the protected functional group reduces the solubility of the propagating polymer chain, which in turn reduces polymer yield and molecular weight. On the other hand, in steric protection cases it is very important to choose the bulky protecting group that cannot only prevent catalyst poisoning, but also can be effectively deprotected. However, the expensive protection and deprotection reactions, with some environmental concerns due to by-products, prohibit any large-scale commercialization. In the third route, some very interesting results were recently reported by using less oxophilic late transition metal catalysts, such as Fe, Ni, Co, and Pd complexes.84±86 Some experimental results show that the copolymerization of -ole®ns
General Approaches in Functionalization of Polyolefins
13
with acrylate monomers can produce copolymers, however, with reduced catalyst activities and polymer molecular weight. The combination of a less oxophilic catalyst and an electronic-protected functional group signi®cantly increases catalyst activity,87 indicating that the lone pair electrons in the heteroatom (such as O and N) still compete with the ole®n insertion during the late transition metal polymerization. The polymers produced usually contain a branched molecular structure with relatively low (or no) melting temperature and crystallinity. There is no example of steric-speci®c polymerization of -ole®n by using late transition metal catalysts. It is very challenging to prepare functionalized i-PP or s-PS polymers by this route. Some detailed experimental results will be discussed in Chapter 4.
3
POST-POLYMERIZATION
Although chemical modi®cation of the preformed polymer seems a logical approach, the reactions in polyole®n homopolymers (PE, PP, EPR, etc.) are very dif®cult due to their inert nature with no obvious reaction site for most chemical modi®cation reactions. The only practical route87±93 is to activate the polymer by breaking some stable C±H bonds and forming free radicals along the polymer chain. The resulting polymeric radicals then undertake chemical reactions, namely addition and coupling, with some chemical reagents coexisting in the system. Scheme 2.3 illustrates the general reaction.
F*
R–O* -Radiation
HCF ⱍ HCH
Coupling H-
HCH Abstraction ⱍ HCH
HC * ⱍ HCH
Plasma
Addition CH2⫽CHF
HC(CH2CH) ⱍ ⱍ n HCH F
F: polar group Scheme 2.3
14
Functionalization of Polyolefins
Typically, a high-energy source (molecular radical, radiation, plasma, etc.), is used to abstract hydrogen atoms from the polymer. Since the stability of the carbon radical decreases in the order tertiary > secondary > primary, the susceptibility of hydrogen abstraction follows the same trend. Accordingly, among the three most investigated polyole®ns ± PE, PP, and ethylene±propylene copolymer (EP) ± PP is the most vulnerable to attack by free radicals, and PE is the least vulnerable, although chain rigidity and crystallinity also hinder the reactions to some extent. In addition to the functionalization reaction, there are many side reactions complicating the products. For example, due to the tertiary carbon radicals formed in PP polymer chain, -scission reaction can easily take place prior to the functionalization reaction. In fact, this degradation chemistry has been used to produce PP resins with controlled rheology (CR) during melt-extrusion.94 The other obvious side reaction is a coupling (crosslinking) reaction between polymeric radicals, which is particularly serious in PE modi®cation reactions. As expected, both degradation and crosslinking are signi®cant for EP copolymers. In addition, various radicals (coexisting with polymeric radicals) in the system produce a signi®cant amount of impurities. Tremendous efforts have been made to minimize the undesirable side reactions. The effects of the initiators, functional monomers, and reaction conditions have been extensively investigated. The detailed experimental results will be discussed in Chapter 5.
4
REACTIVE POLYOLEFINS
Since both direct and post-polymerization processes have only achieved very limited successes and have many drawbacks, it is very interesting to explore the alternative routes to facilitate the functionalization of polyole®ns and to synthesize functional polyole®ns with well-de®ned composition and molecular structure. In the past decade, our group has been focusing on a very powerful approach involving a reactive polyole®n as the ``intermediate'' for functionalization reactions. As illustrated in Scheme 2.4, the reactive monomer is copolymerized with -ole®ns by a Ziegler±Natta or metallocene catalyst. The formed polyole®n copolymer, containing reactive groups, then serves as an ``intermediate'' for the transformation to functional polyole®ns by various reaction mechanisms. The key factor in this approach is the design of a comonomer containing a reactive group that can simultaneously ful®ll the following requirements. First, the reactive group must be stable to metallocene catalysts and soluble in hydrocarbon polymerization media. Second, the reactive monomer should have good copolymerization reactivity with -ole®ns. Third, the reactive group must be facile in the subsequent interconversion reaction to form polar groups under mild reaction conditions. It is particularly effective if the reactive group can serve as an initiator (with suf®cient stability) of the polymerization of functional monomers. In other words, each incorporated reactive group can produce a functional polymer
General Approaches in Functionalization of Polyolefins
α-Olefin + "Reactive" comonomer
Key demands of reactive comonomer: • No catalyst deactivation
Metallocene catalyst
• Effective comonomer incorporation • Versatile transformation reactions • "Living" graft-from polymerization reaction
Polyolefin copolymer containing reactive side groups Graft-from reaction
Functionalization reaction
Polyolefin containing functional side groups
15
Graft copolymer Scheme 2.4
chain containing hundreds of functional groups. With the metallocene technology, the choice of reactive comonomers has dramatically increased. As will be discussed in Chapters 6, 7, and 8, three reactive comonomers ± including borane monomers,68±70,95±106 p-methylstyrene,107±110 and divinylbenzene111 ± have been incorporated to a broad compositional range of polyole®n copolymers with narrow molecular weight and composition distributions. The resulting three type reactive sites, i.e. borane, benzylic protons, and styrene units, pendent along the polyole®n backbone are illustrated in Scheme 2.5.
Polyolefin backbone B R R
CH3
CHCH2 Scheme 2.5
All three reactive sites are very versatile in the subsequent transformation reactions, both in functionalization and graft reactions.112±117 The transformation reactions selectively take place at the reactive sites. In other words, the concentration of functional groups are basically proportional to the concentration of the reactive sites, and the resulting functional polyole®n has well-de®ned molecular structure. As will be discussed in the later chapters, three reactive comonomers provide complementary coverage of the most desirable functional polyole®n compositions and structures. It is very interesting to note that the reactive polyole®n approach has also been extended to the preparation of polyole®n containing a reactive terminal group and polyole®n diblock copolymers. One example is illustrated in Scheme 2.6.
16
Functionalization of Polyolefins
Chain-end unsaturated polyolefin Hydroboration reaction
-Olefins Metallocene catalyst
H-BR2
H-BR2 (CT agent)
Polyolefin homopolymer containing a borane end group graft-from reaction
Functionalization reaction
Polyolefin containing a functional end group
Diblock copolymer
Scheme 2.6
Two reaction mechanisms have been developed for the preparation of boraneterminated polyole®ns, including hydroboration118±120 of terminal-unsaturated polymer and in situ chain transfer reaction121±122 with a dialkylborane reagent during the metallocene polymerization. As will be discussed in Chapter 9, the chain transfer reaction with borane reagent provides a general method to prepare boraneterminated polyole®ns, including all commercial polymers (PE, PP, s-PS, and their co- and terpolymers). In turn, the terminal borane group not only can be converted to various functional groups, but also can be transformed to a free radical initiator for a chain extension reaction to form diblock copolymers. The overall reaction process resembles a transformation from metallocene polymerization to living free radical polymerization via a borane terminal group. A broad range of diblock copolymers, including the combination of a metallocene-prepared segment and a free-radical prepared segment, are now available with relatively well-de®ned molecular structure.
5
POLYOLEFIN BLOCK AND GRAFT COPOLYMERS
Functional polyole®ns with block and graft structures (illustrated in Scheme 2.7) are the most desirable materials, but also the most dif®cult to prepare. Both molecular structures not only offer a large quantity of functional groups, but also preserve the original polyole®n properties, such as crystallinity, melting point, and hydrophobicity. As the established technique for improving the interfacial interaction, block and graft copolymers are the ideal compatibilizers that can dramatically
General Approaches in Functionalization of Polyolefins
Diblock copolymer Polyolefin
Advantages: ⱍ ⱍ ⱍ ⱍ ⱍ ⱍ ⱍ FF FFFFF
–F –F –F –F –F –F
• Providing high concentration of functional groups • Preserving the properties of polyolefins
Graft copolymer Polyolefin
17
• Economical process –F –F –F –F –F –F –F Scheme 2.7
increase the adhesion of polyole®n. In polyole®n blends and composites, the graft/ block copolymers are usually located at the interfaces, the functional polymer segments provide good adhesion to the polar surfaces, and the polyole®n segments can interpenetrate (with entanglements or/and cocrystallization) into pure polyole®n homopolymer domains. In terms of the feasibility of mass-producing polyole®n block and graft copolymers, only a very few linkages (chemical bonds) exist for connecting polymer segments. In other words, the ideal chemical process only requires a few active sites in the polyole®n chain. With the appropriate choice of reactive comonomer or chain transfer agent, it is very possible to produce both block and graft copolymers in an economical fashion.
This Page Intentionally Left Blank
Direct Polymerization Approach
This Page Intentionally Left Blank
3 Direct Polymerization of B- and Si-containing Monomers
1
INTRODUCTION
It would be an ideal case if we could prepare functionalized polyole®ns by a direct polymerization process. In one step, by mixing catalyst with the -ole®nic monomer containing the functional group, the desirable functional polyole®n could be synthesized. Such a process certainly would be the winner of this technology race. However, there are only two functional groups, including borane and silane, with signi®cant experimental evidence to show direct polymerization of functional monomers to polymer by early transition metal catalysis. Most research on silanecontaining polyole®n was done in the late 1950s and early 1960s with a very limited choice of available catalysts (only the ®rst generation of Ziegler±Natta catalysts). In fact, only a few short papers were published with insuf®cient analytical data. On the other hand, the borane case was studied by our laboratory in the late 1980s and early 1990s with a great availability of catalysts, including all the generations of Ziegler± Natta and metallocene catalysts, and the resulting polymers were analyzed in great detail. Before discussing the experimental results, it is very interesting to understand the unique nature of the elements B and Si ± especially their relationship to carbon and transition metal catalysts in the Periodic Table, shown in Fig. 3.1. Line A in the Periodic Table separates the metallic region (left) and non-metallic region (right). Most of the elements in the non-metallic region are used in organic functional groups. On the other hand, line B separates the electron-de®cient elements (Lewis acids) in the left side and electron-rich elements (Lewis bases) on the right side. It is interesting to note that B is the only element that is located in the non-metallic region (right side of line A) and is an electron-de®cient Lewis acid (left side of line B). The Lewis acid nature of borane offers itself a very good chance to coexist with transition metals (Lewis acids). In addition, the size of the boron atom is relatively small, and steric protection can be effectively applied if needed. On the other hand, Si is the only element in the non-metallic region that is electronically neutral to C.
Noble VIIBA Gases
A IA
H 1.0079
3 2
A IIA
4 Li Be 6.941
3
A
•
1 1
24.305
B
B 10.81
B IIIA
B IVA
B VA
B VIA
B VIIA
IBB
VIII
IIBB
IVBA
6
5
• 2
1
IIIBA
9.01218
11 12 Na Mg 22.98977
Sub-group Group Sub-group
VBA
VIBA
•
H
He
Hydride
4.00260
7 8 9 • 10 • Ne C N O F
12.011
14.0067
15.9994 18.998403
20.179
13 14 15 16 17 • 18 • Al Ar Si P S Cl 26.98154
28.0855
30.97376
32.06
35.453
39.948
4
19 20 21 22 23 24 25 䊐 26 27 28 29 30 31 • 32 33 34 35 • 36 • K Fe Ca Sc Ti V Cr Mn Co Ni Cu Zn Ga Ge As Se Br Kr
5
37 38 39 40 41 42 43 䊐 44 45 46 47 48 • 49 50 51 52 53 54 • Rb Sr Y Zr Nb Mo Tc Ru Rh Pd Ag Cd In Sn Sb Te I Xe
6
55 • 56 57 * 72 73 74 75 76 77 78 79 80 • 81 82 83 84 85 86 • Cs Ba La Hf Ta W Re Os Ir Pt Au Hg Tl Pb Bi Po At Rn
39.0983
85.4678
132.9054
7
40.08
87.62
137.33
44.9559
88.9059
138.9055
47.90
91.22
178.49
50.9415
92.9064
180.9479
51.996
95.94
183.85
87 • 88 89 * 104 䊐 105 䊐 106 Fr Ha Ra Ac * Ku (223)
226.0254 227.0278
(261)
(262)
54.9380
(97)
186.207
䊐
55.847
101.07
190.2
58.9332
102.9055
192.22
58.70
106.4
195.09
63.546
107.868
196.9665
65.38
112.41
200.59
69.72
114.82
204.37
118.69
207.2
Acid Base B
(263)
Figure 3.1
72.59
The Periodic Table (based on carbon-12).
74.9216
121.75
208.9804
78.96
127.60
(209)
79.904
126.9045
(210)
83.80
131.30
(222)
Metal Non-metal A
Direct Polymerization of B- and Si-containing Monomers
23
Logically, an -ole®n containing a borane or silane group has a very good chance to be incorporated into the polyole®n using transition metal catalysis. In addition, B and Si are the two elements most close to C with a mainly covalent characteristic of C±B and C±Si bonds. The monomers and polymers containing borane and silane moieties will behave like the corresponding hydrocarbon monomers and polymers with similar solubility and solution viscosity. Therefore, the same reaction conditions and processes for the polymerization of -ole®ns can be directly applied in both cases. The resulting borane- and silane-containing polyole®ns have similar polymer microstructures, i.e. molecular weight, molecular weight distribution, tacticity, etc., to the corresponding polyole®ns. In this chapter, we will focus on the polymerization behaviors of both functional groups using early transition metal (Ziegler±Natta and metallocene) catalysts. The applications of these reactive polyole®ns, especially the borane-containing polymers, will be discussed later. Since only very limited information is available in the case of silane, the detailed discussion will mostly focus on borane-containing polymers.
2
BORANE-CONTAINING MONOMERS
The simplest and most effective route to preparing borane-containing monomers, !-borane--ole®ns, is monohydroboration reaction123,124 of nonconjugated ,!dienes (1,7-octadiene, 1,5-hexadiene, etc.), with a dialkylborane, such as 9borabicyclononane (9-BBN). The major concern in this reaction is the uncontrollable formation of the dihydroborated product as illustrated in Eq. (3.1):
CH2⫽CH–(CH2)n–CH⫽CH2 (excess) Recovered by distillation and reused
H–B
–B
⫽ –B
Unreacted dienes + CH2⫽CH–(CH2)n–CH2–CH2–B
Distillation
Borane-containing monomer
+ B–CH2–CH2–(CH2)n –CH2–CH2–B
(trace) (3.1)
To optimize the reaction condition, a multiple-fold excess of diene was used neat by simply mixing diene with 9-BBN reagent. After hydroboration reaction at
24
Functionalization of Polyolefins
ambient temperature for about one hour, the unreacted diene was recovered by distillation and reused in the next monomer preparation. The borane-containing monomer was usually distilled from residue to assure the purity for polymerization reactions. 11B NMR spectra show a distinctive chemical shift at about 88.2 ppm (vs BF3), indicating a trialkylborane moiety in the monomer. 3 STABILITY OF BORANE MOIETY TO TRANSITION METAL CATALYST The feasibility of polymerizing B-containing monomers by Ziegler±Natta and metallocene catalysts can be deduced from the stability of such catalysts in organoborane solutions.70 Several reactions were used to examine this stability. One of them was homopolymerization of 1-octene by a Ziegler±Natta catalyst with and without the presence of an equimolar amount of Et3B. Gel permeation chromatography (GPC) results in Fig. 3.2 show the absence of any retarding effect on the molecular weight due to the presence of trialkylborane. In addition, Et3B was recovered after polymerization by a distillation process. The same 11B NMR chemical shift (88 ppm vs BF3) was observed in both the original and recovered materials.125 All results imply that no signi®cant reaction takes place in mixing organoborane and the early transition metal catalyst, which suggests that boranesubstituted -ole®ns may be polymerized by Ziegler±Natta and metallocene catalysts to high polymers with desirable steric and regio microstructures.
100
Percent
80
60 (a) (b)
40
(c) 20
0 10
15
20 Retention time (minutes)
25
30
Figure 3.2 GPC curves of poly(1-octene) with (a) 83%, (b) 56%, and (c) 23% conversion; (a) and (c) represent the samples prepared in the presence of Et3B, (b) is for the sample prepared in the absence of borane. (Redrawn from Macromolecules 1990, 23, 378. Copyright 1990 Am. Chem. Soc.)
25
Direct Polymerization of B- and Si-containing Monomers
4
HOMOPOLYMERIZATION OF BORANE MONOMERS65,68
On the basis of stability tests between trialkylborane and early transition metal catalysts, the polymerizability of a series of borane monomers, -ole®n containing !-borane groups, was examined under the same reaction conditions used in regular -ole®n polymerizations. Various Ziegler±Natta and metallocene catalysts, including both homogeneous and heterogeneous catalysts, have been used in this polymerization reaction, as shown in Eq. (3.2):
x CH2CH (CH2)n
Transition metal catalyst
CH2 CH
x
(CH2)n
B R
R, R⬘ are alkyl groups, except H and CH3
R⬘
n >3
(3.2)
B R
R⬘
In most cases, highly viscous polymer solutions were observed in a half hour, indicating a high-molecular-weight polymer and high product yield in a relatively short reaction time. Table 3.1 summarizes the experimental results of several borane monomers (B-7-octenyl-9-BBN, B-5-hexenyl-9-BBN, and B-4-pentenyl-9-BBN) that were polymerized by using isotactic TiCl3 AA/EtAlCl2 and atactic Ti(OR)mCln/i-Bu3Al catalysts. Both B-7-octenyl-9-BBN and B-5-hexenyl-9-BBN show nearly quantitative conversions by using both catalyst systems in 2 h at room temperature, and the molecular weights of the resulting polymers are above one million by GPC measurement (with standard polystyrene calibration curve). The results are very similar to those of high -ole®ns (such as 1-octene) under the same reaction conditions. It is very interesting to note that much weaker polymerization results in the B-4-pentenyl-9-BBN case. Table 3.1
Summary of Ziegler±Natta polymerization of borane monomers Reaction conditions
1
2
Borane polymer
Catalyst / (mmol)
Monomer / (g)
Solvent (ml)
Temp./time ( C)/(h)
Yield/conv. (g)/(%)
Tacticity
(I)/0.5 (I)/0.5 (I)/0.5 (I)/0.5 (I)/0.5 (II)/0.5 (II)/0.5 (I)/0.5
O-BBN/8 O-BBN/8 O-BBN/8 H-BBN/5 H-BBN/5 H-BBN/5 H-BBN/5 P-BBN/6.4
80 80 80 50 50 50 50 50
25/1 25/2 45/1 25/2 25/3 25/1 25/2 25/12
6.0/80 7.3/92 6.8/85 4.5/90 4.7/94 3.5/70 4.1/82 3.0/47
Isotactic Isotactic Isotactic Isotactic Isotactic Atactic Atactic
1
(I): TiCl3 AA/EtAlCl2 1/6, Cat (II) Ti(OR)mCln/i-Bu3Al 1/6. O-BBN: B-7-octenyl-9-BBN; H-BBN: B-5-hexenyl-9-BBN; P-BBN: B-4-pentenyl-9-BBN.
2
26
Functionalization of Polyolefins
During the polymerization, the solution viscosity did not display a noticeable change. Less than 50% monomer conversion was observed in 12 h. The dramatic decrease in reactivity indicated that a spacer with more than three methylene units is necessary to insulate the double bond from the borane moiety. Since the boron is electron de®cient, the electron-withdrawing effect of the borane moiety (I) would decrease the -electron density of the double bond126,127 if it is too close to the borane moiety. Acid±base interaction (II) between borane and ole®n might also be possible if a favorable cyclic structure could be formed as illustrated in Fig. 3.3. In addition, the steric effect of the bulky borane moiety may also hinder the insertion of monomer. All resulting borane-containing polymers are soluble in regular organic solvents, such as toluene and THF. As shown in Fig. 3.4, the 11B NMR spectrum of poly(B-5-hexenyl-9-BBN) which displays a single chemical shift at 88.2 ppm, is the same as that of B-5-hexenyl-9-BBN. CH2
R⬘
(I)
B R
CH2
R
CH
R⬘
B
CH
CH2
CH2
(II)
CH2 Figure 3.3
Two possible deactivation modes.
(a)
(b)
200
150
100 ppm
50
0 BF3⋅OEt2
Figure 3.4 11B NMR spectra comparison between (a) 7-octenyl9-BBN and (b) poly(7-octenyl-9-BBN). (Redrawn from Macromolecules 1988, 21, 865. Copyright 1988 Am. Chem. Soc.)
Direct Polymerization of B- and Si-containing Monomers
27
The coexistence of the same trialkylborane in both monomer and polymer is crucial, and strongly supports two essential facts: the stability and the polymerizability of an alkenylborane with Ziegler±Natta catalysts. The same results were shown in IR spectra (Fig. 3.5) of B-5-hexenyl-9-BBN and poly(B-5-hexenyl-9BBN). After polymerization, all absorption peaks at 3070, 1637, and 905 cm 1, corresponding to the vinyl group, are gone, and the 9-BBN ®ngerprint ranging between 1500 and 950 cm 1 is almost identical. Both poly(B-5-hexenyl-9-BBN) and poly(B-7-octenyl-9-BBN), prepared by isotactic TiCl3 AA/EtAlCl2, are isotactic polymers. On the other hand, the atactic catalyst Ti(OR)mCln/i-Bu3Al results in atactic polymers. Apparently, the borane group at the end of the -ole®n does not interfere with the stereo-speci®c coordination/insertion process, and the tacticity of the polymer was basically governed by the catalyst itself, as in pure high -ole®n cases. The polymer tacticity was examined by using the corresponding hydroxy polymers that are air-stable and convenient to handle in X-ray measurement.
5
COPOLYMERIZATION OF BORANE MONOMERS AND -OLEFINS
Copolymerization is the most effective way to modify polymer. A small percentage of functional groups incorporated into polyole®ns (such as PE, PP, and EP rubber) 2.5
3.0
4.0
5.0
6.0
7.0
8.0 9.0 10
12 14 16
(a)
(b)
4000
3500
3000
2500
2000 1800 1600 1400 1200 1000 800
600
Wavenumber (cm–1)
Figure 3.5 IR spectra comparison between (a) 7-octenyl-9-BBN and (b) poly(7-octenyl-9-BBN). (Redrawn from Macromolecules 1988, 21, 865. Copyright 1988 Am. Chem. Soc.)
28
Functionalization of Polyolefins
can have a profound effect on their physical properties, adhesion, paintability, compatibility, etc. The key question is how to prepare functionalized PE, PP, and EP copolymers with controllable molecular structure, i.e. desirable functional group, concentration and distribution, molecular weight, crystallinity, etc. From the results of stability and homopolymerization studies, borane monomer with a spacer greater than three methylene units between ole®n and borane behaves similarly to the corresponding high -ole®ns. Ideally, we could treat the borane monomer as a high -ole®n in the copolymerization reaction and expect a similar copolymer structure that is governed by the reactivity ratios of two comonomers under each set of the catalyst system. In the Ziegler±Natta heterogeneous catalyst system, the reactivity of monomer is dramatically dependent on the size of the monomer ± the bigger the size of the monomer, the lower the reactivity. Ethylene is generally ®ve times more reactive than propylene, and the reactivity becomes more constant in high -ole®ns. It could be predicted that the copolymerization of ethylene with borane monomer would be very dif®cult in a Ziegler±Natta heterogeneous catalyst system. The borane monomer may have a better chance to be incorporated with propylene and high -ole®ns, such as 1-butene and 1-octene. On the other hand, the use of a homogeneous metallocene catalyst with an open active site could signi®cantly improve the copolymerization ability of borane monomers with ethylene. In this section, all experimental results of copolymerization reactions between borane monomer and -ole®ns, including ethylene, propylene, 1-butene, and 1-octene, will be discussed. The discussion will focus on the reaction conditions, i.e. catalyst system and reaction process, to achieve an individual copolymer with controllable molecular structure. Due to the air-sensitivity of the borane group, most of the analyses were carried out on the corresponding hydroxy or ester polymers after converting borane to hydroxy or ester groups. 5.1
Copolymerization of Ethylene and Borane Monomer128
The copolymerization reaction between ethylene and 5-hexenyl-9-BBN is the best testing ground for studying the copolymerization ability of borane monomer, especially in terms of its dependence on the Ziegler±Natta and metallocene catalysts. Equation (3.3) shows the copolymerization of ethylene and a borane monomer, such as 5-hexenyl-9-BBN, by various catalyst systems. The reactions were carried out in a Parr reactor under N2 atmosphere. Three catalyst systems, including two homogeneous metallocene [Et(Ind)2ZrCl2 and Cp2ZrCl2] with MAO catalysts and one heterogeneous TiCl3 AA/(Et)2AlCl catalyst, were evaluated in the copolymerization reactions. Usually, the reaction was initiated by charging the catalyst solution into the mixture of ethylene and 5-hexenyl-9-BBN, and a constant ethylene pressure was maintained throughout the polymerization process. Almost immediately, white precipitate was observed at the beginning of the reaction. After a speci®c reaction time, the copolymerization was terminated by addition of IPA.
29
Direct Polymerization of B- and Si-containing Monomers
x CH2CH2 + y CH2CH (CH2)4 B
Catalysts: Et(Ind)2ZrCl2/MAO Cp2ZrCl2/MAO TiCl3/EtAlCl2
Catalyst*
3:3
(CH2CH2)x(CH2CH)y (CH2)4 B
Table 3.2 summarizes the experimental results. Overall, the homogeneous zircorocene/MAO catalysts, especially Et(Ind)2ZrCl2/MAO with stained ligand geometry and an open active site, show satisfactory copolymerization results at ambient temperature. Comparing runs A-1 to A-4, the concentration of borane groups in polyethylene is basically proportional to the concentration of borane monomer feed. About 50±60% of borane monomers was incorporated into the PE copolymers after about a half hour. It was very unexpected that the catalyst activity systematically increased with the concentration of borane monomer in the Et(Ind)2ZrCl2/MAO catalyst system. Obviously, no retardation due to the borane groups is shown in these cases. The copolymerization of borane monomers in the Cp2ZrCl2/MAO system (shown in run B-2) is signi®cantly more dif®cult ± only 1.22 mol% of borane monomer incorporated in PE copolymer, even though a high concentration of borane monomer was used. On the other hand, the heterogeneous TiCl3 AA/ Et2AlCl catalyst shows no detectable amount of borane group in the copolymer, as shown in run C-1. Table 3.2 Summary of copolymerization reactions between ethylene (m1) and 5hexenyl-9-BBN (m2) Run Catalyst Comonomers Reaction Temp./time Catalyst activity Borane in no. type1 m1/m2 (psi)2/(g) ( C)/(min) (kg)/(mol hr) copolymer (mol%) A-1 A-2 A-3 A-4 B-1 B-2 C-1 1
I I I I II II III
45/0 45/0.56 45/1.52 45/2.10 45/0 45/5 80/10
30/60 30/30 30/30 30/30 30/70 30/70 60/110
644 1469 2020 2602 337 643 4.0
0 1.25 2.15 2.30 0 1.22 0
Catalysts: Et(Ind)2ZrCl2/MAO (I), Cp2ZrCl2/MAO (II), and TiCl3 AA/Et2AlCl (III); solvent: 100 ml toluene. 2 Corresponding to 0.38 mol/l.
30
Functionalization of Polyolefins
118 116
mV
114 112 110
(a)
108 (b) 106 104 102 34
36
38
40
42
44
46 48 Minutes
50
52
54
56
58
60
Figure 3.6 GPC curves of poly(ethylene-co-5-hexenyl-9-BBN) copolymers containing (a) 0.5 and (b) 1.2 mol% of 5-hexenyl-9-BBN.
Figure 3.6 shows the GPC curves of poly(ethylene-co-5-hexenyl-9-BBN) copolymers containing 0.5 and 1.2 mol% of 5-hexenyl-9-BBN, respectively. The copolymers were prepared by the Et(Ind)2ZrCl2/MAO catalyst. Overall, the copolymers have high molecular weight and narrow molecular weight distribution (Mw/Mn < 3). There is no indication of any negative in¯uence of borane group on the metallocene polymerization. 5.2 Copolymerization of Propylene, 1-Butene, and Borane Monomer129,130 The copolymerization between propylene and borane monomer (5-hexenyl-9-BBN) was carried out by using an isospeci®c heterogeneous Ziegler±Natta catalyst, i.e. TiCl3 AA and Al(Et)2Cl, under inert atmosphere. The polymerization was started by the addition of the catalyst mixture, after aging for a half hour, to a solution of the two monomers in toluene. Almost immediately, white precipitate could be seen in the deep purple solution, which is due to its crystalline structure with high isotactic propylene content. The copolymerization was terminated after a certain reaction time by the addition of isopropanol to destroy the active metal species. Excess
Direct Polymerization of B- and Si-containing Monomers
31
isopropanol was used to ensure the complete coagulation of the polymer from the solution. These borane-containing polymers were isolated from solution by simple ®ltration and then washed repeatedly with isopropanol. The borane concentration of poly(propylene-co-5-hexenyl-9-BBN) was measured by the solution 11B NMR technique, as shown in Fig. 3.7. By using a known amount of triethylborate (chemical shift at 19 ppm) as a reference and comparing the integrated peak areas, the borane content in the copolymer can be quantitatively determined. In a batch copolymerization, by starting with 1/1 propylene/5-hexenyl-9-BBN monomer mole ratio, the initial incorporation of borane monomer in the polypropylene copolymer
(a)
(b)
(c)
115 105 95
85
75
65
55 45 (ppm)
35
25
15
5
–5
Figure 3.7 11B NMR spectra of three poly(propylene-cohexenyl-9-BBN) copolymers containing (a) 1.5 (b) 3.5, and (c) 6.3 mol% borane groups. (Redrawn from Macromolecules 1991, 24, 970. Copyright 1991 Am. Chem. Soc.)
32
Functionalization of Polyolefins
was very small ± only 1.5 mol% in the ®rst 0.1 h, increasing to 3.5 and 6.3 mol% after 1 and 2 h respectively. It is obvious that propylene was preferentially polymerized and consumed. This has an effect on the comonomer ratio over time (decreasing percentage of propylene and increasing percentage 5-hexenyl-9-BBN). The constant change in the monomer feed ratio results in a broad copolymer composition. A sample (PP-OH-A) was obtained when the monomer conversion was nearly complete (16 h) using a 1/1 monomer feed ratio followed by oxidation to convert borane to hydroxy groups, and subjected to fractionation by sequential Soxhlet extractions using various solvents. The solvents were chosen so as to separate by polarity (OH content) and crystallinity (isotacticity and PP sequence length). The sample was extracted with methanol, 2-butanone (MEK), heptane, and xylene. Table 3.3 shows the fractionation results of PP-OH-A, compared to those of sample PP-OH-B prepared by continuous reaction (discussed later). PP-OH copolymers with above 60% alcohol content are soluble in MeOH. The MEK fraction was a rubbery, tacky material indicative of low isotacticity. Due to the low boiling point of heptane, its fraction represents a polymer with intermediate tacticity or with more (and/or random distribution) hexenol units in it that reduce crystal formation. Xylene should dissolve all the remaining highly isotactic polymer. However, a small portion (7.7%) of polymer was insoluble after this extraction process. Overall, the results of a high conversion batch reaction show a broad range of composition distribution, and microstructure in PP-OH-A copolymer. The better way to investigate a copolymerization is to measure the reactivity ratio of the comonomers. The reactivity ratios between -ole®n (r1 k11/k12) and 5-hexenyl-9-BBN (r2 k22/k21) are estimated by the Kelen±TuÈdos method. We obtain r1 70.476, r2 0.028, and r1 r2 1.973 for propylene/5-hexenyl-9-BBN. It is clear that both copolymerization reactions are not ideal cases. The values of r1 r2 are far from unity, and the reactions are favorable for propylene incorporation. In the batch reaction, using isospeci®c Ziegler±Natta catalyst and a ®xed monomer ratio of propylene/5-hexenyl-9-BBN, either a broad distribution of copolymers was obtained using long reaction times or a narrow compositional distribution was obtained using short reaction times, but in extremely low yield. Table 3.3 Fractionation results of hydroxylated polypropylene (PP-OH) hot solvent soluble fractions (wt%) Sample 1
PP-OH-A PP-OH-B2
Methanol
MEK
Heptane
Xylene
Insoluble
42.6 None
18.9 8.5
7.6 14.2
23.1 77.7
7.7 None
1 Sample (PP-OH-A) was prepared by batch reaction with 1/1 propylene/borane monomer mole ratio. 2 Sample (PP-OH-B) was prepared by continuous reaction with 10/1 propylene/borane monomer mole ratio.
Direct Polymerization of B- and Si-containing Monomers
33
Table 3.4 Summary of copolymerization of -olefin and borane monomer by continuous reaction Polymer
Mol% B1 in feed
Mol% OH in polymer
Reaction time
Yield (%)
2
Mv (g/mol)
PP PP-OH PP-OH
0 10 13
0 3 5
2 3 5
93 62 35
2.07 1.78 1.71
230 000 183 000 174 000
1
B 5-hexenyl-9-BBN. intrinsic viscosity.
2
It is feasible to obtain a more uniform copolymer composition by an engineering approach, such as the control of monomer feed ratio during the copolymerization. In other words, the more reactive -ole®n monomer was added gradually in order to keep its concentration constant relative to the borane monomer. The -ole®n was added in decreasing amounts to account for the consumption of borane monomer in the feed. The experimental results are summarized in Table 3.4. The molecular weights of the polymers were determined by intrinsic viscosity measured in a cone/plate viscometer at 135 C in decalin solution. To enhance the solubility of functionalized polymers, the hydroxylated polymers were completely esteri®ed with benzoyl chloride. As shown in Table 3.4, Mv's are high for all samples. This continuous monomer feeding process produces copolymer with a much narrower compositional distribution and higher yield of borane monomer than the corresponding one-shot monomer addition in a batch reactor. In Table 3.3, the PP-OH-B prepared from this continuous process (total mole ratio of propylene and borane monomer of 10/1) yielded no methanol soluble, 8.5 wt% MEK soluble, 14% heptane soluble, and 77.2% xylene soluble fractions. Most of the PP-OH-B is highly isotactic with a semicrystalline microstructure. As the size of -ole®n increases, the borane monomer will be more comparative in the Ziegler±Natta polymerization. Figure 3.8(b) shows the 5-hexenyl-9-BBN (mol%) incorporated in the copolymers during the batch copolymerization of 1-butene and 5-hexenyl-9-BBN,131 using a 1/1 monomer mole ratio and TiCl3 AA/ Et2AlCl catalyst. The ¯uctuation of copolymer composition is still dramatic, but the curve is signi®cantly ¯atter than in the corresponding PP copolymer case. The reactivity order of the monomers is obviously following propylene >> 1-butene > 5hexenyl-9-BBN. The reactivity ratios of 1-butene (r1 7.13) and 5-hexenyl-9-BBN (r2 0.41; r1 r2 2.92) in copolymerization reactions were also estimated by Kelen±TuÈdos method.
5.3
Copolymerization of 1-Octene and Borane Monomer70
We would expect much closer reactivity between 1-octene and 5-hexenyl-9-BBN. In the comparative experiment shown in Fig. 3.8(a), the plot of comonomer
34
Functionalization of Polyolefins 50
(a)
Mol% borane
40
30 (b) 20
10 (c) 0 0
200
100 Time (min)
Figure 3.8 Plots of 5-hexenyl-9-BBN (mol%) incorporated in (a) poly(1-cotene), (b) poly(1-butene), and (c) polypropylene, starting with 1/1 feed ratio.
incorporation is quite ¯at, showing only a small increase in the B-5-hexenyl-9-BBN monomer in the copolymer with increasing monomer conversion. 1-Octene is still slightly more reactive than B-5-hexenyl-9-BBN. Table 3.5 summarizes several experimental results with various comonomer ratios. Due to the very high susceptibility of borane to air oxidation, the borane groups were usually converted to hydroxy and ester groups before measurements. All esteri®ed copolymers were completely soluble in THF and, therefore, were used for molecular weight determination by GPC with a standard polystyrene calibration curve. In general, the copolymers are high-molecular-weight polymers with broad Table 3.5
Summary of copolymerization of 1-octene and B-5-hexenyl-9-BBN Copolymer of 1-octene and B-5-hexenyl-9-BBN
Borane monomer (mol% in feed) 0 25 50 75 100 1
Borane (mol%) 0 15 40 65 100
Mn1 10 329 242 126 66 55
3
Mw/Mn
Tg2( C)
5.1 6.1 7.8 6.0 2.6
GPC curves of the esteri®ed copolymers with standard polystyrene calibration line. Differential scanning calorimetry (DSC) curves of the hydroxylated copolymers.
2
67 51 26 2 15
Direct Polymerization of B- and Si-containing Monomers
35
100 (a)
Percent
80
(b)
60
40
20
0 16
18
20 22 24 Retention time (minutes)
26
28
30
Figure 3.9 GPC curves of esteri®ed copolymer using (a) RI detector and (b) UV detector. (Redrawn from Macromolecules 1990, 23, 378. Copyright 1990 Am. Chem. Soc.)
molecular weight distribution, as expected in Ziegler±Natta polymerization. The decrease in the calculated copolymer molecular weight with the increase in the mole percent borane monomer in the feed may be partially due to the interaction of the functional group with the column, rather than an actual decrease in molecular weight. The employment of phenyl ester in the experiment permits the use of both UV and refractive index (RI) detectors. The UV detector was set at an observation wavelength of 254 nm. Since the chromophore, i.e. the phenyl ester group, is present only in one of the monomers, the comparison of the two chromatogram pro®les (Fig. 3.9) serves as an indication of the homogeneity of the copolymer samples, considering of course a statistical distribution of the ester functionality in the polymer chains. The small shift in one of the curves, with respect to the other, arises because the eluted fractions pass through the UV detector 0.25 min before passing through the RI detector due to the instrument con®guration. The almost exact overlap of the UV and RI pro®les is a strong indication of the homogeneity of the copolymer samples. All the hydroxy copolymers, derived from borane-containing polymers, exhibited a single glass transition temperature (Tg), as shown in Fig. 3.10. The presence of only one glass transition is taken as evidence for the absence of macroscopic phase separation and therefore implies that the copolymer samples are fairly homogeneous. In fact, a linear relationship between the glass transition temperature and the weight fraction of either monomer has been taken as a qualitative indication of the homogeneity and random nature of the copolymer samples.
36
Functionalization of Polyolefins
(e) Endo >
(d) (c) (b) (a)
–125.0
–100.0
–75.0
–50.0 –25.0 Temperature (°C)
0.0
25.0
50.0
Figure 3.10 DSC curves of homopolymers, (a) poly(1-octene) and (e) poly(1-hexen-6-ol), and the hydroxylated copolymers with functional group concentration of (b) 15, (c) 40, and (d) 65 mol%. (Redrawn from Macromolecules 1990, 23, 378. Copyright 1990 Am. Chem. Soc.)
6
POLYMERIZATION OF Si-CONTAINING MONOMERS
Although silane-containing LDPE was commercialized in the late 1960s and has been used as the moisture-crosslinkable material to yield stable products for applications (such as cable shieldings, etc.), the polymer was prepared by free radical copolymerization of ethylene and silane-containing ole®n (such as vinyltrimethoxysilane) at high pressure (150±350 Mpa) and temperature (180±290 C). The preparation of the corresponding silane-containing HDPE and i-PP, using Ziegler±Natta catalysts, is much more dif®cult. Some current commercial products of silane-containing HDPE and i-PP are prepared by a post-polymerization process (chemical modi®cation), which will be discussed in Chapter 5. The earliest report documenting the coordination polymerization of a silanesubstituted ole®n was that of Natta et al. in 1958.57 Allylsilane and trimethylallylsilane (shown in Fig. 3.11) were polymerized to high polymers by using TiCl3/ AlEt3 and TiCl4/AlEt3 initiator systems, respectively. Although no quantitative information of molecular weight and yield was given, both polymers are highly crystalline with isotactic structure and melting points as high as 350±360 C. Apparently, silane moieties (including Si±H group) were shown to be stable to Ziegler±Natta catalysts. Carbonaro, Greco, and Bassi60 demonstrated that the placement of the silyl group directly on the double bond greatly decreases the polymerizability of vinylethyl silane using a VCl3/AlEt3 initiator. Only a low-molecular-weight product was obtained with poor yield (15±20%). The material was, however, isotactic. The
37
Direct Polymerization of B- and Si-containing Monomers Vinylethylsilane
CH2CH ⱍ HSiH ⱍ C2H5 (low reactivity)
Figure 3.11
Trimethylvinylsilane
Allylsilane
CH2CH ⱍ CH3SiCH3 ⱍ CH3
CH2CH ⱍ CH2 ⱍ HSiH ⱍ H
(low reactivity)
Trimethylallylsilane
CH2CH ⱍ CH2 ⱍ CH3SiCH3 ⱍ CH3
Several representive silane-containing monomers.
similar low yield was observed in vinyltrimethylsilane polymerization in the presence of TiCl4/AlEt3.132 Thus, it is apparent that a spacer group between vinyl and silane groups is needed, similar to the borane case. However, a single methylene unit is suf®cient to greatly enhance the polymerizability of silicon-containing monomers. Diallylsilanes of dimethyldiallylsilane and methylphenyldiallylsilane were also homopolymerized by TiCl4/AlEt3133 to high polymers with rubbery and powdery appearances, respectively. As expected in the chain polymerization of symmetrical difunctional monomer, both -ole®n groups involved in the reaction will inevitably result in some degree of crosslinking reaction. However, the major portion of silane polymer is soluble in benzene. In most cases using AlEt3/TiCl4 ratio > 2, all polymers were completely soluble despite most of the difunctional groups engaging the reaction, and only 20±25% of repeating units contained double bonds. Subsequent polymerization±cyclization took place to produce predominant cyclic repeating units as shown in Eq. (3.4):
Z–N catalyst
n
Si
Si
R1 R2
R1 R2
3:4
A number of authors have studied the copolymerization of silane monomers with -ole®ns. Both copolymerization of dimethyldiallylsilane and methylphenyldiallylsilane with propylene were reported by Nametkin et al.134 Longi et al. later reported the copolymerization of allylsilane with propylene and 1-hexene.135 In general, the reactive silane group (especially Si±H) in a polymer is very useful for the formation of crosslinked polyole®ns. However, they are not polar enough to greatly affect the polymer properties, especially in polyole®n copolymers with a small percentage of silane groups. They are also not facile for interconversion to other functional groups.
38 7
Functionalization of Polyolefins
SUMMARY
The direct polymerization of a functional monomer by early transition metal (Ziegler±Natta and metallocene) catalysts is limited to a few monomers containing heteroatoms having Lewis acid or neutral states and forming covalent carbon± heteroatom bonds. Borane and silane groups were shown to be the best examples, exhibiting stability to catalyst and solubility in hydrocarbon media. In the case of silane, only sketchy information is available. However, extensive experimental results have been reported in the case of borane. Basically, borane monomers behave like high -ole®ns, both in terms of their reactivity in Ziegler±Natta and metallocene catalysis, and in the molecular structure of the resulting polymers. The silane groups in polyole®ns can serve as the crosslinking agent. However, they are not polar enough to greatly affect the polymer properties. On the other hand, the interest in borane-containing polyole®ns relates strongly to the versatility of borane groups, which can be transformed to a remarkable variety of functionalities (which will be discussed in Chapter 6) and also can serve as initiators in the preparation of graft and block copolymers (which will be discussed in Chapters 10 and 11). Many new functionalized polyole®ns with various molecular architectures have been obtained based on this chemistry. Most of them would be very dif®cult to obtain by other existing methods.
4 Direct Polymerization of O-, N-, and Cl-containing Monomers
1
INTRODUCTION
Ideally, we would prefer that the polyole®n copolymers have polar functional groups, such as ±OH, ±COOH, NH2, and halides, as these groups would effectively impart very desirable properties (adhesion, compatibility, etc.) to polyole®ns. Unfortunately, as discussed in Chapter 2, the direct polymerization of these functional monomers containing O, N, and halogen heteroatoms by early transition metals (Z±N and metallocene catalysts) is very dif®cult. The nonbonded electron pairs on the N or O of monomers deactivate the acidic active sites and inhibit the polymerization. Considering only a very low concentration of active sites (much less than 1% of monomer) is used in the reaction mixture, any deactivation of the catalyst has a detrimental effect on the polymerization. In early reports, some polar monomers (such as vinyl ether, acrylic ester, acrylonitrile, and vinyl chlorides) have been shown to polymerize by Z±N catalysts. However, these were later proven to be polymerizing not by the metal±carbon insertion, but rather by cationic or free radical propagation.136±139 To circumvent the catalyst deactivation problem, two general research approaches, including (a) protecting the polar functional groups and (b) using less oxophilic late transition metal catalysts, have been studied with some success, as illustrated in Scheme 4.1. The use of the protective approach, by insulating the functional group from the active site during polymerization, has been extensively studied in Ziegler±Natta66,67,71,72,74,77,78,139±147 and recently in metallocene72±76,79±83 catalysis. The reaction involves three steps, including protection of functional groups in the monomer, polymerization, and deprotection. Later, some attention was given to the use of the less oxophilic late transition metal catalyst84±87 in ole®n polymerization with an attempt to prepare functional polyole®ns by a simple copolymerization reaction. The following sections will explain the experimental results, along with the discussion of the advantages and disadvantages of each method.
40
Functionalization of Polyolefins Functional group protection
F
F: –OH, –COOH, NH2, etc. P: protecting group
F P Late transition metals (Ni, Pd, Pt, . . .)
n
F
Deprotection
Early transition metals (Ti, Zr, Hf, V, . . .)
n
F P Scheme 4.1
2
PROTECTION OF FUNCTIONAL GROUPS
The protection of functional groups is the most widely studied method, due to the predominant ole®n polymerization processes using Ziegler±Natta and metallocene catalysts. In fact, the two most important polyole®ns (HDPE and i-PP), which are also the most interesting in functionalization reactions, are exclusively prepared by early transition metal catalysts. The reports trace back to Z±N polymerization studies in the 1960s, and constant research activities in the 1970s and 1980s. In the early 1990s, some interest in this functional approach was also extended to metallocene catalysis due to the enthusiasm about designing a single active site for understanding and minimizing catalyst poison. The ideal protection agent would not only have the capability to prevent the active site from deactivation, but also to provide convenient and effective processes in protection and deprotection reactions. In addition, the protected functional monomer should be very soluble in the polymerization solvents (i.e. hexane, toluene, or monomers), similar to -ole®ns. The homogeneous reaction condition in copolymerization reaction is essential to achieve high comonomer incorporation and high copolymer yield, as well as narrow copolymer composition and molecular weight distributions. The protection methods usually lead to the steric shielding and electronic neutralization of the functional group in the monomer. In addition, the protection strategy has also been extended to the design of catalyst components, i.e. transition metal active sites and cocatalysts in metallocene systems. Many interesting results (discussed later) came from the combined efforts in both monomers and catalysts. Some of the most common protecting agents for functional groups (alcohol, ester, amine, and halides) are listed below: Steric protecting groups: ±CH(CH3)2 Electronic (steric) protecting groups: ±AlRnCl2 n, Al(CH3)3, MAO, ±Si(CH3)3
Direct Polymerization of O-, N-, and Cl-containing Monomers
41
In addition to the external protection agents, it is essential to insulate CC from the heteroatom by a spacer. At least three methylene units are needed to provide suf®cient insulation from any negative effect due to the functional group. Some experimental results and the important features of each functional group protection agent are discussed in the following sections. 3
AMINO-GROUP-CONTAINING MONOMERS
Both steric and electronic routes were reported in the protection of amino groups, and a number of vinyl monomers containing aliphatic and heterocyclic nitrogen have been polymerized using transition metal (Ziegler±Natta and metallocene) catalysts. Key papers addressing polymerization of nitrogen-containing monomers with bulky protecting groups are those of Giannini and co-workers.139,140 They systematically investigated a series of tertiary amines in which the polymerizable terminal vinyl group is separated from the nitrogen atom by a methylene chain from one to nine units in length. Polymerizability of this series of monomers in heterogeneous TiCl3 AA (aluminum activated)/Al(R 0 )2Cl catalyst was evaluated as a function of (a) length of the methylene spacer chain, (b) steric hindrance around nitrogen, and (c) steric encumbrance around aluminum (cocatalyst). Some experimental results are shown in Table 4.1. In general, the polymerization rate of the unsaturated amines is lower than that of -ole®n having a similar structure. Almost no polymer was found when the monomers had a short spacer Summary of Z±N polymerization1 of -olefin-!-tertiary amine
Table 4.1
ⱍ
ⱍ
CH2CH ⱍ (CH2)n ⱍ N R
R
TiCl3⋅AA Al(R⬘)2Cl
ⱍ
ⱍ
(CH2CH)x ⱍ (CH2)n ⱍ N R 1
R
n
CH3
A
3
96
6
C2H5
A
3
240
8
n-C4H9
A
3
120
8
t-C4H9
A
3
22
66
s-C3H7
A
1
90
0
s-C3H7
A
2
137
27
s-C3H7
A
3
14
99
s-C3H7
A
9
15
99
s-C3H7
B
3
96
19
s-C3H7
C
3
96
60
Reaction temperature: 60 C; solvent: benzene. A:CH2±CH±CH2±CH2±CH3; B: C2H5; C: i-C4H9. j
2
CH3
Polymerization Polymer time (h) yield (%)
R 02
R
42
Functionalization of Polyolefins
chain (n 1 and 2). In addition, the monomers containing an amine group (±NR2), where R is ethyl or methyl, almost completely deactivated the catalyst. Only very low yields of oily products were observed. Both methyl and ethyl substitutes failed to suf®ciently shield the nitrogen atom from the catalyst components. However, polymerization effectively occurs and achieves a high polymer and good yield when the aluminum alkyl is Al(C6H13)2Cl and the amine monomer has at least n 3 and R CH(CH3)2. The results clearly show that the polymerization is possible when the nitrogen atom of the monomer and aluminum atom of the organometallic compound used in the catalysis are shielded by bulky branched alkyl groups. For an amine monomer with R methyl or ethyl and n > 3, the polymerization can only take place if they are precomplexed (neutralized) with a stoichiometric amount of the same type of organometallic component used for the initiator preparation. Usually, aluminum alkyl with excess quantity is chosen as the complexing agent. By prior complexing with AlEt2Cl, 5-N,N-dimethylamino-1-pentene and 5-N,N-diethylamino-1-pentene were polymerized to amorphous high polymers by the TiCl3 AA/AlEt2Cl catalyst. The other technique to avoid catalyst poisoning by amino groups, especially secondary NHR groups, is the use of a protecting group, such as ±Si(CH3)3, which provides both steric shielding and electronic neutralization of the basic amine through the d±p bond between Si and N atoms.
CH2CH(CH2)3N
CH(CH3)2 Si(CH3)3
4-I This protected functional monomer (4-I) is very similar to the complexed monomers, except that the electron withdrawing silicon is attached directly to the N atom. In general, the silazanic monomers could only be polymerized if a suf®ciently hindered Al alkyl (in catalyst) was used. After polymerization, the Si±N bond was hydrolyzed in acidic alcohol solutions. The hydrolyzed polymers then contained secondary amino (±NRH) groups. From the polymerization with the above conditions, high-molecular-weight amorphous poly(5-N-isopropylaminopentene) was prepared. This tends to associate in the hydrocarbon solutions. Many researchers followed a similar strategy in protecting amino groups from interaction with Z±N and metallocene catalysts. Langer et al.71 copolymerized 6-N,N-diisopropylamino-1-hexene with propylene by a Z±N catalyst. Polypropylene copolymers thus produced have low concentrations of hindered tertiary amino groups (containing 0.07±0.5 wt% nitrogen), due to the relatively low reactivity of the hindered amine comonomer. Recently, the steric approach was extended to the hindered piperidine group. Nasman and co-workers73 studied the copolymerization of propylene and 1-hexene containing 2,2,6,6-tetramethyl piperidine (4-II) by a fourth generation Z±N catalyst (TiCl4/MgCl2/ED/Al(C2H5)3). Lezzi and co-workers74 reported the copolymerization of ethylene with norbornene containing 2,2,6,6-tetramethyl piperidine (4-III) by the homogeneous Vac3/AlEt2Cl
Direct Polymerization of O-, N-, and Cl-containing Monomers
43
catalyst. Apparently, the functional monomer did not totally inhibit (but retarded) polymerization due to the steric hindrance. However, the precomplexation of 2,2,6,6-tetramethyl piperidine with a stoichiometric amount of triethylaluminum showed signi®cant improvement in copolymer yields.
CH2CH ⱍ (CH2)4 H3C H3C
N H
CH3 CH3
4-II
ⱍ CH2 ⱍ O
CH3
H3C H3 C
N
CH3
R 4-III
Carlini141 extended the complexation technique to 4-vinylpyridine (4-VP). The polymerization of 4-VP, both free and precomplexed with an equimolar amount of aluminum alkyl, was studied using initiator systems combining TiCI3 AA with Al(i-Bu)2Cl, Al(i-Bu)3 or Zn(i-Bu)3. In all cases, the polymerizations were allowed to proceed at room temperature for 30 days. Yield of greater than 90% was obtained when the precomplexed 4-VP adduct was introduced in combination with an aluminum alkyl initiator. However, molecular weights were very low (in the range 1000±3500). The uncompleted 4-VP monomer gave only a 7% yield with aluminum alkyls, and 30% using triisobutylzinc. Carlini attributes the poor yields obtained using free 4-VP to poisoning of the active sites through coordination with the nitrogen electron pair. Collette et al.148 and Datta and Kresge149 applied the chemistry to terpolymerization of ethylene, propylene, and a third norbornene monomer containing a primary amino group in the presence of a Z±N catalyst. The functional group was protected (``masked'') by an aluminum component, such as triethylaluminum (shown in Eq. 4.1).
CH2NH2 + (CH3CH2)3Al
CH2NHAl(CH2CH3)2
4:1
Apparently, the nucleophilicity of N in primary amine is diminished by N±AlR2 moiety, due to strong electron back-donating from N to Al by the d±p bond
44
Functionalization of Polyolefins
as well as some steric shielding effect from alkylaluminum. The resulting aluminum-containing terpolymer precursor was then deprotected (``unmasked'') by hydrolysis to give an EP elastomer containing NH2 groups. Advances in metallocene technology have prompted a new wave of research interest in this direct polymerization of the protected functional monomers. In addition to the unique copolymerization capability to form copolymers with narrow composition and molecular weight distributions, the well-de®ned active site with a broad choice of catalyst and cocatalyst components provides a great opportunity to understand the effects of the protected functional monomer and active site. Waymouth and co-workers ®rst reported the use of per¯uoroborane cocatalysts (in the absence of alkylaluminum cocatalyst) that are stable with the steric-protected amino groups. Both homopolymerization72,76 of 5-(N,N-diisopropylamino)-1pentene and copolymerization reactions with 1-hexene and 4-methyl-1-pentene150 by metallocene catalysts were studied, with various combinations of zirconocene, such as Cp*ZrMe2, rac-EB(THI)ZrMe2, and per¯uroborate, such as B(C6F5)3, [HNMe2Ph][B(C6F5)4] . Although only low-molecular-weight oily products (containing tertiary amino groups) were obtained, the catalyst activity was relatively high. The catalyst activity was sensitive to the nature of the ligands at the transition metal and the substitution pattern at the amine. In the copolymerization reactions, catalyst activities showed linear reduction with the increase of 5-(N,Ndiisopropylamino)-1-pentene concentrations. Overall, the experimental results indicate that the coordination site (Zr±C) is still sensitive to the diisopropylsubstituted amino group. Mulhaupt and co-workers75 reported on a zirconocene/ methylaluminoxane (MAO) catalyst in the copolymerization of ethylene and N,Nbis(trimethylsilyl)-1-amino-10-undecene. It is believed that MAO (with a large quantity in the system) may serve as an in situ complexation agent to protect the catalyst from deactivation. In fact, the catalyst activity is also reversibly proportional to the concentration of the silylated amino groups. Upon hydrolysis of the silylated amino groups, a short-chain branched LLDPE with pendent aminoalkyl groups was obtained.
4 SYNTHESIS OF SYNDIOTACTIC POLYSTYRENE CONTAINING AMINO GROUPS Recently, we were curious about the scope and limitations of this well-de®ned metallocene catalyst system in the polymerization of protected functional monomers. In addition, it is very interesting to prepare functional polyole®ns that can only be prepared by metallocene catalysis. Homopolymerization reactions of several amino-group-substituted styrene monomers151 were conducted by using syndiospeci®c half-sandwich titanocene/per¯uoroborane systems, as shown in Eq. (4.2).
Direct Polymerization of O-, N-, and Cl-containing Monomers
CH2CH ⱍ (CH2)n
Syndiospecific catalyst
(CH2CH)x ⱍ
(CH2)n
N G
HCl
(CH2CH)x ⱍ
(CH2)n
N G
G
45
4:2
N G
H
H
where
n: 0, 1, and 2; G: Si(CH3)3 masking group; Half-sandwich titanocene catalysts: Cp*TiMe3, Cp*TiCl3, IndTiMe3, [2-MeBenz[e]Ind]TiMe3, and [2-Ph-Phen[e]Ind]TiMe3; Cocatalysts: MAO, B(C6F5)3, [Ph3C][B(C6F5)4] , [HNMe2Ph][B(C6F5)4] , and tris(2,2 0 ,2 00 -nona¯uorobiphenyl)borane(PBB).
The research plan was formulated to answer several intriguing questions: (1) (2) (3)
(4)
Is it possible to manipulate an active site with bulky ligands in the catalyst or/and cocatalyst to completely prevent acid±base complexation between the metallocene catalyst system and protected functional group? Is it possible to achieve both high catalyst activity and maintain the stereospeci®c coordination reaction? Since the titanocene cation is generally more oxophilic than the corresponding zirconocene one, is it still possible to homopolymerize the steric-protected functional monomers to a high polymer in such an environment with high functional group concentration? Will the resulting new syndiotactic functional s-PS homopolymers containing primary amino groups be very interesting material in terms of high surface energy, high melting temperatures, and fast crystallization due to strong hydrogen bonding?
Table 4.2 compares the polymerization results of styrene derivatives containing masking amino groups and styrene (control reaction) using Cp*TiMe3/B(C6F5)3 catalyst under similar reaction conditions. Homopolymerization of 4-(dimethylaminomethyl)styrene showed a very low polymer yield, with less than 1% of the catalyst activity seen from the styrene case. The poor catalyst activity may be due to the inadequate steric protection of the amino group by only two methyl groups. On the other hand, all three trimethylsilyl-protected monomers, i.e. 4-[N,N-bis(trimethylsilyl)amino]styrene, 4-[N,N-bis(trimethylsilyl)aminomethyl]styrene, and 4-[N,N-bis(trimethylsilyl)aminoethyl]styrene, exhibited good catalyst activity. Most impressively, 4-[N,N-bis(trimethylsilyl)aminoethyl]styrene, having two methylene spacers between styrene and the protected amino group, showed close to
46
Functionalization of Polyolefins Table 4.2 catalyst1
Polymerization of styrenic monomers using Cp*TiMe3/B(C6F5)3 Syndiotactic3 Tg Tm [Catalyst] Conversion 2 (mM) (wt%) Cat. activity (%) ( C) ( C)
Monomer CH CH 2
ⱍ
CH CH 2
ⱍ
[]4 (dl/g)
0.6
96.5
11151.1
97.1
100 270
0.75
9.5
8.5
96.6
35.8
115 278 0.04 (weak)
5.0
45.0
1580.4
76.5
157 325
0.12
5.0
80.2
2967.4
85.2
152 330
0.29
2.5
90.1
7003.8
90.5
138 320
0.45
ⱍ
CH 2
ⱍ
ⱍ
ⱍ
N Me
Me CH CH 2
ⱍ ⱍ
ⱍ
ⱍ
N Me3Si
SiMe3 CH CH 2
ⱍ ⱍ
CH 2
ⱍ
ⱍ
ⱍ
N SiMe3
Me3Si
CH CH 2
ⱍ ⱍ
(CH 2)2
ⱍ
ⱍ
ⱍ
N Me3Si 1
SiMe3
Polymerization conditions: 35 C for 60 min, Cp*TiMe3/B(C6F5)3 1, [monomer] 1.3 M, toluene solvent 15 ml. 2 Activity in units of kg of polymer/[(mol of Ti) (mol of monomer) h]. 3 Portion of syndiotactic polymer, insoluble in 2-butanone and determined by 13C NMR. 4 Inherent viscosity, [], determined in 1,2,4-trichlorobenzene at 130 C.
47
Direct Polymerization of O-, N-, and Cl-containing Monomers
70% of the reactivity of styrene. The functional group's location away from the ole®n unit is clearly an advantage in minimizing the acid±base complexization. The in¯uence of the functional group on the polymerization reaction was also revealed in the polymer structure. Both the polymer molecular weight and syndiotacticity were parallel to the catalyst activity. About 90% of the poly{4-[N,Nbis(trimethylsilyl)aminoethyl]styrene} obtained was a syndiotactic polymer with high molecular weight (Mw 10.5 104; Mn 4.6 104 g/mol), examined by GPC. The relatively narrow molecular distribution (Mw/Mn 2.3), similar to the corresponding s-PS homopolymer, strongly indicates the negligible effect of the functional group on the catalytic site during the polymerization reaction. Since the masking amino group in 4-[N,N-bis(trimethylsilyl)amino]styrene has some effects on the ole®n polymerization, further study of this system is very interesting in terms of the nature of the catalytic site to the polymerization reaction ± especially the relationship between the bulky ligands in the catalyst or/and cocatalyst and the catalyst activity and syndiotacticity of the resulting polymer. Table 4.3 summarizes the results of 4-[N,N-bis(trimethylsilyl)amino]styrene using various half-sandwich titanocene catalyst and per¯uoroborane cocatalyst systems. Comparing the ®rst four polymerization reactions, using the same
Table 4.3 Syndiospecific polymerization of 4-(N,N-bis(trimethylsilyl)amino)styrene with titanocene catalysts1 Catalyst/ cocatalyst Cp*TiMe3/B(C6F5)3 Cp*TiMe3/ [Ph3C][B(C6F5)4] Cp*TiMe3/ [HNMe2Ph][B(C6F5)4] Cp*TiMe3/PBB IndTiMe3/B(C6F5)3 [2-Me-Benz[e]Ind]TiMe3/ B(C6F5)3 [2-Me-Cp[l]Phen]TiMe3/ B(C6F5)3 Cp*TiMe3/MAO (100) Cp*TiCl3/MAO (1000)
[Catalyst]/ [monomer] Yield Cat. Syndiotactic3 (%) (mM)/(M) (wt%) activity2
Tg4 Tm4 []5 ( C) ( C) (dl/g)
5.0/0.95 5.0/0.95
44 65
2317.9 3424.2
76.5 78.2
160 325 158 325
0.12 0.14
5.0/0.96
40
2107.2
67.8
157 324
0.07
3.0/0.94 5.0/0.95 3.0/0.95
90 40 75
7902.0 2107.2 6585.0
94.5 70.6 85.7
160 325 158 323 160 326
0.28 0.14 0.24
3.0/0.95
86
7550.8
90.2
159 327
0.38
10.0/0.95 10.0/0.96
5 10
131.7 263.4
46.7
158 318
0.08
1 Polymerization conditions: at 35 C for 60 min, [catalyst]/[cocatalyst] 1 except MAO cases, toluene 10 ml. 2 Activity in units of kg of polymer/[(mol of Ti) (mol of monomer) h]. 3 Portion of syndiotactic polymer, insoluble in 2-butanone and determined by 13C NMR. 4 Determined by DSC at a heating rate of 10 C/min. 5 Inherent viscosity, [], determined in 1,2,4-trichlorobenzene at 130 C.
48
Functionalization of Polyolefins
Cp*TiMe3 catalyst and four different borane cocatalysts, the bulky tris(2,2 0 ,2 00 nona¯uorobiphenyl)borane (PBB) shows distinctively higher catalyst activity (more than 2±3 times higher than the others) and the highest content (94.5%) of syndiotactic polymer. The huge (PBB±CH3) anion, associated with the half-sandwich titanocene cation, apparently prevents any acid±base interaction between the active site and the silane-protected amino group. In addition, it does not reduce the capacity of coordination and insertion of -ole®n. In fact, both catalyst activity and syndiotactic polymer content surpasses the levels of the 4-[N,N-bis(trimethylsilyl)aminoethylstyrene case shown in Table 4.2. The similar steric effects were also observed in the catalyst itself. In the second set of comparative reactions, using various titanocene catalysts and the same B(C6F5)3 cocatalyst, signi®cantly higher catalyst activity and content of syndiotactic polymer were observed in the polymerization using a [2-Me-Cp[l]Phen]TiCl3/B(C6F5)3 catalyst having a bulky ligand. It is interesting to note that all catalyst systems using an MAO cocatalyst produced a very low yield of polymer. The failure may be associated with the interaction between MAO and the silane-protected functional group. Overall, with the combination of a dimethylsilane masking group and a selective catalyst system, the styrene derivatives containing amino groups can be homopolymerized to high polymers with high catalyst activity and high content of syndiotactic polymer, similar to the s-PS case. The trimethylsilane-protected amino groups in the polymer were completely converted to primary amino groups by hydrolysis followed by neutralization. Figure 4.1 shows 13C NMR spectra of poly(4-aminostyrene), poly(4-aminomethylstyrene), and poly(4-aminoethylstyrene). All peaks are very sharp and designated to the corresponding carbons in the polymer structures. According to the literature, the presence of single sharp resonance for the quaternary C1 carbon in phenyl groups (at 140.2 ppm in Fig. 4.1(a), 142.1 ppm in Fig. 4.1(b), and 142.5 ppm in Fig. 4.1(c), respectively) shows that these polymers are highly syndiotactic. The same conclusion can be reached by considering two C5 and C6 aliphatic carbons in the polymer backbone:
–C5–C6– ⱍ C1 C2 C2
C3
ⱍ C4
C3
Figure 4.1(a) shows two shape resonances at 45.0 and 41.5 ppm, corresponding to methine C5 and methylene C6 in a high stereoregular environment. In Fig. 4.1(b), in addition to C5 and C6 carbons (peaks at 43.8 and 41.5 ppm, respectively), a new
Direct Polymerization of O-, N-, and Cl-containing Monomers
49
(a)
(b)
(c)
ppm 140
120
100
80
60
40
20
Figure 4.1 13C NMR spectra of (a) poly(4-aminostyrene), (b) poly(4-aminomethylstyrene), and (c) poly(4aminoethylstyrene). (Redrawn from Macromolecules 2000, 33, 5803. Copyright 2000 Am. Chem. Soc.)
resonance at 47.0 ppm corresponds to the methylene carbon in -CH2±NH2 group. In Fig. 4.1(c), two C5 and C6 carbon resonances (peaks at 43.2 and 41.0 ppm, respectively) are accompanied with two new resonances at 44.4 and 30.2 ppm, corresponding to -OH2±CH2±NH2 and -CH2±OH2±NH2, respectively. It is very interesting to note that the sharp resonance feature resembles those of syndiotactic polystyrene, but is very different from those of several reported polystyrene derivatives containing chloro and methoxy groups152 prepared by other catalyst systems.
50
Functionalization of Polyolefins 100 90 (a)
80
Heat flow (mW)
70 60
(b)
50
(c)
40 30 20 10 0 50
100
150
200 250 Temperature (°C)
300
350
400
Figure 4.2 DSC curves of (a) poly(4-aminostyrene), (b) poly(4-aminomethylstyrene), and (c) poly(4-aminoethylstyrene). (Redrawn from Macromolecules 2000, 33, 5803. Copyright 2000 Am. Chem. Soc.)
Figure 4.2 shows DSC curves of poly(4-aminostyrene), poly(4-aminomethylstyrene), and poly(4-aminoethylstyrene). All samples were measured under the same thermal treatment, and the curves were recorded in a second heating cycle. Every sample shows a melting endotherm, indicative of semicrystalline morphology. Their melting temperatures are signi®cantly higher than that of s-PS (270 C) ± about 340 C for poly(4-aminomethylstyrene) and poly(4-aminoethylstyrene), and about 360 C for poly(4-aminostyrene). It is very interesting to note that the endotherm peak becomes very sharp in all three NH2-containing polymers, implying a wellorganized and uniform crystalline structure. The hydrogen bonding between NH2 groups may enhance the crystallization process. In addition to the syndiotactic poly(aminostyrene) homopolymers, we have also extended the protecting chemistry to the preparation of the s-PS copolymers containing amino group. One example is shown in the copolymerization of styrene and 1,3-dimethyl-2-[4-vinylphenyl]-imidazolidine by the syndiospeci®c [2-MeBenz[e]Ind]-TiCl3/MAO and/or [2-Ph-Phen[e]Ind]TiCl3/MAO catalysts, as illustrated in Eq. (4.3). Both catalysts with bulky ligands exhibit high activities in the preparation of poly{styrene-co-(1,3-dimethyl-2-[4-vinylphenyl]imidazolidine)} copolymers. However, the simple CpTiCl3/MAO catalyst showed very low activity with the copolymerization reactions. The steric protection in half-sandwich titanocene catalysts is essential in preventing the catalyst from deactivation.
Direct Polymerization of O-, N-, and Cl-containing Monomers
51
(CH2CH)x(CH2CH)y ⱍ ⱍ
CH2CH ⱍ Syndiospecific catalyst*
+ CH2CH ⱍ
N
N
4:3
HCl/H2O
N
(CH2CH)x(CH2CH)y ⱍ ⱍ
*Half-sandwich titanocene catalyst: [2-Me-Benz[e]Ind]TiMe3 and [2-Ph-Phen[e]Ind]TiMe3
ⱍ
ⱍ C
N
H
O
Figure 4.3 shows the 13C NMR spectrum of poly{styrene-co-(1,3-dimethyl-2[4-vinylphenyl]-imidazolidine)}, containing 3 mol% of 1,3-dimethyl-2-[4-vinylphenyl]-imidazolidine units. In addition to the chemical shifts at 39.8, 53.8, 92.5, and 139.7 ppm, corresponding to the incorporated 1,3-dimethyl-2-[4-vinylphenyl]-imidazolidine units, a single sharp resonance at 145.1 ppm, corresponding to the quaternary carbon C of the phenyl ring, indicates a polymer with
a
b n
c
f
e d
N
h
N m
d
g a
c
b
ppm
180
160
140
g
f
e
120
100
80
60
h
40
20
Figure 4.3 13C NMR spectrum of poly{styrene-co-(1,3-dimethyl-2-[4-vinylphenyl]imidazolidine)} copolymer.
52
Functionalization of Polyolefins
d
c n
b C H m a O
b
d c
a
ppm
10
8
6
4
d
2
e n
b
c
C H m
c
O
a
d e b
a
ppm 200
175
150
125
100
75
50
25
Figure 4.4 (Top) 1H and (bottom) 13C NMR spectra of the s-PS containing aldehyde groups.
high syndiotacticity. Based on the insoluble fraction in 2-butanone, the syndiotactic index of this s-PS copolymer is about 95%. The imidazolidine groups in the poly{styrene-co-(1,3-dimethyl-2-[4-vinylphenyl]-imidazolidine)} copolymer are facile in the interconversion reaction to
Direct Polymerization of O-, N-, and Cl-containing Monomers
53
Heat flow (mW)
aldehyde groups. Usually, the copolymer is dissolved in a mixed solvent (o-dichlorobenzene/toluene 5/1 volume ratio) at elevated temperature. The solution is then cooled down to 60 C, and a few cubic centimeters of 2 N HCl aqueous solution are added into the reaction mixture. After reaction at 60 C for 4 h, the polymer particles are ®ltered, washed, and dried in a vacuum oven to obtain the functionalized s-PS copolymer containing aldehyde groups. Figure 4.4 shows 1H and 13C NMR spectra of a new s-PS copolymer containing 3 mol% aldehyde groups. Both chemical shifts at 9.76 ppm (Fig. 4.4, top) and 194.3 ppm (Fig. 4.4, bottom) clearly show the formed aldehyde (CHO) groups. The interconversion is very complete, without any detectable imidazolidine groups in both 1H and 13C NMR spectra. The same single sharp resonance at 145.1 ppm, corresponding to the quaternary carbon C of the phenyl ring, remains and clearly demonstrates that this functionalized s-PS copolymer with aldehyde groups has highly syndiotactic molecular structure. Figure 4.5 shows DSC curves of new s-PS copolymers containing aldehyde groups. It is very interesting to note that the melting temperatures of s-PS copolymers containing 1.5 and 3 mol% of aldehyde groups are almost identical as that of s-PS homopolymer, at about 270 C. No signi®cant losses in both syndiotacticity and melting point are very important in preserving the most important properties of s-PS polymer in the functionalization reaction.
(c)
(b)
(a) 75
100
125
150
175 200 Temperature (°C)
225
250
275
Figure 4.5 DSC curves of (a) a s-PS homopolymer and two s-PS copolymers containing (b) 1.5 and (c) 3.0 mol% of aldehyde groups.
54 5
Functionalization of Polyolefins
POLYMERIZATION OF OH-CONTAINING MONOMERS
Based on the results of NH2-containing monomers, it is logical to expect slightly easier reaction conditions for the polymerization of protected OH-containing monomers, since the OH group has a generally weaker base than the NH2 group. However, considering the fact that the OH group is only carrying one protecting group, instead of two in NH2 cases, the effectiveness of the protection has to be very dependent on the steric environment at the active site. Both steric and electronic protection was applied in preventing the interaction between catalyst and oxygen. Giannini et al.139,140 protected ±OH groups in 4-penten-1-ol and 10-undecen-1-ol with a Si(CH3)3 group. The resulting siloxanic monomers can only be polymerized to high polymers by a heterogeneous TiCl3 AA catalyst when a bulky hindered aluminum cocatalyst, such as Al(CH2CH(CH3)CH2CH2±CH3)3, is used. After polymerization, the O±Si bond was hydrolyzed in an acidic alcohol solution to give crystalline poly(5-hydroxy-l-pentene) and poly(11-hydroxy-l-undecene). Overall, the polymer yields were low (20±25%). Sivak and Cullo153 extended the chemistry to the copolymerization reactions of ole®n (especially propylene) with siloxanic monomers by using a highly active supported Ziegler±Natta catalyst, TiCl4/MgCl2/ED/AlEt3 (ED: external donor, such as diphenyl dimethoxy silane or phenyl triethoxy silane). The reactions were targeted at the polypropylene copolymers containing a relatively low concentration of functional groups ( < 5 mol%). Usually, the copolymerization reactions were very effective to prepare copolymers with high yield and isotactic PP structure. On the other hand, Nasman and co-workers154 reported the protection of the OH group by precomplexing it with an aluminum component prior to the polymerization. A copolymerization of propylene and 4-(!-alkenyl)-2,6-di-t-butylphenol (pretreated with a stoichiometric molar amount of triethylaluminum) was studied by using a similar TiCl4/MgCl2/AlEt3 catalyst. The results showed that the copolymerization yield, as well as the phenol content of the resulting copolymer, increased with an increasing number of spacers between the double bond and the phenol group. The renewed interest in the direct functionalization approach using a metallocene catalyst also prompted some enthusiasm in the polymerization of OH-containing monomers. Waymouth and co-workers72 reported the homopolymerization of 4trimethylsiloxy-1,6-heptadiene. In the presence of [Cp*2ZrMe]X (X: B(C6F5)4 or CH3(C6F5)3), cyclopolymerization took place rapidly at room temperature to form low-molecular-weight product. The reaction is illustrated in Eq. (4.4). It is interesting to note that the chiral [(EBTHI)ZrMe]X catalysts (EBTHI ethyl-ene-1,2-bis (5-4,5,6,7-tetrahydro-1-indenyl)) are inactive in this polymerization but readily polymerize the more sterically hindered 4-t-butyldimethylsiloxy1,6-heptadiene. Several copolymerization reactions were based on zirconocene/methylaluminoxane (MAO) catalyst systems. The reaction usually involved ethylene and
Direct Polymerization of O-, N-, and Cl-containing Monomers
SiMe3 – [Cp2ZrMe+] B(C6F5 )4 ⱍ O ⱍ
ⱍ O ⱍ SiMe3
Deprotection
ⱍ O ⱍ SiMe3
ⱍ O ⱍ H
ⱍ O ⱍ H
55
4:4
propylene with a small amount of comonomers that had a large spacer between ole®n and functional group, such as 1-hydroxy-10-undecene (4-IV)79,80,155,156 and 6-t-butyl-2-(1,1-dimethylhept-6-enyl)-4-methylphenol (4-V):82
H ⱍ O ⱍ
O ⱍ H 4-IV
ⱍ
4-V
In addition to the silane protecting method, 1-hydroxy-10-undecene was directly copolymerized with ethylene and propylene79,80 using a stereorigid catalyst system, such as Et[Ind]2ZrCl2, Me2Si[Ind]2ZrCl2, etc. However, the catalyst activity decays with the concentration of 1-hydroxy-10-undecene. It is believed that MAO (a large quantity in the system) serves as an in situ protection agent to prevent catalyst deactivation. In fact, the pretreatment of functional monomer with MAO before initiation signi®cantly increases the catalyst activity. The similar pretreatment using trimethylaluminum (TMA)83 also showed better copolymerization results between -ole®ns and OH-containing monomers using zirconocene/MAO catalysts. On the other hand, the hindered phenolic group in 6-t-butyl-2-(1,1-dimethylhept-6-enyl)-4methylphenol exhibited signi®cantly higher stability to zirconocene/MAO catalysts due to good steric protection. In the copolymerization reaction of propylene and 6-tbutyl-2-(1,1-dimethylhept-6-enyl)-4-methylphenol,82 the catalyst activity was markedly higher at Al/phenol molar ratios > 4.4. In fact, the activity surpassed that of the homopolymerization of propylene with high MAO concentration (Al/ Zr > 2500), which resembles the copolymerization reaction with borane-containing monomers discussed in Chapter 3, Section 5. This comonomer effect may be due to the reduction of polymer crystallinity, and therefore an increase of monomer diffusion. The bulky hindered phenol provides the PP polymer with an internal stabilizer, and also signi®cantly reduces its crystallinity and melting temperature.
56 6
Functionalization of Polyolefins
POLYMERIZATION OF COOR-CONTAINING MONOMERS
Compared with the OH group, the protection of the COOH group from interaction with Ziegler±Natta and metallocene catalysts is more demanding. Both oxygen atoms in each ester group have to be insulated from the catalytic site. Usually, the acid group has to be converted to an ester group (COOR) with a bulky R substitution, and then precomplexing the resulting ester group with the Al component of the catalyst system before polymerization. In addition, a long spacer with at least three methylene units between the -ole®n and the ester group in the monomer is needed to prevent negative interference during the coordination/insertion process. Purgett and Vogl66,67 reported a systematic study using a heterogeneous Ziegler±Natta catalyst in the homo- and copolymerization of !-alkenoate derivatives with ethylene and propylene. A most suitable reaction is shown in Eq. (4.5):
R ⱍ CH2CH TiCl3AA/Et2AlCl + CH2CH ⱍ (CH2)8 H3C CO2 H3C
R ⱍ (CH2CH)x(CH2CH)y ⱍ (CH2)8 H3C CO2 H3C
4:5
The polymerization was effective when a bulky 2,6-dimethylphenyl-protected ester group was used in conjunction with the precomplexation of the ester with alkylaluminum compounds before contact with the catalyst. In general, the polymerization reaction requires higher concentrations of catalyst and cocatalyst and longer reaction times (up to few days in homopolymerization) to achieve high polymer yields. The other route taken to minimize the interaction between ester groups and catalysts was the chemical attachment of electron-withdrawing substituents onto the functional groups to form aluminate salts.77,148 Usually, this reaction route allows only a low incorporation of the protected ester-monomer into the polyole®n copolymer. Breslow and colleague157 prepared polypropylene ionomers by copolymerization of propylene and low levels ( < 0.5 mol%) of ethylchloroaluminum-10undecennoate by a heterogeneous TiCl3/Et2AlCl catalyst. The ionic (Cl±Al±O) species in the monomer generally causes great concern during the polymerization process, due to either a premature precipitate (in heterogeneous solution) or an increase in the viscosity beyond stirring means (in homogeneous solution) from ionic association. Datta and Kresge149 improved the process by using a better soluble triisobutylaluminum protecting agent during the terpolymerization of ethylene, propylene, and 5-norbornene-2-diisobutylaluminumcarboxylate. After
Direct Polymerization of O-, N-, and Cl-containing Monomers
57
facile deprotection reaction by hydrolysis, an acid-functionalized EP rubber having only 1±2 mmol% acid groups was obtained. Very few useful results were derived from the metallocene polymerization of ester-containing monomers. Some of the dif®culties may be associated with the impurities in the protected monomer, which may be very sensitive to some metallocene/per¯uoroborane systems. Chien and co-workers83 reported the copolymerization of TMA-protected 10-undecen-1-oic acid with ethylene and propylene by using a rac-[Et(Ind)2ZrC12/MAO system. The acid monomer was pretreated with TMA with a monomer/TMA 1/2 mole ratio. In general, a much lower catalyst activity was observed, equaling about 10% of the corresponding reaction without functional monomers.
7
POLYMERIZATION OF HALOGEN-CONTAINING MONOMERS
Since halogens are part of catalyst systems and some halogen-containing molecules (such as chlorobenzene and tetrachloroethylene)26 are routinely used in Ziegler±Natta polymerization, intuitionally, we would expect that most halo--ole®n monomers should be readily polymerized by Ziegler±Natta or metallocene catalysts. However, many experiments63,158±164 have shown contrary results. In fact, many alkylhalides are known to undergo isomerization or hydrogen halide elimination165 in the presence of Lewis acids. The reaction involves a complex (C±XAl) formation between a Lewis acid, such as aluminum halide, and alkyl halide. The formed alkyl carbonium ion then loses a proton and produces hydrogen halide and ole®n. The reaction rate is strongly dependent on the ionic nature of the complex and varies widely with the reactant. In the !-halo--ole®n cases, the terminal double bond may assist the dehydrohalogenation process by stablizing a carbonium ion, as shown in Eq. (4.6):
CH2 CH2X ⱍ + CH2
AlR
ⱍ ⱍ
CH ⱍ CH2
ⱍ
ⱍ CH2
ⱍ
ⱍ
ⱍ
ⱍ ⱍ
CH ⱍ CH2
CH2 + HX + AlR
ⱍ ⱍ
CH2 CH
ⱍ
CH2 δ+ δ– δ+ δ– CH CH2 X AlR ⱍ ⱍ CH2 CH2 CH2
4:6
CH2
58
Functionalization of Polyolefins
The formed hydrogen halide may further deactivate the catalyst and react with the propagating polymer chain. A considerable effort was aimed at the copolymerization of -ole®ns and vinyl chloride.158±162 However, the preponderance of experimental evidence indicates that the mechanism is radical in most cases.26 Backsai164 developed a technique to decrease the reactivity of the aluminum alkyl toward the halogen by adding a catalytic amount of a third component, a Lewis base such as pyridine. Copolymers of propylene, and 1-hexene with 8-bromo-1-octene have been synthesized. The studies have shown that !-halo--ole®n polymerization is favored by increasing the size of the halogen atom (±I > ±Br > ±Cl > ±F), as well as increasing the number of spacers between the halogen and double bond. Unfortunately, the more desirable (reactive) halogens, such as ±Br and ±I, are not stable when under the processing conditions of polyole®ns, especially at high temperature. Only chloro-containing monomers may be practical. Recently, Def®eux and co-workers81 reported co- and terpolymerization of !-chloro--ole®ns with -ole®ns in the presence of a rac-Et(Ind)2ZrCl2/MAO catalyst. The experimental results showed that 5-chloropent-1-ene failed to either homopolymerize or copolymerize with -ole®ns. 11-Chloroundec-1-ene (4-VI) having a larger spacer was copolymerized with ethylene and terpolymerized with ethylene and propylene under certain conditions, resulting in a relatively low yield, but it could not be homopolymerized. It is very interesting to note that aryl halide appears to be more stable in reaction to the transition metal catalysts due to the lack of a dehydrohalogenation process. Proto and Senatore166 recently reported the copolymerization of ethylene and p-chlorostyrene (4-VII) by using a rac-(isopropylidene)bis(1-indenyl)zirconium-bis(dimethylamide)/MAO catalyst. An alternating ethylene/p-chlorostyrene copolymer was obtained, which has an isotactic arrangement of the p-chlorostyrene units.
Cl Cl 4-VI
8
4-VII
LATE TRANSITION METAL CATALYSIS
In contrast to early transition metal systems, it has been known that late transition metal catalysts are less oxophilic and much more tolerant to polar functional groups containing O and N atoms. In fact, many late transition metal catalysts show no adverse effects in aqueous solution. Unfortunately, in the polymerization of -ole®ns, most late transition metal catalysts often only dimerize or oligomerize
Direct Polymerization of O-, N-, and Cl-containing Monomers
59
monomers due to facile chain transfer reaction, which involves -hydride elimination and subsequent displacement of the growing polymer chain. Recently, Brookhart25,26 reported a unique system involving bulky -diimine ligands, 2,6diacetylpyridine-bis-(2,6-diisopropylaniline) derivatives, which are coordinated to Ni(II), Pd(II), and Fe(II) metal centers to suppress chain transfer reaction and favor propagation reaction, as illustrated in Eq. (4.7): N
+ M
β-hydride elimination
N
+ M
N
N Chain migration
Propagation
H Displacement
4:7 Chain growth
N
Propagation
N
+ M
N
+ M
N
H
Chain migration
Due to the steric requirement, both aryl rings are forced to be perpendicular to the square plane of the -diimine±metal complex. This perpendicular arrangement allows the four isopropyl groups to block the axial binding sites both above and below the square plane, which inhibits the displacement of the growing polymer chain by the ole®n monomer. Several bulky -diimine-coordinated Ni(II), Pd(II), and Fe(II) catalysts not only produce high-molecular-weight polyole®ns, but also show very comparable activities to those of metallocene catalysts. In addition, due to signi®cant chain migration, the polyole®ns produced are branching polymers with the branching density dependent on the catalyst concentration and structure, as well as on the reaction temperature. As expected, a higher reaction temperature enhances the hydride elimination reaction and produces a polymer with higher branching density. Following the development of this unique catalyst system, the copolymerization of -ole®ns with functional monomers, such as methyl acrylate, t-butylacrylate, and methyl vinyl ketone, was investigated. Based on the few experimental results reported, the copolymerization seems to achieve some success at preparing functional copolymers with narrow molecular weight and composition distributions. However, the functional monomers show signi®cantly lower reactivity ( < 1%) than ethylene or propylene. Both the productivity of the copolymerization reaction and the molecular weight of the copolymer are greatly reduced. In addition, most of the copolymers are amorphous, highly branched materials. It is very interesting to gain an understanding of the experimental results from seeing the reaction pathway.
60
Functionalization of Polyolefins
Equation (4.8) illustrates the mechanism in the copolymerization of ethylene and methyl acrylate:
O OR⬘
N N
+
R
M
COOR⬘
2,1-insertion
Chain growth + chain migration
N
N
+ M
N
R
N
O OR⬘
OR⬘
4:8
R
Coordination
+ O M
N N
+ O M
OR⬘
Rearrangement
R
Due to the interaction between the heteroatom and the metal center, the initial 2,1-insertion yields the four-membered chelate complex, which rearranges to the thermodynamically favorable six-membered chelate. The stable six-membered chelate complex is not only the resting state in the catalytic cycle, but also the turnover limiting state that reduces the catalyst activity compared to the corresponding homopolymerization of ethylene. It is interesting to note that the rearrangement also increases the branch size. Recently, Chien and co-workers167 extended the -diimine Ni/MAO catalyst system to study co- and ter-polymerization of ethylene, propylene, and !-hydroxy-ole®n or !-carboxyl--ole®n. Before contact with a catalyst, trimethylaluminum (TMA) was used for precomplexing the functional group in the monomer, with a monomer/TMA ratio of 1/1 and 1/2 for OH and COOH, respectively. In the presence of 5-hexen-1-ol, the polymerization rate showed less effect for the {[bis-N, N 0 (2,6-diisopropylphenyl)imino]acenaphthene}dibromonickel, but caused a 2±5-fold reduction for the {[bis-N,N 0 -(dimesityl)imino]acenaphthene}dibromonickel. Even in this less oxophilic catalyst system, the steric hindrance at the active site still shows some positive effect in preventing catalyst deactivation. Compared to the group 4 metallocene system, the late transition metal system incorporates much better with all three functional monomers, i.e. 5-hexen-1-ol, 10-undecen-1-ol, and 10-undecen-1-oic acid, with much higher catalyst activities.
Direct Polymerization of O-, N-, and Cl-containing Monomers
9
61
SUMMARY
There has been a long and constant scienti®c interest in the direct polymerization of functional (polar) group containing -ole®ns by the transition metal coordination polymerization process. All the challenges boil down to a selection between two coexisting acid±base interactions among a catalyst (acid) and a functional monomer containing two bases (ole®n and heteroatom). The ideal reaction environment is obviously a facile acid±base interaction between an active site and the -electrons of an -ole®n unit (for polymerization), and no acid±base interaction between an active site and a lone pair of electrons in a heteroatom (O, N, X) of a functional group (for catalyst deactivation). In any catalysis system (using both early and late transition metals), to maintain high catalyst activity it is inevitable that some measure of stabilizing heteroatoms has to be provided to subdue the latter interaction. Both research strategies, i.e. the protection of functional group and the usage of less oxophilic transition metal catalyst, have achieved some successes and also shown several limitations. The combination of both measures (including the design of catalyst, cocatalyst, spacer, and functional group) usually results in the best performance, especially in Ziegler±Natta and metallocene catalysis. In addition to some obvious concerns, such as the cost of protection and deprotection processes and the removal of impurities, it is very important to preserve some important polyole®n properties, such as crystallinity, melting temperature, processibility, etc., during the functionalization reaction. The major applications of functional polyole®ns are in the roles of interfacial agents to promote adhesion and compatibility between polyole®n and other polar materials in polyole®n blends and composites.
This Page Intentionally Left Blank
Post-polymerization Approach
This Page Intentionally Left Blank
5 Functionalization by Post-polymerization Processes
1
INTRODUCTION
Chemical modi®cation of the existing commercial polymers (PE, PP, EPR, etc.) presents a very attractive alternative88±93 to achieving the functionalization of polymers without the development of completely new polymerization processes. The approach circumvents the dif®culties of catalyst poisoning by functional groups during the polymerization process. Ideally, the modi®cation reaction only incorporates the desirable functional groups into the polymer chain without changing the polymer structure and properties, such as molecular weight, melting temperature, processibility, etc. Despite the current commercial applications,168±175 there are many concerns in this functionalization approach, namely the ef®ciency and alteration of the molecular structure. Only very limited chemistry is available to activate the completely saturated aliphatic molecular structure. The dif®culty multiplies in polyole®n cases, especially for the highly crystalline PE and PP. They have very limited solubility and are prone to chain scission and/or crosslinking. The current functionalization approach is based on a free radical grafting reaction176±178 that usually starts with a high activation energy step, a hydrogen abstraction reaction produced by breaking a C±H bond in the polymer chain. With coexistence of suitable reagents, such as vinyl monomers, the formed polymeric radicals then react with the reagent and graft vinyl monomers to the polymer chain as illustrated in Eq. (5.1). It is very interesting to note that the acceptance of this post-polymerization approach is strongly related to the reaction processes in polymer melt. The melt process allows the grafting reaction to occur without a solvent.179 In fact, the melt free radical grafting of monomers to polyole®n has become one of the most important applications of reactive extrusion technology. In this chapter, the basic chemistry of the free radical activation mechanism and associated free radical reactions will be discussed in Section 2. The commercial applications of bulk and surface modi®cations will be summarized in Sections 3 and 4, respectively.
66
Functionalization of Polyolefins
R R R R ⱍ ⱍ ⱍ ⱍ CH2CHCH2CHCH2CHCH2CH Hydrogen abstraction
R R R R ⱍ ⱍ ⱍ ⱍ CH2CHCH2CCH2CHCH2CH * nCH2CH ⱍ X
Graft-from reaction
5:1
M:CH2CH ⱍ X
R R R R ⱍ ⱍ ⱍ ⱍ CH2CHCH2CCH2CHCH2CH ⱍ (M)n
2 CHEMICAL MODIFICATION VIA FREE RADICAL REACTION MECHANISM Many high-energy processes were applied in the hydrogen abstraction reaction of polyole®n. Radiation and plasma92,93,180±182 are commonly used in surface modi®cations of polymer ®lms and membranes, and thermal process176±179 with external peroxide is commonly used in polymer bulk modi®cations. In bulk modi®cation, it is usually more demanding to minimize the disturbance of polyole®n structure, because further processing and/or reacting with other polymers, glass ®ber, ®llers, etc., is needed to produce the ®nal product. Equation (5.2) illustrates the general free radical grafting scheme involving thermal decomposition of an organic peroxide (RO±OR) in the presence of polyole®n and functional monomers (M). The alkoxy radical (R±O*) is known to be the most effective and easily formed species for hydrogen abstraction, which is usually prepared by in situ homolytic cleavage of the labile oxygen±oxygen bond of an organic peroxide (RO±OR) at elevated temperatures. Azo initiators, such as 2,2 0 -azobisisobutyronitrile (AIBN), are rarely used in free-radical grafting reactions due to their low hydrogen abstracting capacity. The R±O* radicals thus formed may follow two completely different reaction pathways, with one leading to undesired homopolymer by initiating the polymerization (ki) of the coexisting functional monomers (M), and the other one to the desired hydrogen abstraction (kH) of polyole®n. The ability of R±O* radicals to create polymeric carbon radicals (C*) along the polyole®n backbone is
Functionalization by Post-polymerization Processes
67
essential for free radical grafting reaction. Since the stability of the carbon radical decreases in the order tertiary > secondary > primary, the susceptibility of hydrogen abstraction follows the same trend. Accordingly, among the three most investigated polyole®ns, PP, ethylene±propylene copolymer (EP), and PE, PP is the most vulnerable to attack and to form tertiary C* radicals, and PE is the most resistant to attack and to formation of secondary C* radicals.183,184
ROOR 2RO* R
ⱍ
CH2CH
kH
M ki2
R ⱍ *MCH nM kp2
R ⱍ (M)xCH
M
nM
ki
kp
(M)x Homopolymer
Hydrogen abstraction
R ⱍ CH2C * R β-scission ⱍ ks CCH2 + R ⱍ *CH
M
M: CH2CH ⱍ X ki1
Crosslinking
kc
R ⱍ CH2C ⱍ M * nM kp2
5:2
R ⱍ CH2C ⱍ CCH2 ⱍ R
R ⱍ CH2C ⱍ (M)x
In addition to the desirable grafting (ki1), the polymeric C* radicals face two possible side reactions, chain scission (ks) and crosslinking (kc), and the outcome is strongly dependent on their polymer structures. When the polymeric C* radical reacts with a functional monomer (M) by addition reaction, this monomer is then grafted onto the polymer chain. The free radical may propagate (kp1) to react with more functional monomers and form a longer graft side chain. In other words, this graft-from reaction produces a graft copolymer having a polyole®n backbone and several functional polymer side chains. Crosslinking (kc) is the most undesirable reaction, especially in melt, because it not only consumes the reactive sites, but also
68
Functionalization of Polyolefins
reduces the processibility of the ®nal product. On the other hand, chain scission (ks) reduces the polymer molecular weight and transfers the C* radical to one of the polymer broken chain ends. The terminal polymeric radical then engages in the grafting reaction by initiating (ki2) and propagating (kp2) with functional monomers (M) to produce a diblock copolymer, containing a polyole®n and functional polymer segments. The severe chain scission is always a major concern that can signi®cantly reduce the polymer's mechanical properties, as well as its compatibility with the parent polymer. In other words, the resulting functionalized polymer may only have limited capability as an interfacial agent to promote interactions between polyole®ns and other materials. Along with the above-mentioned free radical reactions, the propagating C* radicals should be completed with various termination and transfer processes. Some may undergo chain transfer with hydrogen atoms of the same or another polymer chain, which form new polymeric C* radicals that repeat the grafting cycle to yield more grafts. It is clear that this free radical grafting process involves a complicated reaction sequence, and the ®nal functionalized polymer structure is very dif®cult to control. The ef®ciency of the free radical grafting reaction and the ®nal polymer structure are determined by several reaction constant ratios. The higher the kH/ki ratio, the more the polymeric C* radicals will be formed and the higher the grafting observed should be. In addition to choosing a free radical initiator with high hydrogen abstraction capability, the functional monomer itself is equally important, which affects not only graft ef®ciency, but also the extent of ungrafted impurities. The functional monomer should have a high ki1/ki or ki2/ki ratio. In other words, the favorable functional monomer should have a higher reactivity toward the C* radical than the R±O* radical. However, the ®nal polymer structure is largely governed by the ks/kc/ki1 ratio. Generally, the secondary C* radicals (in PE cases) have a high kc value and the resulting product has some degree of crosslinking.185 On the other hand, the tertiary C* formed in PP has a high ks value and the resulting functionalized PP has a reduced molecular weight.186,187 In fact, the molecular weight of PP is usually reversibly proportional to the extent of its functionalization. The values of kp1 and kp2 are also important parameters in measuring the incorporation of functional monomers by a polymeric C* radical.
3
BULK FUNCTIONALIZATION BY REACTIVE PROCESS
Various conditions, including solution, suspension, and melt, have been used in the free radical grafting reaction of polyole®ns. For economic and environmental reasons, melt grafting is a highly preferred industrial process for modifying polyole®n bulk. The reaction usually takes place in a batch mixer or a screw extruder by mixing molten polyole®n with a free radical initiator and functional monomers. Three important features in this reactive process are high reaction temperature, high viscosity, and heterogeneity in mixing hydrophobic polyole®n, peroxide, and
Functionalization by Post-polymerization Processes
69
functional polar monomers. These features give rise to tremendous challenges in controlling the reaction process, especially to achieving a homogeneous polymer product with a desirable composition and structure. Owing to the absence of a solvent, an elevated temperature (at least 30 C above the melting temperature of polyole®n) is needed to reduce the polymer melt viscosity and provide suf®cient mixing between the polymer and chemical reagents. The high temperature in turn effects all the anticipated reaction steps, especially the ones (peroxide decomposition and hydrogen abstraction) requiring high activation energies. Generally, the increase in temperature is favorable to the grafting reaction. However, it cannot be ignored that it is possible this will in¯uence the side reactions (shown in Eq. 5.2) and depolymerization188 of the grafted polymer. Some common functional monomers, such as maleic anhydride, have ceiling temperatures close to the grafting temperature. Furthermore, the hal¯ives (t1/2) of most peroxides employed at the grafting temperature ( 200 C) are usually very short, being in the order of minutes or seconds. They are also very sensitive to temperature ± an increase of 5±10 C can reduce the hal¯ife of a peroxide by a factor of two. Obviously, it is very important to understand each reaction step and further optimize the reaction condition to achieve the control of functionalized polymers. Despite numerous attempts, the combination of chemical and physical complications renders systematic studies dif®cult. Many experimental results reported in the patent literature only provide technological information with limited scienti®c details. 3.1
Maleic-anhydride-grafted PP Polymers (PP-MA and PP-SMA)
Although many functional monomers have been used for the melt grafting reaction, the most investigated monomer is maleic anhydride (MA).189 MA-modi®ed polyole®ns are the most widely used commercial functionalized polyole®ns. The double bond of maleic anhydride is reactive to free radicals and has a low tendency to homopolymerize due to a steric reason (low ceiling temperature). Therefore, the homopolymerization184,185 during the melt grafting reaction at elevated temperatures could be minimized. Most importantly, the anhydride group is a very reactive and ef®cient coupling agent190±195 with glass ®bers, ®llers, and functional polymers (such as polyamides, etc.). All these advantages, plus their cost effectiveness, make the MA-modi®ed polyole®ns very attractive and useful commercial products. In the late 1960s, Ide et al.196 performed the pioneering work on the melt grafting of MA onto an isotactic PP in a Brabender mixer using BPO (benzoyl peroxide) or DCP (dicumyl peroxide) as a free radical initiator. They showed that the MA grafting reaction was a very rapid process, and both peroxides gave virtually the same grafting yields at 185 C. However, a relatively high concentration of peroxide, up to 1.8 phr (parts per hundred resin), is needed to reach suf®cient conversion (30±40%) of MA monomers. A signi®cant PP chain degradation was observed with the intrinsic viscosity [] changing from 1.96 dl/g of the virgin PP to a low value of 0.7±0.85 dl/g at 135 C in tetrahydronaphthalene.
70
Functionalization of Polyolefins
Hogt,197 among others,198±200 carried out a similar grafting reaction in a corotating twin screw extruder (screw speed of 200 rpm and temperature 200±240 C), using a DTBPIB initiator (1,3-bis(t-butylperoxyisopropyl)benzene). He found that although an increase in the peroxide concentration increased the MA grafting yield, a generally low grafting MA yield was observed with extensive PP degradation. In addition, the product is usually colored, despite the removal of unreacted monomers and impurities by vacuum ventilation at near the die. The coloration is from the presence of some MA oligomers. It is interesting to note that a higher grafting yield (up to twice) is usually observed in the batch mixer, compared to the extruder, which may be associated with reaction time and oxygen presence in the mixer. The mixer is usually a semi-open system, whereas the extruder is better sealed against air. Lambla and co-workers201 reported a melt grafting reaction using a powdery or porous PP. The polymer was premixed with MA monomer and (2,5-di-t-butylperoxy)-2,5-dimethylhexane (DTBPH with t1/2 at 190 C 1 min) initiator before feeding the mixture into a corotating twin screw extruder. The thought was that higher absorption capacity of liquid monomer/peroxide in powdery or porous PP might facilitate the grafting reaction and also reduce the monomer loss. The grafting reaction was operated between 180 and 200 C with screw speed of 80 rpm for 3 min. As shown in Fig. 5.1, only a very low percentage of MA monomers (< 10%) was grafted onto PP and the resulting MA-modi®ed PP (PP-MA) polymers had the grafted MA (MAg) concentration was below 0.3 phr. Only a very small effect was shown in varying the MA concentration. Although the increase in peroxide concentration increases the MA grafting yield, the polymer signi®cantly reduced its molecular weight. The grafted MAg was usually inversely proportional to the resulting polymer molecular weight. However, when styrene (St) is added as a comonomer, the MA grafting yields were increased signi®cantly even at very low peroxide concentrations (as shown in Fig. 5.1). For example, with 0.05 phr DTBPH charged, 40% of the MA introduced was already grafted onto PP. The MA grafting yield did not level off but increased linearly with increasing MA concentration, regardless of the peroxide concentration. The contribution is attributed to the formation of a charge transfer complex (CTC) between the electron accepting MA and the electron donating St. This complex enhances the initiation of a free radical grafting (polymerization) reaction and results in poly(styrene-altmaleic anhydride) (SMA alternating copolymer).202±206 Almost no notable difference between the mixer and twin screw extruder results was observed in this MA/St grafting reaction. In general, the modi®ed PP-SMA was less degraded with relatively high molecular weight and high MA incorporation. Due to the low level of the incorporated MA units, very little convincing and consistent experimental evidence was reported to describe precise microstructures of PP-MA. The more recent 13C NMR results184 studied by Heinen et al., using 13 C-enriched MA to enhance the NMR signal, offer signi®cant structure information. The predominant graft structure of PP-MA prepared by reactive extrusion consists of a reduced PP polymer chain with several single grafted MA units and a terminal MA unit. On the other hand, the PP-SMA polymer may have a less
Functionalization by Post-polymerization Processes
71
1.6 1.4
[MA]g (phr)
1.2 1.0 MA–St system
0.8 0.6
MA system
0.4 0.2 0.0 1.0
1.5
2.0 [MA]i (phr)
2.5
3.0
Figure 5.1 Grafting yield of [MA]g onto a powdery and stabilized PP as a function of MA concentration with or without St with various amounts of DTBPH. [DTBPH]: * 0.05 phr; & 0.1 phr; ~ 0.2 phr; ! 0.5 phr.
degraded PP chain, and most of the incorporated MA units are in SMA segments located along the PP chain, as illustrated in Scheme 5.1.
PP-MA
PP-SMA
CH3 ⱍ CH2C ⱍ O O O CH3 ⱍ CH2C
PP
O
PP
(S-alt-MA)n
S-alt-MA:
CH3 ⱍ CH2CH O
CH3 ⱍ CH2C (S-alt-MA)n
CH2CH O O O Scheme 5.1
O
72
Functionalization of Polyolefins
3.2
Maleic-Anhydride-Modi®ed PE (PE-MA)
In free radical grafting reactions of PE polymers, the rate constant of each reaction step (in Eq. 5.2) signi®cantly differs from those in PP cases. Crosslinking becomes an important concern in the free-radical-modi®ed PE products. The unstable secondary polymeric radicals in PE (PE*) are much more prone to the free radical coupling reaction between two PE* polymers. In addition, the lower grafting temperature may also encourage the coupling reactions between two PE-MA* propagating radicals and between PE* and PE-MA*. The degree of side reactions is very sensitive to the radical concentration and reaction temperature. In the late 1960s, Porejko et al.207 reported the pioneering work on a free radical MA grafting reaction of LDPE polymers in dilute (1% polymer) xylene solution at 110 C using a BPO initiator. They observed high concentration (> 40%) of MA units incorporated into partially crosslinked PE-MA products, and the intensity of brown color (due to MA oligomers) in the PE-MA polymer is proportional to the MA concentration. Basically, the grafting reaction is consistent with the MA graft-from oligomerization184 accompanied with crosslinking (Eq. 5.2). The MA oligomer chains are formed and located along the PE backbone as illustrated in Scheme 5.2.
O
O
O PE-MA O n
O
O Scheme 5.2
Higher reaction temperatures (> 110 C), closer to the ceiling temperature of MA polymerization (150 C at [MA] 5M), rapidly prevented the oligomerization of MA monomers and reduced the MA incorporation. Gaylord and Mehta208 studied the melt grafting process of PE with an MA monomer (10 wt% of PE) and a free radical initiator (2 wt% of PE) in a Brabender at 140 or 180 C. Considerable crosslinking occurred when MA/BPO mixture was added to the molten LDPE at 140 C. The crosslinking reaction was reduced by introducing an electron donating compound, such as dimethylformamide (DMF), into the reaction mixture. DMF shows good reactivity with a propagating PE-MA* radical, and therefore reduces the propagating radicals and free radical coupling reactions. In fact, crosslinking was largely eliminated with the presence of a signi®cantly high concentration of DMF ([BPO]/[DMF] 5/3). However, the same free radical reduction process also reduces the incorporation of MA units into the
Functionalization by Post-polymerization Processes
73
PE chain. The resulting PE-MA contains only 0.3 wt% of incorporated MA units. In general, there is very little detailed information on the kinetics of this melt grafting reaction and the resulting PE-MA structure, due primarily to the low concentration of incorporated MA units.
3.3
Maleic-Anhydride-Modi®ed EP (EP-MA)
Since the ethylene±propylene copolymer (EPR) contains a nearly 1/1 ratio of ethylene and propylene units in the polymer chain, both crosslinking and degradation should occur in the free radical grafting reaction. The few competing reactions will produce a mixed product containing both soluble and insoluble materials, and the ratio of the two portions should be very dependent on the reaction conditions. Many experimental conditions,209 including initiator, temperature, procedures, etc., were studied to determine the extents of crosslinking and degradation that accompany the grafting reaction. A typical example was produced by Gaylord210 by mixing EPR with dicumyl peroxide (DCP) and MA monomer in a Brabender mixer at 180 C for 10 min. Some experimental results are summarized in Table 5.1. In general, both the MA grafting ef®ciency and content were very low, similar to those in PP cases. The percentage of the insoluble (crosslinked) portion is very sensitive to the reaction condition. In the presence of 5 wt% MA (based on EPR), the increase of DCP (0.25±1.0 wt% to EPR) decreased the amount of cyclohexane-insoluble gel from 65 to 27%. The soluble portion, containing about 1 wt% MA units, also reduced its intrinsic viscosity []. The hydrogen abstraction at the facile tertiary C±H and subsequent chain degradation in propylene units appear to increase with the peroxide concentration. When the DCP concentration was kept constant at 0.5 wt% while the charged MA concentration increased from 5 to 20 wt%, the amount of cyclohexane-insoluble polymer decreased from 43 to 13%. The [] of the cyclohexane-soluble EP-MA increased. Higher MA concentration clearly enhanced grafting reaction and reduced both chain degradation and crosslinking reactions. Table 5.1
Summary of maleic-anhydride-modified EP-MA products
Reaction conditions1 MA (wt%) DCP (wt%) 5 5 5 10 20 1
0.25 0.5 1.0 0.5 0.5
Cyclohexane-soluble (%)
[] (dl/g)2
33 56 73 66 76
1.10 0.56 0.89 1.26 1.56
Starting EPR [] 1.42 dl/g. Toluene, 30 C.
2
MA (wt%) Insoluble (%) 0.9 1.0 1.3 0.7 0.6
65 43 27 43 13
74
Functionalization of Polyolefins
Maleic-anhydride-modi®ed EP (EP-MA) can also be prepared by ene reaction211 by reacting EPDM with MA (without initiator) at high temperature, as illustrated in Eq. 5.3:
O
ⱍ
CH2 EP
O
O O
CH2CH ⱍ CH2 ⱍ CH O ⱍ HC O ⱍ CH3 O
5:3
O
+
CH2CH ⱍ CH2 ⱍ HC ⱍ CH
ene reaction
CH ⱍ CH3 + O
EP
CH2CH ⱍ CH2 ⱍ CH
EP
The reaction takes place with the mechanism corresponding to a concerted electronic-transfer type reaction without any free radical species. The graft ef®ciency and MA content usually are lower than those seen from the free radical process. However, the undesirable crosslinking and chain scission can be largely prevented in the EP-MA product.
4
SURFACE FUNCTIONALIZATION
Surface modi®cation of polyole®ns (®lm, sheet, ®ber, and membranes) has attracted great interest, since it presents a simple and effective process (economic in many cases) to improve surface properties (such as adhesion, printability, biocompatibility, wettability, hydrophilicity, weatherability, etc.). Ideally, the reaction only takes place on the outer surface without altering the bulk properties of the polymer.92,182,212 The depth of the modi®ed layer is very dependent on the reaction processes and conditions. In most applications, the modi®ed polyole®n (namely, PE and PP) is directly used as the product or as a laminated layer in a composite structure. In other words, no further processing (by solution or melt) is necessary. Therefore, the reduction of processibility (due to crosslinking) during the surface modi®cation is not a major concern. However, the polymer chain degradation resulting in the loss of mechanical strength on surface and/or in bulk is a major issue. Chemical grafting of functional monomers (acrylic and methacrylic)213±218 or polar reagents (halogen, oxygen, halogens, ozone, etc.)219±222 onto the polyole®n surface is the most effective and widely studied method. Most surface grafting
Functionalization by Post-polymerization Processes
75
reactions follow a similar reaction mechanism (free radical activation then grafting), shown in Eq. (5.2). The major differences from bulk cases are the reaction conditions, because surface reaction is less concerned in chemical mixing and crosslinking reactions. Most surface reactions allow a broad choice of activation (hydrogen abstraction) methods and chemical reagent delivery systems. The hydrogen abstraction can be achieved by many free radical formation methods, including chemical, radiation, plasma, ¯ame, etc. Chemical reagents are introduced to the polyole®n surfaces by vapor or solution absorption processes. The gas phase procedures are largely preferred because of the ease of handling and disposing of a smaller quantity of reagents. 4.1
Chemical Process
Many organic reactions of alkanes have been applied to the chemical modi®cations of PE and PP surfaces. One of the most documented methods is an oxidation reaction using various soluble chromium and manganese oxidants,223±225 such as chromium trioxide in sulfuric acid and potassium permanganate in sulfuric acid. The principle functional groups formed on the surface are carboxylic acids and ketones. However, the use of toxic reagents and the need to dispose of the acid solution containing a transition metal make these oxidation reactions dif®cult to accept in commercial applications. Free radical grafting reactions of PE and PP ®lms and ®bers in solution phase were also studied by many researchers.226±229 A typical example from Gabara and Porejko230 is the MA grafting reaction of a LDPE ®lm in an acetic anhydride solvent using BPO as the initiator. The reaction was carried out in air at temperatures of 80±110 C. As expected, a large amount of ungrafted MA homopolymer is formed in solution as well as on the ®lm surface. After the ®lm was extracted with boiling acetone to remove the ungrafted homopolymer, a signi®cant weight increase of the brown ®lm indicates the formation of a grafted MA polymer. The MA-modi®ed PE ®lms are crosslinked ± even those ®lms containing a low percentage of MA. 4.2
Radiation and Photo Processes
A radiation-induced free radical grafting reaction has also been widely used. Highenergy radiations, such as electron beam and gamma radiation, have been applied to initiate functionalization reactions. Usually, high-energy sources penetrate deep into the polymer bulk where there is no or very low concentration of functional monomers or polar reagents. Crosslinking and degradation in bulk, as well as the ungrafted homopolymer on the surface, are major disadvantages. To minimize homopolymerization, polyole®n was pre-irradiated in vacuum before introducing monomers. For example, acrylonitrile was grafted onto PP by addition of the monomer to a -radiated PP ®lm.231 The extent of grafting was proportional to the
76
Functionalization of Polyolefins
trapped polymeric radicals, and it decreased with the increase of the time between the irradiation and the monomer exposure. PP ®ber was also grafted with methacrylates232 after -radiation. The ®rst was a liquid phase grafting produced by immersing irradiated PP ®ber into the liquid monomer. Butyl and propyl methacrylates were grafted to a very high extent. Second was a gas phase grafting produced by exposing the irradiated PP sample to the vapor of the monomer; the grafting rate varied linearly with the vapor pressure of the monomer. The results showed that the rate of grafting was much less for vapor-exposed PP than for liquidimmersed PP, due to good swelling in the liquid phase. The pre-irradiation process was extended to many other grafting reactions215,233±237 in which various monomers, such as maleic anhydride, acrylic acid, acrylamide, methyl methacrylate, vinyl chloride, and vinyl acetate, were used in the surface modi®cation of PE and PP ®lms and ®bers. Odor238 proved by X-ray diffraction studies that the grafting process of vinyl monomers onto PP ®bers mainly took place in the amorphous phase. Low-energy UV radiation was also used in the free radical grafting reactions. The modi®cations are mainly limited to the surface of the substrate. Usually, the simple exposure of PE and PP ®lms to UV light in the presence of a vinyl monomer requires high dosages and short wavelengths (< 300 nm) to produce enough polymeric radicals for a grafting reaction. One way to improve the low quantum yield is to add a photosensitizer that absorbs the light and transfers the energy to activate the polymer surface. Tazuke and Kimura239 reported a successful grafting reaction by UV light on the surfaces of PE and PP ®lms that were precontacted with a solution containing a photosensitizer (such as benzophenone) and functional monomers (acrylic acid, acrylamide, etc). Although the surface of the PE or PP ®lm was completely covered with functional polymers, a large portion of them was ungrafted homopolymer. 4.3
Corona and Plasma Processes
Surface modi®cation of the biaxially drawn PP ®lm by corona (electric discharge under atmosphere pressure) treatment240 is a successful commercial process giving good ®lm adhesion and durability. On the other hand, cold plasma (an electric glow discharge under reduced atmospheric pressure) treatment provides a versatile method for producing polymer surfaces with low and high surface energy, depending mostly on the choice of treatment conditions. Table 5.2 summarizes some experimental results182 using various cold plasmas (O2, CO, CO2, N2, NO, NO2, CF4, and SF6). The contact angle with the distilled water varied widely (from near 0 to more than 100 ) on the plasma-treated samples. Carbon dioxide plasma produces the surface with the highest wettability, and CF4 plasma offers the surface with the lowest surface tension. The ionized gas, composed of positively charged species, free radicals, and electrons, is very reactive to PE and PP chains, namely starting with hydrogen abstraction to form polymeric radicals and then moving on to grafting reaction,
Functionalization by Post-polymerization Processes
77
Table 5.2 Effects of plasma treatments on hydrophilicity of PE and PP surfaces Contact angle in distilled water ( ) Polymer
Plasma
Before treatment
After treatment
PE PE PE PE PE PE PE PE PP PP PP PP PP
CF4 SF6 NO2 N2 O2 NO CO CO2 CF4 SF6 N2 CO2 O2
103 103 102 102 102 102 102 102 104 104 94 94 95
121 117 52 48 46 28 16 8 123 117 50 45 23
degradation, crosslinking, transfer reaction, etc., discussed in Eq. (5.2). In the oxygen-derivative gases, the major functional groups on surface are acid, ester, and ether. In the nitrogen-derivative gases, major nitrogen functional groups are amino, amide, nitrile, and imide groups. In the ¯uorine-derivative gases, the active F atoms exchange with H atoms on the polymer surface. Some plasma treatments,241±246 especially involving inert gases (N2, Ar), are used to generate polymeric radicals on PE and PP surfaces ± a similar objective to that of the pre-irradiation process. The formed radicals then allow contact with the vinyl monomers (such as acrylic acid) by either vapor absorption or dipping into a monomer solution. The subsequent graft-from polymerization achieves a high content of functional groups on the surface.
5
SUMMARY
Functionalization of commercial polyole®ns (PE, PP, and EPR) by postpolymerization processes has been extensively studied in both melt (bulk) and solid (surface) conditions. In general, the surface modi®cations provide some improved properties of polyole®n ®lm and ®ber products. The modi®cation of bulk polymers has faced great challenges with only limited success. The high-energy functionalization processes have to be dealt with by carefully balancing the needed functionalization and the undesirable side reactions. In addition, the combination of chemical and physical dif®culties makes it very hard to understand each reaction in detail.
78
Functionalization of Polyolefins
Most of the studied functionalization reactions involve the free radical reaction mechanism (Eq. 5.2), with a very dif®cult step of hydrogen abstraction by removing hydrogen from a completely saturated polymer backbone. The high activation energy of this C±H bond breaking reaction requires severe reaction conditions, such as high temperature, radiation, and plasmas, and is usually not a clean and selective reaction. Multiple free radical species (other than polymeric radicals) usually coexist in the system and inevitably produce many undesirable impurities (including ungrafted homopolymers), which not only reduce the overall grafting ef®ciency, but also have negative effects to the product. Unfortunately, some of them are also dif®cult to remove by simple puri®cation processes. The other major concern in the grafting reaction is the instability of the formed polymeric radicals. Many comperative side reactions, namely degradation of the polymer backbone and/ or crosslinking between the polymer chains, can easily alter the basic polymer structure. It is very clear that a new technology is needed to prepare functionalized polyole®ns with good control of the reaction products. As will be discussed in the following chapters, a very effective functionalization process has been discovered in our laboratory by using ``reactive'' polyole®ns instead of the inert PE and PP homopolymers. The reactive groups, such as borane, p-methylstyrene, and dienes, located along the polyole®n backbone or at the chain end, provide the selective sites for functionalization reactions under mild reaction conditions. A broad range of functionalized polyole®ns, including block and graft copolymers, have been prepared and will be discussed later. Some of them may easily replace the inert PE, PP, and EPR in current commercial functionalization processes.
Reactive Copolymer Approach
This Page Intentionally Left Blank
6 Functionalization via Reactive Polyole®ns Containing Borane Groups
1
INTRODUCTION
As discussed in Chapters 3, 4, and 5, both direct and post-polymerization processes achieve only partial success, with many limitations. It is clear that there is a fundamental need for developing a new chemistry that can address the challenge, especially by using a commercially acceptable process to prepare functional polyole®ns with desirable functional groups and molecular structure. The approach of ``reactive'' polyole®ns (brie¯y discussed in Chapter 2, Section 4) may provide an attractive alternative. Basically, the functionalization approach combines the advantages of both direct and post-polymerization processes (and avoids their disadvantages). The reaction involves two steps, starting with direct copolymerization of -ole®n and a comonomer containing a ``reactive'' group that is completely stable with the catalytic site, and that can be effectively incorporated into polyole®n by existing commercial processes. In turn, the formed polyole®n contains ``reactive'' sites that can be selectively interconverted to the desirable functional groups under mild reaction conditions. The second step may also be accomplished during the melt process (reactive extrusion). Several ``reactive'' comonomers have been identi®ed. In this chapter, we will focus on the reactive polyole®n containing borane groups. Three major advantages of the borane approach are: (a) The stability of borane groups to Ziegler±Natta and metallocene catalysts. Most trialkylboranes are acidic and do not interact with either the acidic catalyst or the cocatalyst. No protection reaction is required before polymerization. (b) The solubility of the borane monomers and polymers in the hydrocarbon solvents (hexane and toluene) used in Ziegler±Natta and metallocene polymerization. Soluble monomers and propagating polymer chains are essential to obtain a high-molecular-weight polymer and to assure the optimum copolymerization conditions.
82
Functionalization of Polyolefins
(c) The versatility of borane groups. The borane groups incorporated into the polyole®n are very reactive in comparison to their corresponding organoborane molecules. Many borane conversion reactions developed by Brown123 can be directly applied to the polymer cases.
R
I2/NaOH
R
ⱍ (CH2CH)x(CH2CH)y ⱍ (CH2)n ⱍ I
R
ⱍ ⱍ H O /NaOH (CH2CH)x(CH2CH)y 2 2 (CH2CH)x(CH2CH)y ⱍ ⱍ (CH2)n (CH2)n H2NCl ⱍ ⱍ (I) B O ⱍ H
R
O
O
O * * O
R
ⱍ (CH2CH)x(CH2CH)y ⱍ (CH2)n ⱍ
O ⱍ
MA + S
O
(II)
O
ⱍ
B
O
ⱍ (CH2CH)x(CH2CH)y ⱍ (CH2)n ⱍ
R
O
O2
ⱍ (CH2CH)x(CH2CH)y ⱍ (CH2)n ⱍ NH2
R
MMA R
ⱍ (CH2CH)x(CH2CH)y ⱍ (CH2)n ⱍ
O
PMMA
ⱍ (CH2CH)x(CH2CH)y ⱍ (CH2)n ⱍ
O
P(S-alt-MA) MMA: methyl methacrylate PMMA: poly(methyl methacrylate) S: styrene P(S-alt-MA): Styrene/maleic anhydride alternating copolymer
6:1
Functionalization via Reactive Polyolefins
83
As illustrated in Eq. (6.1), many polyole®ns containing functional groups (such as hydroxy and halides) have been prepared via the interconversion of boranecontaining polymers (I). This process is particularly suitable in the preparation of hydroxylated polymers, due to the quantitative conversion under mild reaction conditions. In addition, the borane groups in the polyole®n (I) can also be quantitatively oxidized by oxygen at ambient temperature to form polymeric radicals (II) that very effectively react with ole®nic molecules, especially the ones containing polar groups.
2
BORANE-CONTAINING POLYOLEFINS
As shown in Chapter 3, a whole composition range of borane-containing polymers ± from thermoplastic (Tm similar to those of HDPE and i-PP) to elastomer (Tg < 55 C) ± have been prepared by the copolymerization process using commercially available transition metal catalysts. The commonly used borane moiety is 9-BBN, due to its availability in large quantity. In addition, the 9-BBN-containing monomers and copolymers are stable for a long period of time (6 months in dry-box) or at elevated temperatures (>200 C) as long as O2 is excluded. Furthermore, 9-BBN provides an asymmetric trialkylborane structure that is crucial in many selective functionalization reactions to achieve a complete interconversion. PE copolymers were usually prepared by metallocene catalysts with constrained ligand geometry. The resulting PE-B copolymer has narrow molecular weight and composition distributions. On the other hand, in the preparation of boranecontaining PP, PB (polybutene) and PO (polyoctene) most of the copolymerization reactions were carried out by Ziegler±Natta catalysts due to the better isotactic polymer structure (in the PP and PB cases) and favorable comonomer incorporation with good yield (in the PO case). In addition, some borane-containing polymers can also be prepared by hydroboration reaction of the unsaturated polymers, including commercial EPDM and butyl rubber, with dialkylborane reagents. The detailed results will be discussed in Chapter 8.
3
HYDROXYLATION REACTIONS
As shown by Brown,123 organoborane is one of the most versatile reagents in organic synthesis. With it, many functional groups can be obtained in high yields under mild reaction conditions ± the most effective being a hydroxylation reaction by NaOH/H2O2. In general, the reactivity of the borane group in polyole®n resembles that of the corresponding organoborane molecule. The ¯exible sidechain-bearing borane group, combining the excellent reactivity of the borane group and a good physical af®nity between the borane and the oxidation reagent, provides
84
Functionalization of Polyolefins
favorable reaction conditions in both homogeneous and heterogeneous conditions. Several functional polyole®ns obtained from this route are discussed below. 3.1
Poly(hexen-6-ol) and Poly(octen-8-ol)65
The borane-containing homopolymers, including poly(hexenyl-9-BBN) and poly(octenyl-9-BBN), discussed in Section 3.4, were converted into the corresponding poly(hexen-6-ol) and poly(octen-8-ol) polymers by reacting with NaOH/ H2O2 in THF solvent at 50 C for 2 h. During the oxidation reaction, the solution changes from homogeneous to heterogeneous due to the insolubility of hydroxyated polymer in THF. However, both poly(hexen-6-ol) and poly(octen-8-ol) are soluble in 1-propanol at room temperature. It is very interesting to note that these polyalcohols are isotactic polymers with a comb-like molecular structure. The X-ray diffraction pattern (Fig. 6.1) of poly(octen-8-ol) shows a strong re¯ection at Bragg angle 2 20 , corresponding to Ê . The spacing between parallel side chains, analogous to spacing (d) of 4.4 A the re¯ection maximum on X-ray patterns of n-paraf®ns, indicates a helix polymer chain arrangement in the crystalline domains. The crystalline structure is also revealed by DSC measurement with a strong endotherm peak at 110 C. Furthermore, the polymer prepared has a primary alcohol at the end of each side chain. In addition to the good mobility of this polar group ± very desirable for interacting with other materials ± the primary hydroxy group is very stable up to very high temperatures. The dehydration temperature (> 300 C) is well above the processing temperature ( 200 C), and is very different from that (170 C) of poly(vinyl (a)
150 (b)
Intensity (cps) × 10
120 (CH2)6 (CH2)6 (CH2)6 ⱍ ⱍ ⱍ O O O ⱍ ⱍ ⱍ H H H isotactic poly(1-octen-8-ol)
90
60 A1
A1 A1
30
0 0
7
14
21
28
35
42
49
56
63
70
Two-theta (degrees)
Figure 6.1 (a) X-ray diffraction of poly(1-octen-8-ol). (Redrawn from Macromolecules 1988 21, 865. Copyright 1988 Am. Chem. Soc.) (b) Structure.
Functionalization via Reactive Polyolefins
85
alcohol), which is decomposed during the polymer processing. The poor thermostability of poly(vinyl alcohol) is namely due to the dehydration of the secondary alcohols and the driving force favorable to form a conjugated structure. 3.2
Polyethylene Containing OH Groups (PE-OH)128,247
The borane groups (both primary and secondary) in polyethylene were completely converted to the corresponding hydroxy groups by reacting with NaOH/H2O2 reagents at 40 C for 3 h. The reaction was carried out by suspending boranecontaining polyethylene (PE-B) powders in a solution, such as toluene. In addition to the reactivity and mobility of the borane groups, their effective interconversion in heterogeneous phase must be due to their high surface area. In the semicrystalline PE structure, the ¯exible borane-bearing side chains must position themselves in the amorphous phases or on the surfaces of particles, which allows effective contact with the reaction reagents. The hydroxylation reaction usually doesn't show any detectable side reaction. The same high polymers with narrow molecular weight and composition distributions were observed in all metallocene-prepared polymers. It is very interesting to note that the hydroxylated polyethylene (PE-OH) prepared from PE-B (I) is structurally similar to that of linear low-density polyethylene (LLDPE) as shown in Scheme 6.1. (CH2CH)x(CH2CH)y ⱍ CH2 ⱍ CH2 ⱍ CH2 ⱍ LLDPE CH3
(CH2CH)x(CH2CH)y ⱍ CH2 ⱍ CH2 ⱍ CH2 ⱍ LLDPE-OH CH2 ⱍ O ⱍ H
Scheme 6.1
The only difference is an OH end group located at each side chain. GPC and DSC curves of PE-OH copolymers, obtained from a homogeneous Et(Ind)2ZrCl2/MAO catalyst system, show narrow molecular weight and composition distributions. Both melting point (Tm) and crystallinity (c) of the copolymer (summarized in Table 6.1) decrease with the side chains. The higher the density of side chain, the lower the Tm and c. The PE-OH sample (A-4) containing 2.3 mol% of 1-hexen-6-ol, with an average of one side chain per 43 ethylene units, only possesses 23.3% crystallinity (less than half of PE) with Tm 105.9 C (30 C lower than PE). The crystallinity
86
Functionalization of Polyolefins Table 6.1 A summary of thermal properties and crystallinity of poly(ethylene-co-1-hexen-6-ol) copolymers Sample no.
1-hexen-6-ol (mol%)
Tm ( C)
Hf (66 cal/g)
c1 (%)
Control A-1 A-2 A-3 A-4
0 1.25 1.70 2.15 2.30
136.8 119.0 116.7 106.1 105.9
35.6 22.3 21.5 16.5 15.4
53.9 33.8 32.5 24.9 23.3
1 c (Hf/Hf ) 100, where Hf and Hf (66 cal/g) are the fusion enthalpies of copolymer and fold-chain polyethylene, respectively.
(c) of the copolymer was calculated from fusion enthalpy according to the equation given in Table 6.1.
3.3
Polypropylene Containing OH Groups (PP-OH)129,130,248
It is interesting to extend this functionalization chemistry to other polyole®ns, polypropylene, poly(1-butene), and poly(1-octene). In the borane-containing PP and poly(1-butene) cases, the oxidation reaction conditions were carried out in heterogeneous reaction conditions, similar to that of the PE-B case. On the other hand, the conversion of borane-containing poly(1-octene) copolymer (amorphous) was carried out in a homogeneous solution using a good solvent, such as THF. All reactions were completed under mild reaction conditions, which may be due to the excellent reactivity of borane moieties and the good af®nity between borane (acid) and NaOH/ H2O2 reagent (base). As discussed in Chapter 3, the borane-containing polypropylene, poly(1-butene), and poly(1-octene) are usually prepared by the commercially available heterogeneous Ziegler±Natta catalysts, such as the isospeci®c TiCl3 AA/Et2AlCl catalyst. The borane monomer behaves like a high -ole®n in the Ziegler±Natta polymerization ± the bigger the size of the monomer, the lower the reactivity. The monomer reactivity follows the trend of propene > 1-butene > 1-octene 5hexenyl-9-BBN. In a batch reaction of propylene and 5-hexenyl-9-BBN (see Section 5.2), with periodic addition of propene monomer to account for the consumption of the borane monomer, the resulting isotactic PP-B copolymer has a unique ``brush-like'' structure. After oxidation reaction of the borane groups, the resulting PP-OH copolymer inherits the ``brush-like'' structure, as shown in Scheme 6.2. Due to the signi®cant difference in comonomer reactivity ratios and the ¯uctuation of propylene concentration during the copolymerization reaction, the side
Functionalization via Reactive Polyolefins
87
CH3 CH3 CH3 CH3 CH3 ⱍ ⱍ ⱍ ⱍ ⱍ CH2CHCH2CHCH2CHCH2CHCH2CHCH2CHCH2CHCH2CH ⱍ ⱍ ⱍ (CH2)4 (CH2)4 (CH2)4 ⱍ ⱍ ⱍ O O O ⱍ ⱍ ⱍ F F H H H
F
F F
F F F
F
PP crystalline domain F: OH
Scheme 6.2
chains (with functional groups) in the polymer are concentrated at the polymer chain end. The functionalized PP with the ``brush-like'' structure has many interesting physical properties. This type of polymer structure not only possesses a desirable concentration of functional groups with good mobility and high surface area, but also preserves the original physical properties ± such as crystallinity, melting point and thermal stability ± of pure PP. Figure 6.2 compares the DSC curves of PP-OH polymers with PP. Almost the same melting temperature was observed, even in the PP-OH containing 3 mol% of bulky functional groups. The PP-OH with a long polypropylene sequence is essential for many applications, especially the surface modi®cation of PP ®lms and particles. PP-OH polymer has to be able to secure both interfaces, including PP/PP-OH and PP-OH/substrate. The PP sequence in PP-OH can cocrystallize with pure PP, and the mobile functional group located at the end of each side chain preferentially moves away from the hydrophobic PP crystalline matrix to the surfaces, which can then interact with the substrate by polar group interactions. The cocrystallization of PP and PP-OH was revealed by DSC analysis, as shown in Fig. 6.3. A PP ®lm and PP-OH copolymer were melted together and the DSC endotherm was observed over the normal melting range of these two components. Upon cooling, crystallization occurred at a temperature of approximately 110±130 C, and the peak's shape and location seems to be the net result of individual PP and PP-OH crystallization. However, when the sample was reheated after cooling, cocrystallization of PP and PP-OH was shown by the presence of a single melting peak intermediate below the melting region of PP, as shown in Fig. 6.3(b). The other study directly used the laminated sample containing PP and PP-OH layers. As shown in Fig. 6.3(c), the melting endotherm of the PP/PP-OH laminate sample consists of three peaks. The peak in the middle is at approximately the same position as the previously identi®ed cocrystallization peak in Fig. 6.3(b), whereas the other two peaks are indicative of pure PP and PP-OH. Accordingly, cocrystallization occurs at the PP/PP-OH interface in the laminate. From the polymer processing viewpoint, it is very important to know the thermal stability of the ``brush-like'' PP-OH copolymer. Figure 6.4 compares the TGA
88
Functionalization of Polyolefins
14
Heat flow (mW)
12 10 (a) 8 6
(b)
4
(c)
2 0 –25
0
25
50
75 100 Temperature (°C)
125
150
175
2.0
2.0
1.8
1.8
1.6
1.6
1.4
1.4
(a)
1.2
1.2
1.0
1.0 Cocrystallization
0.8
0.8
(b)
0.6
0.6 0.4
0.4 (c)
0.2 0.0 100
Figure 6.3 laminate.
Heat flow (W/g)
Heat flow (W/g)
Figure 6.2 DSC curves of (a) PP and two PP-OH polymers containing (b) 0.5 and (c) 3 mol% of OH groups.
110
120
0.2 130
140 150 Temperature (˚C)
160
170
0.0 180
DSC curves of (a) PP, (b) PP/PP-OH blend, and (c) PP/PP-OH
curves of PP-OH with 5 mol% hydroxy groups and two homopolymers, isotactic PP and poly(1-hexen-6-ol), under O2 and argon atmospheres. The thermal stability of PP-OH is higher than that of pure i-PP, which may be attributed to the high thermal stability of poly(1-hexen-6-ol) due to the primary hydroxy groups. The combination of higher thermal stability and similar melting temperature in PP-OH, compared to those of i-PP, implies that the similar processing conditions used in the i-PP case can be applied to PP-OH.
Functionalization via Reactive Polyolefins
89
100
Weight %
(c) (b) (c) 50
(a)
(a)
(b)
0 40 90 140 190 240 290 340 390 410 490
40 90 140 190 240 290 340 390 410 490
Temperature (°C)
Figure 6.4 TGA curves of (a) isotactic PP, (b) PP-OH with 5 mol% hydroxy groups, and (c) poly(1-hexen-6-ol), under O2 (left) and argon (right) atmospheres. (Redrawn from Macromolecules 1993, 26, 3019. Copyright 1993 Am. Chem. Soc.)
3.4
Poly(1-octene) Containing OH Groups (PO-OH)249
Due to the similar reactivity of borane monomer (B-5-hexenyl-9-BBN) and 1-octene in Ziegler±Natta polymerization using heterogeneous TiCl3 AA catalyst, the composition of poly(1-octene-co-1-hexen-6-ol) is basically controlled by the comonomer feed ratio. High OH content in poly(1-octene) copolymer can be easily prepared with narrow composition distribution. Figure 6.5 compares IR and DSC curves of three copolymers containing 25 mol% (PO-OH-31), 50 mol% (PO-OH-11), and 75 mol% (PO-OH-13) of 1-hexen-6-ol, and two homopolymers, poly(1-octene) and poly(1-hexen-6-ol). Only one Tg transition in each sample is observed, which reveals the absence of macroscopic phase separation, and therefore implies that the copolymer samples are fairly homogeneous. In fact, a linear relationship between the Tg and the weight fraction of either monomer has been taken as a qualitative indication of the homogeneity and random nature of the copolymer samples. 3.5
EPDM Containing OH Groups (EP-OH)250
The most convenient route in the preparation of EP-OH is to start the hydroboration reaction with a commercial EPDM elastomer containing a few mol% of diene (1,4hexadiene or 5-ethylidene-2-norbornene) adducts. The hydroboration reaction is made very effective by adding 9-BBN into the EPDM/THF solution. With a suf®cient amount of 9-BBN, all double bonds in the EPDM can be completely reacted at room temperature to form hydroborated polymer. In turn, the incorporated borane
90
Functionalization of Polyolefins
(a)
(b) (e) (c) Endo >
(d) (d) (c) (e)
(b) (a)
–125 –100 –75
–50
–25
Temperature (°C)
0
25
50 4000 3500 3000 2500 2000 18001600 14001200 1000 800 Wavenumber (cm–1)
Figure 6.5 Comparisons of DSC curves (left) and IR spectra (right) and of the homopolymers and copolymers (a) poly(1-octene), (b) PO-OH-31, (c) PO-OH-11, (d) PO-OH-13, and (e) poly(1-hexen-6-ol). (Redrawn from Macromolecules 1990, 23, 378. Copyright 1990 Am. Chem. Soc.)
groups in the EPDM are very easily converted to OH groups under mild reaction conditions, similar to PO-OH cases. Two common EP-OH polymers are shown in Scheme 6.3. (CH2CH)x(CH2CH)y(CH2CH)z (CH2CH)x(CH2CH)y ( ⱍ CH2 ⱍ CH2 ⱍ HCOH ⱍ CH3
)z ⱍ
O ⱍ H
Scheme 6.3
Both hydroboration and oxidation reactions are very clean ± the same polymer molecular weight and molecular weight distribution are observed before and after reactions. Since the OH concentration, dictated by unsaturation content, is very low (a few percent) in the polymer, the EP-OH polymer exhibits similar low Tg and elastic properties.
Functionalization via Reactive Polyolefins
4
91
SELECTIVE AUTO-OXIDATION REACTION OF ALKYL-9-BBN251,252
As discussed above, the incorporated borane groups in the polyole®n serve as the reactive sites for selective functionalization under mild reaction conditions. Instead of the conversion of one borane to one functional group, it is highly desirable to convert the borane group to a polymeric free radical species that can initiate free radical graft-from reaction of the functional monomers, such as MA and MMA, shown in Eq. (6.1). Such a reaction scheme dramatically increases the ef®ciency and versatility of the borane group. The functional group introduced to the polymer is predetermined by the monomer that is used. In other words, a whole range of functionalized polyole®ns (PE, PP, EP, etc.) containing various functional groups, such as acid, ester, amine, amide, and ether, and a high concentration of functional groups can be prepared by this reaction mechanism. In addition, a single borane group can incorporate thousands of functional groups into a polyole®n. The key step lies in the interconversion of the borane group to a polymeric radical. Although the free radical polymerization by the oxidation adducts of trialkylborane ± such as triethylborane and tributylborane253±255 with oxygen at room temperature ± were reported in the late 1950s, the experimental results showed only a very low percentage of borane groups participating in the polymerization reactions. The general sense gained from the results was that the borane oxidation reaction256±258 was very complicated and led to many intermediates being produced. During the application of this borane/O2 free radical initiation system to graft-from reactions, it is essential to have a selective and clear oxidation reaction at the B-alkyl group that is part of polymer side chain. In our laboratory, we have systematically investigated the selective auto-oxidation reaction251,252 of asymmetric borane moieties, including small ethyl-9-BBN molecule and its corresponding 9-BBN containing polypropylene (PP-9-BBN). The comparative oxidation reactions were monitored by 13C, 11B nuclear magnetic resonance (NMR), and electron spin resonance (ESR) techniques. It was concluded that the oxygen insertion into alkyl-9-BBN prefers the B-C linear alkyl group to the B-C bicyclic ring as shown in Eq. (6.2):
CB
O2
CB
COOB
(I")
COB (III)
(II) COB
(IV)
C O
2
B
6:2
+ HOB
The stable double chair-form structure may prevent an unfavorable ring strain increase due to the insertion of a peroxyl group into the chair-form structure.
92
Functionalization of Polyolefins
However, in the small ethyl-9-BBN molecule case, the unstable peroxylborane (II) that is formed engages in further intermolecular reactions to form complicated products, including some double- and triple-oxidized adducts, i.e. BO2R and BO3 species. The same intermolecular reactions were largely prevented in the PP-9-BBN case due to the presence of isolated borane in the PP solid matrix. In the presence of free radical polymerizable monomers (MMA, maleic anhydride, etc.), the peroxylborane (II) that is formed reacts with monomers at room temperature and initiates the free radical polymerization. ESR measurement provides direct evidence of the radical species. As shown in Fig. 6.6, a singlet ESR signal was observed in the O2-oxidized PP-9BBN sample, which has the g-value of 2.0155, corresponding to peroxy (ROO*) or alkoxy (RO*) radicals. The radical appears to be very stable for many days. Upon addition of the MMA monomers, the ESR signal changes to a hyper®ne splitting pattern, indicative of a propagating PMMA carbon radical (R*). Because the peroxyl radical (ROO*) is incapable of initiating the free radical polymerization, the stable radical formed in the O2 oxidation of PP-9-BBN must be the alkoxyl radical (RO*). It is very interesting to note that the homolytic cleavage of peroxyborane may be enhanced by the formation of a stable borinate radical (B-O*), as illustrated in Scheme 6.4. Compared with the known stable nitroxide radicals, such as the 2,2,6,6-tetramethyl-1-piperidinyloxy (TEMPO) radical,259 in which the radical is stabilized [⫻103]
[⫻103]
4
6 4
2 2 0
0 –2
–2 –4 –4
–6
–8
–6 3275 (a)
3300 3325
3350 3375 [G]
3400 3425
3450
3275 (b)
3300
3325 3350 3375 [G]
3400 3425 3450
Figure 6.6 ESR spectrum of (a) the O2-oxidized PP-9-BBN and (b) upon addition of MMA monomers.
Functionalization via Reactive Polyolefins
O
B
O
Borinate
93
N
Nitroxide
Scheme 6.4
by electron-donating of the lone-pair electrons in the p-orbital of nitrogen to the free radical, the radical in the borinate group may demonstrate a reverse stabilization mechanism by back-donating electron density to the empty p-orbital of boron.
5
MALEIC-ANHYDRIDE-MODIFIED POLYOLEFINS260
As discussed, the stable polymeric radicals, formed during the O2 oxidation reaction of the borane-containing polyole®n, can initiate the free radical graft-from polymerization. In the case of the maleic anhydride (MA) monomer, only a single MA unit can be incorporated into the polymer (as shown in Eq. (6.1)) by each polymeric radical formed due to the low MA homo-polymerization capability. The MA units in the resulting MA-modi®ed polyole®n are basically controlled by the borane units in the polymer. However, when styrene (S) coexists with MA, the graft-from polymerization that takes place will produce an alternating styrene/maleic anhydride (SMA) copolymer. In other words, multiple MA units can be incorporated into polyole®ns by a single borane group. The resulting polymer has a graft molecular structure with a polyole®n backbone and several SMA side chains. Table 6.2 summarizes the experimental results of several maleic-anhydridemodi®ed PP polymers that are compared with the starting poly(propylene-co-5hexenyl-9-BBN) copolymer, having 0.8 mol% 9-BBN units. Both oxidation and Table 6.2
Summary of MA- or SMA-grafted PP copolymers Reaction conditions1
Product properties
Sample
MA (g)
ST (g)
Temp. ( C)
Time (min)
MA (wt%)
Mv (105)
Tm ( C)
Tg ( C)
PP-B PP-MA-1 PP-MA-2 PP-g-SMA-1 PP-g-SMA-2 PP-g-SMA-3 PP-g-SMA-4
Ð 1 1 1 1 1 2
Ð 0 1 1 1 1 2
Ð 20 20 45 45 45 45
Ð 4 10 1 4 16 16
Ð 1.2 1.8 2.4 11 16 30
3.7 3.7 3.8 Ð Ð Ð Ð
152 152 152 153 154 155 155
Ð Ð Ð Ð 217 215 215
1
1 g hydroborated PP, 20 ml benzene.
94
Functionalization of Polyolefins
graft-from reactions were carried out by suspending PP-B ®ne powders in a solvent at low reaction temperature (0±30 C). Despite heterogeneous mild reaction conditions, the graft reaction is apparently very effective. All borane groups were oxidized and reacted with the MA reagent in 10 min. The molecular weights of the MA-grafted copolymers (PP-MA-1 and PP-MA-2) are almost the same as that of the starting materials, which indicates that there is no signi®cant structure change in the PP backbone during the whole modi®cation process. The functionalization reaction selectively occurred in the borane group. This is a signi®cant departure from traditional free radical modi®cation that is usually accompanied with extensive side reactions, including PP backbone degradation, as discussed in Chapter 5. Adding styrene in the graft-from reaction signi®cantly increases the MA graft content due to the graft-from copolymerization of styrene and maleic anhydride. As shown in Table 6.2, the incorporated MA concentration is basically proportional to the reaction time and monomer concentration. Figure 6.7 compares the IR spectra of
8
(c)
7
Absorbance
6 5 4 (b) 3 2 1
(a)
0 4000
Figure 6.7 g-SMA-3.
3500
3000
2500 2000 Wavenumber
1500
1000
500
IR spectra of (a) PP-3, (b) PP-g-SMA-2, and (c) PP-
Functionalization via Reactive Polyolefins
95
70.0 65.0 (d) Heat flow (mW)
60.0 55.0 50.0
(c)
45.0 40.0
(b)
35.0 30.0 25.0 50.0
(a)
100.0
150.0 200.0 Temperature (°C)
250.0
300.0
Figure 6.8 DSC curves of (a) PP-B, (b) PP-g-SMA-2, (c) PPSMA-3, and (d) PP-SMA-4.
PP-B, PP-g-SMA-2, and PP-g-SMA-3. Two new absorption peaks at 1860 and 1780 cm 1 correspond to two C O vibrational stretching modes in succinic anhydride. With an increase in the grafting reaction time, the MA content increases. The propagating radical is quite stable in the grafting reaction, and a high molecular weight for the SMA side chain can be achieved. In fact, the Tg of SMA in the PP-gSMA copolymers was clearly observed in the DSC curves shown in Fig. 6.8. With an increase in the grafting reaction time, the intensity of the Tg transition peak increases. It is clear that the SMA graft length in PP-g-SMA increases, which results in the formation of a separate and bigger SMA domain size in the PP-g-SMA. This chemistry can be directly applied to all borane-containing polyole®ns, including the ones prepared by hydroboration of unsaturated polymers, such as commercial EPDM and butyl rubber containing diene comonomers. The experimental results regarding unsaturated polymers will be discussed in Chapter 8.
6
APPLICATIONS
The functionalized polyole®n is a very useful material for improving several desirable properties of polyole®ns, such as PE, PP, EP, etc. In this section, a PP-OH sample with brush-like structure (discussed in Section 3.3) will be used in the
96
Functionalization of Polyolefins
surface modi®cations of bulk PP matrix. The resulting PP with hydrophilic surface was found to be very useful for many applications. 6.1
Adhesion Studies of PP/Al and PP/Glass Laminates261,262
There is almost no interaction between PP and most materials with polar surfaces, such as the glass ®ber and aluminum foil (Al) used in polymer composites. The PPOH, having a brush-like structure (shown in Scheme 6.2) is used as the surface modi®er to improve PP adhesion. Ideally, the ¯exible OH groups interact with the substrate surface, and the PP segment of PP-OH (having a similar Tm to the PP) cocrystallized with the PP. Both drawn and undrawn PP ®lms (commercial products) were laminated with PP-OH treated substrates. As shown in Table 6.3, the peel test results are compared with those obtained from standard acid-etched samples that involve surface modi®cation of both PP and Al by a dichromate-sulfuric solution followed by the same lamination procedure. PP/Al laminates bonded by PP-OH were found to exhibit an extraordinary 7±10-fold increase in peel strength over acidetched samples. Peeled Al and PP surfaces reveal typical hydrophobic PP surfaces with water drop contact angles from 130 to 140 . The adhesive energy is clearly higher than the cohesive energy. The same results were revealed in SEM studies. As shown in Fig. 6.9, both peeled Al and PP (drawn) surfaces show similar morphologies. It is clear that cohesive failure occurs and gives rise to the high peel strengths observed for the PP/Al laminate. When peeling Al from well-bonded PP, the failure path appears to propagate within the PP layers. This behavior also helps explain that the peel strength of drawn PP/Al laminates is greater than those of undrawn PP samples. The same adhesion results were observed in the PP/glass laminates.262 Table 6.4 shows the 90 peel test results of several PP/glass laminates, including both undrawn and drawn PP samples with various glass surfaces. Most of the specimens have high peel strength, especially the drawn PP/acidetched E-glass laminates, with PP-OH interfacial agent and hot press process (about 1200±1500 N/m). For comparison, a control experiment involving laminated acidetched PP ®lm with acid-etched E-glass, without the use of PP-OH, only shows very low adhesion (too low to show any signi®cant number in the 90 peel test). All Table 6.3
Peel strength of PP/Al laminates261
Sample Acid etched Undrawn PP/Al Drawn PP/Al PP-OH solution cast Undrawn PP/Al Drawn PP/Al
Peel strength (N/m) 126 130 675 1155
Functionalization via Reactive Polyolefins
Figure 6.9
97
Scanning electron micrograph of the peeled surfaces of drawn PP/Al laminate. Table 6.4
Peel strength of PP/glass laminates262 Peel strength (N/m)1 Undrawn PP
Glass surfaces E-glass Cleaned using acetone and hydrated Acid etched Quartz Cleaned using acetone and hydrated
Drawn PP
(a)
(b)
(a)
(b)
223 416
327 1092
297 520
415 1373
362
529
488
672
1
(a) Solution casting method with 20 mm PP-OH thickness; (b) Hot press process with 25 mm PP-OH thickness.
results indicated the important role of the PP-OH layer as the interfacial agent between the glass and the PP ®lm. The results of SEM and the contact angles of water drops on the peeled surfaces all indicate cohesive failures in the PP/PP-OH matrix. It is believed that the strong adhesion between the PP-OH and the glass surface is primarily due to the chemical bonding. The re¯ection IR studies provide the experimental evidence of the chemical reaction between free Si±OH groups on the
98
Functionalization of Polyolefins
Absorbance
(a)
(b)
(c)
2600
2800
3000
3200
3400
3600
3800
4000
Wavenumber (cm–1)
Figure 6.10 Re¯ection IR spectra of (a) glass surface, (b) PP-OH Ê ) coated on glass, and (c) PP-OH/glass after (thickness about 1000 A thermal treatment.
glass surface and the hydroxy groups in PP-OH. Figure 6.10 compares the IR spectra of two samples ± a silica glass surface, and coated PP-OH/glass before and after thermal treatment. The IR absorption of the glass surface in Fig. 6.10(a) shows two absorptions ± the broad band between 3000 and 3700 cm 1 is due to H2O and vicinal Si-OH absorption modes, and the small peak at 3745 cm 1 corresponds to the free Si±OH groups on the Ê , several additional surface. After PP-OH coating with a thickness of about 1000 A 1 absorption peaks between 3000 and 2800 cm , corresponding to saturated hydrocarbon, appear in Fig. 6.10(b). At this stage, the PP-OH thin ®lm is mostly in physical interaction with the glass surface and can be redissolved in xylene solvent at an elevated temperature. After the thermal treatment of PP-OH/glass at 140 C for 2 h, the peak intensity at 3745 cm 1 is almost completely diminished, as shown in Fig. 6.10(c). In addition, the PP-OH thin ®lm becomes completely insoluble in xylene even up to 150 C. Both results strongly suggest an effective chemical reaction of free silanol groups and primary hydroxy groups to form ±Si±O±C± bonds, which may be attributed to the high adhesion at the PP-OH/glass interface. 6.2
Hydrophilic PP Membranes263±265
A similar surface modi®cation by PP-OH was also applied in the preparation of hydrophilic PP membranes, which are very desirable materials. PP has been used
Functionalization via Reactive Polyolefins
99
extensively in ®ltration membrane applications because PP is chemically stable and mechanically sturdy, and can be used at relatively high temperatures. However, hydrophobic PP is nonwettable by water, and is impermeable to ions in an aqueous solution unless a high pressure is applied. Moreover, PP membranes, as well as other hydrophobic membranes, are characterized by a ¯ux decline that is caused by fouling266,267 due to solute absorption and pore blocking. Several surface modi®cation approaches (similar to the post-polymerization processes discussed in Chapter 5) were reported, which treated the PP membranes with a sulfonating agent268 or plasma,269 and by grafting of hydrophilic acrylic monomers,267,270 with very limited success. A series of hydrophilic PP/PP-OH asymmetric membranes were constructed and evaluated,263±265 with the pore size gradually enlarged from a relatively dense surface to a more open surface. Each membrane comprises a polypropylene (PP) matrix and hydroxylated polypropylene (PP-OH) located on the surfaces of the membrane, including pores. The preparation of asymmetric PP/PP-OH membranes involves a ®lm casting process, in which a homogeneous heated (170 C) solution of PP/PP-OH/solvent/fugitive hydrophilic additive is poured onto a preheated Te¯on plate (90 C). Several polar hydrophilic molecules, including phenol, polyethylene glycol (PEG), and polyvinylpyrrolidone (PVP), were used as the fugitive additives (pore-forming agents). After ®lm casting at an elevated temperature, the thin ®lm was allowed to pre-evaporate until the gel ®lm was formed on the exposed surface. The solution with a thin gel layer was then covered with a glass plate to provide a controlled, slow evaporation and cooling. After a completely gelled ®lm was obtained, the resulting ®lm was immersed in a bath of hot water (80 C) for a few days to remove the hydrophilic additives. The fresh hot water was changed often. The resulting ®lm, now an asymmetric membrane, was stored in a distilled water bath before measuring its performance. Table 6.5 summarizes the process conditions and some physical properties of three typical PP/PP-OH membranes. Figure 6.11 shows the scanning electron micrographs of two surfaces on an A-1 PP/PP-OH membrane, with a ®lm thickness of about 60 mm. The dense, active Table 6.5
Summary of three aymmetric PP/PP-OH membranes263 Physical properties Preparation conditions
Pore size (mm)
Sample
PP/PP-OH (g)/(g)
Additive1 (g)
Xylene (g)
Contact angle2
Dense layer
Open layer
Tensile strength (Mpa)
A-1 A-2 A-3
4/4 4/4 4/4
Phenol/8 PEG/2 PEG/1 PVP/1
392 392 520
79 79 90
< 0.1 0.13±0.5 0.5±1.5
10±25 1±2 2±8
7.5 13.2 4.8
1
PP (Mw 230 000 g/mol), PP-OH (3 mol% of OH group; Mw 183 000 g/mol), PEG (Mw 1000 g/mol), and PVP (Mw 10 000 g/mol). 2 The contact angle of the dense skin layer.
100
Functionalization of Polyolefins
Figure 6.11 Scanning electron micrographs of two surfaces on A-2 PP/PPOH membrane: (top) surface in contact with Te¯on; (bottom) surface exposed to air.
Functionalization via Reactive Polyolefins
101
surface with the pore size < 0.1 mm is from the PP/PP-OH ®lm in contact with a Te¯on surface. On the other hand, a pore size of as high as 10 mm was observed on the support surface that was open to air. In general, the pore size and density on surfaces are strongly related to the processing conditions, the combination of additives (concentration and molecular weight), and the pre-evaporation conditions. The ®nal pore structure in the membrane is the result of fugitive additives in the cast ®lm. The additives diffuse from the surface that is in contact with Te¯on to the surface that is open to air during the pre-evaporation process. The PP/PP-OH membranes typically have a water permeability of about 7±25 l/m2 per hour under a pressure of 2 kg/cm2, and the contact angles (measured by water droplet) on the surface of the membrane are about 80 . A high percentage of the hydroxy groups in PP-OH copolymer must be located on the surfaces of the membrane, including the surface of the pores. The resulting asymmetric PP/PP-OH membranes were evaluated in the ultra®ltration studies by measuring solute permeability coef®cient, percent rejection, and stability in long-term use. The examination was performed at 25 C with a pressure of 2 kg/cm2. The concentration of solute ± including water, polyethylene glycol, and dextran with various molecular weights ± is about 2000 ppm in water. All PP/PPOH membranes showed good selectivity. The rejection molecular weight was correlated to the pore size on the dense layer ± the larger the size, the larger the rejection molecular weight. Figure 6.12 shows the sharp selectivity in both A-1 and A-3 membranes. 100
Rejection (%)
80
60 (a) (b) 40
20
0
102
103 104 Molecular weight
105
Figure 6.12 Ultra®ltration studies for (a) A-1 and (b) A-3 membranes.
102
Functionalization of Polyolefins
For the A-1 membrane, the rejection molecular weight is about 7500 g/mol for PEG and l0 000 g/mol for dextran. For the A-3 membrane, the rejection molecular weight increases to 20 000 g/mol for PEG and about 40 000 g/mol for dextran. As expected, the ¯ux is strongly related to the pore size in the dense layer. In general, a relatively high ¯ux is observed in each PP/PP-OH membrane, including sample A-1 with a highly dense layer. In addition to the macroscale structural factors, such as pore structure and thickness, some microscale (molecular) factors may also in¯uence mass transfer rates. The ¯exible hydroxy groups located at the end of each side chain in the PP-OH polymer may provide advantages, such as thermodynamic transitions related to macromolecular relaxation phenomena in transporting hydrophilic mass through the pore structures. Only a single membrane was used in each set of the evaluation study. In addition, each evaluation was carried out for a long period without signi®cant changes in rejection (%) and ¯ux. In fact, the used membrane was reexamined by SEM measurement after the completion of the evaluation studies. No detectable change in membrane morphology, including pore size and shape, was observed. The signi®cant improvement in the antifouling property was clearly attributed to the hydroxy groups located on the surfaces of the PP/PP-OH membrane. 6.3
Immobilized Catalysts on Polyole®ns271±280
Another application of functionalized polyole®ns that has been studied is the immobilization of soluble catalysts so that catalysts can be recovered and reused for many reaction cycles. This process becomes very important when the reactions require a large quantity of toxic catalyst, such as in the oligomerization of ole®ns. Semicrystalline PE and PP are ideal support materials that possess several essential properties, such as good chemical and physical stability, and excellent processibility to allow various forms (®lm, ®ber, particle, and porous structures) with high surface area. The PP-OH polymer cocrystallized on the surface of PP provides the anchor sites for immobilizing a soluble catalyst with chemical bonds. One example (shown in Eq. (6.3)) is the immobilization reaction between PP-OH and Lewis acids, such as EtAlCl2 and BF3:
PP crystalline domain OH
+
EtAlCl2
OAlCl2
BF3
OBF2
6:3
n-BuLi
O– Li+ +
The immobilization reaction was very effective in heterogeneous condition at room temperature, by mixing PP-OH with EtAlCl2 or PP-O Li with BF3 to produce
Functionalization via Reactive Polyolefins
103
Table 6.6 Summary of PIB prepared using PP±O±AlCl2 catalysts in hexane Run
Temp. ( C)
Time (min)
Mn1 (g/mol)
PDI1
Yield (%)
A-1 A-2 A-3 A-8 A-10 A 0 -1 A 0 -2 A 0 -3 A 0 -14 A 0 -15 C5-O-AlCl2 catalyst C5-O-AlCl2 catalyst
25 25 25 25 0 25 25 25 25 0 25 0
90 60 20 15 15 180 120 60 180 300 15 15
1050 1150 1150 1180 4540 1370 1660 1230 1320 5450 1180 5450
2.0 1.6 1.8 1.5 2.6 3.0 2.6 2.4 2.6 2.6 2.3 2.6
100 100 100 100 100 100 100 65 100 76 100 100
1
GPC measurement.
chemically bonded ±OAlCl2, ±OBF2 groups, respectively, on the surfaces of the polypropylene support. The polymer-supported catalyst was then used as the Lewis acid catalyst in the carbocationic oligomerization of isobutylene. After the reaction, polyisobutylene (PIB) was obtained by ®ltering out the supported catalyst, then removing the solvent and unreacted monomers under vacuum. The recovered catalyst was mixed with another isobutylene/hexane solution, followed by the same separation and recovery processes. This reaction cycle was repeated a number of times. Table 6.6 shows a typical result of PIB obtained by using a supported PP±O±AlCl2 catalyst.271 The A 0 runs were carried out by using a regenerated supported catalyst used in the A runs. Typically, the monomer (isobutylene) to catalyst (±OAlCl2) ratio was about 500 to 1. In most reaction cycles, the conversion from monomer to polymer was complete within 15 min. The same catalyst activity was maintained for many reaction cycles. Both the PIB product and catalyst activity results were very similar to those seen from control reactions using a homogeneous C5±O±AlCl2 catalyst. The high and stable catalyst activity in the PP/PP-OH support system must be due to a high and stable surface area, which could be a consequence of small particle size, the crystallinity of polyole®n, and the ¯exibility of the side chain between the catalyst and polymer backbone.
7
SUMMARY
The borane approach provides a very effective method for the functionalization of polyole®n with well-controlled polymer structure. This general method can be
104
Functionalization of Polyolefins
applied to all polyole®ns and to most important functional groups. The advantages of this chemistry lie in a unique combination of traits, including: (a) Easy incorporation of borane groups into polyole®n. In direct polymerization, the borane containing -ole®ns behaves like -ole®ns and can be effectively incorporated into polyole®n structures. This facile chemistry is due to a unique combination of (i) the stability of borane moiety to transition metal catalysts, (ii) the solubility of borane compounds in the hydrocarbon solvents (hexane and toluene) used in transition metal polymerizations, and (iii) advances in metallocene catalysis that allow the effective incorporation of high monomers. In post-polymerization, the hydroboration reactions are very effective due to the excellent reactivity of boronhydride to ole®nic units in the unsaturated polyole®n, and the good solubility of borane compounds in the hydrocarbon solvents used in the reactions. (b) The versatility of borane groups, which can be modi®ed for a remarkable variety of functionalities, including use as a living radical initiator for graft-from reactions.
7 Functionalization via Reactive Polyole®ns Containing p-Methylstyrene Groups
1
INTRODUCTION
This chapter discusses a second reactive comonomer, p-methylstyrene (p-MS), that can be effectively incorporated into polyole®ns by transition metal catalysts and can be subsequently interconverted to various desirable functional groups, such as ±OH, ±NH2, ±COOH, anhydride, and halides. Preferably, the reactive groups can be effectively transformed to ``stable'' anionic initiators for ``living'' anionic graft-from polymerization reactions, as illustrated in Eq. (7.1). The resulting graft copolymers, containing a polyole®n backbone and several well-de®ned functional polymer side chains, offer not only a high concentration of functional groups, but also preserve the desirable polyole®n properties (such as crystallinity, melting point, and glass transition temperature). The major advantages of p-MS are its commercial availability, easy incorporation into polyole®ns, and versatility281,282 in the functionalization chemistry under various reaction mechanisms ± including free radical, cationic, and anionic processes. The benzylic protons are known to be facile in many chemical reactions, such as halogenation,283±286 metallation,287,288 and oxidation289,290 to form desirable functional groups at the benzylic position under mild reaction conditions. In addition, the benzylic protons can also be interconverted to a stable anionic initiator for ``living'' anionic graft-from polymerization.291,292 In general, the preparation of p-MS-containing polyole®n copolymers has been greatly enhanced by the exciting metallocene technology.293±296 The well-de®ned metallocene catalyst with constrained ligand geometry, having a spatially opened catalytic site, allows the effective incorporation of large comonomers.297 One new commercial polymer, poly(ethylene-co-styrene) copolymer,298,299 has certainly bene®ted from this technology, which also assures success in the preparation of poly(ethylene-co-p-methylstyrene) copolymers.
106
Functionalization of Polyolefins
R
ⱍ
R
ⱍ (CH2CH)x(CH2CH)y ⱍ
CH2CH + CH2CH ⱍ ⱍ
CH3
ⱍ
Co(II)/O2
Metallocene or Ziegler–Natta catalyst
R
COOH
R
ⱍ CH2 ⱍ Br
ⱍ (CH2CH)x(CH2CH)y ⱍ NBS/BPO
R
ⱍ (CH2CH)x(CH2CH)y ⱍ
Maleic anhydride /BPO
ⱍ (CH2CH)x(CH2CH)y ⱍ
ⱍ
CH3
ⱍ CH2 ⱍ O
s-BuLi/TMEDA
O
O
R
CI-Si(CH3)3
R
ⱍ (CH2CH)x(CH2CH)y ⱍ
ⱍ (CH2CH)x(CH2CH)y ⱍ ⱍ
CO2
CH2Li O
MMA
R
ⱍ CH2 ⱍ Si(CH3)
R
ⱍ CH2 ⱍ COOH
ⱍ (CH2CH)x(CH2CH)y ⱍ
R
ⱍ (CH2CH)x(CH2CH)y ⱍ ⱍ
CH2 PMMA
ⱍ (CH2CH)x(CH2CH)y ⱍ ⱍ (CH2)3 ⱍ OH
7:1
Functionalization via Reactive Polyolefins
107
2 POLY(ETHYLENE-CO-p-METHYLSTYRENE) COPOLYMERS107,108,300,301 2.1
Copolymerization of Ethylene and p-Methylstyrene (p-MS)
In a typical copolymerization, the reaction starts by the addition of the metallocene catalyst mixture to a solution of the two monomers in solvent under an inert gas atmosphere. The appearance of the reacting polymer solution is very dependent on the quantity of p-MS used. In the high p-MS cases, a homogeneous solution was observed throughout the copolymerization reaction. In the low p-MS cases, a slurry solution with white precipitates was observed right at the beginning of the reaction. The precipitation is obviously due to the crystallinity of the copolymer, which has long ethylene sequences. The copolymer was isolated by ®ltering, washed completely with MeOH, and dried under vacuum at 50 C for 8 h. Table 7.1 summarizes the copolymerization results by using three metallocene catalysts, including a simple Cp2ZrCl2 catalyst and two bridged ones (Et(Ind)2ZrCl2 and [C5Me4(SiMe2NtBu)]TiCl2). Table 7.1 p-MS301
Run no. p-363 p-365 p-356 p-358 p-361 p-371 p-357 p-360 p-362 p-375 p-270 p-377 p-378 p-267 p-379 p-380 p-383
Summary of the copolymerization reactions1 between ethylene and
Ethylene/ p-MS Catalyst (mmol)2 (psi)/(M) I/17 I/17 II/17 II/17 II/17 II/17 II/17 II/17 II/17 II/17 III/10 III/10 III/10 III/10 III/10 III/10 III/10
45/0.678 45/0.678 45/0.085 45/0.678 45/1.36 45/2.03 45/0.085 45/0.678 45/1.36 45/2.03 45/0 45/0.447 45/0.912 45/0.447 45/0.912 45/1.82 10/1.82
Solvent/ temp. ( C)
Yield (g)
Catalyst efficiency (kg P/(mol M h))
Hexane/50 Toluene/50 Hexane/50 Hexane/50 Hexane/50 Hexane/50 Toluene/50 Toluene/50 Toluene/50 Toluene/50 Toluene/30 Hexane/30 Hexane/30 Toluene/30 Toluene/30 Toluene/30 Hexane/30
24.2 8.44 6.5 21.9 18.9 20.8 5.70 15.1 19.5 19.4 4.4 12.0 15.5 13.0 17.4 24.2 15.9
1423.5 496.5 382.4 1288.2 1111.8 1223.5 335.3 888.2 1147.1 1141.2 440.0 1200.0 1550.0 1300.0 1740.0 2420.0 1590.0
p-MS in Conversion copolymer of p-MS (mol%) (%) 2.20 1.84 1.83 5.16 7.20 8.94 1.30 4.76 6.36 8.49 Ð 13.5 22.6 10.9 21.6 32.8 40.0
26.2 7.7 47.2 51.1 29.0 25.4 29.9 32.8 27.0 22.8 Ð 90.3 81.3 83.8 86.9 75.8 54.6
1 45 psi ethylene 0.309 mol/l in toluene, 0.424 mol/l in hexane at 50 C; 0.398 mol/l in toluene, 0.523 mol/l in hexane at 30 C. 10 psi ethylene 0.116 mol/l in hexane at 30 C. 2 I: Cp2ZrCl2/MAO; II: Et(Ind)2ZrCl2/MAO; III: [(C5Me4)SiMe2N(t-Bu)]TiCl2/MAO. [Reproduced with permission from Kinetic and Microstructure Studies of Poly(ethylene-cop-Methylstyrene) Copolymers Prepared by Metallocene Catalysts with Constrained Ligand Geometry, T. C. Chung and H. L. Lu, J. polym. Sci: Part A: Polym. Chem. 1998, 36, 1017. Copyright 1998 John Wiley & Sons]
108
Functionalization of Polyolefins
The copolymerization ef®ciency follows the sequence of [C5Me4(SiMe2NtBu)]TiCl2 > Et(Ind)2ZrCl2 > Cp2ZrCl2, which is directly relative to the spatial opening at the active site. Between the comparable runs p-363/p-358 and p-365/p-360, Cp2ZrCl2 (I) and Et(Ind)2ZrCl2 (II) catalysts produced the copolymers with 2.2/5.16 and 1.84/4.76 mol% of p-MS concentrations, respectively, and 26.2/51.1 and 7.72/ 32.8% overall p-MS conversions, respectively. On the other hand, the [C5Me4(SiMe2NtBu)]TiCl2 catalyst offers the best reactions of the three catalysts in the preparation of PE-p-MS copolymers. In run p-377, about 90% of the p-MS was incorporated into the copolymer in 1 h. In run p-383, the reaction produced copolymer containing 40 mol% of p-MS, which is close to the ideal 50 mol% (as will be discussed in the reactivity ratio and copolymer microstructure studies, the consecutive insertion of p-MS in all three catalyst systems is almost impossible). A high concentration of p-MS in the copolymer and the high p-MS conversion were achieved by catalyst (III), which must provide excellent spatial freedom for the p-MS to have access to the propagating chain end. In general, the catalyst activity systematically increases with the increase of p-MS content, which was also observed in the 1,4-hexadiene copolymerization reactions248 and could be a physical phenomenon relative to the improvement of monomer diffusion in the lower crystalline copolymer structures. In the [C5Me4(SiMe2NtBu)]TiCl2 case, the catalyst activity attains a value of more than 2.4 106 g of copolymer per mole of Ti per hour in run p-380, which is about six times the value for the homopolymerization of ethylene in run p-270 under similar reaction conditions. It is very interesting to note that a very small solvent (hexane and toluene) effect in relation to the catalyst (III) activity was observed in the comparative runs (p-377/p-267) and (p-378/p-379), despite the signi®cant difference in the beginning reaction conditions (heterogeneous in hexane and homogeneous in toluene). However, the solvent effect is very signi®cant in both the catalyst (I) and (II) systems. All comparative reaction pairs (p-363/p-365, p-356/357, p-358/p-360, p-361/p-362, and p-371/p-375), carried out under similar reaction conditions, consistently show a higher p-MS incorporation into the hexane solution. The explanation of this solvent effect is not clear. Figure 7.1 shows the typical GPC curves of the homo- and copolymers prepared by the [C5Me4(SiMe2NtBu)]TiCl2 catalyst. The uniform molecular weight distribution in all samples, with Mw/Mn 2±3, implies the single-site polymerization mechanism. In fact, the GPC curves show a slight reduction of molecular weight distribution in the copolymers, from Mw/Mn 2.86 in PE to 1.68 in PE-p-MS containing 18.98 mol% of p-MS. Similar narrow molecular distribution results were also observed in the copolymers prepared by the Et(Ind)2ZrCl2 catalyst. The better diffusibility of monomers in the copolymer structures (due to lower crystallinity) may help to provide the ideal polymerization condition. It is very interesting to note that the average molecular weight of the copolymers remained very high throughout the composition range, which may be attributed to the relatively high reactivity of p-MS. Figure 7.2 shows the comparison of DSC curves between the PE homopolymer and PE-p-MS copolymers prepared by the [C5Me4(SiMe2NtBu)]TiCl2 catalyst.
1.40 18.98 mol% 9.82 mol% 1.08 mol% PE (0 mol%)
1.20
dwt / [d(log MW)]
1.00 0.80 0.60 0.40 0.20 0.00 6.30
5.80
5.30
4.80
4.30
3.80
Log MW
Figure 7.1 GPC curves of PE and three PE-p-MS copolymers, containing 1.08, 9.82, and 18.98 mol% of p-MS units, prepared using the [C5Me4(SiMe2NtBu)]TiCl2 catalyst. [Reproduced with permission from Kinetic and Microstructure Studies of Poly(ethylene-co-p-Methylstyrene) Copolymers Prepared by Metallocene Catalysts with Constrained Ligand Geometry, T. C. Chung and H. L. Lu, J. polym. Sci: Part A: Polym. Chem. 1998, 36, 1017. Copyright 1998 John Wiley & Sons]
8.0 (a) (c)
7.0 (b)
(b) 5.0
DSC
Heat flow (W/g)
6.0 (c)
(a)
4.0 (d) 3.0 (e)
2.0 1.0
(f) 50
75 100 125 Temperature (°C)
150
– 40 – 20
0 20 40 60 Temperature (°C)
80 100
Figure 7.2 DSC curves; (left) (a) PE and PE-p-MS containing (b) 1.08, (c) 2.11, (d) 5.40, (e) 9.82, and (f) 18.98 mol% of p-MS; (right) PE-p-MS containing (a) 18.98, (b) 32.8, and (c) 40 mol% of p-MS. [Reproduced with permission from Kinetic and Microstructure Studies of Poly(ethylene-co-p-Methylstyrene) Copolymers Prepared by Metallocene Catalysts with Constrained Ligand Geometry, T. C. Chung and H. L. Lu, J. polym. Sci: Part A: Polym. Chem. 1998, 36, 1017. Copyright 1998 John Wiley & Sons]
110
Functionalization of Polyolefins
Even a small amount (1 mol%) of p-MS comonomer incorporation has a signi®cant effect on the crystallization of polyethylene. The melting point (Tm) and crystallinity (c) of the copolymer are strongly relative to the density of the comonomer ± the higher the density, the lower the Tm and c. Only a single peak is observed throughout the whole composition range, and the melting peak completely disappears at 10 mol% of p-MS concentration. The removal of crystallinity provides the opportunity to obtain elastic properties in many thermoplastic polymers. Figure 7.2 (right) shows the glass transition temperature (Tg) of PE-p-MS copolymers with 18.98, 32.8, and 40 mol% p-MS, respectively. The lowest observed Tg is 5.7 C in the copolymer containing 18.98 mol% p-MS. With the increase of p-MS, the Tg of copolymer systematically increases. It is clear that the PE-p-MS copolymer is un®t to serve as a good elastomer due to the high Tg of poly(p-methylstyrene). To achieve an elastic polymer, the system requires a third monomer that can provide low Tg. The detailed results of elastic poly(ethylene-ter-propylene-ter-p-MS) and poly(ethylene-ter-1-octene-ter-p-MS) terpolymers will be discussed later. 2.2
p-Methylstyrene Reactivity Ratio300,301
The best way to investigate comonomer reactivity is to measure the reactivity ratio. To obtain meaningful results, a series of experiments were carried out by varying the monomer feed ratio and comparing the resulting copolymer composition at low monomer conversion (< 10%) rates. The reactivity ratios between ethylene (r1 k11/k12) and p-MS (r2 k22/k21) are estimated by the Kelen±TuÈdos method. In [C5Me4(SiMe2NtBu)]TiCl2 cases, high r1 (r1 19.6 and 21.4 for 20 and 60 C, respectively) and low r2 (r2 0.04 and 0.08 for 20 and 60 C, respectively) indicate the strong tendency for ethylene consecutive insertion and very low possibility for continuous p-MS insertion. The values of r1 r2 are close to unity at both temperatures, which suggests nearly ideal random copolymerization reactions and a very small probability of ®nding two adjacent p-MS units in the polymer chain. In other words, the p-MS units will be homogeneously distributed in the polymer chain. In Et(Ind)2ZrCl2 cases, the copolymerization reactions exhibit even higher r1 (r1 > 60), which is very strongly favorable for ethylene incorporation, and almost no possibility of p-MS consecutive insertion (r2 0). The less opened active site in the Et(Ind)2ZrCl2 catalyst may sterically prohibit p-MS consecutive insertion. It is very interesting to compare the reactivity of p-MS with styrene and methylstyrene isomers. The consumption of comonomer during the reaction is a useful way to understand the dynamic of copolymerization reaction. Figure 7.3(a) shows comparative plots of p-MS and styrene incorporation vs reaction time in the batch copolymerization reaction of ethylene (29 psi) and comonomer (0.356 mol/l), using [C5Me4(SiMe2NtBu)]TiCl2 catalyst at 60 C. The signi®cantly better p-MS incorporation starts in the beginning of the copolymerization reaction. Figure 7.3(b) compares the incorporated comonomer concentration (mol%) vs comonomer ratio. Each copolymerization reaction was carried out at 40 C for
Functionalization via Reactive Polyolefins 100
22
90
p-MS or styrene concentration in the copolymers (mol%)
(a) p-MS Styrene
80
Conversion (%)
70 60 50 40 30 20 10 0
111
(b)
20
p-MS Styrene
18 16 14 12 10 8 6 4 2 0
0
10
20 30 40 50 60 Polymerization (min)
70
0.0
1.0 2.0 3.0 4.0 Ratio of p-MS or styrene to ethylene in the feed
5.0
Figure 7.3 Plots of (a) comonomer conversion vs reaction time and (b) comonomer concentration in copolymer vs comonomer ratio, during the copolymerization reactions of ethylene with p-methylstyrene and with styrene using the [C5Me4(SiMe2NtBu)]TiCl2 catalyst.
15 min by using the [C5Me4(SiMe2NtBu)]TiCl2 catalyst. In every monomer feed ratio, p-MS consistently shows a more than 30% higher incorporation rate than seen in the corresponding styrene. The signi®cantly higher p-MS incorporation must be due to the electronic donation of a p-methyl group, which is favorable in the ``cationic'' polymerization mechanism. Sterically, the methyl group at parasubstitution doesn't effect the monomer insertion. The p-MS incorporation was also compared with its derivatives, o-MS and m-MS, with consistently better incorporation. Both isomers (o-MS and m-MS) failed to receive the full bene®ts of the electronic and steric effects that existed in p-MS.
3 3.1
POLY(PROPYLENE-CO-p-METHYLSTYRENE) COPOLYMERS302 Metallocene Copolymerization
The copolymerization of propylene and p-MS was investigated by using both metallocene and Ziegler±Natta catalysts. Table 7.2 summarizes the copolymerization results of using two of the most common isospeci®c metallocene catalysts, {SiMe2[2-Me-5-Ph(Ind)]2}ZrCl2 (A) and Et(Ind)2ZrCl2 (B), respectively. Overall,
112
Functionalization of Polyolefins
Table 7.2 A summary of copolymerization of propylene and p-MS using {SiMe2[2-Me-5-Ph(Ind)]2}ZrCl2 (A) and Et(Ind)2ZrCl2 (B) Polymerization conditions1
Copolymer composition
Productivity Monomers Cat./Conc. MAO Temp. C3/p-MS (KgPP/ [P] [E] [p-MS] Tm (psi)/(ml) (mmol)/(ml) (Al/Zr) ( C) (mol Zr. atm.h)) (mol%) (mol%) (mol%) ( C) 75/1.5 75/1.5 75/1.5 75/1.5 75/3.0 75/1.5 75/3.0 145/1.5 145/1.5 145/3.0 75/1.5 75/1.5 75/3.0 75/5.0 75/1.5 (130P/7E)2 75/1.5 (115P/20E)3 145/3.0
A/100 A/100 A/100 A/100 A/100 A/100 A/65 A/50 A/50 A/50 B/100 B/100 B/100 B/100 B/100
2000 5000 8000 5000 5000 5000 5000 2000 2000 2000 2000 2000 2000 2000 2000
30 30 30 60 60 80 100 80 100 80 30 60 60 60 30
5.8 5.8 3.9 29.0 17.4 106.3 59.5 88.0 244.0 36.0 42.5 108 38.7 7.7 1005.3
99.1 99.2 99.0 99.1 98.3 98.8 97.7 99.4 99.1 98.8 99.9 99.2 99.1 98.7 82.2
Ð Ð Ð Ð Ð Ð Ð Ð Ð Ð Ð Ð Ð Ð 17.6
0.9 0.9 1.0 0.9 1.7 1.2 2.3 0.6 0.9 1.2 0.1 0.8 0.9 1.3 0.2
Ð4 Ð4 Ð4 Ð4 Ð4 143 128 148 142 Ð4 142 124 123 119 103
B/100
2000
30
1314.7
72.5
27.2
0.3
75
B/50
2000
80
340.0
99.2
Ð
0.8
116
1
Polymerization in 100 ml toluene. Ethylene and propylene premixed (130 psi propylene and 7 psi ethylene) and charged at 75 psi. 3 Ethylene and propylene premixed (115 psi propylene and 20 psi ethylene) and charged at 75 psi. 4 Bimodal Tm. 2
these metallocene catalysts were not very effective in producing PP-p-MS copolymers as measured by catalyst activity and p-MS incorporation. Although the resulting copolymers show a narrow molecular weight distribution, the copolymer molecular weights are signi®cantly reduced compared to the homopolymer. Compared to the results seen in the homopolymerization of propylene, the presence of even a small amount of p-MS in these cases shows severely reduced catalyst activity. These results are completely opposite to those observed in the corresponding ethylene/p-MS copolymerization reactions, which show a signi®cant increase in catalytic reactivity. The drastic differences may be due to the steric jamming phenomenon in the crossover k21 reaction (from p-MS to propylene), as illustrated in Eq. (7.2). It is well known that in metallocene catalytic polymerization, the insertion of the styrene monomer is predominantly secondary (2,1-insertion),303±305 while the 1,2insertion of the propylene monomer is dominant. Once the propagating PP chain has a chance to react with the p-MS monomer (k12 reaction) via 2,1-insertion, the bulky
Functionalization via Reactive Polyolefins L k12 2,1-insertion
TiCHCH2CH2CH ⱍ ⱍ CH3 L H3C (II) ⱍ
L
TiCH2CH ⱍ CH3 L (I)
113
(Very slow reaction) k21
1,2-insertion
7:2
L
TiCH2CHCHCH2CH2CH ⱍ H3C L (III) ⱍ
p-phenyl group in the last unit of the growing chain (II) is adjacent to the central metal atom and blocks the upcoming 1,2-insertion of a propylene unit (i.e. k21 reaction). Since the homopolymerization (k22) of p-MS via the metallocene coordination mechanism is known to be near zero,300 the metallocene active site at the p-MS unit dramatically slows the propagation process. Therefore, a small amount of p-MS signi®cantly reduces the catalyst reactivity. The reaction kinetic must follow the sequence of k11 > k12 k21 > k22 0, and the most active reaction is the consecutive propylene insertion (k11 reaction). An approach to increasing catalyst activity and p-MS incorporation is to alter the p-MS insertion from a 2,1- to a 1,2-sequence (i.e. reduce the steric jamming in the k21 reaction). With an increase in the reaction temperature, both polymer yield and p-MS incorporation signi®cantly increase, which may be attributed to this insertion sequence change. However, the increase in catalyst activity is at the expense of molecular weight due to faster -hydrogen elimination. It is interesting to note that by increasing the temperature, the formation of p-MS homopolymer is reduced. The ``cationic'' polymerization mechanism, responsible for the homopolymerization of p-MS via the MAO cocatalyst, is certainly unfavorable at high temperatures. Another approach to overcoming the reduced catalyst activity is to add a small amount of ethylene. The sluggish propagating chain end of the p-MS unit, a dif®culty in both the k21 and k22 reactions, allows the insertion of ethylene, which re-energizes the propagation process. As shown in Table 7.2 (runs de®ned in notes 2 and 3), catalyst activity was dramatically improved, producing results better than those for homopolymerization. The experimental results strongly support the steric jamming theory in the metallocene copolymerization of propylene and styrenic monomer. Unfortunately, the incorporation of ethylene units into PP copolymers signi®cantly reduces their melting points and crystallinities, which are crucial to some applications. 3.2
Ziegler±Natta Copolymerization
The other approach for improving the copolymerization reaction of propylene and p-MS is to employ isospeci®c Ziegler±Natta catalysts that can only proceed via
114
Functionalization of Polyolefins
Table 7.3 Summary of copolymerization reactions between propylene (m1) and p-MS (m2) in the presence of (MgCl2/TiCl4/ED/AlEt3) catalyst1 Copolymer properties m1/m2 (psi)/(mmol)
Cat. activity (kg/(mol Ti atm h))
p-MS (mol%)
Mw (103 g/mol)
Mw/Mn
Tm ( C)
Hf (J/g)
29/0 29/33.9 29/68.6
658 645 650
0 0.4 0.6
109.9 168.2 202.2
4.41 5.54 6.23
158.4 154.9 154.3
59.9 57.1 56.2
1
Polymerization conditions: catalyst 17.4 mmol Ti, Al/Ti 90, ED/Ti 6, 50 C, 100 ml toluene.
1,2-insertion in both the propylene and p-MS incorporations. Table 7.3 summarizes the experimental results for PP-p-MS copolymers produced using a commercial MgCl2 supported titanium catalyst containing an internal electron donor (MgCl2/ TiCl4/ED/AlEt3). Overall, the presence of the p-MS monomer does not retard the activity of either catalyst system. Polymer productivity is very similar in the reactions with and without the presence of p-MS. It is unexpected that the molecular weight of the copolymer slightly increases with the increase of p-MS concentration, although the molecular weight distribution broadens slightly. A high-molecular-weight PP-p-MS (Mw 202 000 g/mol) was obtained even with relatively low propylene pressure. It is very interesting to note that this catalyst system produces no detectable homopolymer of p-MS during the copolymerization reaction. Acidity of the active Ti species is reduced by coordination with the internal electronic donor (ED).
4 POLYOLEFIN ELASTOMERS CONTAINING p-METHYLSTYRENE GROUPS109,306 As discussed for poly(ethylene-co-p-methylstyrene) copolymers, despite the completely amorphous structure, the lowest Tg observed in this type of copolymer was about 5 C, which is too high to be useful in most elastomer applications. For many commercial applications, the most desirable elastomers are those with low Tg (< 45 C) and ``reactive'' sites (such as p-MS units) that can effectively form crosslinking networks and produce stable residues. In ethylene±propylene cases, this means preparing a random terpolymer containing a close to equal molar ratio of ethylene/propylene and some p-MS ``reactive'' units. With the unprecedented capability of metallocene technology in copolymerization reactions, it is also very interesting to expand the polyole®n elastomer to new classes containing high -ole®ns, such as 1-octene (instead of propylene), which can effectively prevent the crystallization of small consecutive ethylene units and provide low Tg properties.
Functionalization via Reactive Polyolefins
4.1
115
Poly(ethylene-ter-propylene-ter-p-methylstyrene) (EP-p-MS)
Table 7.4 summarizes the terpolymerization reactions of ethylene, propylene, and p-MS, using a [C5Me4(SiMe2NtBu)]TiCl2/MAO metallocene catalyst. Overall, the metallocene catalyst shows excellent activity (3.4±5.4 106 g polymer/mol Zr per hour at 50 C) in all reactions, with comparative reactivities between ethylene and propylene and good incorporation of p-MS. The incorporation of p-MS seems quite insensitive to the ethylene/propylene feed ratio. In both comparative sets of runs p-120 vs p-119 and p-128 vs p-127, with a constant p-MS concentration in each comparative run (0.05 and 0.03 mol/l, respectively) and varying ethylene/propylene feed ratios, the incorporation of p-MS is very constant at about 1.6±1.8 and 0.6±0.65 mol%, respectively. In general, the molecular weights of these terpolymers are very high. Comparing runs p-116 vs p-120 and p-117 vs p-119, with the same amount of ethylene and propylene feeds and with and without p-MS, only a small reduction in molecular weight arises from the incorporation of p-MS. It is very interesting to note that the replacement of p-MS with styrene in the same reaction conditions signi®cantly lowers the molecular weight of poly(ethylene-ter-propylene-ter-styrene). The results may be attributed to the relatively comparable reactivity of p-MS with ethylene and propylene. The electronic donation of the p-methyl group in p-MS is favorable in this cationic coordination polymerization mechanism. The glass transition temperature (Tg) was examined by DSC. Figure 7.4 shows several DSC curves of EP-p-MS terpolymers (samples p-112, p-118, and p-120 in Table 7.4). Each curve has only a sharp Tg transition in a ¯at baseline, without any Table 7.4 A summary of terpolymerization1 of ethylene, propylene, and p-MS using [C5Me4(SiMe2NtBu)]TiCl2/MAO306 Run no.
Monomer feed [E]/[P]/[p-MS] (mol/l)
Catalyst activity (kg/(mol Ti h))
Copolymer [E]/[P]/[p-MS] (mol %)
p-116 p-117 p-107 p-115 p-110 p-111 p-112 p-118 p-113 p-120 p-119 p-128 p-127
0.13/0.28/0 0.12/0.35/0 0.10/0.43/0 0.06/0.60/0 0.10/0.43/0.1 0.10/0.43/0.3 0.10/0.43/0.5 0.12/0.35/0.3 0.08/0.50/0.3 0.13/0.28/0.05 0.12/0.35/0.05 0.14/0.25/0.03 0.13/0.28/0.03
4.9 103 4.7 103 4.8 103 4.9 103 5.4 103 4.9 103 5.3 103 4.0 103 3.5 103 4.1 103 4.0 103 4.4 103 3.8 103
53.9/46.1/0 41.8/58.2/0 36.2/63.8/0 13.0/87.0/0 37.0/58.5/4.5 39.2/52.9/7.9 40.3/48.6/11.1 46.4/43.6/10.0 32.4/59.2/8.4 54.4/43.8/1.8 46.1/52.3/1/6 56.3/43.1/0.6 50.7/48.6/0.7
1
Tg ( C) 49.4 43.7 35.5 16.4 20.3 19.1 9.1 20.7 12.4 45.8 41.0 48.6 45.9
Mw (g/mol)
Mn (g/mol)
300 400 284 300 Ð Ð 184 100 194 200 188 900 198 200 183 200 237 700 195 800 269 400 244 300
134 300 120 800 Ð Ð 90 600 97 600 80 900 74 900 91 800 107 500 75 100 104 300 85 500
polymerization conditions: 100 ml toluene; [Ti] 2.5 10 6 mol; [MAO]/[Ti] 3000; 50 C; 15 min. (Reproduced with permission from Macromolecules 1998, 31, 2028. Copyright 1998 Am. Chem. Soc.)
116
Functionalization of Polyolefins
DSC
(a)
(b)
(c)
– 60
– 40
– 20
0 20 40 Temperature (°C)
60
80
Figure 7.4 Comparison of DSC curves of three EP-p-MS terpolymers having ethylene/propylene/p-MS mole ratio of (a) 40.3/48.6/11.1, (b) 46.4/43.6/10.0, and (c) 54.4/43.8/1.8. (Redrawn from Macromolecules 1998, 31, 2028. Copyright 1998 Am. Chem. Soc.)
detectable melting point. The Tg is clearly a function of the propylene and p-MS contents. Comparing the ethylene±propylene copolymers (without p-MS units) (runs p-116, p-117, p-107, and p-115), the Tg transitions are linearly proportional to the propylene contents and level off at 50 C with the composition 50% of propylene content (similar results were reported for the EPDM case). The Tg transition signi®cantly increases with the incorporation of p-MS into ethylene/ propylene copolymers. It is very interesting to compare runs p-116 and p-120, both with ideal 54 mol% ethylene content and only a very small difference in p-MS/ propylene mole ratios (0/46 vs 1.8/44), the Tgs are 50 and 45 C, respectively. Overall, the composition of EP-p-MS material with low Tg< 45 C is very limited, only to the polymers with < 2 mol% of p-MS content. Despite the random terpolymer structure and the ideal (55/45) ethylene/propylene ratio, a further increase of p-MS raises the Tg of the terpolymer to > 40 C. Obviously, the high Tg of both propylene (Tg of PP 0 C) and p-MS (Tg of poly(p-MS) 110 C) components preclude EP-p-MS from producing some desirable elastomers containing both a high content of ``reactive'' p-MS and a low Tg (< 45 C) transition. 4.2
Poly(ethylene-ter-1-octene-ter-p-methylstyrene) (EO-p-MS)
Table 7.5 summarizes the terpolymerization reactions of ethylene, 1-octene, and p-MS, using a [C5Me4(SiMe2NtBu)]TiCl2/MAO metallocene catalyst. Overall, the
Table 7.5 Summary of terpolymerization1 of ethylene, 1-octene, and p-MS using [C5Me4(SiMe2NtBu)]TiCl2/MAO metallocene catalyst306 Monomer concentration in feed (mol/l) Run no. 3
p-465 p-4663 p-470 p-471 p-472 p-473 p-477 p-476 p-475 p-478 p-474 p-3964 p-3835 1
Copolymer composition (mol%)
E2
1-Oct
p-MS
Yield (g)
[E]
[O]
[p-MS]
0.25 0.25 0.20 0.40 0.40 0.40 0.40 0.40 0.40 0.40 0.40 0.52 0.13
0.89 0.89 0.80 0.80 0.80 0.40 0.20 0.80 0.60 0.40 0.60 0.38 0
0 0.13 0.10 0.10 0.20 0.20 0.20 0.40 0.40 0.40 0.15 0.91 1.82
7.0 5.2 7.3 10.1 9.8 8.2 6.4 9.0 9.1 7.9 9.0 5.6 15.9
41.4 40.0 54.2 61.1 60.3 59.6 80.2 63.4 67.2 73.3 64.7 64.0 60.0
58.6 54.5 43.0 36.0 36.3 34.0 14.1 29.3 24.7 18.5 31.3 18.1 0
0 5.6 2.7 2.9 4.4 6.4 5.7 7.3 8.1 8.1 4.0 17.7 40.0
Tg ( C) 61.8 51.3 56.2 58.1 55.7 50.1 37.3 50.3 48.2 44.7 55.7 25.8 38.3
Mw (g/mol)
Mn (g/mol)
PD
134 515 75 496 173 989 219 752 182 185 208 920 227 461 202 085 205 124 246 300 224 476 106 954 Ð
60 614 36 045 74 703 96 802 77 497 86 812 91 490 96 035 93 763 122 306 102 617 55 825 Ð
2.2 2.1 2.3 2.3 2.4 2.4 2.5 2.1 2.2 2.0 2.2 1.9 Ð
Polymerization conditions (unless speci®ed otherwise): 100 ml of toluene; [Ti] 2.5 10 6 mol; [MAO]/Ti 3000; 50 C; 30 min. Solubility of ethylene: 0.25 mol/l for 29 psi in hexane at 60 C, 0.20 mol/l for 2 bar, 0.40 mol/l for 4 bar in toluene at 50 C. 0.52 mol/l for 45 psi in hexane at 30 C, 0.13 mol/l for 10 psi in hexane at 30 C. 3 Solvent: 100 ml of hexane; 60 C. 4 Solvent: 100 ml of hexane; 30 C; [Ti] 2.5 10 5 mol; [MAO]/[Ti] 2000; ethylene pressure 45 psi. 5 Solvent: 100 ml of hexane; [Ti] 10 10 6 mol; [MAO]/[Ti] 1500; 30 C; 60 min. (Reproduced with permission from Macromolecules 1998, 31, 2028. Copyright 1998 Am. Chem. Soc.) 2
118
Functionalization of Polyolefins
polymer molecular weight is quite high (Mw 200 000 g/mol) and is not signi®cantly dependent on the content of p-MS. The molecular weight distributions (Mw/Mn) < 2.5, similar to most metallocene-based homo- and copolymers, indicate single-site reaction with good comonomer reactivities. It is interesting to note that the incorporation of 1-octene shows some reduction in high p-MS concentration conditions (comparative runs p-471 vs p-476 and p-475 vs p-474). Following the enchainment of p-MS, subsequent insertion of ethylene is faster than that of 1-octene, possibly due to steric hindrance at the active site. In run p-478, using the same comonomer concentration, the resulting terpolymer having an ethylene/ 1-octene/p-MS mole ratio of 9/2/1 indicates the comonomer reactivity sequence ethylene > 1-octane > p-MS. The thermal transition temperature of PO-p-MS terpolymer was examined by DSC studies. Figure 7.5 compares the DSC curves for two PO-p-MS terpolymers (runs p-472 and p-478) and one poly(ethylene-co-p-MS) (run p-383). Comparing the curves for runs p-472 and p-383, both having the same ethylene ( 60 mol%) content but different 1-octene/p-MS ratios, the Tg changes from > 30 C in run p-383 (with no 1-octene) to < 55 C in run p-472 (with 35 mol% 1-octene). It is very interesting to note that the EO-p-MS sample with even up to 8 mol% of p-MS still shows Tg < 45 C, which is very different from the
DSC (mw)
(a)
(b)
(c)
– 80
– 60
– 40
– 20 0 20 Temperature (°C)
40
60
80
Figure 7.5 DSC curves for two PO-p-MS terpolymers (a) runs p-472 and (c) p-478 and (b) one poly(ethyleneco-p-MS) (run p-383). (Redrawn from Macromolecules 1998, 31, 2028. Copyright 1998 Am. Chem. Soc.)
Functionalization via Reactive Polyolefins
119
corresponding EP-p-MS copolymer discussed in the previous section. These results clearly show the advantages of the 1-octene comonomer (over propylene), which assures the formation of an amorphous polyole®n elastomer with low Tg and high p-MS content. The DSC curve of the terpolymer (p-478) contains a very weak crystalline peak at 5 C. Apparently, a total concentration of 1-octene and p-MS of more than 30 mol% may be necessary to completely eliminate the crystallization of ethylene sequences.
5 FUNCTIONALIZATION OF P-METHYLSTYRENE CONTAINING POLYMERS107±109,291,307 The major research interest concerning the incorporation of p-MS into polyole®ns is due to its versatility283±292 in accessing a broad range of functional groups. The benzylic protons are ready for many chemical reactions, such as halogenation, oxidation, and metallation. Most of functionalization reactions take place exclusively at the p-CH3 position. Therefore, the polymer backbone is untouched during functionalization and the extent of functionalization is governed by the concentration of p-MS groups. 5.1
Lithiation and Subsequent Transformation Reactions108
The lithiation and subsequent reactions provide a convenient way to access many functional polymers. Equation (7.3) shows some of the reactions that were studied in PE-p-MS copolymers. The lithiation reaction was carried out by mixing PE-p-MS powder with an excess alkyllithium, such as sec-BuLi and n-BuLi, in the presence of TMEDA at 60 C for a few hours. TMEDA functioned as a dissociation reagent of butyllithium. The unreacted reagents can be easily removed by a few cycles of ®ltration and washing the lithiated PE powders with a hydrocarbon solvent, such as cyclohexane or hexane. It is very important to note that the heterogeneous condition allows the easy removal of excess reagent. Some of the lithiated polymer was converted to organosilane-containing polymer by reacting with a slight excess of chlorotrimethylsilane. Figure 7.6 compares the 1H NMR spectra of the starting PE-p-MS, containing 0.9 mol% of p-MS, and the resulting trimethylsilyl-containing PE copolymers, which had been metallated by either sec- or n-BuLi/TMEDA, respectively, under the same reaction conditions. In Fig. 7.6(a), in addition to the major chemical shift at 1.35 ppm corresponding to CH2, there are three minor chemical shifts around 2.35, 2.5, and 7.0±7.1 ppm corresponding to CH3, CH, and aromatic protons in p-MS units, respectively. After the functionalization reaction, Fig. 7.6(b) and (c) show the reduction of peak intensity at 2.35 ppm and no detectable intensity change at either
120
Functionalization of Polyolefins
(CH2CH2)x(CH2CH)y ⱍ
(CH2CH2)x(CH2CH)y ⱍ
ⱍ
ⱍ
CH3
CH2
ⱍ
Si(CH3)3 ClSi(CH3)3
ⱍ
COOH
CO2
O
(CH2CH2)x(CH2CH)y ⱍ
(CH2CH2)x(CH2CH)y ⱍ P(φ)2Cl
ⱍ
ⱍ
CH2Li
(CH2)3 ⱍ
OH
CH3-O-9-BBN
(CH2CH2)x(CH2CH)y ⱍ
(CH2CH2)x(CH2CH)y ⱍ
ⱍ
7:3
ⱍ
CH2
CH2
B
P
ⱍ
ⱍ
ⱍ
ⱍ
NaOH/H2O2
(CH2CH2)x(CH2CH)y ⱍ ⱍ
CH2 ⱍ
OH
of the 2.5 and 7.0±7.1 ppm chemical shifts. In addition, two new peaks at 0.05 and 2.1 ppm, corresponding to Si±(CH3)3 and -CH2±Si, are observed. Overall, the results indicate a ``clean'' and selective metallation reaction at the p-methyl group. Benzylic protons in the backbone are totally protected from attack by lithiation reagents. The integrated intensity ratio between the chemical shift at 0.05 ppm and the chemical shifts between 7.0 and 7.1 ppm, and the number of protons both chemical shifts represent, determines the ef®ciency of the metallation reaction. The n-BuLi/TMEDA converted only 24 mol% of p-MS to benzyllithium. On the other hand, the sec-BuLi/TMEDA was much more effective, achieving 67 mol% conversion under similar reaction conditions. Apparently, the metallation
Functionalization via Reactive Polyolefins
121
(c)
(b)
(a)
9.0
8.0
7.0
6.0
5.0
4.0
3.0
2.0
1.0
0.0
ppm
Figure 7.6 1H NMR spectra of (a) PE-p-MS containing 0.9 mol% of p-MS and two corresponding trimethylsilyl derivatives prepared via lithiation reactions using (b) n-BuLi/TMEDA and (c) sec-BuLi/TMEDA reagents.
reaction was not inhibited by the insolubility of polyethylene, so most p-MS units must be located in the amorphous phases that are swellable by the appropriate solvent during the reaction. Besides silylation, lithiated PE-p-MS can be readily converted into a variety of other functional groups. The COOH-containing polymer was synthesized by bubbling CO2 gas through a THF slurry of the lithiated PE-p-MS copolymer. It was observed that the color (yellow) was gradually fading. After complete decoloration, the reaction was terminated by methanol. The resulting polymer was completely soluble in xylene or tetrachloroethane at elevated temperatures. Figure 7.7(a) shows the 1H NMR spectrum of the carboxylated polymer. There are two new peaks
122
Functionalization of Polyolefins
(c)
(b)
(a)
9.0 8.5 8.0 7.5 7.0 6.5 6.0 5.5 5.0 4.5 4.0 3.5 3.0 2.5 2.0 1.5 1.0 0.5 0.0 ppm
Figure 7.7 1H NMR spectra of three functionalized PE polymers containing (a) -CH2COOH, (b) -(CH2)3±OH, and (c) -CH2±OH.
around 3.65 ppm and 7.1±7.4 ppm, corresponding to benzylic protons adjacent to the carboxylic acid (±COOH) group and p-CH2COOH-substituted aromatic protons, respectively. The IR spectrum also demonstrated the existence of a carbonylic acid group in the polymer. The degree of the carboxylation reaction (the ratio between [p-CH2COOH] in the copolymer and [p-CH3] in the starting copolymer) estimated from the ratio of integrated peak areas at 3.65 ppm and 7.0±7.4 ppm is 60.3%. From results described in the previous section, we know that the ef®ciency of a lithiation reaction by s-BuLi/TMEDA is about 60%. Accordingly, the transformation reaction from the lithiated polymer to COOH-containing polymer is near completion.
Functionalization via Reactive Polyolefins
123
On the other hand, by bubbling ethylene oxide gas through the lithiated PE± p-MS/THF suspension solution, a hydroxylated PE containing p-(1-hydroxypropyl)styrene units was obtained. The 1H NMR spectrum of the resulting polymer is shown in Fig. 7.7(b), with three new peaks at 3.71, 2.73, and 2.17 ppm corresponding to three types of CH2 protons between the aromatic ring and the OH group. The ef®ciency of the functionalization reaction calculated from the ratio of the integrated peak areas between 3.7 ppm and 7.0±7.1 ppm is 76.3%. It was a very clean and ef®cient reaction. The reaction between methoxyl-9-BBN and n-BuLi/TMEDA-lithiated PE-p-MS in THF at room temperature gave a borane-containing polymer. It has been well demonstrated in Chapter 6 that the borane-containing polymer is a very versatile intermediate for further functionalization reactions and graft/block copolymerization reactions. As one example, the borane-containing polymer was converted to a benzylic-alcohol-containing polymer by hydrolyzing in the presence of NaOH/ H2O2 in THF. Figure 7.7(c) shows the 1H NMR spectrum of the resulting polymer. Two new peaks around 4.7 ppm and 7.1±7.4 ppm correspond to the benzylic protons next to the OH group and the aromatic protons of benzylic alcohol. The extent of the functionalization estimated from the ratio of integrated peak areas around 4.7 ppm and 7.0±7.4 ppm is 22.1%. Since the lithiation ef®ciency estimated by converting to a silane group is 24%, the ef®ciency of the reaction between lithiated polymer and methoxy-9-BBN is 92% (assuming a complete hydrolysis reaction of boranecontaining polymer). Amidation of PE-p-MS was achieved by the reaction between a sec-BuLi/ TMEDA-lithiated PE-p-MS and phenyl isocyanate (Ph±N=C=O) in benzene at 60 C for 1 h. The chemistry involves a nucleophilic substitution reaction of benzyllithium to Ph±N=C=O and a hydrogen rearrangement reaction. The IH NMR spectrum of the amide-group-containing copolymer shows two new peaks at 3.71 ppm and 7.1±7.5 ppm corresponding to benzylic protons adjacent to the amide group, aromatic protons in the amide group, and aromatic protons in the amidesubstituted styrene unit, respectively. The extent of the functionalization estimated from the ratio of integrated peak areas between benzylic protons in backbone (2.56 ppm) and benzylic protons in the unreacted p-CH3 group (2.35 ppm) is 49.8%. The PPh2-containing polymer was also achieved through the reaction between a sec-BuLi/TMEDA-lithiated PE-p-MS and chlorodiphenylphosphine in THF at room temperature. The1H NMR spectrum of the resulting polymer shows three additional peaks at 3.6, 7.45, and 7.70 ppm corresponding to benzylic protons next to the PPh2 group and the aromatic protons of two phenyl groups bonded to phosphine. The degree of the functionalization estimated by 1H NMR is 56.7%. A successful functionalization reaction not only achieves the needed functionalities but also preserves the desirable properties, such as molecular weight, crystallinity, melting temperature, etc. of polyole®ns. Figure 7.8 compares the GPC curves of PE-p-MS and the resulting trimethylsilyl-containing PE-p-MS. It is clear that the molecular weight and molecular weight distribution of the copolymer before and after functionalization are almost identical within the experimental error.
(a)
0.8
(b)
dwt/d(logM)
0.6
0.4
0.2
0 5.6
5.4
5.2
5
4.8
4.6
4.4 4.2 log MW
4
3.8
3.6
3.4
3.2
3
Figure 7.8 GPC curves of (a) PE-p-MS copolymer with 1 mol% of p-MS and (b) the resulting trimethylsilyl-containing copolymer with 60% modi®cation.
9 8
Heat flow (w/s)
7
(a) (b)
6 (c)
5 4 3 2 1 80
90
100
110 120 Temperature (°C)
130
140
150
Figure 7.9 DSC curves of (a) PE-p-MS with 1 mol% of p-MS and two functionalized copolymers containing (b) trimethylsilyl and (c) carboxylic acid groups, both with 60% functionalization.
Functionalization via Reactive Polyolefins
125
Figure 7.9 compares DSC curves of a PE-p-MS and the two resulting functionalized copolymers, containing trimethylsilyl and carboxylic acid groups, respectively. It is signi®cant that only very slight differences exist in both Tm and crystallinity before and after functionalization. Apparently, lithiation and subsequent transformation reactions exclusively take place at the p-CH3 groups of the p-MS units in the copolymer without attacking the polymer backbone. The above examples are by no means exhaustive, as many other functionalities can be achieved by this approach. In addition, the same chemistry can be easily extended to other polyole®ns, such as PP-p-MS. Overall, the lithiated PE-p-MS can be very ef®ciently converted to a variety of functional polyethylene copolymers under relatively mild reaction conditions. The high ef®ciency of the transformation reactions may be attributed to the chemical immobilization of the lithium reagent and the relatively high stability of the benzyllithium. 5.2
Free Radical Maleic Anhydride Grafting Reaction307
Maleic-anhydride- (MA) modi®ed polyole®ns are among the most important functional polyole®ns in industry due to the low cost of maleic anhydride and the high activity of the anhydride group. Unfortunately, the current free radical MA grafting reactions involve many undesirable side reactions (crosslinking, chain scission, etc.) as illustrated in Eq. (7.4): PE
CH2CH2
PE
CH2CH *
(Crosslinking)
PP
CH3 ⱍ CH2CH
*OR
PP
CH3 ⱍ CH2C * (Chain scission)
PE PP
PE PP
CH2CH
CH2CH
7:4 CH3
O
O
O
PE PP
CH2CH
CH2 O
O
O
CH2 *
(Stable radical)
126
Functionalization of Polyolefins
The p-MS groups in polyole®ns, PE-p-MS, PP-p-MS, EP-p-MS, and EO-p-MS, provide the selective sites for free radical grafting reaction. Therefore, the resulting MA-grafted copolymers have a well-controlled molecular structure (i.e. molecular weight and MA content). To enhance the selectivity on p-MS units, the MA grafting reactions of semicrystalline PE-p-MS and PP-p-MS copolymers are usually carried out in suspension (heterogeneous) reaction conditions. Since the p-MS side groups are only located in the amorphous phase, the swollen amorphous domains provide the physical contacts between the p-MS groups, initiator, and MA reagent. On the other hand, the secondary CH2 (backbone) or tertiary CH (backbone) units in the crystalline domains are largely intact. Table 7.6 summarizes the results of MA modi®cation of PE-p-MS copolymer in suspension conditions at 75 C. Benzene was used as the solvent and BPO as the initiator. The MA graft content was determined by IR spectrum. Two commercial PE polymers (HDPE and LDPE) and poly(ethylene-co-styrene) (PE-S) copolymer were also modi®ed under identical Table 7.6
Summary of MA modification1 of PE homo- and co-polymers MA-modified PE
Sample
[Comonomer] (mol%)
BPO (wt%)
MA (wt%)
Mv ( 104 g/mol)
Tm ( C)
HDPE HDPE-g-MA-1 HDPE-g-MA-2 LDPE LDPE-g-MA-1 LDPE-g-MA-2 PE-S PE-S-g-MA-1 PE-S-g-MA-2 PE-S-g-MA-3 PE-p-MS-1 PE-p-MS-1-g-MA-1 PE-p-MS-1-g-MA-2 PE-p-MS-1-g-MA-3 PE-p-MS-2 PE-p-MS-2-g-MA-1 PE-p-MS-2-g-MA-2 PE-p-MS-2-g-MA-3 PE-p-MS-3 PE-p-MS-3-g-MA-1 PE-p-MS-3-g-MA-2 PE-p-MS-3-g-MA-3
Ð Ð Ð Ð Ð Ð [S] 5 [S] 5 [S] 5 [S] 5 [p-MS] 2 [p-MS] 2 [p-MS] 2 [p-MS] 2 [p-MS] 5 [p-MS] 5 [p-MS] 5 [p-MS] 5 [p-MS] 10 [p-MS] 10 [p-MS] 10 [p-MS] 10
Ð 0.2 0.5 Ð 0.2 0.5 Ð 0.2 0.5 1.0 Ð 0.2 0.5 1.0 Ð 0.2 0.5 1.0 Ð 0.2 0.5 1.0
Ð 0 0 Ð 0.4 1.1 Ð 1.0 1.6 2.3 Ð 1.5 2.4 3.2 Ð 2.9 3.7 6.4 Ð 5.6 7.9 8.8
14.2 14.3 14.2 4.38 4.52 Crosslinking2 3.54 3.61 Crosslinking2 Crosslinking2 6.10 6.23 6.40 6.41 3.83 3.94 4.06 4.07 3.18 3.23 3.35 3.63
135 135 135 115 115 Ð 122 122 Ð Ð 130 129 128 128 117 117 116 117 97 96 97 97
1
Reaction condition: 5 g polymer, 2 g MA, 50 ml benzene, 75 C, 3 h. Crosslinking is indicated by the poor solubility of polymer in decalin at 135 C. [Reproduced with permission from Synthesis of Maleic Anhydride Grafted Polyethylene and Polypropylene with Controlled Molecular Structures, Bing Lu and T. C. Chung, J. polym. Sci: Part A: Polym. Chem. Ed.. 2000, 38, 1337. Copyright 2000 John Wiley & Sons] 2
Functionalization via Reactive Polyolefins
127
conditions and used as control reactions to study the role of p-MS in the modi®cation reaction. It is clearly shown that the MA-modi®ed PE-p-MS copolymers (PE-p-MS-g-MA) had higher MA graft contents than the MA-modi®ed PE (HDPE and LDPE) and PES polymers. In addition, all MA-modi®ed PE-p-MS copolymers showed intrinsic viscosities similar to those of the corresponding starting copolymers, strongly indicating that there was no detectable crosslinking in the MA-modi®ed PE-p-MS copolymers. As the initiator concentration or p-MS concentration in the copolymers increases, the MA graft content increases (for instance, compare the PE-p-MS-1, 2, and 3 series). In the control reactions, the HDPE sample showed almost no graft reaction under identical conditions, possibly due to its high crystallinity and thus less swelling at 75 C. The LDPE samples showed some graft reactions at a low initiator concentration, but had extensive crosslinking side reactions at a high initiator concentration. It is interesting to compare the modi®cation of PE-p-MS copolymers with PE-S copolymers. Under identical reaction conditions and comonomer contents, the MA-modi®ed PE-p-MS-2 copolymers consistently had higher MA-grafted contents than the modi®ed PE-S-1 copolymers. In addition, the MA-modi®ed PE-pMS-2 copolymers showed no crosslinking in all the reactions, but the modi®ed PES-1 had crosslinking at high initiator concentrations. This strongly indicates the advantage of p-MS units ± due to the steric effect, the backbone tertiary benzylic proton has less reactivity than the pendant p-methyl group. In general, the melting point of the PE-p-MS-g-MA copolymer is similar to that of the starting PE-p-MS copolymer, indicating that the crystalline phase did not change signi®cantly during the modi®cation. This may be attributed to the suspension process, which keeps the crystalline phase untouched during the reaction. Table 7.7 summarizes the MA modi®cation of PP-p-MS copolymers containing 0.8 and 1.5 mol% p-MS units in suspension conditions at 125 C. Biphenyl was used as the solvent and dicumyl peroxide (DCP) was the initiator. A commercial PP polymer was also modi®ed under identical conditions and used for the control reactions. The MA-modi®ed PP-p-MS copolymers (PP-p-MS-g-MA) also show much higher MA graft contents than the commercial PP homopolymer. In addition, the PP-p-MS-g-MA copolymer exhibits an intrinsic viscosity that is similar to the corresponding starting PP-p-MS copolymer. This indicates no detectable side reaction, whereas the modi®ed PP homopolymer shows severe degradation, especially at a high initiator concentration. Comparing the PP-p-MS-1 and PP-p-MS-2 sets, higher p-MS and initiator concentrations result in higher MA graft content. To identify the grafting point, MA-modi®ed copolymers were analyzed by 1 H-NMR. Figure 7.10 shows the 1H-NMR spectra of the PE-p-MS-g-MA and PP-pMS-g-MA copolymers, respectively. There are three new peaks at identical shifts, 2.7, 3.5, and 3.7 ppm, in each MA-modi®ed sample. The peak at 2.7 ppm corresponds to methylene protons of -CH2-MA, and two new peaks at 3.5 and 3.7 ppm are assigned to the methylene and methine protons of succinic anhydride, respectively. Therefore, the NMR study clearly indicates that the MA grafting reaction
128
Functionalization of Polyolefins Table 7.7 Summary of MA modification1 of PP homo- and copolymers MA-modified PP Polymers Sample
[p-MS] (mol%)
Initiator (wt%)
PP PP-g-MA-1 PP-g-MA-2 PP-p-MS-1 PP-p-MS-1-g-MA-1 PP-p-MS-1-g-MA-2 PP-p-MS-2 PP-p-MS-2-g-MA-1 PP-p-MS-2-g-MA-2
0 0 0 0.8 0.8 0.8 1.5 1.5 1.5
Ð 1.0 2.0 Ð 1.0 2.0 Ð 1.0 2.0
(10
Mv 5 g/mol)
2.14 1.32 0.61 1.60 1.57 1.53 1.31 1.30 1.26
MA (wt%)
Tm ( C)
Ð 0.7 1.1 Ð 1.6 2.2 Ð 2.1 2.8
162 161 158 157 156 155 153 152 152
1 Reaction conditions: 5 g polymer, 2 g MA, 50 g biphenyl, DCP initiator, 125 C, 3 h. [Reproduced with permission from Synthesis of Maleic Anhydride Grafted Polyethylene and Polypropylene with Controlled Molecular Structures, Bing Lu and T. C. Chung, J. polym. Sci: Part A: Polym. Chem. Ed. 2000, 38, 1337. Copyright 2000 John Wiley & Sons]
took place on the p-methyl groups of the copolymers. In the past, it was very dif®cult to observe the resonance of protons of MA on MA-grafted polymers, possibly due to the dipolar broadening of resonance near the graft points, which have restricted mobility.308±310 Thus, the observance of MA protons in both the PE-p-MS-g-MA and PP-p-MS-g-MA copolymers must be attributed to the fact that they are on the ¯exible pendant p-methyl groups and have high mobility. It is very interesting to note that some short-chain MA oligomers may also form in both PE-p-MS-g-MA and PP-p-MS-g-MA samples. Comparing the integrated peak intensities between 2.3 (-CH3) and 2.7 ppm (-CH2-MA) in Fig. 7.10(b) with the corresponding protons, about 60% (0.5 mol%) of the p-MS units were involved in the free radical grafting reaction. However, 2.2 wt% ( 1.0 mol%) of MA units were observed in PP-p-MS-g-MA, averaging about 2 MA units per p-MS-activated site. The low grafting temperatures (75 and 125 C for PE-p-MS and PP-p-MS, respectively) may offer favorable reaction conditions for the oligomerization308,311 of MA monomers. 5.3
Halogenation Reactions
A bromination reaction was carried out in CCl4 using benzoyl peroxide (BPO) as a free radical initiator and N-bromosuccinimide (NBS) as a bromination reagent. The reaction was performed at re¯uxing temperature under a nitrogen atmosphere in a dark environment and a 1.5/1 ratio of NBS to p-CH3. A slightly brown polymer powder was obtained. On the other hand, a chlorination reaction was performed by using 1,2-azobisisobutyronitrile (AIBN) as a free radical initiator and an excess amount of sulfuryl chloride (SO2Cl2) as a chlorination reagent in CCl4 at 60 C.
Functionalization via Reactive Polyolefins
129
7.5
7.0
6.5
6.0
5.5
5.0
4.5
4.0 3.5 ppm
3.0
2.5
2.0
1.5
1.0
0.5
0.0 –0.5
7.5
7.0
6.5
6.0
5.5
5.0
4.5
4.0 3.5 ppm
3.0
2.5
2.0
1.5
1.0
0.5
0.0 –0.5
Figure 7.10 copolymers.
1
H-NMR spectra of (top) PE-p-MS-g-MA and (bottom) PP-p-MS-g-MA
Figure 7.11 shows 1H NMR spectra of the brominated and chlorinated PE-p-MS polymers. Two major new peaks at 4.54 ppm and 7.34 ppm, corresponding to the protons in benzylbromide (-CH2Br) and the aromatic protons in the p-bromomethylstyrene unit, are observed in Fig. 7.11(b). It is clearly shown that the bromination reaction predominantly takes place at the p-CH3 position. The degree of bromination estimated from the integrated peak areas between the p-bromomethyl
130
Functionalization of Polyolefins *
*
Figure 7.11 polymers.
1
H NMR spectra of (a) chlorinated and (b) brominated PE-p-MS
protons at 4.54 ppm and the aromatic protons at 7.0±7.4 ppm is 56.8%. In Fig. 7.11(a), two peaks at 4.63 and 7.34 ppm correspond to p-chloromethyl protons (-CH2Cl) and the aromatic protons in -CH2Cl. The peaks at 3.95 and 1.80 ppm are attributed to the protons of ±CHCl±CH2± and ±CHCl±CH2± or ±CH2C()Cl±CH2±, respectively, in the backbone. This seems to suggest that, in addition to the p-CH3 position, chlorination also remarkably takes place at the methylene groups in the polymer backbone. The degree of chlorination at p-CH3 estimated from the integrated peak areas between the p-chloromethyl protons at 4.63 ppm and the aromatic protons at 7.0±7.4 ppm is only 30%. Figure 7.12 compares the DSC (top) and GPC (bottom) curves of the starting PEp-MS copolymer and the resulting brominated and chlorinated copolymers. It is clear that both bromination and chlorination do not signi®cantly in¯uence the melting point and crystallinity of the copolymer. Both reactions predominantly take
Functionalization via Reactive Polyolefins
131
5.0 4.5 (a)
Heat flow (W/g)
4.0 3.5
(b) (c)
3.0 2.5 2.0 1.5 1.0 50
75
100 125 Temperature (°C)
150
1.0 (a) 0.8
(b)
(c)
dwt /d(logM)
0.6
0.4
0.2
0.0 5.8
5.6
5.4
5.2
5.0
4.8 4.6 4.4 Log Mw
4.2
4.0
3.8
3.6
Figure 7.12 DSC (top) and GPC (bottom) curves of (a) PE-p-MS copolymer with 3.13 mol% of p-MS and the resulting (b) brominated (56.8%) and (c) chlorinated (30% for p-CH3) copolymers.
place in the amorphous phases. However, the GPC results show a decrease in the molecular weight (Mw 51 762, Mn 19 356, Mw/Mn 2.67) of the brominated sample compared to the starting copolymer (Mw 96 706, Mn 35 192, Mw/ Mn 2.75). This may be due to a very low degree of bromination taking place at the benzylic protons in backbone, which causes the formation of tertiary radical and polymer chain scission. On the other hand, for the chlorination reaction ± despite
132
Functionalization of Polyolefins
a remarkably high degree of chlorination in the polymer backbone ± the molecular weight and molecular weight distribution of the chlorinated polymer is almost identical to that of the starting copolymer. The tertiary benzylchloride may be stable enough to prevent the polymer from chain scission. The brominated and chlorinated poly(ethylene-co-p-MS) copolymers can be further converted to a variety of functionalized copolymers. The chemistry has been extensively demonstrated in references 312 and 313. However, it is worth noting that the brominated copolymers are not very stable, particularly when at elevated temperature and exposed to light. A crosslinking reaction was observed when the polymer solution was maintained at over 120 C in air for a certain time. Therefore, special care is needed during the processing and further functionalization. 5.4
Oxidation Reaction
Oxidation of the PE-p-MS copolymer was carried out by bubbling oxygen through the polymer solution containing a solvent mixture of chlorobenzene and acetic acid (3/1 in volume) and a catalyst system of cobalt (II) acetate tetrahydrate (CoAc2 4H2O) and sodium bromide (NaBr) at 105 C for 3 h. The resulting polymer was precipitated by methanol. The oxidation product is soluble in xylene or 1,1,2,2tetrachloroethane at an elevated temperature. The 1H NMR spectrum of the product shows that several new peaks appear in the range of 7.2±8.0 ppm, which correspond to the aromatic protons of p-COOH- and/or p-CHO-substituted styrene units in the copolymer. Apparently, the oxidation 8 7
Heat flow (W/g)
6 5
(a)
(b)
4 3 2 1 0 50
75
100 Temperature (˚C)
125
150
Figure 7.13 DSC curves of (a) the starting PE-p-MS copolymer and (b) the carboxylated polymer.
Functionalization via Reactive Polyolefins
133
reaction of the p-CH3 groups in poly(ethylene-co-p-MS) by Co(III)Ac2Br is consistent with the results shown in the literature.314 Figure 7.13 compares the DSC curves of the starting PE-p-MS copolymer and the resulting oxidized copolymer. It is interesting to observe that both the Tm and Hf (123.4 C and 98.3 J/g) of the oxidized polymer are slightly higher than that of the starting copolymer (Tm 120.2 C, Hf 77.9 J/g). This suggests that the oxidation selectively takes place at p-CH3, otherwise the crystallization would be hindered instead of being enhanced when it takes place in the backbone. The CHO and COOH functional groups may induce relatively strong intra- or intermolecular interactions, such as hydrogen bonding, which would increase the rigidity of the polymer and consequently increase the Tm and enhance crystallization.
6
SUMMARY
This chapter summarizes the synthesis, properties, and applications of ``reactive'' polyole®n copolymers containing p-methylstyrene (p-MS) groups. The combination of the copolymerization ability of metallocene catalyst and the electronic donating effect of p-MS provides a favorable reaction condition for both co- and terpolymerization reactions between p-MS and -ole®ns (ethylene, propylene, 1-octene, etc.). A broad composition range of co- and terpolymers ± such as poly(ethyleneco-p-methylstyrene) (PE-p-MS), poly(ethylene-ter-propylene-ter-p-methylstyrene) (EP-p-MS) and poly(ethylene-ter-1-octene-ter-p-methylstyrene) (EO-p-MS) ± with properties ranging from semicrystalline thermoplastics to completely amorphous elastomers have been prepared with narrow molecular weight and composition distributions. In turn, the p-MS groups in polyole®ns are very versatile intermediates and can be interconverted to many desirable functional groups, such as ±OH, ±NH2, ±COOH, maleic anhydride, silane, and halides. In addition, the p-MS groups in polyole®ns can be easily metallated to produce polymeric anions for living ``anionic'' graft-from polymerization. As will be discussed in Chapter 11, many graft copolymers ± such as PE-g-PS, PE-g-PMS, PE-g-PMMA, PE-g-PAN, PP-g-PS, PP-g-PMS, PP-g-PB, PP-g-PMMA, and PE-g-PAN ± have been prepared with relatively well-de®ned molecular structure.
This Page Intentionally Left Blank
8 Functionalization via Reactive Polyole®ns Containing Unsaturated Groups
1
INTRODUCTION
A classic method of activating polyole®ns is the introduction of unsaturation. Commercial EPDM (ethylene/propylene/diene terpolymers)3 is a good example that contains several mol% of diene units, such as 5-ethylidene-2-norbornene (ENB), 1,4-hexadiene, and dicyclopentadiene (DCP). The diene units in EPDM are essential for the subsequent crosslinking and functionalization reactions to form the desirable rubber products. In the past, commercial EPDM polymers were mostly prepared by vanadiumbased homogeneous Ziegler±Natta catalysts.315 Diene conversion was usually low due to the rapid reduction of the active sites. To prevent the undesirable crosslinking reaction during the polymerization, most of diene monomers used had to be asymmetric and contain only one polymerizable ole®n unit, either an -ole®n or a constrained cyclo-ole®n. As discussed, the preparation of polyole®n co- and terpolymers has been greatly enhanced by single-site metallocene catalysis. The same bene®ts are clearly shown in the copolymerization of -ole®n and diene monomers. The well-de®ned constrained metallocene catalysts that have a spatially opened catalytic site allow the effective incorporation of large diene monomers, even styrenic dienes,316 with much smaller -ole®ns, such as ethylene and 1-propene. In addition, it is possible to design a metallocene active site with a speci®c stereo opening that exhibits excellent selectivity in ole®n enchainment. This selectivity broadens the choice of dienes (even symmetric dienes) and provides high diene incorporation into polyole®ns without causing branching or/and crosslinking side reactions. The large availability of unsaturated polyole®ns,111,247,248,317±324 from thermoplastics (PE and PP) to elastomers (EP), offers an excellent opportunity for choosing them as the starting material in the preparation of functional polyole®n and graft copolymers. In addition to the classic organic ole®n reactions, some reactive ole®n
136
Functionalization of Polyolefins
species are capable of serving as initiators or monomers in graft reactions. The subsequent polymerization reaction in the presence of unsaturated polyole®n can produce a polyole®n graft copolymer with several polymer side chains (these can be polar functional polymers if free radical polymerization is used). This reaction process offers a very convenient route in the preparation of functional polyole®n graft copolymers (this will be discussed in Chapter 11).
2
DIENE MONOMERS
Many diene monomers were studied in the co- and terpolymerization reactions with -ole®ns. They can be summarized into four areas, including symmetric aliphatic,319,324 asymmetric aliphatic,247,248,317,318 cyclic,320,323 and aromatic dienes,111,322 as shown in Scheme 8.1. Cyclic dienes
Aliphatic dienes Symmetric
Asymmetric
1,3-butadiene 1,4-hexadiene 1,5-hexadiene 4-methyl-1,4 hexadiene 1,7-octadiene 1,9-decadiene 1,11-dodecadiene
Aromatic dienes
2-methylene-5-norbornene divinylbenzene 5-vinyl-2-norbornene p-(3-butenyl)styrene dicyclopentadiene
Scheme 8.1
In most polymerization reactions, the conjugated 1,3-butadiene was found to drastically reduce catalyst activity as well as the copolymer molecular weight. The symmetric aliphatic dienes, such as 1,5-hexadiene, caused concerns about cyclization and crosslinking reactions. The preferred dienes in polymerization reactions are obviously asymmetric dienes that contain a polymerizable external or strained ole®n and a stable internal ole®n, and the resulting polyole®n has several pendent internal double bonds. However, from the functionalization and graft reaction standpoint, it is very desirable to have a pendent external (instead of internal) double bond that has higher chemical reactivity. In other words, the dienes with two external double bonds, or one strained double bond and one external double bond, are very desirable. However, the polymerization reaction of either monomer has to be very selective, with good control of only one double bond engaging in the polymerization. Four important types of diene monomers ± 1,4-hexadiene, 5-vinyl-2-norbornene, p-(3-butenyl)styrene, and divinylbenzene (shown left to right in Scheme 8.2) ± will be discussed in detail to show the scope and limitations of diene monomers in both polymerization and functionalization reactions.
Functionalization Containing Unsaturated Groups
CH ⱍ CH3
ⱍ CH
ⱍ CH
CH2
CH2CH ⱍ
CH2
CH2
ⱍ CH
CH2CH ⱍ (CH2)2 ⱍ
CH2
ⱍ CH
CH2CH ⱍ ⱍ CH
CH2CH ⱍ CH2 ⱍ CH
ⱍ CH
CH ⱍ CH3
CH2CH ⱍ (CH2)2 ⱍ
CH2CH ⱍ CH2 ⱍ CH
137
CH2
CH2
Scheme 8.2
As expected, the asymmetric non-conjugated 1,4-hexadiene exhibits the most trouble-free incorporation into polyole®ns, especially using metallocene catalysts with constrained ligand geometry. On the other hand, the copolymerization reactions between -ole®ns and the other three diene monomers containing two reactive ole®n sites have to be dealt with using great care. Some selective initiators have to be used to prevent double enchainments. Among them, the copolymerization of p-divinylbenzene (DVB), having two identical styrenic units, is particularly limited to only a few catalysts that can completely avoid branching or/and crosslinking reactions. It is very interesting to note that the resulting DVB copolymers containing pendent styrene units are very versatile for many reactions, including graft reactions by anionic, cationic, free radical, and transition metal coordination polymerizations. 3
CO- AND TERPOLYMERIZATION OF -OLEFINS AND DIENES
Several experimental results will be discussed in depth to explain the general reaction conditions and the resulting unsaturated polyole®n. The discussion will emphasize the metallocene copolymerization that provides excellent diene incorporation, and the resulting copolymers with narrow molecular weight and composition distributions. 3.1
1,4-Hexadiene
The copolymerization of ethylene and 1,4-hexadiene247was carried out in an autoclave under an N2 atmosphere. Usually, the reaction was initiated by charging a
138
Functionalization of Polyolefins
catalyst solution into the mixture of 1,4-hexadiene and ethylene, and maintaining a constant ethylene pressure throughout the polymerization process. The copolymerization was terminated by addition of dilute HCl/CH3OH solution. Table 8.1 compares the experimental results of using three catalyst systems ± one heterogeneous TiCl3 AA/(Et)2AlCl catalyst (I) and two homogeneous metallocene catalysts, Cp2ZrCl2 (II) and Et(Ind)2ZrCl2 (III) with MAO. Overall, both metallocene catalysts show satisfactory reactivities at ambient temperatures. The catalyst activity also systematically increases with the concentration of comonomers (1,4-hexadiene), as seen in the previous borane and p-MS cases. In fact, the catalyst activity in the Et(Ind)2ZrCl2/MAO system attains a value of more than 5 106 g of copolymer per mole of Zr per hour in run I-8, which is about eight times the value for the homopolymerization of ethylene in run I-1 under similar reaction conditions. About 80% of 1,4-hexadiene monomers were incorporated into the PE copolymer after a 2 h reaction time in run I-7. On the other hand, the Ziegler±Natta catalyst, TiCl3 AA/(Et)2AlCl, performs very poorly in the incorporation of 1,4-hexadiene in PE, which must be due to the extreme difference in the comonomer reactivities. It is very important to note that the single site Et(Ind)2ZrCl2/MAO catalyst produces unsaturated PE copolymers with narrow molecular weight and composition distributions. Figure 8.1 compares the GPC and DSC curves of a PE homopolymer and three poly(ethylene-co-1,4-hexadiene) copolymers containing 1.8, 2.7, and 4.6 mol% dienes, respectively. The copolymer molecular weight is reduced in the presence of 1,4-hexadiene comonomers. However, the molecular weight Table 8.1 Summary of copolymerization reactions between ethylene (m1) and 1,4-Hexadiene (m2)1 Copolymer Run Feed m1/m2 Reaction temp/ Cat. activity Diene Mw 10 no. (mol/l) time ( C)/(min) (kg/(mol h)) (mol%) (g/mol) I-1 I-2 I-3 I-4 I-5 I-6 I-7 I-8 II-1 II-2 II-3 II-4 III-1 1
0.38/0 0.38/0.061 0.38/0.122 0.38/0.244 0.38/0.244 0.38/0.244 0.38/0.244 0.38/0.488 0.38/0 0.38/0.244 0.38/0.488 0.38/0.244 0.38/0.244
30/60 30/60 30/60 30/30 30/60 30/90 30/120 30/60 30/70 30/70 30/70 65/60 30/70
644 1048 1480 3828 2320 1960 1550 5120 326 606 1278 1960 0.67
0 1.8 2.7 3.4 3.3 3.5 4.0 4.6 0 2.0 2.8 2.2 0
250 68 43 71 41 42 43 50 291 65 47.5 117
3
Mw/Mn Tm c ( C) (%) 3.6 4.1 4.3 3.7 2.8 2.7 3.2 2.5 6.6 3.3 2.0 3.3
137 121 116 107 110 108 108 107 134 120 114 121
Catalysts: Et(Ind)2ZrCl2/MAO (I), Cp2ZrCl2/MAO (II), and TiCl3 AA/(Et)2AlCl (III).
54 40 31 27 30 27 31 30 65 40 34 49
Functionalization Containing Unsaturated Groups
139
0.84 0.82 0.80 Volts
0.78
(d) (c)
0.76 (b)
(a)
0.74 0.72 0.70 0.68
30 32 34 36 38 40 42 44 46 48 50 52 54 56 58 60 Minutes
(b) (c)
(a)
Heat flow (mW)
(d)
50
75
100 125 Temperature (°C)
150
175
Figure 8.1 GPC (top) and DSC (bottom) curves of (a) HDPE and three poly(ethylene-co-1,4-hexadiene) copolymers, containing (b) 1.2, (c) 2.7, and (d) 4.6 mol% 1,4-hexadiene. (Redrawn from Macromolecules 1994, 27, 7533. Copyright 1994 Am. Chem. Soc.)
distribution remains relatively narrow over a broad range of compositions. The DSC curve of each copolymer shows only a single melting peak that systematically shifts to lower temperatures with increasing diene incorporation. 3.2
5-Vinyl-2-norbornene
The asymmetric 5-vinyl-2-norbornene (VNB) has two reactive double bonds, but they have different reactivity levels, depending on the nature of the catalyst. Both Ziegler±Natta and metallocene catalysts were evaluated in the ethylene and VNB copolymerization reactions.323 Only a very low concentration (<1 mol%) of VNB
140
Functionalization of Polyolefins
Table 8.2 Summary of copolymerization of ethylene (M1) and 5-vinyl-2norbornene (M2)1 Feed M1/M2 (mol/l)
Reaction time (min)
Monomer conversion (%)
Cat. activity (kg of copolymer/g of Zr)
VNB in copolymer (mol%)
[] of copolymer2 (dl/g)
0.11/0 0.11/0.28 0.11/0.33 0.11/0.39 0.11/0.44
20 45 45 45 45
100 49 29 28 20
14.0 7.1 4.1 4.6 3.5
0 6 11 10 14
2.38 0.50 0.37 0.32 0.29
1 Polymerization conditions: [Cp2ZrCl2] 5 10 ene 25 ml; temperature 35 C. 2 Measured in 1,2,4-trichlorobenzene at 135 C.
5
M; [MAO] 7.5 10
2
M; tolu-
was incorporated into PE by the Ziegler±Natta catalyst. However, by changing the catalyst to Cp2ZrCl2/MAO, the VNB incorporation signi®cantly increases. In addition, the enchainment of VNB occurred mostly through the more strained endocyclic double bond, leaving vinyl pendent groups in the copolymer. Table 8.2 summarizes the experimental conditions and results. By adjusting the feed ratio, a copolymer with a high concentration (14 mol%) of VNB has been prepared. No crosslinking was observed in any of the resulting copolymers. 3.3
p-(3-Butenyl)styrene
p-(3-Butenyl)styrene is a very distinctive diene monomer because of its two double bonds, -ole®n and styrenic ole®n, which can engage in polymerization reactions under very different catalyst systems. By choosing TiCl4/Mg(C2H5O)2/ED/ (i-C4H9)3Al (ED: ethyl benzoate) catalyst, the polymerization selectively takes place on the -ole®n unit and produces a copolymer with many pendent styrene units,322 as shown in Scheme 8.2. Unfortunately, only a few experimental results have been reported, and these have incomplete analytic data. In one example, ethylene/propylene/p-(3-butenyl)styrene terpolymer was prepared, by mixing ethylene (3 kg/cm2), propylene (7 kg/ cm2), and p-(3-butenyl)styrene (10 mmol) in 600 ml of toluene at 70 C. The supported TiCl4/Mg(C2H5O)2/ED catalyst (0.015 mmol Ti) with a triisobutylaluminum cocatalyst (3 mmol) produced 222 g of ethylene/propylene/p-(3-butenyl)styrene terpolymer in 1 h. The terpolymer has a p-(3-butenyl)styrene content of 0.15 mol% and an ethylene/propylene mole ratio of 35/65. Overall, the incorporation of p-(3-butenyl)styrene is very small when a Ziegler±Natta catalyst is used. 3.4
Divinylbenzene
The incorporation of symmetric divinylbenzene (DVB) into polyole®n is the most challenging area in these sorts of copolymerization effort. Most polymerization
Functionalization Containing Unsaturated Groups
141
methods, including free radical, cationic, and anionic processes, result in polymers with crosslinking molecular structures. Several reports325 discussed the copolymerization of -ole®n (ethylene and propylene) and DVB by Ziegler±Natta catalysts. The resulting copolymers were very inhomogeneous, showing broad composition and molecular weight distribution. The DVB content in the ethylene and propylene copolymers was below 0.3 mol% (1 wt%) and the overall DVB conversion was only a few percent in each case. In addition, the catalyst activity was inversely proportional to the concentration of DVB in the monomer feed. The extent of side reactions was not reported, and may be very dif®cult to determine due to the very low concentration of DVB in the copolymer. Machida et al.326 reported the copolymerization reaction of ethylene and DVB using metallocene catalysts. In general, the copolymers obtained have a long-chain branched molecular structure. In other words, some of the incorporated DVB units engaged in double enchainments with both ole®n groups. It is interesting to note that some of the copolymers contain a very high degree of unsaturation, with an unsaturation/divinylbenzene (TUS/DOU) ratio up to 200. Most of the unsaturated groups must be derived from the polymer side chains during the termination process. In our laboratory, we have been investigating the metallocene catalysts that can prepare -ole®n/DVB copolymers with a linear copolymer structure.111 In general, the metallocene catalysts that have non-bridged ligand geometry, such as Cp2ZrCl2 with a small open active site, incorporate a very low concentration of DVB units. On the other hand, the metallocene catalysts with constrained ligand geometry, such as [C5Me4(SiMe2NtBu)]TiCl2 with a large open active site, engage in serious double enchainments with both vinyl groups in the DVB monomer, which results in copolymers with branching or/and crosslinking structures. However, there is a small group of metallocene catalysts with narrowly de®ned openings at active sites, which can effectively incorporate DVB comonomers and produce an -ole®n/DVB copolymer with a linear copolymer structure. In other words, the catalyst can effectively incorporate DVB into the polymer through single enchainment, and there is no reactivity with the styrenic units already existing in the polymer, as illustrated in Eq. (8.1): +
R
ⱍ
CH2CH
+ CH2CH
CH
Specific constrained metallocene catalyst
R
ⱍ
(CH2CH2)x(CH2CH)y(CH2CH)z
ⱍ
Cp⬘ + C A– Y M Cp
(I) CH
CH2CH2
8:1
CH2
X
CH2
The ideal -ole®n/DVB copolymer would have a linear polymer structure with narrow molecular weight and composition distributions, and the unsaturation/ divinylbenzene (TUS/DOU) mole ratio would be nearly unity.
142
Functionalization of Polyolefins
Table 8.3 Summary of terpolymerization1 of ethylene, propylene, and divinylbenzene using a Cp2ZrCl2/MAO catalyst Monomer feed
Terpolymer composition (mol%)
Mixed C2/C3 (psi) DVB (mol/l) Yield (g) 60/40 60/40 60/40 60/40 60/40
0 0.214 0.357 0.714 1.290
8.11 5.10 4.84 3.60 3.56
C2
C3
DVB
TUS/DOU
55.9 63.4 68.8 74.8 79.5
44.1 35.7 30.3 23.8 19.0
0 0.46 0.88 1.4 1.5
Ð 0.78 0.85 0.86 0.85
Tg ( C) 52.1 48.7 42.5 Ð2 Ð2
1
Polymerization condition: [cat] 2.5 10 6 mol; [MAO]/[Zr] 3000; solvent: 100 ml hexane; temperature: 50 C; pressure: 30 psi; time: 15 min. 2 Crystallinity existed.
In a typical reaction, ethylene and propylene (with a desirable pressure ratio) were premixed in a reservoir. The premixed gas (with a desirable pressure) was introduced to a 300 ml autoclave that contained DVB and diluent. The terpolymerization started as soon as a metallocene catalyst was added to the reactor, and the reaction continued at a designated reaction temperature (between 50 and 80 C) and constant gas pressure. The reaction time varies between 15 min and 1 h, depending on the catalyst and monomer concentrations, reaction temperature, monomer conversion, and desired molecular weight. Table 8.3 summarizes the reaction conditions and results from using a Cp2ZrCl2/MAO catalyst. Overall, the incorporation of DVB units was very low, with TUS/DOU mole ratio < 1. The poor incorporation of styrenic monomers in this catalyst system is not surprising. The low value of the TUS/DOU mole ratio indicates that some branching reaction may also be taking place. In addition, the molecular weight of the terpolymer was generally very low ± about 10 000 g/mol. Table 8.4 summarizes the experimental results from using an Et(Ind)2ZrCl2/MAO catalyst system. Much higher incorporation of the DVB comonomers was observed in all reactions ± up to 20 mol% of DVB units in the EP-DVB terpolymer has been prepared. In addition, all of the terpolymers were completely soluble (after polymerization reaction) in common organic solvents, such as hexane, toluene, and tetrahydrofuran (THF). In general, the prepared EP-DVB terpolymers have a relatively well-de®ned molecular structure with narrow molecular weight and composition distributions. Figure 8.2 shows a 1H NMR spectrum of an EP polymer containing 2.9 mol% of DVB units. Four chemical shifts were observed at near 5.3, 5.8, 6.8, and 7.0± 7.4 ppm (with the integrated peak intensity ratio near 1 : 1 : 1 : 4), corresponding to three individual vinyl protons and four aromatic protons in the pending styrene unit. The integrated peak intensity ratio indicates that the mole ratio of unsaturation/ divinylbenzene moieties (TUS/DOU) is nearly equal. Small deviations may be associated with contamination by a small quantity of ethylstyrene in the DVB monomer feed, as well as with the sensitivity of the NMR instrument.
Functionalization Containing Unsaturated Groups
143
Table 8.4. Summary of terpolymerization1 of ethylene, propylene, and divinylbenzene using an Et(Ind)2ZrCl2/MAO catalyst Monomer feed
Terpolymer composition (mol %)
Mixed DVB Cat. activity C2/C3 (psi) (mol/l) (kg/(mol Zr h)) 40/60 40/60 40/60 40/60 60/40 60/40
0.1 0.3 0.6 1.2 0.3 0.6
2780 2000 1650 1230 1810 1710
C2
C3
DVB
TUS/DOU
56.4 59.3 62.4 56.0 67.8 65.3
42.5 39.0 32.9 22.9 30.6 30.3
1.1 1.7 4.7 21.1 1.6 4.4
95.0 93.2 97.0 98.0 97.0 95.2
Tg ( C) 50.6 47.3 31.3 21.6 36.4 29.0
1
Polymerization conditions: [cat] 2.5 10 6 mol; [MAO]/[Zr] 3000; solvent: 100 ml hexane; temperature: 50 C; pressure: 30 psi; time: 15 min.
Figure 8.2 DVB units.
1
H NMR spectrum of an EP polymer containing 2.9 mol% of
Figure 8.3 compares the GPC curves of the EP copolymer (E: 64.8 mol%, P: 35.2 mol%) and the EP-DVB terpolymer (E: 63.7 mol%, P: 33.4 mol%, DVB: 2.9 mol%) prepared under identical reaction conditions, except for the addition of the DVB monomer. Both polymers have similar molecular weight distributions (Mw/Mn ~2), indicating nearly ideal reaction conditions. However, it is interesting to note a signi®cantly higher molecular weight in the terpolymer. The presence of DVB monomers may have some effect on the chain transfer reaction. In fact, a signi®cant reduction of chain end unsaturation was observed in the DVB containing terpolymers, as examined by a FTIR spectrum. The DSC results also showed a homogeneous terpolymer microstructure with no detectable melting point (Tm) and
144
Functionalization of Polyolefins (a)
(b)
90 80 70
mV
60 50 40 30 20 28
Figure 8.3 terpolymer.
30
32
34
36
38 40 Minutes
42
44
46
48
50
GPC curves of (a) EP copolymer and (b) EP-DVB
Table 8.5 A summary of terpolymerization1 of ethylene, propylene, and divinylbenzene using [C5Me4(SiMe2NtBu)]TiCl2/MAO catalyst Monomer feed
Terpolymer composition (mol%)
Mixed DVB Cat. activity C2/C3 (psi) (mol/l) (kg/(mol Ti h)) 50/50 50/50 50/50
0.1 0.3 0.6
453 400 421
C2
C3
DVB
TUS/DOU
48.6 48.8 41.3
44.8 37.1 28.8
6.6 14.1 29.9
0.42 0.44 0.52
Tg ( C) n.d.2 n.d.2 n.d.2
1
Polymerization conditions: [cat] 2.5 10 6 mol; [MAO]/[Zr] 3000; solvent: 100 ml hexane; temperature: 50 C; pressure: 30 psi; time: 15 min. 2 Cannot be determined.
a sharp glass transition temperature (Tg) with a ¯at baseline in each terpolymer (shown in Table 8.4). Table 8.5 summarizes the reaction condition and results of the terpolymerization of ethylene, propylene, and DVB by a [C5Me4(SiMe2NtBu)]TiCl2/MAO catalyst. In general, the incorporation of DVB into the ethylene±propylene terpolymer was very effective using a [C5Me4(SiMe2NtBu)]TiCl2/MAO catalyst with even up to a 30 mol% DVB content. However, most of the terpolymers produced are not completely soluble. Some insoluble gel particles are indicative of a certain degree of crosslinking reaction during the polymerization. The DSC curve of each terpolymer shows no clear thermal transition and an uneven baseline. The mole ratio of unsaturation/divinylbenzene moieties (TUS/DOU) in soluble fraction is well below
Functionalization Containing Unsaturated Groups
145
unity, indicating that some secondary side reactions (double enchainment) happened at the pending styrene groups in the copolymer. Overall, the experimental results clearly show that the terpolymerization reaction involving DVB monomers is strongly dependent on the catalyst system used. To prepare a linear EP/DVB terpolymer, the constrained geometry catalyst system with a speci®c opening at the active site (such as Et(Ind)2ZrCl2) is needed to effectively incorporate DVB monomers without causing branching and/or crosslinking problems during the polymerization process.
4
FUNCTIONALIZATION OF UNSATURATED POLYOLEFINS
Numerous ole®n reactions have been applied in the modi®cation of the unsaturated polyole®ns. As expected, the reactivity is very dependent on the nature of the double bond and reaction conditions. In general, the polymer reaction is very demanding due to the limited solubility of the polymer and its intolerance to side reactions, especially uncontrolled crosslinking reactions. These are usually not a problem in small molecule cases. Clean and effective chemical reactions are essential to achieve a functional polyole®n with well-de®ned molecular structure. Several modi®cation reactions will be discussed to illustrate the general functionalization routes. 4.1
Hydroboration Reaction
Hydroboration reaction is the most clean and effective ole®n chemistry, and is readily adopted in the modi®cation of unsaturated polyole®ns containing both internal and external double bonds. An example of hydroboration of poly(ethyleneco-1,4-hexadiene)247 was carried out by suspending the polymer powder in a THF solvent. Despite the heterogeneous conditions, both dialkylborane reagents, 9borabicyclononane (9-BBN) and lithium dimethylborohydride, reacted with all the internal double bonds in 3 h at 65 C. The same reaction results were also observed in poly(propylene-co-1,4-hexadiene) cases.248 In general, the borane-containing polymers are stable for long periods of time as long as O2 is excluded. However, the borane-containing polymer is very versatile for many reactions, as illustrated in Eq. (8.2). Figure 8.4 shows the 1H NMR spectra of PE-OH, PE-OSi(CH3)3, and PE-NH2, which was derived from poly(ethylene-co-1,4-hexadiene) containing 1.8 mol% diene units. The chemical shift at 5.5 ppm (in Fig. 8.4(a)), corresponding to the internal unsaturation, disappears to the limit of NMR sensitivity and exhibits a secondary alcohol methine peak, appearing at 3.6 ppm (in Fig. 8.4(b)). The hydroxy groups in PE-OH can be further converted to silane groups by reacting with chlorotrimethylsilane. As shown in Fig. 8.4(c) silation was complete, with the disappearance of the secondary alcohol methine peak at 3.6 ppm and the appearance
R
ⱍ (CH2CH)x(CH2CH)y ⱍ
CH3
CH ⱍ CH3 R
ⱍⱍ
HB
CH2 ⱍ CH
CH3
R
ⱍ (CH2CH)x(CH2CH)y ⱍ
HB
ⱍ (CH2CH)x(CH2CH)y ⱍ
CH2 ⱍ CH2 ⱍ HCB ⱍ CH3
ⱍ (CH2CH)x(CH2CH)y ⱍ
O
CH2 ⱍ CH2 ⱍ HCOOB ⱍ CH3
ⱍ ⱍ
CH2 ⱍ CH2 CH 3 ⱍ HCB ⱍ CH3 CH3
R
O2
CH2 ⱍ CH2 ⱍ HCNH2 ⱍ CH3
R
ⱍ (CH2CH)x(CH2CH)y ⱍ
CH2 ⱍ CH2 ⱍ HCOH ⱍ CH3
O
R
O
ⱍ (CH2CH)x(CH2CH)y ⱍ
CH2 ⱍ CH2 ⱍ HCO ⱍ CH3
O
R
ⱍ (CH2CH)x(CH2CH)y ⱍ
NaOH/H2O2
O
NH2OSO3H
O
8:2
Functionalization Containing Unsaturated Groups
147
(d) *
*
*
(c) *
(b) *
*
(a) *
* 8.0
7.0
6.0
5.0
4.0 ppm
3.0
2.0
1.0
0.0
Figure 8.4 1H NMR spectra of (a) poly(ethyleneco-1,4-hexadiene, containing 1.8 mol% dienes, and its corresponding functional polymers, (b) PEOH, (c) PE-OSi(CH3)3, and (d) PE-NH2.
of Si±CH3 at around 0.3 ppm. Apparently, hydroboration and oxidation reactions were not inhibited by the insolubility of polyethylene. The ef®ciency of the reaction must be due to the high surface area of the reactive sites. While some of the polyethylene segments are crystallized, the side chains containing the double bonds are expelled into the amorphous phases, which are swellable by the appropriate solvent during the reaction. In addition, the high reactivities in both the hydroboration and oxidation reactions certainly enhance the ef®ciency of the functionalization. All three B±C bonds are quantitatively converted to B±OH and C±OH groups.
148
Functionalization of Polyolefins
The amination reactions of trialkylboranes are usually not quantitative,327 as only one or two B±C bonds in the trialkylborane molecule can be interconverted to C±NH2 groups. The borane adducts after partial interconversion become too stable to continue the amination process. To assure that the B±C bond joined to the polymer side chain is interconverted, selective amination is essential. Lithium dimethylborohydride was therefore used to provide two blocking methyl groups during the amination reaction. As shown in Fig. 8.4(d), the aminated PE shows only two new chemical shifts at around 2.8 ppm, corresponding to two secondary amino methine (CH±NH2) groups located at the 4 and 5-position of 1,4-hexadiene unit. The amino group containing PE is completely soluble in xylene at an elevated temperature. Figure 8.5 compares the DSC curves before and after hydroxylation reaction of poly(ethylene-co-1,4-hexadiene), containing 1.8 mol% dienes. Both samples underwent the same thermal history. There is no signi®cant change in their crystalline structure after functionalization. The GPC study also shows no changes in the molecular weight and molecular weight distribution of copolymers. The borane reagents seem to provide a clean and effective route to the functionalization of polyethylene copolymers despite the semicrystalline structure. The concentration of functional groups in PE is essentially controlled by the quantity of borane reagents and the percentage of double bonds in the PE copolymer. Similar experimental results were also observed in PP and EPDM cases that were carried out in heterogeneous and homogeneous conditions, respectively.
20 18 (a)
16 Heat flow (mW)
14 12 10 8 6 (b)
4 2 0 50
75
100 125 Temperature (°C)
150
175
Figure 8.5 DSC curves of (a) poly(ethylene-co-1,4-hexadiene), containing 1.8 mol% dienes, and (b) its corresponding PE-OH. (Redrawn from Macromolecules 1994, 27, 7533. Copyright 1994 Am. Chem. Soc.)
Functionalization Containing Unsaturated Groups
4.2
149
Crosslinking Reactions
One major advantage of the -ole®n and divinylbenzene (DVB) copolymer is the existence of many pending styrene groups along the backbone that are very reactive in many chemical reactions, including free radical, cationic, anionic, and transition metal coordination processes. The pending styrene moieties can serve as the reactive sites for selective functionalization reactions to produce the functionalized polyole®ns, or serve as the monomers for subsequent graft reactions that produce polyole®n graft copolymers having a polyole®n backbone and other polymer side chains (discussed in Chapter 11). In this section, we will focus on the application of the DVB units in crosslinking reactions. A technology has long been desired in the industry for developing a facile and controllable method to produce crosslinked polyole®n products. The DVB units in polyole®ns provide an attractive new possibility. Both UV and thermal processes have been studied for many -ole®n/DVB copolymers, which may engage in [2 2] and [2 4] reactions, respectively, as shown in Eq. (8.3):
EP
ⱍ ⱍ CHCH2
Thermal
UV
EP ⱍ
EP
8:3
ⱍ CHCH2 ⱍ ⱍ CH2CH ⱍ EP
ⱍ
EP
In general, EP/DVB terpolymers became completely insoluble after heating the sample at 150 C for 1 h. On the other hand, the UV (source: PC-2, 5±10 mW/cm2) crosslinking reaction took place at room temperature. The radiated sample was subjected to extraction by boiling toluene to examine the gel content after radiation. Table 8.6 shows the experimental results of a EP/DVB terpolymer containing 2.3 mol% of DVB units. Figure 8.6 shows the in situ FTIR spectra during the UV radiation of a EP/DVB copolymer containing 2.3 mol% DVB units. The C=C stretching mode at 1630 cm 1 gradually disappears, with several new peaks appearing between 800 and 1000 cm 1 corresponding to the saturated C±C±H defraction modes.
150
Functionalization of Polyolefins Table 8.6 Gel content of EP/DVB terpolymers after UV radiation1 Radiation time (min)
Gel Content2 (wt%)
5 15 30 60 120 240
23.4 35.0 76.3 100 100 100
1
UV source: PC-2, 5±10 mW/cm2. After boiling toluene extraction for 12 h.
2
5
Absorbance
4
3
2
1
2000
1800
1600
1400
1200
1000
800
600
Wavenumber
Figure 8.6 FTIR during UV radiation.
4.3
spectrum
of
EP/DVB
terpolymer
Maleation Reaction
The thermal [2 4] cyclolization reaction of -ole®n/DVB polymers was also extended to the preparation of maleic-anhydride-modi®ed polyole®ns. The pendent styrene moieties in the polymer are very reactive in thermal maleation reaction, as illustrated in Eq. (8.4). Thermal maleation reaction usually took place by mixing a DVB-containing polymer, maleic anhydride, and a free radical inhibitor in a xylene solution. Under nitrogen conditions, the solution was heated to 140 C for 2 h. The maleated polymer was isolated by precipitation in isopropanol. Repeated washings with isopropanol
Functionalization Containing Unsaturated Groups
EP
ⱍ
O
O
O
ⱍ
O
EP
ⱍ CHCH2
151
8:4
O O
and acetone were performed before drying the resulting polymer under vacuum. Figure 8.7 compares the IR spectra of three maleated ethylene/1-octene/DVB terpolymers with 0.86, 1.20, and 11.5 wt% incorporated MA, respectively, and a starting ethylene/1-octene/DVB terpolymer. The strong C=O absorption band (between 1800±1600 cm 1) in Fig. 8.7(d) indicates an almost quantitative reaction between the pendent styrene units and the MA reagent. The other route in the preparation of maleic-anhydride-modi®ed polyole®ns is the application of a suitable borane reagent (such as 9-BBN) that can easily be incorporated into polyole®n via hydroboration reaction of the unsaturated polyole®n, as illustrated in Eq. (8.2). As discussed in Chapter 6, the 9-BBN groups in a polymer can be selectively oxidized by oxygen to form stable polymeric radicals, which can then initiate a free radical graft reaction in the presence of monomers. In the case of the maleic anhydride (MA) monomer, only a single MA unit can be incorporated into the polymer by each polymeric radical due to MA's low capability for homopolymerization. Therefore, the MA concentration is basically controlled by the borane units in the polymer. Table 8.7 summarizes the experimental results of several MA-modi®ed PP polymers,260 which are compared with the corresponding
1.0
Absorbance
0.8
(d)
0.6
(c)
0.4
(b)
0.2
(a)
4000
3500
3000
2500 2000 Wavenumber
1500
1000
500
Figure 8.7 IR spectra of (a) a starting ethylene/1-octene/DVB terpolymer and three maleated ethylene/1-octene/DVB terpolymers containing (b) 0.86, (c) 1.20, and (d) 11.5 wt% of incorporated MA units.
152
Functionalization of Polyolefins
Table 8.7
Summary of MA-grafted PP polymers Reaction conditions1
Product properties
Sample
[HD] (mol%)
MA (g)
ST (g)
Temp. ( C)
Time (min)
MA (wt%)
Mv (105)
Tm ( C)
PP-1 PP-1-MA PP-2 PP-2-MA PP-3 PP-3-MA-1 PP-3-MA-2
0.5 0.5 1.1 1.1 2.4 2.4 2.4
Ð 1 Ð 1 Ð 1 1
Ð 0 Ð 0 Ð 0 1
Ð 20 Ð 20 Ð 20 20
Ð 4 Ð 4 Ð 4 16
Ð 0.4 Ð 0.8 Ð 1.6 5.3
9.7 9.7 7.6 7.5 3.7 3.7 3.8
156 156 154 154 152 152 152
1
Reaction conditions: 1 g hydroborated PP-HD, 20 ml benzene. Table 8.8 Summary of EP-MA polymers derived from an EPDM terpolymer1 Run no.2 EP-MA-1 EP-MA-2 EP-MA-3 EP-MA-4 EP-MA-5
9-BBN/double bond
MA (wt%)
1.0 0.8 0.4 0.2 0.1
10.5 8.6 3.9 2.1 1.1
1 EPDM: E/P 75/25, ENB 5 mol% from Uniroyal Chemical Company. 2 Oxidation and grafting reaction at room temperature.
three starting poly(propylene-co-1,4-hexadiene) copolymers (PP-1, PP-2, and PP3), having 0.5, 1.1, and 2.4 mol% 1,4-hexadiene (HD) units, respectively. Comparing the corresponding MA-modi®ed PP polymers (PP-1-MA, PP-2-MA, and PP-3-MA), the MA graft content is proportional to the 1,4-hexadiene content in the PP copolymer. The molecular weight of the MA-grafted copolymers is almost the same as that of the starting materials, which indicates that there is no signi®cant structural change of the PP backbone during the whole modi®cation process. This is a signi®cant departure from the traditional free radical modi®cation that has extensive side reactions, including PP backbone degradation, as discussed in Chapter 4. We also saw in Chapter 4 that EPDM containing pending ole®nic units quickly forms crosslinked material under free radical reaction conditions. Commercial MAmodi®ed EPDM is normally produced by ene reaction,328 which requires a high temperature (> 240 C) and results in products with a very low MA content. The low graft ef®ciency is attributed to the low reactivity of ene reaction in the hindered ole®n unit. Investigating the borane/O2 approach to EPDM polymers is therefore very interesting. Borane-modi®ed EPDM (EP-B) was prepared by hydroboration of EPDM with a 9-BBN reagent, then the EP-B was reacted with oxygen in the presence of an MA reagent at ambient temperature. Table 8.8 summarizes the results of MA-modi®ed EPDM polymers (EP-MA). All reactions were started with the same EPDM polymer containing 5 mol% ENB units and used various quantities of
Functionalization Containing Unsaturated Groups
153
9-BBN. The 9-BBN/double bond ratio was used to control the concentration of MA units in the EP polymer. Basically, the MA graft content is dependent on the amount of 9-BBN introduced into the polymer. A high MA concentration (>10 wt%) can be incorporated into the EPDM polymer, and the molecular weight and distribution of the EP-MA polymer remain similar to that of the original EPDM seen in the GPC measurement. 4.4
Epoxidation Reaction
Both the internal and external double bonds in the unsaturated polyole®n were also effectively converted to epoxide groups. One example is poly(ethylene-co-5-vinyl-2-norbornene),323 as illustrated in Eq. (8.5):
m-Chloroperbenzoic acid/toluene
CH CH2
8:5
O
The copolymer, containing 10.5 mol% of VNB, was reacted with m-chloroperbenzoic acid in toluene at 55 C for 3 h. An almost quantitative level epoxidation of the vinyl groups was observed without detectable polymer chain degradation. 5
SUMMARY
Functionalization of polyole®n via unsaturated polyole®n offers a valuable chemical route. With a proper choice of the unsaturation moiety and functionalization reaction, it is possible to prepare the desired functional polyole®n with a well-controlled molecular structure. The crucial link is the preparation of an unsaturated polymer with facile unsaturation moieties. The advance of metallocene technology provides the essential solution to empower this chemical approach. With the designed metallocene catalysts, most diene monomers can be effectively incorporated into polyole®n without causing the undesirable branching and crosslinking side reactions. Among them, divinylbenzene (DVB) is the most challenging and interesting one, and is well known to produce crosslinked polymer products in all other polymerization processes. A broad range of linear -ole®n/DVB co-, ter-, and tetrapolymers are available. The incorporated DVB units resemble the styrene monomer and are very facile in many reactions. As will be discussed in Chapter 11, the pendent styrene units not only can be converted to functional groups, but also can engage in other polymerization reactions to form graft copolymers containing a polyole®n backbone and several functional polymer side chains.
This Page Intentionally Left Blank
9 Synthesis of Polyole®ns with a Terminal Functional Group
1
INTRODUCTION
A polymer containing a terminal functional group presents a unique opportunity to serve as a building block for constructing multisegmented polymers ± even polymers with complex supermolecular structures. In polyole®ns, the opportunity is even more interesting due to the lack of functionality and the dif®culty in preparing polyole®n block/graft copolymers and long-chain branched polymer structures. In general, it is always a scienti®c challenge to prepare polymers with a terminal functional group. This chapter summarizes three general methods used in the preparation of polyole®ns with a terminal functional group. They are based on (i) living transition metal coordination polymerization, (ii) in situ chain transfer reaction during transition metal coordination polymerization, and (iii) chemical modi®cation of the polymer having an unsaturated chain end. In general, the chain transfer reaction is the most convenient and effective method for introducing a functional group to the polymer chain end. Metallocene catalysis, with a well-de®ned reaction mechanism, provides the crucial link to achieving high ef®ciency.
2
LIVING POLYMERIZATION APPROACH
In the late 1970s, Doi et al.329±332 reported a living transition metal coordination polymerization of propylene using a homogeneous Ziegler±Natta catalyst, such as V(acac)3/Al(C2H5)2Cl, at a very low reaction temperature (< 65 C). The resulting syndiotactic PP (s-PP) polymers had a narrow molecular weight distribution and the polymer molecular weight was proportional to the polymerization time. As expected, the control termination reaction of a living polymer can result in a polymer chain having a desirable terminal group. One reaction is the termination by
156
Functionalization of Polyolefins
iodine, as shown in Eq. (9.1):
CH3
CH3
CH2CHV* + I2
CH2CHI + V*I
9:1
Iodine reacts quantitatively with the vanadium±carbon bond to yield an iodine± polymer bond. In the 1990s, several new transition metal complexes (metal: Sm, Co, Ni, etc.) were also shown to be useful for living polymerization333±337 of -ole®ns at much higher temperatures. Brookhart et al.338 applied the chemistry to prepare a variety of functional-group-terminated polyethylenes, as shown in Eq. (9.2):
Me5
Me5 Co L
H
+
CH3 +
Co L
Ar
X
L
H
+ Ar
9:2
Me5
Me5 Co
Ar a Ph b p-C6H4Cl c p-C6H4OAc d p-C6H4CF3 e Ar = C6F5
nCH2CH2
Co L
H
+
H2
PE-Ar
(CH2CH2)n–1CH2CH2Ar
The Co(III) with -aryl-substituted, -agostic complexes, processes a functional group and initiates the living polymerization of ethylene. This chemistry introduced the functional group at the onset of the polyethylene chain, and the polymer obtained had a narrow molecular weight distribution. In general, the scope of this living polymerization approach is limited to a few special catalyst systems and reaction conditions. There are no reports of the application of the chemistry in preparing functional-group-terminated i-PP and s-PS polymers. In addition, this route is very dif®cult to apply in the commercial processes due to low catalyst ef®ciency. By the very nature of the living reaction process, each catalyst only produces a polymer chain.
Synthesis of Polyolefins with a Terminal Functional Group
3
157
CHAIN TRANSFER REACTION APPROACH
In my view, the in situ chain transfer reaction during the polymerization process is an ideal route for preparing polymers with a terminal functional group. With the appropriate chain transfer agent (CT), it is possible to have a chain transfer reaction taking place without changing the polymerization rate. Each catalyst can produce multiple polymer chains as usual, and the polymer chain produced has a terminal group (the residue of the chain transfer agent). Usually, the polymer's molecular weight is inversely proportional to the [CT]/[monomer] ratio. In fact, chain transfer to hydrogen is a routine practice in industry to control the polymer's molecular weight and to produce polymers with a completely saturated polymer chain. If a similar reaction kinetic could be applied to the chain transfer agent containing a functional group, the resulting polymer should have a terminal functional group. Thus, this chemistry is a potential drop-in technology with only a small modi®cation to current polymerization conditions, and might also be applicable to all polyole®ns (PE, PP, EP, s-PS, etc.). Several interesting chain transfer agents have been studied in the past two decades. Zinc-alkyl339±344 and aluminum-alkyl345 were reported to act as coinitiator and chain transfer agent in heterogeneous Ziegler±Natta polymerization of propylene. Recently, two chain transfer agents, silane and borane containing Si±H346±349 and B±H121,122 moieties, respectively, have been successfully applied in metallocenemediated polymerization. The chemistry is very general and applicable to most -ole®ns and their mixtures. The resulting silane- and borane-terminated polyole®ns, having relatively well-de®ned molecular structures, were used as intermediates in the preparation of more elaborate polymer structures. It is very interesting to note that the chain transfer reaction has also been shown in styrenic monomers (such as styrene and styrene derivatives) under some polymerization conditions. 3.1
Zinc±Alkyl-Terminated Polyole®ns
For a long time, zinc±alkyl has been known to act in dual roles as a coinitiator (alkylation agent) and as a chain transfer agent in Ziegler±Natta polymerization. The addition of a large amount of Zn(C2H5)2340±342 into the TiCl3/AlEt3-propylene polymerization system causes a marked decrease in the polymer's molecular weight. The resulting Zn-alkyl-terminated PP was converted into various functional groups, such as hydroxy, carboxyl, halogen, or vinyl groups, with approximately 50% yield. In the early 1990s, Soga and co-workers343,344 studied several bis(!-alkenyl)zinc compounds, such as bis(3-butenyl)zinc, bis(7-octenyl)zinc, and (4-methyl-3pentenyl)zinc, in propylene polymerization in an attempt to prepare polypropylene with a terminal alkylzinc group on one chain end and an ole®nic group on the other. As expected in a heterogeneous isospeci®c TiCl3 catalyst system with multiple active sites, the reaction was very complicated, and formed a mixture of polymers with zinc groups on the side chains and chain ends. In fact, bis(!-alkenyl)zinc serves
158
Functionalization of Polyolefins
three roles in the reaction ± coinitiator, comonomer, and chain transfer agent. Overall, the polymer has broad molecular weight and composition distributions. 3.2
Silyl-terminated Polyole®ns
Marks and co-workers346,347 reported that organosilanes (containing Si±H groups) serve as chain transfer agents in lanthanocene-mediated polymerizations to afford silyl-terminated ethylene polymers and copolymers, including ethylene/1-hexene and ethylene/styrene. The chemistry was expanded to group 4 metallocene catalysts348,349 and a broad range of monomers, including ethylene, propylene, 1-hexene, and styrene. The primary silane (PhSiH3) was shown to be an ef®cient chain transfer agent in catalyst systems with an open active site, such as a quasimetallocene cation with a weak coordinating anion. An example is the polymerization of propylene in the presence of a PhSiH3 chain transfer agent and an [Me2Si(Me4C5)tBuN]TiMeB(C6F5)4 -constrained geometry catalyst. As shown in Fig. 9.1, the molecular weight of the resulting silyl-terminated atactic PP (PP-b-SiPhH2) is inversely proportional to [PhSiH3]. It is interesting to note that only one of three Si±H moieties in PhSiH3 involves a chain transfer reaction in the [Me2Si(Me4C5)tBuN]TiMeB(C6F5)4 catalyst system. However, the remaining two Si±H moieties in PP-b-SiPhH2 are still reactive in other catalyst systems having higher open active sites, such as (Me5C5)TiMe2B(C6F5)4 . In one study, PP-b-SiPhH2 was used as a polymeric chain transfer agent in the presence of styrene and a single Cp±Ti system, (Me5C5)TiMe2B(C6F5)4 . A diblock copolymer of atactic polypropylene±syndiotactic polystyrene was produced. 3.3
Borane-terminated Polyole®ns
In our laboratory, we investigated organoboranes (containing B±H moiety) as the chain transfer agents.121,122 The research was formulated with several intriguing questions and objectives in mind. With the known reactivity of Si±H groups to some metallocene catalytic sites, it is very interesting to study similar chain transfer reactions with B±H groups, and their reactivity differences. It is especially interesting to investigate the B±H chain transfer reaction in the catalyst systems, such as {SiMe2[2-Me-4-Ph(Ind)]2}Zr Me2B(C6F5)4 , in which the chain transfer reactions with H±H and Si±H groups are ineffective. With the knowledge of facile ligand exchange of B±H with many metal±alkyl groups, it is very possible that the B±H group has a higher reactivity with M±C active sites. In addition, the resulting boraneterminated polyole®ns would be very desirable, and could serve as the intermediates for preparing a broad range of polar group-terminated polyole®ns as well as diblock copolymers containing both polyole®n and a functional polymer segment. There are two concerns in the B±H chain transfer reaction, namely, hydroboration reaction of the B±H group to -ole®n monomers and ligand exchange reaction between borane and the aluminum-alkyl coinitiator. These reactions are not an issue
Synthesis of Polyolefins with a Terminal Functional Group
159
50
40
Me Si Ti+-H Me N
H
Si
n R
n R
Mn × 10–3
Si H 30 Me Me
+
H n
Ti
Si N
R 20
10
0 0
5
10
15
20
25
30
35
1/[PhSiH3] (M–1)
Figure 9.1 Relationship of PP-b-SiPhH2 number average molecular weights to [PhSiH3] for the [Me2Si(Me4C5)tBuN]TiMeB(C6F5)4 mediated propylene polymerization. Inset: catalytic cycle for this process. [Reproduced with permission from J. Am. Chem. Soc. 1998, 120, 4019. Copyright 1998 Am. Chem. Soc.]
in silane cases. It is known that the borane compounds containing B±H groups usually form a stable dimer (unreactive to ole®ns) in hexane and toluene solvents that are used in metallocene polymerizations. On the other hand, we can use a per¯uoroborate coinitiator instead of MAO or aluminum±alkyl compounds, which are stable with most alkylborane compounds. Therefore, the B±H moiety in dialkylborane can engage a chain transfer reaction by ligand exchange between B±H and M±C (M: transition metal), as illustrated in Eq. (9.3):
B
PE
H H
CZr+(Cp*)2A–
B kp
ktr PE
CZr(Cp*)2
nCH2CH2
(Cp*)2Zr+HA–
B H
A–: [MeB(C6F5)3]–
PE
CB
9:3
160
Functionalization of Polyolefins
Table 9.1 Metallocene-activated ethylene polymerization1 in the presence of 9-BBN as the chain transfer agent Run 1 2 3 4 5 6 7 8 9 10 11 12
Catalyst [Cp2 ZrMe][MeB(C6F5)3] [Cp2 ZrMe][MeB(C6F5)3] [Cp2 ZrMe][MeB(C6F5)3] [Cp2 ZrMe][MeB(C6F5)3] [Cp2 ZrMe][MeB(C6F5)3] [Cp2 ZrMe][MeB(C6F5)3] [Cp2 ZrMe][MeB(C6F5)3] [Cp2 ZrMe][MeB(C6F5)3] [Cp2 ZrMe][B(C6F5)4] [Cp2 ZrMe][B(C6F5)4] [(Ind)2ZrMe][MeB(C6F5)3] [Cp2ZrMe][MeB(C6F5)3]
9-BBN Reaction Yield Catalyst2 Mn3 (mM) time (min) (g) activity (10 3) Mw/Mn 0 3.0 4.5 7.5 7.5 12.0 18.0 23.4 4.5 7.5 7.5 7.5
3 3 3 3 5 3 3 3 3 3 3 3
3.50 2.01 2.05 1.45 2.02 1.20 0.75 0.25 2.00 1.51 1.90 2.50
2333 1333 1366 1033 1333 800 500 167 1333 1000 1267 1667
85.2 76.0 55.5 42.2 45.8 19.4 8.9 3.7 59.4 46.2 43.9 46.9
2.0 2.4 2.9 2.6 2.6 2.7 3.2 4.0 2.6 2.5 2.3 2.1
1
Pressure 1 atm; [ethylene] 0.11 M. Catalyst activity kg of PE/(mol of catalyst atm h). 3 By GPC in 1,2,4-trichlorobenzene vs polystyrene standards. (Reproduced with permission from J. Am. Chem. Soc. 1999, 121, 6763. Copyright 1999 Am. Chem. Soc.) 2
Each polymer chain will have a terminal borane group if the chain transfer reaction is the dominant termination reaction. Table 9.1 summarizes the experimental results of ethylene polymerization in the presence of 9-BBN using two metallocene catalysts, [Cp*2 ZrMe][MeB(C6F5)3] and [Cp2 ZrMe][B(C6F5)4] (Cp* 5-C5H5, 5-Me5C5). To maintain the constant comonomer feed ratio, the reactions were carried out with rapid mixing and a short reaction time. After a 3±5 min reaction time, the polymer solution was quenched with anhydrous/anaerobic MeOH, and the resulting borane-terminated polyethylene (PE-t-B) was washed with anhydrous/anaerobic THF to remove excess 9-BBN and then dried at 50 C in a high vacuum. For analytical studies, a PE-t-B polymer was usually oxidized by NaOH/H2O2 to form a hydroxy-terminated polymer (PE-t-OH). Basically, higher concentrations of a 9-BBN chain transfer agent produced lower molecular weights of the resulting polyethylene. The polymer molecular weight distribution is generally narrow, which is consistent with single site polymerization processes. The catalyst activity was also depressed in the presence of 9-BBN, which may re¯ect the competitive coordination at metallocene active sites between monomer and chain transfer agents. Figure 9.2 shows the plot of polymer molecular weight (Mn) vs the mole ratio of ethylene/9-BBN for the comparative runs 2±8 in Table 9.1. The polymer's molecular weight is almost linearly proportional to the molar ratio of [ethylene]/[9-BBN]. It is clear that the chain transfer reaction with 9BBN (with rate constant ktr) is the dominant termination process, and competes with the propagating reaction (with rate constant kp). The degree of polymerization (Xn)
Synthesis of Polyolefins with a Terminal Functional Group
161
100
Mn × 10–3 (g/mol)
80
60
40
20
0 0
10 20 30 [ethylene] / [9-BBN] (mol/mol)
40
Figure 9.2 Plot of the average molecular weights (Mn) of PE-t-B polymers vs the mole ratio of [ethylene]/ [9-BBN] in the feed (runs 2±8 in Table 9.1). (Redrawn from J. Am. Chem. Soc. 1999, 121, 6763. Copyright 1999 Am. Chem. Soc.)
follows a simple comparative equation Xn kp[ole®n]/ktr[9-BBN] with a chain transfer constant ktr/kp 1/75. The existence of a borane group at the PE chain end is also evidenced by the corresponding hydroxy-terminated PE (PE-t-OH). The 1H NMR spectrum121 of a low-molecular-weight PE-t-OH (Mn 3700 g/mol) shows a major chemical shift at 1.30 ppm corresponding to CH2 in the PE backbone, which is accompanied by several weak peaks at 0.97 ppm (chain end CH3), 1.58 ppm (±CH2CH2±OH), 2.25 ppm (±OH), and 3.62 ppm (±CH2±OH). The peak intensity ratio of OH : CH2± O : CH3 1 : 2 : 3 (2%) indicates the exclusive production of hydroxy-terminated polyethylene. It is interesting to note that there is no detectable vinyl group associated with the conventional chain transfer process (via -H elimination). The same results were also observed in the 13C NMR spectrum with chemical shifts corresponding to ±CH2±OH ( 62.99) and chain end CH3 ( 13.85) groups. These ®ndings strongly indicate the in situ chain transfer to 9-BBN moiety during the catalytic polymerization of ethylene. The existence of a borane group at the PE chain end is further supported by its ability in the chain extension reaction to form diblock copolymers, which will be discussed in Chapter 11. Similar B±H chain transfer reactions can be applied to other metallocenemediated ole®n polymerizations. One example is the synthesis of borane-terminated syndiotactic polystyrene (s-PS-t-B). Table 9.2 shows several [Cp*TiMe2][MeB(C6F5)3] -mediated polymerizations of styrene122 with the presence of 9-BBN.
162
Functionalization of Polyolefins
Table 9.2 [Cp2 TiMe2][MeB(C6F5)3] -activated styrene syndiospecific polymerization in the presence of 9-BBN as a chain transfer agent1 Run 1 2 3 4 5 6 7 8
9-BBN (mmol)
Reaction time (min)
Yield (g)
Catalyst activity2
Syn.3 (%)
Tm ( C)
[]
Mw (g/mol)
25 50 100 150 200 300 400 500
3 3 3 3 3 3 3 3
1.3 1.4 1.3 1.1 0.8 0.5 0.3 0.1
2600 2800 2600 2200 1600 1000 600 200
97.0 96.2 95.8 95.2 93.4 94.7 95.0 93.2
270 271 270 271 270 272 270 270
1.270 0.806 0.580 0.373 0.340 0.188 0.116 0.087
460 000 240 000 150 000 80 000 70 000 30 000 15 000 10 000
1
Styrene 10 ml, catalyst concentration 10 mmol. Catalyst activity kg of s-PS/(mol of Ti h). 3 Syndiotactic index was determined by 13C NMR. Reproduced with permission from Macromolecules 1999 32, 8689. Copyright 1999 Am. Chem. Soc. 2
To minimize mass transfer and maintain the constant 9-BBN/styrene ratio, the reactions were carried out with rapid mixing and a short reaction time. The polymer solution was quenched with anhydrous/anaerobic MeOH, and the resulting borane-terminated s-PS (s-PS-t-B) was washed with anhydrous/anaerobic THF to remove excess 9-BBN. The effects of the chain transfer reaction are clearly revealed by the reduction of the polymer's molecular weight in the presence of 9-BBN. Basically, the higher the concentration of 9-BBN chain transfer agent used, the lower the molecular weight of the resulting polyethylene. The catalyst activity was somewhat depressed if a high concentration of 9-BBN was presented in the system, which may re¯ect the competitive coordination at the titanocene active sites between monomers and chain transfer agents. It is very interesting to note that the syndiotacticity and melting temperatures of all s-PS-t-B polymers are similar (shown in Table 9.2) to those of the s-PS polymer. The 9-BBN chain transfer agent did not interfere with the regioand stereo-selective insertion process. Some of the s-PS-t-B polymer was oxidized by NaOH/H2O2 reagents at 40 C for 6 h to form a hydroxy-terminated polymer (s-PS-t-OH). For the 1H NMR study, the terminal hydroxy group was further reacted with Cl±Si(CH3)3 to form silaneterminated s-PS (s-PS-t-O-Si(CH3)3). Figure 9.3 shows the 1H NMR spectrum of an s-PS-t-O-Si(CH3)3 sample with a molecular weight of about 15 000 g/mol. There are two sets of major peaks at 1.45 and 2.10 ppm corresponding to the CH2 and CH protons in the s-PS backbone, and at 6.82 and 7.21 ppm corresponding to the protons of the phenyl groups that have a syndiotactic arrangement. In addition, two minor peaks at 0.15 and 3.81 ppm correspond to the protons of the ±O±Si(CH3)3 and ±CH()±OSi(CH3)3 groups, respectively.
Synthesis of Polyolefins with a Terminal Functional Group
8
6
4 ppm
2
163
0
Figure 9.3 1H NMR spectrum of s-PS-t-O-Si(CH3)3 (solvent: C2D2Cl4; temp.: 110 C).
The combination of the peak intensity ratio between the silane and phenyl groups and the molecular weight of the polymer indicates that most of the s-PS polymers contain a terminal silane group. The presence of silane terminal group in each s-PS polymer implies an effective in situ chain transfer reaction occurred as well as oxidation and silation reactions, despite heterogeneous reaction conditions. Most of the borane and silane terminal groups are located in the amorphous phases, and are readily accessible for the reaction. It is very important to broaden the scope of this chemistry to the copolymerization reactions. In addition to understanding the generality of B-H chain transfer reaction, the chemistry could provide a very valuable route to the functionalization of new polyole®n copolymers, such as metallocene-LLDPE and ethylene/styrene copolymers, which are becoming important commercial materials in new polyole®n applications. In the ethylene copolymerization, the constrained geometry [C5Me4(SiMe2NtBu)TiMe] catalyst with larger spatial opening at the catalytic site was a preferred choice, providing favorable copolymerization conditions for all ethylene copolymerization with many large size monomers, such as 1-propene, 1-butene, 1-octene, styrene, and p-methylstyrene. In the B-H chain transfer reaction, it is very interesting to examine the effects of the potential acid-base interaction between borane and imido ligand in the [C5Me4(SiMe2NtBu)TiMe] catalyst used to prepare boraneterminated polyole®n copolymers. Table 9.3 summarizes the experimental results of two representative sets of ethylene/1-octene and ethylene/styrene copolymerization reactions by using the 9-BBN chain transfer agent and the [C5Me4(SiMe2NtBu)TiMe][MeB(C6F5)3] complex. In general, both resulting 9-BBN terminated poly(ethylene-co-1-octene)
Table 9.3 Summary of Ethylene/1-Octene and Ethylene/Styrene Copolymerization Reactions by Using [C5Me4(SiMe2NtBu)TiMe][MeB(C6F5)3] Complex in the Presence of 9-BBN Chain Transfer Agent Reaction Conditiona
Run
F-1 F-2 F-3 F-4 F-5 G-1 G-2 G-3 G-4 G-5 a
Catalyst
Time/Temp (min/ C)
Ethylene (atm)
Comonomer/ (M)
9-BBN (mM)
Activity (kg P/molTi atm h)
Comonomerb Incorp (mol%)
Mnc (10 3)
Mw/Mnc
3/50 3/50 3/50 3/50 3/50 5/50 5/50 5/50 5/50 5/50
1 1 1 1 1 1 1 1 1 1
O/0.32 O/0.32 O/0.32 O/0.32 O/0.32 S/0.87 S/0.87 S/0.87 S/0.87 S/0.87
0 30 60 90 120 0 30 60 90 120
263 269 251 206 202 351 351 363 486 525
36.3 35.7 36.8 37.6 39.0 22.6 22.4 19.3 28.5 28.0
44.6 40.8 27.3 22.0 20.2 41.1 38.7 19.6 9.9 7.7
2.8 2.4 2.5 2.7 2.8 2.1 2.4 2.4 2.6 2.6
Catalyst concentration 3 mM; O: 1-octene; S: styrene; [ethylene] 0.11 M; 20ml toluene; 50 C. by 1H-NMR. c by GPC in THF vs polystyrene standard. b
Copolymer Structure
Synthesis of Polyolefins with a Terminal Functional Group
165
(EO-t-9-BBN) and poly(ethylene-co-styrene) (ES-t-9-BBN) copolymers show similar trends ± decreasing the polymer molecular weight with increasing B-H chain transfer agent ± discussed in the homopolymerization reactions. Compared to the control runs (F-1 and G-1) (without 9-BBN chain transfer agent), the copolymerization reaction shows similar comonomer incorporation and catalyst activity with the presence of 9-BBN. There are no detected effects due to acid-base interaction between borane and imido ligand in the [C5Me4(SiMe2NtBu)TiMe][MeB(C6F5)3] complex. Copolymer with very low molecular weight (few thousand) has been obtained, and the molecular weight distribution is generally narrow, which is consistent with single site polymerization processes. Figure 9.4 shows the plot of the molecular weight (Mn) of both EO-t-9-BBN and ES-t-9-BBN copolymers (in Table 9.3) versus 1/[9-BBN]. The linear relationship indicates the B-H chain transfer reaction (ktr) as the dominant termination process. The estimated chain transfer constants (ktr/kp), based on the combined monomer concentration, are 1/29 and 1/24 for EO-t-9-BBN and ES-t-9-BBN, respectively. Both of the chain transfer constants in the copolymerization reactions are signi®cantly higher than that (1/75) of the corresponding ethylene homopolymerization. The comonomers clearly slow down the propagating process, especially in ethylene/ styrene case. The cationic nature of the catalyst site is known to proceed 45
40
35
Mn, (10–3)
(a) 30 (b)
25
20
15
10
5 5
10
15
20
25
30
35
1/[9-BBN], (mol/L)–1
Figure 9.4 The plots of the molecular weight (Mn) of (a) EO-t-9-BBN and (b) ES-t-9-BBN copolymers versus 1/[9-BBN] (runs in Table 9.3).
166
Functionalization of Polyolefins
Figure 9.5 13C (DEPT-135) NMR spectrum of a hydroxy-terminated poly(ethylene-co-styrene), with an inset of chemical shift assignments.
2,1-insertion28 of styrene, which may result in a less active propagating site due to the interaction between the cationic Ti+ center and the adjacent phenyl group at the polymer chain end. The ``slow-down'' propagating chain end (after styrene insertion) increases the vulnerability to chain transfer reaction with 9-BBN. It is logical to predict the majority of the chain end structures will have a styrene unit right before the incorporation of borane terminal group. To increase the NMR sensitivity for determining the end group, the terminal borane group was converted to a hydroxy group. Figure 9.5 shows the 13C (DEPT-135) NMR spectra of a hydroxy-terminated poly(ethylene-co-styrene) (ES-t-OH) that was converted from an ES-t-9-BBN polymer (sample G-4 in Table 9.3; styrene content 28.5 mole%; Mn 9,900 g/mole; Mw/Mn 2.6). The inset shows the chemical shift assignments. Along with all the expected chemical shifts, corresponding to the aliphatic and aromatic carbons in the polymer chain, there is only one chemical shift (at 51.36 ppm) corresponding to the end group ±C()H±OH. The results clearly explain the chain transfer mechanism in the copolymerization reactions. 3.4
Styrene-derivative-terminated Polyole®ns350
Recently, we have also discovered a new polymerization process to prepare styrene derivative terminated polyole®ns. The process involves contacting -ole®n (> C3)
Synthesis of Polyolefins with a Terminal Functional Group
167
with styrene derivative and hydrogen simultaneously in the presence of some metallocene catalysts containing a speci®c bridged Cp* ligand. Ironically, the desirable metallocene catalysts usually show very poor styrene incorporation in the copolymerization reaction between propylene and styrene. The reaction mechanism involved in forming the styrene-derivativeterminated polyole®n may be exempli®ed by the polymerization of propylene in the presence of p-methylstyrene (p-MS) and hydrogen chain transfer agents using a rac-Me2Si[2-Me-4-Ph(Ind)]2ZrCl2/MAO catalyst, as illustrated in Eq. (9.4): CH3 (III)
CH3 ⱍ CHCH2CH2CHZr+A–
PP
CH2CH
ktr1
H2 2,1-insertion
ktr2
CH3 CH3 ⱍ
CH3 ⱍ CHCH2Zr+A– (II)
PP PP (IV)
CH3 ⱍ ⱍ CHCH2CH2CHZr+A– H
kp
H 1,2-insertion
Zr+HA– (I) (CH3)
PP
9:4
CH2CH2
CH3
CH3 ⱍ nCH2CH
(V)
During the polymerization of propylene (with 1,2-insertion manner) the propagation Zr±C site (II) can also react with p-methylstyrene (with 2,1-insertion manner) to form p-methylstyrene-terminated polypropylene (III). The catalytic Zr±C site in compound (III) becomes inactive to both propylene and p-methylstyrene302 due to the combination of steric hindrance between the active site (Zr±C) and the incoming monomer (propylene with 1,2-insertion), and the formation of a complex between the adjacent phenyl group and the Zr ion. On the other hand, with the presence of hydrogen the dormant Zr±C site (III) can react with hydrogen to form PP-t-p-MS (V) and regenerate a Zr±H species (I) that is capable of reinitiating the polymerization of propylene and continuing the polymerization cycle. Overall, the polymerization process resembles a sequential chain transfer reaction ± ®rst to styrene (or styrene derivative) and then hydrogen ± during the metallocenemediated propylene polymerization. This process not only produces PP with a
168
Functionalization of Polyolefins
terminal p-MS unit, but also maintains high catalyst activity. The molecular weight of the resulting polymer is proportional to the [propylene]/[p-MS] ratio. Table 9.4 summarizes the results of p-methylstyrene-terminated polypropylene (PP-t-p-MS) using a rac-Me2Si[2-Me-4-Ph(Ind)]2ZrCl2/MAO catalyst in the presence of p-MS and hydrogen chain transfer agents. The systematic study was conducted to evaluate the effect of hydrogen and p-MS concentrations on the catalyst activity and polymer molecular weight. All comparative reaction sets show that hydrogen is necessary to complete the chain transfer reaction to p-methylstyrene during the rac-Me2Si[2-Me-4-Ph(Ind)]2ZrCl2/MAO-mediated polymerization of propylene. In general, the change of hydrogen concentration does not affect the molecular weight and molecular weight distribution of the resulting p-MS-terminated polypropylene polymers. However, a suf®cient quantity of hydrogen, which increases with the increase of [p-MS], is needed to maintain high catalyst activity and p-MS conversion. Figure 9.6 compares the GPC curves of PP-t-p-MS polymers with a PP homopolymer. All polymers were prepared under the same reaction conditions, except for varying the quantity of the p-MS chain transfer agent. An appropriate concentration of hydrogen was also introduced in each chain transfer reaction to assure the completion of the polymerization cycles. The GPC curves show the systematic reduction of the polymer molecular weight, along with the increase of the p-MS concentration. The low-molecular-weight Table 9.4 Summary of PP-t-p-MS polymers prepared1 by the combination of a rac-Me2Si[2-Me-4-Ph(Ind)]2ZrCl2/MAO catalyst and p-MS/hydrogen chain transfer agents
Run Control Control 1 2 3 4 Control 5 6 7 8 Control 9 10 11 1
p-MS Mn p-MS H2 Yield Catalyst p-MS in PP conversion PDI (%) ( 10 3) (Mw/Mn) (mol%) (M) (psi) (g) activity2 1 0 2 0.0305 0.0305 0.0305 0.0305 0.0305 3 0.076 0.076 0.076 0.076 0.076 4 0.153 0.153 0.153 0.153
0 0 2 6 12 35 0 6 12 20 35 0 12 20 35
26.94 0.051 3.80 8.04 12.04 24.67 0 0.91 1.69 8.81 10.52 0 0.35 3.81 4.41
86 208 163 12 160 25 728 38 528 78 944 Ð 2 912 5 408 28 192 33 664 Ð 1 120 12 192 14 112
0 0.16 0.14 0.15 0.15 0.13 Ð 0.40 0.41 0.43 0.41 Ð 0.66 0.61 0.63
Ð 0.05 8.30 18.83 28.19 50.05 Ð 2.33 4.33 23.65 26.86 Ð 0.72 7.26 8.67
77 600 59 700 55 500 54 800 55 400 34 600 Ð 27 600 25 900 20 500 25 800 Ð 10 000 11 700 9 700
Reaction conditions: 50 ml toluene, propylene (100 psi), [Zr] 1.25 10 [Zr] 3000, temperature 30 C, time 15 min. 2 Catalyst activity kg PP/(mol catalyst h).
6
2.9 3.4 2.4 2.5 2.3 2.8 Ð 2.1 2.3 2.3 2.3 Ð 2.0 2.0 1.9
mol/l, [MAO]/
Synthesis of Polyolefins with a Terminal Functional Group
169
200
180
160 (a) 140
(b) (c) (d)
(e)
120
100
20
22
24
26 Minutes
28
30
32
Figure 9.6 GPC curves of (a) PP with molecular weight (Mn) of 77.6 103 and several PP-t-p-MS polymers with Mn of (b) 54.8 103, (c) 25.9 103, (d) 10 103, and (e) 4.6 103 g/mol. (Redrawn from J. Am. Chem. Soc. 2001, 123, 4871. Copyright 2001 Am. Chem. Soc.)
PP-t-p-MS (Mn as low as a few thousandths) that has been prepared is a wellde®ned polymer with narrow molecular weight distribution (Mw/Mn 1.7). In fact, the molecular weight distribution of PP-t-p-MS gradually reduced along with the reduction in the polymer molecular weight. All experimental results clearly indicate effective chain transfer reactions. The kinetic constants during the polymerization can be obtained from Fig. 9.7, which shows the plot of the PP-t-p-MS molecular weight to the mole ratio of [propylene]/[p-MS]. The linear relationship between Mn and [propylene]/ [p-MS] clearly indicates that the chain transfer reaction to p-MS (with rate constant ktr) is the dominant termination process, and competes with the propagating reaction (with rate constant kp). The degree of polymerization (Xn) follows the equation Xn kp[propylene]/ktr[styrene] with a chain transfer constant ktr/kp 1/6.36. Figure 9.8 shows the 1H NMR spectra of PP-t-p-MS (Mn 4600). In addition to the peaks between 0.9 and 1.7 ppm corresponding to protons in the PP chain, there are three peaks at 2.7, 7.1, and 2.35 ppm corresponding to (±CH2±C6H5±CH3), respectively, located at the polymer chain end. The polymer molecular weight calculated from the chain end group is quite consistent with that of the GPC result. The terminal p-methyl group in PP-t-p-MS is very reactive in many chemical reactions, as discussed in Chapter 7. In addition, the benzylic proton can be easily metallated with butyl lithium to form a benzylic anion at the chain end, which can
170
Functionalization of Polyolefins 60 000 50 000
Mn (g/mol)
40 000 30 000 20 000 10 000 0 0
20
40 60 80 100 [propylene]/[p-methylstyrene] (mol/mol)
120
Figure 9.7 Plot of PP-t-p-MS molecular weight to mole ratio of [propylene]/[p-MS].
ppm
9
8
Figure 9.8
7
6
5
4
3
2
1
1
H NMR spectra of PP-t-p-MS (Mn 4600).
then carry out living anionic polymerization of styrene and methyl methacrylate to produce polyole®n diblock copolymers. The experimental results will be discussed in the next chapter. The same chain transfer reaction scheme can be applied to many styrene derivatives, including the ones containing masked functional groups. One styrene derivative that does not require a masking group is p-chlorostyrene (p-Cl-St), which can be directly incorporated to the polyole®n chain end in the presence of hydrogen
Synthesis of Polyolefins with a Terminal Functional Group
171
Table 9.5 Summary of PP-t-p-Cl-St polymers prepared1 by the combination of a rac-Me2Si[2-Me-4-Ph(Ind)]2ZrCl2/MAO catalyst and p-Cl-St/hydrogen chain transfer agents Example
p-Cl-St (M)
H2 (psi)
Yield (g)
Catalyst activity2
p-Cl-St in PP (mol%)
Mn (10 3)
PDI (Mw/Mn)
Control 1 1 2 Control 2 3 Control 3 4
0.144 0.144 0.144 0.289 0.289 0.433 0.433
0 6 20 0 20 0 27
0.11 1.36 7.56 0 4.44 0 8.48
352 4 402 24 192 0 15 712 0 27 200
0.12 0.13 0.12 Ð 0.24 Ð 0.39
54 600 45 300 46 100 Ð 18 700 Ð 8 200
1.9 1.9 2.1 Ð 2.1 Ð 1.9
1 Reaction conditions: 50 ml toluene, propylene (100 psi), [Zr] 1.25 10 [Zr] 3000, temperature 30 C, time 15 min. 2 Catalyst activity kg PP/(mol catalyst h).
6
mol/l, [MAO]/
gas. Table 9.5 summarizes the experimental results of p-chlorostyrene-terminated PP (PP-t-p-Cl-St) polymers that were prepared by using a rac-Me2Si[2-Me-4Ph(Ind)]2ZrCl2/MAO catalyst in the presence of p-Cl-St and hydrogen chain transfer agents. The reactions were also compared with the corresponding control runs to understand the effects of hydrogen. All three comparative reaction sets show the necessity of hydrogen to complete the chain transfer reaction to p-chlorostyrene during the rac-Me2Si[2-Me-4Ph(Ind)]2ZrCl2/MAO-mediated polymerization of propylene. In general, the change of hydrogen concentration does not affect the molecular weight and molecular weight distribution of the resulting polymer. A suf®cient quantity of hydrogen, increasing along with the [p-Cl-St], is needed to achieve high catalyst activity. Overall, the chain transfer reaction to p-Cl-St (with rate constant ktr) competes with the propagating reaction of propylene (with rate constant kp). The degree of polymerization (Xn) follows a simple comparative equation Xn kp[propylene]/ktr [p-Cl-St] with a chain transfer constant ktr/kp 1/21.2. The polyole®n containing a terminal benzylic halide group is a very interesting material. The reactive benzylic halide end group can easily convert to many other functional groups, or serve as a coupling site for graft-onto reactions. It can also be used as a free radical imitator, especially with the combination of CuBr, for atom transfer radical polymerization. 4 CHEMICAL MODIFICATION OF CHAIN-END UNSATURATED POLYMER This reaction scheme involves two steps, including (i) the preparation of the chainend unsaturated polymer, and (ii) functionalization of the chain-end unsaturated group. The method is very much limited to polypropylene, due to the availability of the chain-end unsaturated polymer.
172 4.1
Functionalization of Polyolefins
Synthesis of Chain-End Unsaturated Polypropylene (u-PP)
Two methods were reported in the preparation of u-PP polymers with good control of chain-end unsaturation and polymer molecular weight. The ®rst one is a thermal degradation process that was carried out by mixing a molten isotactic PP polymer (with high molecular weight and Tm 164 C) with a high-temperature-decomposed peroxide reagent (proton-extraction agent) in a Brabender mixer. The thermal degradation reaction is very effective, even with relatively low concentrations of peroxide and a low reaction temperature. Figure 9.9 compares the GPC and IR curves of the initial PP and the resulting u-PP polymers.120,351 With increasing chain degradation, the molecular weight distribution of the u-PP is gradually reduced ± an indication of statistic degradation process. The polymer chain was degraded 10 times on average within 8 min. The molecular weight of the resulting polymer is inversely proportional to the peroxide concentration and reaction time. The chain-end vinylidene group can be detected by IR spectrum with a weak absorption peak at 888 cm 1. It is very dif®cult to quantify the ole®n concentration based on the peak intensity alone; however, it was reported that each thermal degradation generates a vinylidene unit at one broken chain end. Statistically, the product is a mixture352 containing predominantly PP chains with one vinylidene chain-end unit, some PP chains having both vinylidene chain-end units, and only a few PP chains without a vinylidene group, especially after an extended degradation reaction. The other method of producing u-PP is the metallocene polymerization of propylene with some metallocene catalysts, which operate the termination reaction by -hydrogen elimination.353 Two of the most common isospeci®c metallocene catalysts, {SiMe2[2-Me-5-Ph(Ind)]2}ZrCl2 and Et(Ind)2ZrCl2, were studied under various reaction conditions. The high isospeci®c {SiMe2[2-Me-5-Ph(Ind)]2}ZrCl2 catalyst, suppressing -hydrogen elimination, produces high-molecular-weight PP with concentrations of ole®nic units below the detectable level in all reaction conditions. On the other hand, Et(Ind)2ZrCl2 clearly shows the termination proceeding with -hydrogen elimination. The higher the polymerization temperature, the lower the molecular weight of the polymer and the higher the chain-end unsaturation concentration. The resulting u-PP polymers have narrow molecular distributions and relatively low melting temperatures (Tm < 140 C). 4.2
Hydroboration and Oxidation
The functionalization of the vinylidene end group of u-PP has been studied by using various ole®nic chemistry, including hydroboration, hydroalumination, and hydrosilation. Due to the extremely low concentration of double bond units and the insolubility of polymer in many organic solvents that dissolve the reactive reagents, many chemical modi®cations have only limited success. By far, the most versatile method is hydroboration reaction by boron hydride reagents, such as
Synthesis of Polyolefins with a Terminal Functional Group
173
(a) (b) (c) (d) (e) 112
mV
110
108
106
104
102 20
25
30
35 40 Minutes
45
50
55
0.28
Absorbance
0.26 0.24
(e)
0.22
(d)
0.20
(c) (b) (a)
0.18 920
915
910
905
900 895 890 Wavenumber
885
880
875
870
Figure 9.9 GPC (top) and IR (bottom) spectra of (a) starting PP (Mn 100 000 g/mol, Mw/Mn 6 and Tm 164 C) and the corresponding thermally degraded PP with molecular weight (Mn) of (b) 48 103, (c) 37 103, (d) 31 103, and (e) 23 103 g/mol. (Redrawn from Macromolecules 1999, 32, 2525. Copyright 1999 Am. Chem. Soc.)
9-borabicyclononane (9-BBN) and BH3/THF. Both the reactivity and solubility of dialkylborane in hydrocarbon media (which swells the u-PP polymer) are advantages for this reaction. By stirring u-PP powder with a slight excess of 9-BBN in THF at 55 C for 5 h, the hydroboration reaction is almost quantitative in suspension condition. Due to the mobility, the polymer chain end (with vinylidene group)
174
Functionalization of Polyolefins
prefers to position itself on the surface or in the amorphous phase, where it is swelled by borane/THF solution. The resulting borane-terminated PP (PP-b-B) is very useful intermediate for interconversion to various polar-group-terminated PPs, as illustrated in Eq. (9.5): PP
CH3 ⱍ CCH2 NaOH/H2O2
PP
CH3 ⱍ CHCH2OH (II)
CH3 ⱍ H2NOSO3H CHCH2BR2
PP
CH3 ⱍ CHCH2NH2
PP
CH3 ⱍ CHCH2BCl2
HBR2
PP (I)
O2
PP
BCl3
CH3 ⱍ CHCH2O* *OBR2
(VI)
9:5 (IV)
(i) 1-butylazide (ii) KOH
MA
CH3 O ⱍ CHCH2O
O
PP (VII)
(III)
PP
CH3 ⱍ CHCH2NH(C4H9) (V)
O
As shown in the borane chemistry, the ef®ciency of the interconversion of borane to other functional groups is very dependent on the borane moieties. In trialkylborane compounds, all three B-alkyl bonds engage the oxidation reaction by NaOH/ H2O2 to obtain hydroxy-terminated polymer (II). Most borane moieties are suitable for this reaction. However, only one or two B-alkyl bonds engage in amination reaction.123 An asymmetric borane moiety, such as H±B(CH3)2 with two B±CH3 blocking groups, has to be applied to achieve NH2-terminated PP (III) with high yield. The reaction usually takes place by hydroborating u-PP with dimethylborane that was prepared in situ from lithium dimethylboronhydride. The dimethylboraneterminated PP was then quantitatively aminated using hydroxyamine-o-sulfonic acid. The other reported amination route involves the disproportional reaction of the terminal trialkylborane group with boron trichloride to obtain a dichloroboraneterminated PP (IV), which was then brought into contact with 1-butylazide to obtain the PP (V) having a 1-butylamino group at the chain end. The most facile functionalization chemistry may be the oxidation reaction of PP-t-B with oxygen.251 Upon the controlled oxidation reaction, the borane end group is converted to a ``stable'' polymeric alkoxyl radical (associated with the boroxyl radical) (VI), that initiates a free radical reaction with most free-radical polymerizable monomers. The functional group and its concentration incorporated
Synthesis of Polyolefins with a Terminal Functional Group
175
to PP are basically governed by the monomer feed. In other words, this free radical graft-from reaction (via borane end group) is capable of preparing a broad range of functional-group-terminated PP polymers with a controlled functional group concentration. One example is maleic-anhydride-terminated PP (VII). The polymeric radical at the PP chain end was contacted with maleic anhydride both with and without the presence of styrene.120,351 Due to MA's low tendency for homopolymerization,354 a very low concentration of MA (possibly only a single MA unit) was incorporated into the PP structure to form MA-terminated PP (PP-t-MA). On the other hand, with the presence of styrene in the PP-B/MA mixture, the free radical graft-from polymerization takes place to extend the PP chain end with an alternating styrene and MA (SMA) copolymer.355 In other words, a diblock copolymer PP-b-SMA is obtained, containing both PP and SMA segments. The MA concentration in the copolymer is governed by the molecular weight of the SMA segment. The detailed experimental results will be discussed in Chapter 12.
5
SUMMARY
Despite the experimental dif®culties, signi®cant progress has been made in the preparation of polyole®ns having a terminal functional group. The combination of metallocene catalysis and in situ chain transfer reaction is by far the most effective and convenient route. The chemistry can be easily extended to many polyole®n structures, including co- and terpolymers. This drop-in technology needs only very minor modi®cations of current commercial polymerization processes. In fact, it would be very convenient and bene®cial to replace the current hydrogen chain transfer agent with borane or p-MS/H2 chain transfer agents. Such a process not only allows the control of polymer molecular weight and chain-end saturation, it also provides a reactive functional group at the polymer chain end for further reactions. Both borane and p-MS-terminated polyole®ns open the door to preparing diblock copolymers containing a polyole®n and a free-radical or anionic prepared functional (polar) polymer, which will be discussed in Chapter 10. The whole polymerization process resembles a continuous reaction with a middle transformation of metallocene to living free radical or anionic polymerization to prepare two completely distinctive polymer blocks. From the commercial viewpoint, living polymerization is a very expensive process. One catalyst site can only produce one polymer chain. On the other hand, the chain transfer reaction (resembling the commercial process of chain transfer reaction to H2) maintains extremely high catalyst productivity.
This Page Intentionally Left Blank
Polyole®n Block and Graft Copolymers
This Page Intentionally Left Blank
10 Synthesis of Functional Polyole®n Diblock Copolymers
1
INTRODUCTION
Diblock copolymer is the most effective interfacial agent in polymer blends and composites.356,357 Usually, only a small quantity (as low as 1%) of a suitable diblock copolymer is needed to change the incompatible binary blends to a uniform microphase morphology with strong interfacial adhesion. The issue is particularly important in polyole®n due to the lack of interaction with other materials ± including polymers, pigments, ®llers, glass ®ber, and metals. Obviously, it is very desirable to develop a facile chemistry to prepare polyole®n diblock copolymers containing functional polymers. In polymer chemistry, most diblock copolymers have been prepared by living polymerization processes, namely anionic,358,359 cationic,360,361 and recently metathesis362,363 with sequential monomer additions. As discussed in Chapter 9, the living coordination polymerization of -ole®n by transition metals is very limited. In addition, the catalyst's sensitivity to functional groups further constrains the chemistry's ability to produce a functional polyole®n diblock copolymer. In this chapter, we will discuss the ®ve most studied methods of preparing polyole®n diblock copolymers containing a polyole®n and functional polymer segments. They include (i) living Ziegler±Natta polymerization, (ii) transformation reaction from Ziegler±Natta to living free radical polymerization, (iii) transformation from Ziegler±Natta to living anionic polymerization, (iv) transformation reaction from anionic to Ziegler±Natta polymerization, and (v) coupling reaction between two polymers. Most methods result in products with an intimate mixture of homopolymers and perhaps some diblock copolymer. However, some methods (especially the second and third ones) offer simple reaction processes and produce polyole®n diblock copolymers with a broad range of copolymer composition and well-designed molecular structure.
180 2
Functionalization of Polyolefins
LIVING ZIEGLER±NATTA (METALLOCENE) POLYMERIZATION
The application of living transition metal coordination polymerization with sequential monomer addition in the preparation of functional polyole®n diblock copolymers is very rare. Simultaneously ful®lling the requirements of living coordination polymerization and of catalyst compatibility with functional groups is very dif®cult. Only a few experimental results have indicated catalyst systems capable of carrying out living coordination polymerization, as well as other polymerization mechanisms upon changing the reaction condition. Doi et al.331,332 ®rst reported the synthesis of s-PP-bPMMA diblock copolymer by a soluble V(acac)3/AlEt2Cl catalyst. As illustrated in Eq. (10.1), after syndiospeci®c coordination polymerization of propylene at 78 C, the living polypropylene chain end (C±V) was transformed to a free radical (C*) by raising the reaction temperature to 25 C for the polymerization of MMA. CH3 ⱍ CH2CH
CH3 ⱍ s-PPCH2CHV+3
V(acac)3/Al(C2H5)2Cl –78°C MMA –78°C
25°C
10:1 CH3 CH3 ⱍ ⱍ s-PPCH2CHCH2C . V+2 ⱍ CO ⱍ OCH3
MMA 25°C
s-PP
PMMA
In the early 1990s, Yasuda et al.333 reported a new organolanthanide complex, LnRCp2 (Ln Sm, Yb, Lu; R H, CH3), that is capable of serving the dual functions of polymerizing both ethylene and polar monomers (MMA, CL, etc.) with ``living'' manners. The living ole®n coordination polymerization takes place at ambient temperature. Upon addition of polar monomers, the polymerization process changes to other mechanisms, such as ring-opening or group-transfer mechanisms, as illustrated in Eq. (10.2): OCH3 CH3 ⱍ ⱍ C O C SmCp*2 CH2 ⱍ O ⱍ PMMACH2CC ⱍ ⱍ CH3 OCH3
Cp*2SmH
20°C MMA
PECH2CH2SmCp*2 20°C
PE
10:2
O
CL
PE
ⱍ
CH2CH2
PCLC(CH2)5OSmCp*2
The living polyethylene chain end (C±Sm) engages the group-transfer polymerization of MMA or the ring-opening polymerization of lactones. Several polyethylene diblock copolymers were prepared with good control of copolymer
Functional Polyolefin Diblock Copolymers
181
structure and composition. However, the experimental results are very much limited to polyethylene copolymers. Overall, the combination of living transition metal coordination polymerization and transformation reaction to other mechanisms is a very special chemistry, but very limited in scope. Only a few living propagating (C±M) chain ends are capable of changing to other polymerization mechanisms without deactivation by polar functional groups. So far, the results have not covered some of the most desirable functional polyole®n diblock copolymers containing an isotactic polypropylene or syndiotactic polystyrene block and a functional polymer block. 3 TRANSFORMATION FROM METALLOCENE TO LIVING FREE RADICAL POLYMERIZATION It is very clear that the transformation from the coordination ole®n polymerization mechanism to the other mechanism of polymerizing functional monomers is the best route to prepare a functional polyole®n diblock copolymer having two very distinctive (polar and unpolar) polymer blocks. Each polymer block is prepared by the most suitable polymerization mechanism, and the combined capability provides good control of the copolymer's structure and composition. The major hurdle is certainly centered at the transformation reaction ± how to design a reaction scheme that can effectively transform the active site from transition metal ole®n coordination polymerization to the functional monomer polymerization. When considering the control of copolymer structure and composition, the transformation from metallocene to living free radical polymerization is the logical scheme. In our laboratory, we have discovered a very effective and convenient method of transforming metallocene to living free radical polymerization mechanism via a borane-terminated polyole®n.118±122 As illustrated in Eq. (10.3), the borane-terminated polyole®n can be effectively prepared by chain transfer reaction during the metallocene polymerization or hydroboration of a chain-end unsaturation: CH3 ⱍ CH2CH + HB
CH3 ⱍ CCH2
PP
Metallocene
HB
catalyst
PP
CH3 ⱍ CHCH2B
(I)
10:3
O2
(II) PP
CH3 ⱍ CHCH2OOB
(III) PP
CH3 ⱍ CHCH2O* *OB MMA
PP
CH3 ⱍ CHCH2O
PMMA
(IV)
182
Functionalization of Polyolefins
The borane terminal group can then be quantitatively transformed to a living free radical initiator for the living polymerization of functional (polar) monomers. This chemistry is applicable to most metallocene catalysts and all ole®ns (ethylene, propylene, 1-hexene, 1-octene, styrene, cyclic ole®ns, and mixtures) and free radical polymerizable polar monomers, such as methacylates and acrylates. Two routes, in situ chain transfer reaction to B±H group during metallocene polymerization and hydroboration of chain-end unsaturated polymer, to prepare borane terminated PE, i-PP, and s-PS have been discussed in Chapter 9, Sections 3.3 and 4.2, respectively. The terminal borane group in polymer (I) can be spontaneously oxidized to a peroxide (B±O±O±C) moiety (II) even at very low temperature ( 65 C). Due to the unfavorable ring strain increase by inserting oxygen into the C±B bonds in the bicyclic ring of 9-BBN, which destroys the stable double chair-form structure, the oxidation reaction selectively takes place at the C±B bond363,364 in the linear alkyl group to produce peroxyborane (C±O±O±B). The peroxyborane (II) behaves very differently from regular benzoyl peroxides, and consequently decomposes by itself even at ambient temperature. The decomposition reaction follows the homolytical cleavage of peroxide to generate an alkoxy radical (C±O*) and a borinate radical (B±O*). The alkoxyl radical (C±O*), located at the end of polyole®n chain, is very reactive and can then be used for the initiation of radical polymerization with the presence of free radical polymerizable monomers. On the other hand, the borinate radical (B±O*), stabilized by the empty p-orbital of boron through back-donating electron density, is too stable to initiate polymerization. However, the borinate radical may form a weak and reversible bond with the growing chain end during the polymerization reaction. Upon the dissociation of the electron pairs in the resting state, the growing chain end can then react with monomers to extend the polymer chain to form diblock copolymer (IV). Overall, the reaction process resembles a transformation reaction from metallocene coordination polymerization to living free radical polymerization via a borane group at the polymer chain end. It is interesting to note that the reaction involves only one borane group per polymer chain. The whole reaction process provides an ultimate test for examining the ef®ciency of the borane reagent in the chain extension process. 3.1
Polyethylene Diblock Copolymers121
Borane-terminated polyethylene (PE-t-B) polymers with narrow molecular weight distribution, were prepared by metallocene catalysts, [Cp2 ZrMe] [MeB(C6F5)3] and [Cp2 ZrMe] [B(C6F5)4] (Cp* 5-C5H5, 5-Me5C5), in the presence of 9-BBN. The PE-t-B polymer was then subjected to oxidation reaction by oxygen in the presence of free radical polymerizable MMA monomers. The resulting reaction mixture was carefully fractionated by Soxhlet extraction using boiling THF to remove any PMMA homopolymer. In most cases, only a very small amount (< 10%) of PMMA homopolymer may be initiated by the radical in a bicyclic ring, instead of a polymeric radical, due to the small non-selective oxidation reaction of alkyl-9-BBN. The
Functional Polyolefin Diblock Copolymers
183
generally insoluble fraction (but soluble in 1,1,2,2-tetrachloroethane, 1,2,4-trichlorobenzene at elevated temperatures) is PE-b-PMMA diblock copolymer. Table 10.1 summarizes the experimental results of two sets of chain extension reactions using two starting PE-t-B polymers with Mn 19.4 103 g/mol and 42.7 103 g/mol, respectively. The experimental result in both comparative sets clearly shows the increase of MMA conversion and the content of PMMA in the diblock copolymer with the increase of the reaction time. The extent of chain extension reaction is basically proportional to the reaction time, which clearly indicates the living free radical polymerization of MMA. High concentration of PMMA in the diblock copolymer can be achieved with narrow molecular weight distribution. It is very interesting to note the remarkable ef®ciency of the borane terminal group, with only one unit per polymer chain. Figure 10.1 compares the GPC curves of the PE-b-PMMA diblock copolymer (sample in row 3 of Table 10.1) and the starting PE-t-B polymer (I). The polymer Table 10.1
Summary of PE-b-PMMA diblock copolymers
Reaction conditions PE-t-B (g)
1
(I)/3 (I)/3 (I)/3 (II)/5 (II)/5 (II)/5 (II)/5
Temp/time O2/MMA ml/mol ( C)/(h) 25/6 25/12 25/24 25/2 25/6 25/12 25/24
1.9/1.87 1.9/1.87 1.9/1.87 1.4/1.87 1.4/1.87 1.4/1.87 1.4/1.87
THF extraction Yield (g)
Insol. (%)
Sol. (%)
3.98 6.65 8.05 6.12 6.78 8.74 10.24
96 93 90 97 95 93 89
4 7 10 3 5 7 11
Diblock copolymer Mn PE/PMMA (10 3) Mw/Mn mole ratio 49.2 62.5 90.3 47.5 61.8 76.3 97.6
2.1 2.4 2.0 2.3 2.4 1.9 2.9
1
100 : 43 100 : 62 100 : 102 100 : 3 100 : 13 100 : 22 100 : 36
PE-t-B (I): Mn 19.4 103 g/mol, Mw/Mn 2.7; PE-t-B (II): Mn 42.7 103 g/mol; Mw/ Mn 2.2.
dw/d(log Mw)
1.2
(a)
(b)
0.8
0.4
0.0 4.00
5.00 log Mw
6.00
Figure 10.1 GPC curve comparison between (a) PE-t-B polymer (Mn 19 400 and Mw 38 800 g/mol) and (b) the corresponding PE-b-PMMA (Mn 90 300 and Mw 243 800 g/mol). (Solvent: trichlorobenzene; temp.: 135 C.) (Redrawn from J. Am. Chem. Soc. 1999, 121, 6763. Copyright 1999 Am. Chem. Soc.)
184
Functionalization of Polyolefins
molecular weight increases several times from Mn 19 400 to 90 300 g/mol, which is consistent with the 1H NMR results of a [PE]/[PMMA] mole ratio of nearly 1/1 in the PE-b-PMMA copolymer. The monochromatic increase of the copolymer molecular weight, with only a slightly broadening molecular weight distribution, clearly point to the existence of a borane group at each PE chain end and a living radical polymerization of MMA in the chain extension process. Figure 10.2 compares the 1H NMR spectra of a starting PE-t-B polymer (the borane group was converted to hydroxy group) and two PE-b-PMMA diblock copolymers. The new peak at 3.58 ppm, corresponding to methoxyl groups (CH3O) in PMMA, increased its intensity with the reaction time.
(c)
(b)
(a) 10
9
8
7
6
5
4
3
2
1 ppm
Figure 10.2 1H NMR spectrum of (a) PE-t-OH (Mn 3700) containing a terminal hydroxy group and two PE-b-PMMA diblock copolymers containing (b) 13 mol% and (c) 50 mol% PMMA. (Solvent: C2D2Cl4; temp.: 110 C.)
Functional Polyolefin Diblock Copolymers
185
Polypropylene Diblock Copolymers118,119
3.2
Borane-terminated PP (PP-t-B) was prepared by both chain transfer and hydroboration routes. PP-t-B polymer was then subjected to an oxidation reaction by oxygen in the presence of free radical polymerizable monomers, such as methyl methacrylate (MMA), ethyl methacrylate (EMA), vinyl acrylate (VA), butyl acrylate (BA), maleic anhydride (MA), and styrene. Many PP diblock copolymers have been prepared that are otherwise very dif®cult to obtain by other existing methods. Table 10.2 summarizes the reaction conditions and experimental results of several PP-b-PMMA copolymers. All reactions were started from the same PP-t-B sample (with Mn 13 000 and Mw/Mn 1.48), which was prepared by hydroboration reaction of a chain-end unsaturated PP. After chain extension, each reaction product was extracted for 24 h with re¯uxing acetone and heptane using a Soxhlet apparatus to give three fractions. The overall conversion of diblock copolymer from PP-t-B is usually between 50 and 80%, which is very encouraging considering there is only one borane initiator in each polymer chain and some ( > 20%) of the polymer chains have no borane group. Slightly better results, with fewer homopolymers, were achieved using benzene solvent and may be due to the slow diffusion of oxygen in the reaction media, which offers better selectivity in the oxidation of PP-9-BBN. Figure 10.3 compares the GPC curves of a PP-b-PMMA diblock copolymer (sample in row 3 of Table 10.2) and the starting PP homopolymer. The molecular weight more than doubles from Mn 13 000 to Mn 29 000 g/mol and the molecular weight distribution (MWD) slightly increases from 1.48 to 1.69. The GPC results are consistent with the 1H NMR measurements that show about a 1/1 mole ratio between PP and PMMA in the copolymer. The same radical chain extension was also applied to other free radical polymerizable monomers, including ethyl methacrylate (EMA), vinyl acrylate (VAc), butyl acrylate (BA), and styrene. The products were subjected to the same vigorous fractionalization processes. Figure 10.4 shows the 1H NMR spectra of PP-b-PEMA, PP-b-PBA, and PP-b-PS copolymers, containing 27, 34, and 11 mol% of radical polymerized monomers, respectively. All reactions were run under similar conditions, ambient temperature, and THF solvent. It is very interesting to note that the borane-mediated radical polymerization is particularly effective with the polar monomers. The rate of propagation may be enhanced due to the interaction between Table 10.2
Summary of PP-b-PMMA diblock copolymers Reaction conditions
1
Fractionalization results
PP (g)
MMA (g)
Solvent (ml)
Oxygen (ml)
Acetone (g)
Heptane (g)
Insoluble (g)
MMA in copolymer (mol%)
0.5 0.5 13.3
5 5 80
THF/5 Benzene/5 THF/80
0.44 0.44 10.8
0.25 0.16 Ð
0.18 0.08 Ð
0.72 0.90 Ð
33 28 45
1
PP-t-B sample (Mn 13 000, Mw/Mn 1.48).
186
Functionalization of Polyolefins 0.642 0.640 0.638
Volts
0.636
(b)
(a)
0.634 0.632 0.630 0.628 0.625 0.620 36
Figure 10.3 copolymer.
38
40
42
44
46
48 50 Minutes
52
54
56
58
60
GPC curves of (a) PP and (b) the corresponding PP-b-PMMA
the boron moiety in the propagating end group and the heteroatom (O, N, Cl) in the polar monomer, a ``Coordination'' radical addition process, as illustrated below:
R⬘
R⬘
C* *OB
3.3
C
OR
H
CC
H
O H
s-PS Diblock Copolymers122
The new route for synthesizing polyole®n diblock copolymers was also extended to syndiotactic polystyrene (s-PS). In a typical example, an s-PS-t-B polymer prepared by the chain transfer reaction shown in Chapter 9 was suspended in benzene and subjected to an oxidation reaction by oxygen in the presence of free radical polymerizable monomers, i.e. methyl methacrylate (MMA) and n-butyl methacrylate (BMA). Figure 10.5 shows the 1H NMR spectra of several corresponding s-PS-t-B products, including (a) s-PS-t-O-Si(CH3)3, (b) s-PS-b-PMMA (Table 10.3, sample 3), and (c) s-PS-b-PBMA (Table 10.3, sample 4), respectively. The peak at 3.78 ppm is assigned to the methoxyl groups (OCH3) in the PMMA block, and the peak at 4.15 ppm is assigned to the methylene groups (OCH2) in the PBMA block.
Functional Polyolefin Diblock Copolymers
187
* d-benzene
* (c)
* d-toluene * * * (b)
* d-toluene
(a)
* **
7.5 7.0 6.5 6.0 5.5 5.0 4.5 4.0 3.5 3.0 2.5 2.0 1.5 1.0 0.5 ppm
Figure 10.4 1H NMR spectra of (a) PP-b-PEMA with 27 mol% EMA, (b) PP-b-PBA with 34 mol% BA, and (c) PP±b-PS with 11 mol% styrene.
Apparently, many polar polymer blocks can be incorporated in the s-PS diblock copolymers. Table 10.3 summarizes the reaction conditions and experimental results of diblock copolymers. In general, both PMMA and PBMA blocks increase along with the reaction time, and chain extension reactions continue even after 12 h. Diblock copolymers with a nearly 1/1 mole ratio of [styrene]/[MMA] or [styrene]/[BMA] have been prepared, despite the heterogeneous reaction conditions. These effective and long-lived chain extension reactions again provide strong evidence of the
188
Functionalization of Polyolefins
(a)
(b)
(c)
8
6
4 ppm
2
0
Figure 10.5 1H NMR spectra of (a) s-PS-t-O-Si(CH3)3 and diblock copolymers, (b) s-PS-b-PMMA, and (c) s-PS-b-PBMA. (Solvent: C2D2Cl4; temp.: 110 C.) (Redrawn from Macromolecules 1999, 32, 8689. Copyright 1999 Am. Chem. Soc.)
borane group's existence at the polymer chain end and the living free radical chain extension process. Figure 10.6 compares two DSC curves of s-PS-t-B (I) and the corresponding s-PS-b-PMMA (sample 3 in Table 10.3). Two identical thermal transition temperatures, including a Tm near 270 C and a Tg near 100 C for the s-PS polymer, were observed in both samples. A clear new Tg at about 130 C (corresponding to the high-molecular-weight PMMA polymer) is shown in Fig. 10.6(b). Both polymer segments in the s-PS-b-PMMA copolymer must have long consecutive (undisturbed) sequences to form separate domains.
Functional Polyolefin Diblock Copolymers Table 10.3
189
Summary of s-PS-b-PMMA and s-PS-b-PBMA diblock copolymers Reaction conditions
Diblock copolymer
1
s-PS-t-B / Temp./time O2 Polar monomer Yield s-PS block Polar block Blocks (g) ( C)/(h) (ml) (mol) (g) Mn ( 10 3) Mn ( 10 3) mole ratio (I)/5 (I)/5 (I)/5 (I)/5 (I)/5 (I)/5 (II)/7 (II)/7 (II)/7
25/6 25/12 25/24 25/10 25/20 25/24 25/6 25/12 25/24
1.9 1.9 1.9 1.9 1.9 1.9 2.4 2.4 2.4
MMA/1.87 MMA/1.87 MMA/1.87 BMA/1.87 BMA/1.87 BMA/1.87 MMA/1.87 MMA/1.87 MMA/1.87
6.25 7.56 9.98 6.72 7.56 8.94 7.56 8.13 8.98
15.0 15.0 15.0 15.0 15.0 15.0 70.0 70.0 70.0
2.0 5.0 14.0 7.1 15.0 18.3 3.0 8.0 17.0
100 : 14 100 : 35 100 : 97 100 : 35 100 : 72 100 : 89 100 : 4.5 100 : 12 100 : 25
1
s-PS-t-B (I): Mn 15 103 g/mol, Mw/Mn 2.2; s-PS-t-B (II): Mn 70 103 g/mol; Mw/Mn 2.1. 90 88 86
Heat flow (mW)
84 82 80 78 (a)
76 74 72
(b)
70 68 75
125
175 225 Temperature (°C)
275
Figure 10.6 DSC curves of (a) s-PS (I) and (b) s-PS-b-PMMA. (Redrawn from Macromolecules 1999, 32, 8689. Copyright 1999 Am. Chem. Soc.)
4 TRANSFORMATION FROM ZIEGLER±NATTA TO LIVING ANIONIC POLYMERIZATION As discussed in Chapter 9, p-methylstyrene-(p-MS) terminated polyole®ns have been prepared by the combination of metallocene catalysis and chain transfer reaction to p-MS/H2. The terminal p-MS group provides the active site to transform
190
Functionalization of Polyolefins
metallocene polymerization to living anionic polymerization,350 as illustrated in Eq. (10.4):
(I) PP
CH2CH2
CH3
s-BuLi/TMEDA (II) PP
CH2CH2
+ CH2 Li
10:4
Styrene (III) PP
CH2CH2
CH2
PS
The metallation reaction of p-MS-terminated polypropylene (PP-t-p-MS) was carried out under heterogeneous reaction conditions by suspending the powder form of PP in cyclohexane. An excess of s-BuLi/TMEDA was used to ensure a complete reaction. Usually, the reaction mixture was stirred at room temperature for a few hours before removing the polymer powder from the solution by ®ltration and washing. To examine the ef®ciency of the reaction, some of the metallated polymer was terminated with Cl±Si(CH3)3. Figure 10.7 compares the 1H NMR spectra of the silated product and the starting PP-t-p-MS (Mn 25.9 103 g/mol; PDI 2.3). In Fig. 10.7(a), in addition to the three major peaks at 0.95, 1.35, and 1.65 ppm corresponding to the CH3, CH2, and CH protons in the PP backbone, there are three minor peaks at 2.35, 2.60, and 7.0±7.1 ppm corresponding to -CH3, ±CH2±-CH3, and aromatic protons, respectively, in the p-MS terminal group. After the metallation and silation reactions, two new peaks appear at 0.05 and 2.15 ppm and correspond to the two types of protons in the new terminal -CH2±Si(CH3)3 group. These are shown in Fig. 10.7(b). Based on the peak intensities, the overall conversion is about 85%. Apparently, the metallation reaction is not hindered by the heterogeneous reaction conditions. Most of the terminal -CH3 groups may be located on the surface or in the amorphous phases and thus be readily accessible for the metallation reaction.
Functional Polyolefin Diblock Copolymers
191
(b)
(a)
ppm
9
Figure 10.7
8 1
7
6
5
4
3
2
1
H NMR spectrum of (a) the starting PP-t-p-MS and (b) the silated product.
192
Functionalization of Polyolefins
Most of the lithiated PP-t-p-MS (II) was used to prepare diblock copolymers. By mixing polymer powder with styrene monomer in cyclohexane solvent, the living anionic polymerization took place to produce PP-b-PS diblock copolymer. After the reaction, the product was vigorously extracted by re¯uxing THF to remove any PS homopolymer. In all cases (including high-molecular-weight PP-t-p-MS polymer cases), the soluble PS homopolymer fraction was negligible, as shown in Table 10.4. Basically, the PP-b-PS product increases its PS content and molecular weight as more styrene is introduced during the reaction. Not all of the styrene was converted in the relatively short reaction time under non-polar solvent conditions. Figure 10.8 (top) shows the 1H NMR spectra of PP-b-PS (Mn 44.0 103 g/mol; PDI 2.34), and (bottom) compares the GPC curve of PP-b-PS (b) with the starting PP-t-p-MS (Mn 25.9 103 g/mol; PDI 2.3) (a). Despite the doubling of polymer molecular weight, the molecular weight distribution (PDI) remains very constant. The quantitative analysis of the diblock copolymer composition was calculated using the ratio of two integrated intensities between the aromatic protons ( 6.4±7.3 ppm) in the PS segment and the methylene protons ( 1.35± 1.55 ppm) in the PP segment, along with the number of protons both chemical shifts represent. Overall, the transformation from metallocene to living anionic polymerization was very effective ( > 80%), and the molecular structure of the copolymer could be easily controlled by the starting p-MS-terminated polyole®n and the quantity of monomer introduced during the living anionic chain extension reaction. This diblock reaction bene®ts from the known and well-de®ned living anionic polymerization process, and offers a complementary method to the previous one, i.e. transformation from metallocene to living free radical polymerization. The combination provides a powerful tool to prepare a broad range of polyole®n diblock copolymers containing a metallocene-prepared polyole®n block and an anionic or free-radical-prepared polymer block.
Table 10.4
Summary of PP-b-PS diblock copolymers
Reaction conditions
1
PP-b-PS products
PP1 (g)
Styrene (g)
Time (h)
Yield (g)
THF soluble (g)
PP/PS (mol)
Mn (g/mol)
Mw/Mn
0.73 0.73 0.69 0.76
0.9 2.7 4.5 9.0
1.0 1.0 3.0 3.0
0.94 1.54 2.77 3.60
n.g. n.g. n.g. n.g.
86.9/13.1 60.6/34.4 52.2/47.8 48.8/51.2
Ð 54.0 Ð 117.5
Ð 2.34 Ð 2.25
PP-t-p-MS sample (Mn 25.9 103 g/mol; Mw/Mn 2.3).
Functional Polyolefin Diblock Copolymers
ppm
9
8
7
6
5
4
3
2
193
1
200
mV
180 160 (b)
(a)
140 120
100
22
23
24
25
26 27 Minutes
28
29
30
31
Figure 10.8 (Top) 1H NMR spectra of PP-b-PS diblock copolymer and (bottom) GPC curves of the starting PP-t-p-MS (a) and the resulting PP-b-PS (b) polymers.
5 TRANSFORMATION FROM LIVING ANIONIC TO ZIEGLER±NATTA POLYMERIZATION Considerable efforts were devoted to the study of transformation reaction from living anionic polymerization to Ziegler±Natta polymerization366,367 as a route for
194
Functionalization of Polyolefins
preparing polyole®n diblock copolymers. The chemistry (as illustrated in Eq. 10.5) is based on the idea that the living anionic-polymerized polymer with active alkylithium moiety can serve as an alkylation agent (co-initiator) of transition metal halide: CH2CH ⱍ
BuLi
PS
– CH2CH Li+ Anionic polymerization ⱍ (excess)
TiCl4 PS
CH2CHTiCl3 + Mixtures ⱍ
10:5
CH2CH2 Ziegler–Natta polymerization PS
PE
The resulting Ziegler±Natta catalytic site, then located at the polymer chain end, initiates the polymerization of -ole®ns to produce diblock copolymer containing an anionic prepared polymer block and a polyole®n block. Some obvious limitations were shown in this sequential anionic/Ziegler±Natta polymerization route. Due to the sensitivity of the Ziegler±Natta catalyst to polar groups, the ®rst anionic-polymerized block has to be a hydrocarbon polymer. It is very dif®cult to prepare a functional polymer by this route. In addition, a large excess of reducing agent (including alkylithium) in comparison to the transition metal is usually needed to form an active Ziegler±Natta catalyst. In other words, only a small portion of the anionic-prepared polymers involves the subsequent Ziegler±Natta polymerization. Therefore, the major portion of the product is simply a mixture of homopolymers.
6
COUPLING REACTION
Although a coupling reaction presents a simple route for linking two molecules, the chemistry in polyole®n cases is very dif®cult. In addition to the dilute concentration of the reactive end group, the kinetic of mixing between the high polyole®n and polar polymer is very slow due to high (solution or melt) viscosity and incompatibility between two very different polymers. It is also very dif®cult to control the stoichiometry of two reactive groups, which is essential for an effective coupling reaction. Furthermore, the availability of functional-group-terminated polyole®ns is simply very limited.
Functional Polyolefin Diblock Copolymers
195
Doi et al. reported a coupling reaction between living polystyrillithium and an iodine-terminated syndiotactic polypropylene to produce a diblock copolymer.368 We also studied a coupling reaction between maleic-anhydride-terminated PP and nylon,120 as illustrated in Eq. (10.6):
PS
PP
10:6
O
O
O
Nylon
PP
O + H2N
PP
PS
N
Nylon
PP
CH3 ⱍ – CH2CHI + Li+ CHCH2 ⱍ
O
Neither of these diblock copolymer structures was proven unequivocally. Despite the choice of reaction pairs with good reactivities, the product inevitably contains some unattached homopolymers, although they are not shown in the morphology of the polymer blends.
7
SUMMARY
It is both a scienti®c challenge and an industrial goal to develop a facile method of preparing functional polyole®n diblock copolymers containing a polyole®n block and a functional polymer block. The ®rst three methods, involving the reaction sequence of preparing a polyole®n block before a functional polymer block, are clearly the better choices. The living Ziegler±Natta polymerization is mostly useful for the preparation of polyethylene diblock copolymers, due to the limitation of applicable catalyst systems and reaction conditions. On the other hand, the methods of transformation from metallocene to living free radical or living anionic polymerizations are very general, covering the whole range of metallocene catalysts and copolymer compositions. In addition to the effectiveness and convenience of the transformation reaction, the diblock copolymer obtained has a relatively wellde®ned molecular structure and a very low content of homopolymer. The molecular weight of the polyole®n block is basically controlled by the concentration of the chain transfer agent, with a molecular weight distribution (Mw/Mn) of about 2. The second functional polymer block is prepared by living polymerization, and its molecular weight is governed by the monomer concentration and reaction time.
This Page Intentionally Left Blank
11 Synthesis of Functional Polyole®n Graft Copolymers
1
INTRODUCTION
Beside the block copolymer structure, the other segmental polymer is the graft copolymer. Functional polyole®n graft copolymers, containing a polyole®n backbone and several functional polymer side chains, can provide the needed interactive properties, as well as preserve desirable polyole®n properties (crystallinity, melting temperature, etc.). As expected, a polyole®n graft copolymer is also an effective interfacial agent, serving as the compatibilizer in polyole®n blends and composites. In chemistry, it is very important for developing a facile route to prepare polyole®n graft copolymers containing functional polymers, such as PMMA, PVA, and PCL. Unfortunately, the chemistry to prepare polyole®n graft copolymers is very limited. Numerous approaches, based on post-polymerization process, have been employed in forming polyole®n graft copolymers. Ionizing radiation (X-rays,
-rays, and e-beams) in the presence of air, ozone, UV with accelerators, and free radical initiators369,370 have all been used to form polymeric peroxides. When heated in the presence of monomers, some polymeric peroxides initiate graft polymerization. However, these high-energy reactions lead to many side reactions, such as crosslinking and chain cleavage resulting in diminished processibility and mechanical properties, respectively. In most cases, the structure and composition of copolymers are dif®cult to control given the considerable amounts of ungrafted homopolymers. On the other hand, there have been no reports of attempts using the direct polymerization process, which might involve a functional macromonomer as a comonomer in the copolymerization reaction. Catalyst poisoning by functional groups, along with the possible synthesis dif®culty and low reactivity of such a macromolecule, would certainly prohibit such attempts. By far, the most effective method for preparing functional polyole®n graft copolymers is the chemistry involving the intermediate reactive polyole®ns (shown in the next page) developed in our laboratory:
Functionalization of Polyolefins
R
ⱍ B
198
R
ⱍ
ⱍ
ⱍ CH3
ⱍ CHCH2
Three reactive groups ± borane,371±375 p-methylstyrene,107±109 and divinylbenzene350 ± located along the polyole®n side chains provide the selective sites for the subsequent graft polymerization of functional monomers. It is interesting to note that they have served as initiator, monomer, or termination agent during the graft polymerization. The combination of these three reactive polyole®n approaches offers a comprehensive coverage of polyole®n graft copolymers with well-de®ned molecular structure, i.e. controlled graft density and graft length.
2 LIVING RADICAL GRAFT-FROM REACTION VIA BORANE REACTIVE SITES As discussed in Chapters 3 and 8, borane groups can be effectively incorporated into polyole®n side chains via both direct and post processes involving metallocene catalysts. Borane-containing polyole®ns with relatively well-de®ned molecular structure, i.e. controlled borane content and narrow molecular weight and composition distributions, are available. It is interesting to note that the direct copolymerization usually produces a primary borane group (A) and the postpolymerization results in the secondary borane groups (B) in the polyole®n copolymers, as illustrated below:
CH3
CH3
ⱍ
(CH2CH)x(CH2CH)y ⱍ
ⱍ
(CH2CH)x(CH2CH)y ⱍ
CH2
CH2
CH2
CH2
ⱍ
ⱍ
ⱍ
ⱍ
HCB
CH2
(A)
ⱍ
CH2 ⱍ
(B)
ⱍ
CH3
B
As illustrated in Eq. (11.1), the alkyl-9-BBN side groups in polyole®ns can then be selectively oxidized at the aliphatic C±B group for graft-from polymerization. The formed peroxyborane (B±O±O±C) initiates a living radical polymerization in the presence of free radical-polymerizable monomers, such as methacrylates, vinyl acetate, and acrylonitrile, at ambient temperature. This living polymerization
Synthesis of Functional Polyolefins
199
process minimizes the undesirable chain transfer reaction and termination (coupling and disproportional) reaction between the two growing chain ends, which result in the formation of homopolymers and crosslinked material. In most cases, the resulting graft copolymers are completely soluble and processible. The graft length (PMMA side chain) is basically controlled by the MMA concentration and reaction time. Some interesting graft polymers ± including PE-g-PVA, PE-g-PMMA, PP-gPMMA, PP-g-PMA, PP-g-PVA, EP-g-PMMA, and butyl-g-PMMA ± have been synthesized with controllable compositions and molecular microstructures. The following examples explain the general experimental conditions and results.
Polyolefin-bonded alkyl-9-BBN CB O2
CO* *OB (I)
2.1
MMA
CH3 ⱍ COCH2C* *OB ⱍ CO ⱍ (II) O ⱍ CH3
11:1
PP Graft Copolymers374,375
In a typical example, 2 g of a borane-containing polypropylene copolymer prepared by direct polymerization and containing 0.5 mol% borane monomer was placed in a suspension of 12 g dry, degassed, and uninhibited MMA in a sealed, opaque ¯ask. The reaction was initiated by injecting dry O2 into the reaction ¯ask. After a certain reaction time, the reaction was terminated by recovering the unreacted MMA under vacuum. The white solid was extracted with acetone (which can effectively separate a mixture of i-PP and MMA homopolymers) in a Soxhlet apparatus for 24 h. Table 11.1 summarizes the experimental results of PP-g-PMMA copolymers, which are obtained from both polymer (A) and (B) containing primary and secondary alkyl-9-BBN, respectively. Comparison among runs A-1 to A-4 shows the relationship of the addition of oxygen to the graft ef®ciency. Even though the ®nal stoichiometry of oxygen to boron should be 1 : 1, the best results in this heterogeneous reaction system are realized when the O2 is introduced slowly so that O << B at any time. Excess O2 is not only a poison in free radical polymerizations, but also leads to over-oxidation to boronates and borates, which are poor free radical initiators at room temperature. The polarity of the solution also affects the graft reaction. THF is a very good solvent in this reaction. Nonpolar solvents, such as benzene, slow down the graftfrom reaction, which may be due to the solubility of O2 in the solvent. In run A-5,
200
Functionalization of Polyolefins
Table 11.1
Summary of PP-g-PMMA graft copolymers
Run no.
9-BBN in PP (mol%)
O2 (ml/h)
Monomer/ solvent
Reaction time (h)
A-1 A-2 A-3 A-4 A-5 B-1 B-2
0.5 0.5 0.5 0.5 0.5 1.7 1.7
1.5/12 3.0/1 1.4/3 6 (at once) Diffusion Diffusion 1/12
MMA (neat) MMA (neat) MMA/THF MMA (neat) MMA/THF MMA (neat) MMA/benzene
48 2 12 48 48 24 12
MMA in PP (mol%) 66 6 52 1.5 12 18 13
2.0 (d)
Absorbance
1.5
(c) 1.0
(b) 0.5
(a) 0.0 3400 3200 3000 2800 2600 2400 2200 2000 1800 1600 1400 1200 1000 800 Wavenumber
Figure 11.1 IR spectra of (a) PP-OH with 0.5 mol% OH groups and three PP-gPMMA graft copolymers containing (b) 20, (c) 52, and (d) 66 mol% MMA. (Redrawn from Macromolecules 1993, 26, 3467. Copyright 1993 Am. Chem. Soc.)
the oxygen was introduced by diffusion of air through the rubber septum which was tightly installed on the top of the reactor. The insuf®cient O2 in this process leads to a low percentage of PMMA formation. It is very interesting to note that a signi®cant difference was observed from using primary and secondary borane groups in the graft-from reactions. Despite the higher borane concentration in polymer (B), polymer (A) shows a higher concentration of PMMA grafted to PP. In addition, a signi®cantly high concentration ( > 30%) of the PMMA homopolymer was isolated
Synthesis of Functional Polyolefins
201
by using polymer (B). The high grafting ef®ciency in polymer (A) must be attributed to the higher reactivity of the unhindered polymeric primary C±B bond, as opposed to the more sterically hindered secondary C±B bonds. Figure 11.1 compares the IR spectra of three PP-g-PMMA graft copolymers and a corresponding PP-OH. All of them were derived from the same copolymer (A), which contained 0.5 mol% primary borane monomers and was prepared by a direct copolymerization reaction. PP-OH polymer was obtained by converting borane groups to hydroxy groups, and all graft copolymers were prepared under the same reaction conditions except for reaction time. The IR spectrum of PP-OH is basically indistinguishable from that of i-PP because of the extremely low concentration of hydroxy groups. On the other hand, the strong absorption band at 1730 cm 1, corresponding to ester groups, clearly shows the existence of PMMA in the graft copolymers. A high concentration ( > 65 mol%) of PMMA can be incorporated into PP by a small quantity (0.5 mol%) of borane groups. The PMMA content increases with the reaction time, which indicates the living free radical polymerization process, and the graft length is basically controlled by the reaction time and monomer concentration. 2.2
EP Graft Copolymers364,365
Similar graft-from reactions have been extended to other polyole®ns. Satisfactory results were obtained in both homogeneous (PE, PP, and PB cases) and heterogeneous (PO, EP, and butyl rubber cases) reaction conditions. In one homogeneous case, a commercial EPDM rubber was used to prepare EP-g-PMMA graft copolymers. Figure 11.2 compares the 1H NMR spectra of the resulting EP-g-PMMA copolymers and the starting EPDM rubber, poly(ethylene-ter-propylene-ter-1,4hexadiene). The chemical shift at 3.6 ppm in Figs 11.2(b) and (c) corresponds to the methyl groups (CH3O) in PMMA. The chemical shifts between 2.1 and 0.7 ppm include all of the protons in EP and ®ve of the protons in the methyl group located on the PMMA backbone. The copolymer composition was calculated by the ratio of the two integrated intensities at 3.6 and 0.7±2.1 ppm and the number of protons both chemical shifts represent. Figures 11.2(b) and (c) indicate 28 and 52 mol% PMMA in EP-g-PMMA copolymers, respectively. The DSC curve of the EP-g-PMMA copolymer (with a 50/50 composition) shows two glass transition temperatures (Tg), 47 and 130 C, corresponding to the starting EPDM rubber and PMMA homopolymer, respectively. This result indicates a clear phase separation between the EP backbone and the PMMA side chains. The copolymer backbone must have enough consecutive sequences of EP backbone units to form separate domains. The side chain must be a high-molecular-weight polymer with a microstructure similar to that of PMMA homopolymer. The clear phase separation with hard (polar) and soft (nonpolar) domains is a very interesting molecular structure. In fact, most of the graft copolymers behave like thermoplastic elastomers.
202
Functionalization of Polyolefins
(c)
(b)
(a)
9
8
7
6
5
4
3
2
1
0
ppm
Figure 11.2 Comparison of the 1H NMR spectra of (a) EPDM and two resulting EP-g-PMMA copolymers containing (b) 28, and (c) 52 mol% PMMA. (Redrawn from Macromolecules 1994, 27, 26. Copyright 1994 Am. Chem. Soc.)
Synthesis of Functional Polyolefins
203
3 LIVING ANIONIC GRAFT-FROM REACTION VIA p-METHYLSTYRENE REACTIVE SITES As discussed in Chapter 7, p-methylstyrene-(p-MS) containing polyole®ns have been prepared via direct copolymerization reactions. The copolymers include a wide range of compositions and narrow molecular weight and composition distributions if a metallocene catalyst is used. Usually, a low concentration (< 1 mol%) p-MS in the copolymer is preferred for the preparation of graft copolymer because the resulting graft copolymer will have low graft density and long graft length. The p-MS groups in polyole®n can be effectively metallated at ambient temperature. The lithiated PEp-MS and PP-p-MS copolymers contain several polymeric anions, and provide a unique opportunity for the preparation of PE and PP graft copolymers, as illustrated in Eq. (11.2): CH3 ⱍ (CH2CH)x(CH2CH)y ⱍ
s-BuLi
CH3 ⱍ (CH2CH)x(CH2CH)y ⱍ
ⱍ
–ⱍ CH2 +
CH3
nM M: styrene, p-methylstyrene, butadiene, methyl methacrylate, acrylonitrile
Li
CH3 ⱍ (CH2CH)x(CH2CH)y ⱍ
11:2
ⱍ
CH2 (M)n
The initiation sites are homogeneously distributed in the polymer chain. The living anionic graft-from reactions were generally carried out at ambient temperature by suspending the lithiated PE-p-MS or PP-p-MS copolymer in cyclohexane in the presence of the monomers. After the polymerization reaction, the crude product was vigorously extracted to remove any ungrafted homopolymer. In all cases, the homopolymer fraction was negligible. The insolubility of the crystalline PE and PP backbones at room temperature allows for maximum removal of the unreacted lithiation reagent after the metallation reaction. The combination of pure lithiated polymer and the living graft-from polymerization minimizes the formation of ungrafted homopolymer during the graft-from polymerization.
204
Functionalization of Polyolefins
PE-g-PS and PE-g-PMS Graft Copolymers376
3.1
Figure 11.3 shows 1H NMR spectra of three resulting PE-g-PS graft copolymers containing 25.6, 38.1, and 43.8 mol% polystyrene, respectively. They were prepared from the same starting PE-p-MS polymer, containing 0.9 mol% of p-MS units. In contrast to the starting PE-p-MS, three additional chemical shifts arise in the graft
(c)
(b)
(a)
9
8
7
6
5
4
3
2
1
0
ppm
Figure 11.3 1H NMR spectra of PE-g-PS graft copolymers with (a) 25.6, (b) 38.1, and (c) 43.8 mol% polystyrene. (Redrawn from Macromolecules 1997 30, 1272. Copyright 1997 Am. Chem. Soc.)
Synthesis of Functional Polyolefins
205
copolymers at 1.55, 2.0, and 6.4±7.3 ppm corresponding to CH2, CH, and aromatic protons in polystyrene. The quantitative analysis of the copolymer composition was calculated by the ratio of two integrated intensities between the aromatic protons ( 6.4±7.3 ppm) in the PS side chains and the methylene protons ( 1.35± 1.55 ppm) and the number of protons both chemical shifts represent. The same anionic graft-from polymerization was also applied to the p-MS monomer. Figure 11.4 shows the 1H NMR spectra of three polyethylene-g-poly(p-methylstyrene) (PE-g-PMS) copolymers with various concentrations of PMS
(c)
(b)
(a)
9
8
7
6
5
4
3
2
1
0
ppm
Figure 11.4 1H NMR spectra of PE-g-PMS graft copolymers with (a) 13.8, (b) 27.9, and (c) 37.2 mol% poly(p-methylstyrene). (Redrawn from Macromolecules 1997, 30, 1272. Copyright 1997 Am. Chem. Soc.)
206
Functionalization of Polyolefins
Table 11.2
Summary of PE-g-PS and PE-g-PMS graft copolymers
Reaction conditions1 Run PE -Li no (g)
ST/p-MS (g)
Starting material3 A1 1.0 ST/1.7 A2 1.0 ST/2.9 A3 1.0 ST/4.0 B1 1.0 ST/1.4 B2 1.0 ST/2.3 B3 1.0 ST/4.2 C1 0.58 p-MS/0.50 C2 0.60 p-MS/1.43 C3 0.60 p-MS/2.60
Graft copolymers2 Yield Graft mol% Graft Graft Tg Tm Hf J/g of graft (J/g of PE) (g) (wt%) density length ( C) ( C) 2.30 3.30 3.90 2.00 2.50 4.42 0.92 1.60 2.10
25.6 38.1 43.8 20.5 27.4 47.8 13.8 27.9 37.2
0 (56.1) (69.6) (74.3) (48.9) (58.4) (77.3) (40.3) (62.0) (71.4)
0 0.8 0.8 0.8 2.0 2.0 2.0 2.0 2.0 2.0
0 23.4 41.4 52.2 6.99 10.5 23.9 4.10 11.6 17.5
Ð 103.6 103.8 103.0 84.6 91.9 103.7 Ð 80.0 102.0
127.8 121.9 120.3 120.9 126.6 123.2 120.3 Ð 121.4 121.4
199.0 (199.0) 60.6 (138.0) 41.7 (137.2) 31.1 (121.0) 99.6 (195.0) 80.4 (193.1) 25.4 (111.9) Ð 70.7 (186.0) 34.2 (119.6)
1
Solvent: anhydrous cyclohexane (30 ml/g PE); reaction time: 1 h; reaction temperature: 25 C. 2 Graft density: no. of graft/1000 carbons of backbone; graft length: 103 g/mol. 3 The starting material was poly(ethylene-co-p-methylstyrene) with 0.90 mol% Å w 100 000 g/mol, M Å w/M Å n 2.5. p-MS, M
side chains. They were also prepared from the same PE-p-MS copolymer containing 0.9 mol% p-MS units. Four new chemical shifts at around 1.55, 1.95, 2.35, and 6.4± 7.0 ppm ± corresponding to CH2, CH, CH3, and aromatic protons in PMS side chains ± are observed in all three PE-g-PMS copolymers. The concentration of PMS was determined by the intensity ratio of two chemical shifts, 1.35±1.55 ppm for methylene protons and 6.4±7.0 ppm for aromatic protons in the side chains, and the number of protons both chemical shifts represent. Figures 11.4(a), (b), and (c) indicate three PE-g-PMS copolymers containing 13.8, 27.9, and 37.2 mol% of PMS respectively. Table 11.2 summarizes the experimental conditions and results of three sets of comparative graft-from reactions, all started from the same PE-p-MS copolymer, containing 0.9 mol% p-MS. The runs A1±3 are the same three graft copolymers shown in Fig. 11.3 and the runs C1±3 are the same graft copolymers shown in Fig. 11.4. Both the yield and graft compositions (PS or PMS content) of the graft copolymer are basically proportional to the quantity of monomers used in the graftfrom reaction. The good control of graft copolymer formation is obviously due to the living anionic polymerization, which effectively converts monomers to the grafted side chains. The graft density is de®ned as the number of grafted side chains per 1000 repeating methylene units of the polyethylene backbone. Since this process involves living anionic polymerization and fast initiation, it is reasonable to assume that each benzylic lithium produces one PS side chain and that the side chains have a narrow molecular weight distribution. Therefore, the graft density is the same as the
Synthesis of Functional Polyolefins
207
density of the benzylic anions and can be calculated by a simple equation: Graft density
mol p-MS (efficiency of metallation) 10 2
In sets A, B, and C, the graft densities are about 0.8, 2, and 2, which on average correspond to about 2.5, 6, and 6 side chains per PE backbone, respectively. Under the same assumption of living anionic graft-from polymerization, the graft length (de®ned as the average molecular weight of the side chain) can be estimated by the following equation: Graft length (weight of graft copolymer) [weight of starting P(E-co-p-MS
g=mol mol benzylic lithium The molecular weight of the side chain is inversely proportional to the degree of metallation and proportional to the quantity of monomer used in the graft-from polymerization. Overall, this chemistry provides a very useful route for preparing PE-g-PS and PE-g-PMS copolymers with relatively well-de®ned molecular structures, i.e. relaÅ w/M Å n 2.5) of the backbone and tively narrow molecular weight distribution (M well-de®ned side chains. All the important factors in a graft copolymer ± including graft density, graft length and copolymer composition ± can be controlled during the reaction processes. Figure 11.5 compares the DSC curves of the starting PE-p-MS and three PE-g-PS copolymers (runs B1±3 in Table 11.2), which have identical graft densities and different graft lengths. All samples were prepared with the same weight and were given the same thermal treatment by heating at 20 C/min. In Fig. 11.5(a), the endotherm peak at 127.8 C is clearly due to the melting point (Tm) of crystalline PE. It is interesting to note that as the PS graft length increases, the Tm of PE only slightly decreases. Sample B3, containing 47.8 mol% (77.3 wt%) PS, still shows the Tm at 120.3 C. On the other hand, the heat of fusion (Hf J/g of graft copolymer) is very dependent on the PS or PMS graft chain length, as shown in Table 11.2. In samples B1, B2, and C1, with graft chain lengths < 12 103 g/mol, the heat of fusion after normalizing with the content of PE (Hf J/g of PE) in each case is very similar to that of the pure PE sample, despite relatively high PS or PMS contents (48.9, 58.4, and 62.0 wt%, respectively). The simple dilution effect seems to govern the heat of fusion in the graft copolymers. However, in samples A1±3, B3, and C3, with the graft length > 17 103 g/mol, the additional disorder associated with long grafted side chains becomes signi®cant. So far, there is no theoretical explanation for the graft chain length effect on the crystallinity of the PE backbone. On the other hand, the glass transition temperature (Tg) of the PS side chains is clearly observed in Figs 11.5(b), (c), and (d). As the molecular weight of PS increases, so does the Tg. The Tg becomes constant when the graft length exceeds 15 103 g/mol.
208
Functionalization of Polyolefins Tm = 120.3°C
12
Tm = 127.8 °C Tm = 123.1 °C
11
(d)
Heat flow (W/g)
Tg 103.77 °C Tm = 126.6 °C
10
(c)
9
Tg 91.89 °C
8 (b) Tg 84.56 °C
7
(a)
6 5 50
75
100 125 Temperature (°C)
150
175
Figure 11.5 DSC curves of (a) the starting PE-p-MS and three resulting PE-g-PS graft copolymers containing (b) 20.5, (c) 27.4, and (d) 47.8 mol% polystyrene. (Redrawn from Macromolecules 1997, 30, 1272. Copyright 1997 Am. Chem. Soc.)
3.2
Synthesis of PE-g-PMMA and PE-g-PAN Graft Copolymers115
The lithiated PE-p-MS polymer was also used as the starting material to initiate graft-from polymerization of methyl methacrylate (MMA) and acrylonitrile (AN) in THF or cyclohexane solvents. The crude polymer products were subjected to solvent fractionation. In most cases, less than 10% homopolymer was obtained. Table 11.3 summarizes the experimental results of PE-g-PMMA and PE-g-PAN. Overall, the graft-from polymerization reactions of MMA and AN are less ef®cient compared with those of styrene and p-MS. A polar solvent like THF gives poor yield at both 0 C and 25 C, while a nonpolar solvent such as cyclohexane gives much better results. Polar solvents may increase the nucleophilicity of the carbanion, which results in more side reactions. It is very interesting to note that the anionic polymerization of polar monomers using butyllithium as an initiator cannot achieve a high polymer at ambient temperature. Usually, very low reaction temperatures ( < 20 C) are required. In the lithiated PE-p-MS case, the formed polymeric benzylic lithium is much more stable, and therefore minimizes the side reactions. As can be seen from Table 11.3, graft-from polymerization reactions of MMA and AN in the nonpolar solvent cyclohexane are fairly effective and a suf®ciently long graft length can be achieved even at ambient temperature.
Synthesis of Functional Polyolefins Table 11.3
209
Summary of PE-g-PMMA and PE-g-PAN coplymers
Lithiated polymer (g)
Monomer (g)
Solvent
Temp. ( C)
Time (h)
Graft copolymer (g)
1.0 1.0 1.0 0.8 0.8 1.2 1.2 1.0
MMA/3.7 MMA/3.4 MMA/5.0 MMA/4.0 MMA/4.0 AN/3.0 AN/3.6 AN/3.0
THF THF THF Hexane Hexane THF THF Hexane
0 0 25 25 0 25 25 25
1.5 15 5 5 5 1 16 16
1.86 2.66 1.90 3.08 2.21 1.60 2.25 2.99
PMMA or PAN content (mol%) 20.0 31.8 20.1 44.4 33.0 15.0 32.0 51.2
[Reproduced with permission from Synthesis of PP Graft Copolymers via Anionic Living Graft-from Reactions of Polypropylene Containing Reactive p-Methylstyrene Units, H. L. Lu and T. C. Chung, J. Polym. Sci., Polymer. Chem. Ed. 1999, 37, 4176. Copyright 1999 John Wiley & Sons]
3.3
PP-g-PB, PP-g-PS and PP-g-PMS Graft Copolymers377
A similar anionic graft-from polymerization was applied to the lithiated PP-p-MS polymer. Figure 11.6 shows the 1H NMR spectra of PP-g-PS and PP-g-p-MS copolymers, both prepared from the same starting PP-p-MS containing 0.6 mol% p-MS units. Compared with the 1H NMR spectrum of the starting PP-p-MS, Fig. 11.6(a) shows three additional chemical shifts at 1.55, 2.0, and 6.4±7.3 ppm corresponding to CH2, CH, and aromatic protons in polystyrene. The copolymer composition was calculated using the ratio of two integrated intensities between the aromatic protons ( 6.4±7.3 ppm) in the PS side chains and the methylene protons ( 1.35±1.55 ppm) and the number of protons both chemical shifts represent. On the other hand, the 1H NMR spectra in Fig. 11.6(b) shows four new chemical shifts around at 1.55, 1.95, 2.35, and 6.4±7.0 ppm corresponding to CH2, CH, CH3, and aromatic protons in the PMS side chains. The concentration of PMS was determined by the intensity ratio of the two chemical shifts 1.35±1.55 ppm for the methylene protons and 6.4±7.0 ppm for the aromatic protons in the side chains. Table 11.4 summarizes the experimental conditions and results of the PP-g-PS and PP-g-PMS graft copolymers. Overall, the graft-from reactions are quantitative, and almost complete monomer conversion can be achieved within a few hours. The graft content increases with the monomer concentration and reaction time. Since the graft-from reaction involves a living anionic polymerization with fast initiation, it is reasonable to assume that each benzylic lithium produces one polymer side chain and these side chains have similar molecular weight (the experimental evidence will be discussed later). The graft density, de®ned as the number of grafted side chains per 1000 carbons in the PP backbone, is the same as the density of benzylic anions. On the other hand, the graft length (de®ned as the average molecular weight of the
210
Functionalization of Polyolefins
(b)
(a)
9
8
7
6
5
4
3
2
1
0
ppm
Figure 11.6 1H NMR spectra of two graft copolymers: (a) PP-g-PS containing 43 mol% PS and (b) PP-g-PMS containing 48 mol% PMS. [Reproduced with permission from Synthesis of PP Graft Copolymers via Anionic Living Graft-from Reactions of Polypropylene Containing Reactive p-Ms Units, H. L. Lu and T. C. Chung, J. Polym. Sci., Polymer. Chem. Ed. 1999, 37, 4176. Copyright 1999 John Wiley & Sons] Table 11.4
Summary of PP-g-PS and PP-g-PMS graft copolymers
Reaction conditions1 PP Li (g) 1.5 1.5 1.0 1.0 1.0 1.0 1.0 1
Graft copolymers
Graft Graft ST/p-MS length (g) Time Yield content Graft2 Tg (h) (g) (mol%) density (103 g/mol) ( C)
Tm ( C)
Hf J/g of graft (J/g of PP)
ST/2.3 ST/4.5 ST/3.0 p-MS/3.1 p-MS/3.1 p-MS/3.1 p-MS/3.1
155.9 154.3 157.0 Ð 156.5 154.2 156.5
29.7 (49.5) 19.7 (45.9) 15.6 (45.2) Ð 33.3 (53.3) 19.8 (49.5) 12.8 (46.1)
1.0 1.0 2.3 0.5 1.0 2.0 4.0
2.5 3.5 2.9 1.3 1.6 2.5 3.6
21.9 35.3 43.0 12.0 17.6 34.8 48.1
1.5 1.5 1.5 2.5 2.5 2.5 2.5
8.68 17.4 24.7 3.36 5.04 12.6 21.8
103.3 105.0 108.5 Ð 111.5 111.9 115.5
Solvent: anhydrous cyclohexane (30 ml/g PP); reaction temperature: ambient temperature. Graft density: no. of graft/1000 carbons of backbone.
2
Synthesis of Functional Polyolefins
211
4.0 3.9 3.8
Heat flow (mV)
3.7 3.6
(c)
3.5 3.4 3.3 3.2
(b)
3.1 3.0
(a)
2.9 2.8 50
75
100 125 Temperature (°C)
150
175
Figure 11.7 DSC curves of three PP-g-PS graft copolymers with (a) 21.9, (b) 35.3, and (c) 43.0 mol% PS contents. [Reproduced with permission from Synthesis of PP Graft Copolymers via Anionic Living Graft-from Reactions of Polypropylene Containing Reactive p-Ms Units, H. L. Lu and T. C. Chung, J. Polym. Sci., Polymer. Chem. Ed. 1999, 37, 4176. Copyright 1999 John Wiley & Sons]
side chain) is inversely proportional to the degree of metallation and proportional to the quantity of monomer used in the graft-from polymerization. Basically, both graft density and graft length in the graft copolymers can be predetermined in this graftfrom reaction. Figure 11.7 compares the DSC curves of three PP-g-PS graft copolymers with the same estimated graft density (1.5/1000 C) and different graft lengths of 8.68, 17.4, and 24.7 103 g/mol, respectively. Each sample shows two clear phase transitions, one glass transition temperature (Tg) which varies with the molecular weight of the polystyrene side chain, and one melting point (Tm) which is relatively constant and is similar to that of the starting PP-p-MS. It is signi®cant to observe a sharp Tg transition and that the Tg increases as the calculated graft length increases. In fact, the Tg correlates well with the Fox±Flory equation13 (Tg Tga K/Mn), where Tga 373 K and K 1.2 105. These results strongly indicate the presence of side chains with narrow molecular weight distribution and a living graft-from polymerization. Figure 11.8(a) shows the 1H NMR spectra of a PP-g-polybutadiene (PP-g-PB) graft copolymer containing 79 mol% butadiene units. In addition to three chemical shifts corresponding to the PP backbone, there are several new chemical shifts corresponding to the polybutadiene side chains. The allylic protons in the PB side chains are shown in the chemical shifts between 2.0 and 2.5 ppm. Coexisting
212
Functionalization of Polyolefins
(b)
(a)
9
8
7
6
5
4
3
2
1
0
ppm
Figure 11.8 1H NMR spectra of (a) PP-g-PB containing 79 mol% butadiene units and (b) PP-g-PMMA containing 26 mol% PMMA. [Reproduced with permission from Synthesis of PP Graft Copolymers via Anionic Living Graft-from Reactions of Polypropylene Containing Reactive p-Ms Units, H. L. Lu and T. C. Chung, J. Polym. Sci., Polymer. Chem. Ed. 1999, 37, 4176. Copyright 1999 John Wiley & Sons]
1,2- and 1,4-microstructures are demonstrated in the chemical shifts at 5.03 ppm, corresponding to two external ole®nic protons (=CH2); at 5.68 ppm, corresponding to internal protons (±CH=) in the 1,2-structure; and at 5.51 ppm, corresponding to internal protons (±CH=) in the 1,4-structure. The relatively integrated peak areas of 1.0±1.8 ppm, 5.03 ppm, 5.51 ppm, and 5.68 ppm, and the number of protons each peak represents, quantitatively give the total PB graft content and the mole ratio of 1,2 and 1,4 microstructures. The predominant 1,2-addition of butadiene in the graft-from reaction must be due to the existence of a polar chelating agent TMEDA (from the lithiation reaction) at the Li active site. In general, the PB content in the graft copolymer estimated by the 1H NMR measurement is in good agreement with that estimated by the weight increase in the graft-from reaction.
Synthesis of Functional Polyolefins Table 11.5 Run no.1 185D 185G 218D 185E
213
Summary of PP-g-PMMA and PP-g-PAN graft copolymers
PP Li (g)
Monomer (g)
CH2 (ml)
Yield (g)
Graft3 (mol%)
Tg ( C)
Tm ( C)
1.00 0.68 0.95 1.00
MMA/4.0 MMA/5.4 MMA/5.0 AN/3.0
30 20 15 30
1.70 2.20 3.10 2.30
23.0 38.5 48.7 50.0
71.8 79.2 96.2 Ð
155.8 157.0 156.8 Ð
1
Reaction conditions: ambient temperature, 18 h. Cyclohexane. 3 PMMA or PAN content in the graft copolymers, estimated by weight. 2
3.4
PP-g-PMMA and PP-g-PAN Graft Copolymers377
Anionic graft-from polymerizations of polar monomers, including methyl methacrylate (MMA) and acrylonitrile (AN), initiated by the lithiated PP-p-MS copolymer were also studied. Both graft copolymers (PP-g-PMMA and PP-g-PAN) were prepared using the same lithiated PP polymer, containing about 0.3 mol% benzyl lithium active sites. Figure 11.8(b) shows the 1H NMR spectrum of a resulting PP-gPMMA copolymer. In addition to the chemical shifts corresponding to the polypropylene backbone, the new chemical shift at 3.6 ppm corresponds to methyl groups (CH3±O) in the PMMA side chains. The quantitative analysis of the copolymer composition was calculated using the ratio of two integrated intensities between 3.6 ppm and 1.1 to 2.1 ppm and the number of protons both chemical shifts represent. Figure 11.8(b) indicates 26 mol% of PMMA in the PP-g-PMMA copolymer. Table 11.5 summarizes the experimental results. Overall, up to 50 mol% of PMMA and PAN has been incorporated into the side chains of the graft copolymers with very minimal PMMA or PAN homopolymer. Both graft density and graft length can be controlled by varying the active site and monomer concentrations. As discussed before, the side reaction that occurred in the initiation step between the butyllithium and the polar group was largely avoided by using a stable benzylic lithium initiator during the anionic graft-from polymerization. The Tg of the PMMA graft also increases as the graft length increases. On the other hand, the melting point of the PP backbone remains unchanged as the PMMA content increases. Apparently, both the PP backbone and the PMMA graft domains are also well phase separated. 4
RING-OPENING GRAFT-FROM REACTION
Ring-opening polymerization provides a convenient route for preparing polyole®n graft copolymer containing polycondensation polymer side chains. Since the functionalized polyole®ns are available, it is possible to use the functional group as an initiator for ring-opening reactions. One example is an anionic ring-opening
214
Functionalization of Polyolefins
reaction of -caprolactone (-CL) from a hydroxylated PP (PP-OH). In a typical example, a ®ne powder form of PP-OH (Mv 183 000 g/mol), containing 1.4 mol% hexenol units was metallated with excess n-butyllithium to form the lithium alkoxide. A toluene slurry of the powdery solid was then reacted with a 3-molar equivalent of diethylaluminum chloride for 12 h to form the PP-aluminum alkoxide. The monomer, e.g. caprolactone, was then added to a slurry of the PP-OAlEt2 in toluene. The reaction was terminated (after 24 h at room temperature) by the addition of MeOH, and the polymer was isolated via precipitation into acidi®ed MeOH. The polymer mass was extracted with hot acetone in a Soxhlet apparatus under N2 for 48 h to remove any -CL homopolymer. Results from such reactions to produce PP-g-PCL copolymers are summarized in Table 11.6. Runs 1±4 all started with the same PP-OH copolymer containing 1.4% hexenol monomer units. The weight percent PCL in the resulting graft copolymer increased linearly with increasing -CL in the feed. The relatively long reaction time can be explained by the heterogeneous reaction conditions. The diffusion of -CL into the PP matrix would be the rate-limiting step for the reaction. In fact, runs 2 and 4 differ in the reaction time of 24 and 60 h, but resulted in copolymers with 33 and 59 wt% PCL respectively. The amount of homo-PCL produced, i.e. the acetone soluble fraction, also increased with the -CL feed. Any residual aluminum alkyl not covalently bound to the polymer could easily initiate the homopolymerization of the -CL. Excess aluminum alkyl associated with the bound Al±Os may not be completely washed out by the nonpolar solvents. Run 5 started with commercial grade propylene and 1,4hexadiene copolymer containing approximately 1.6% unsaturated monomer units. The polymer was hydroborated with 9-BBN and oxidized to give functionalized PP with the secondary alcohols on either the 4 or 5 position of the hexadiene branch. The secondary alcohol can also be converted to the secondary aluminum alkoxide, which is recognized as an active initiator in the graft-from reaction of -CL. Figure 11.9 shows the 1H NMR spectra of three PP-g-PCL graft copolymers. The chemical shifts at 4.1 ppm and 2.3 ppm correspond to methylene groups (CH2±O) and (CH2±C=O), respectively, in PCL. The chemical shifts at 1.9, 1.6, and 1.1 ppm correspond to methine, methylene and methyl groups in polypropylene. The Summary of PP-g-PCL copolymers and reaction conditions 375
Table 11.6
Reaction conditions
Products
Run
PP-O-AlEt2 (g)
-CL (g)
Time (h)
Acetone soluble (g)
Acetone insoluble (g)
-CL in graft (wt%)
1 2 3 4 5
2 2 2 2 2
2.169 4.299 8.243 4.571 6.525
24 24 24 60 24
0.366 0.430 2.581 2.102 1.387
2.252 2.763 3.922 4.120 3.603
16.8 32.9 56.8 59.5 45.4
Reproduced with permission from Macromolecules 1994, 27, 1313. Copyright 1994 Am. Chem. Soc.
Synthesis of Functional Polyolefins
215
(c)
(b)
(a)
5.5
5.0
4.5
4.0
3.5
3.0 2.5 ppm
2.0
1.5
1.0
0.5
0.0
Figure 11.9 1H NMR spectra of PP-g-PCL graft copolymers containing (a) 17, (b) 33, and (c) 57 wt% PCL in d10-o-xylene at 120 C. (Redrawn from Macromolecules 1994, 27, 1313. Copyright 1994 Am. Chem. Soc.)
216
Functionalization of Polyolefins
quantitative analysis of the copolymer composition was calculated by the ratio of two integrated intensities between 4.1 ppm and 2.0 to 1.0 ppm and the number of protons both chemical shifts represent. Figures 10.9(a), (b) and (c) indicate 17, 33, and 57 wt% PCL, respectively, in PP-g-PCL copolymers. High concentrations of PCL can be incorporated into the side chains of polypropylene despite the heterogeneous reaction conditions. Figure 11.10 shows the DSC curves of three PP-g-PCL copolymers containing 17, 33, and 57 wt% PCL. All samples were given the same thermal treatment by heating in a Mettler hot stage at 180 C for 15 min before cooling quiescently. Two distinctive crystalline structures are formed in all of the graft copolymers. It is clear that the high melting peak at about 150 C is due to the PP backbone, and that the low temperature through at about 50±60 C is due to the PCL side chains. Both phases are clearly separated. In fact, slight increases in the melting point (Tm) and heat of fusion (H) of the PCL with the increase of the PCL content may be due to the increases in its molecular weight and crystallinity.
5
GRAFT REACTIONS VIA DIVINYLBENZENE SITES
One major advantage of the -ole®n/divinylbenzene copolymers is the existence of many pendent styrene groups along the backbone, which are very reactive in many chemical reactions ± including free radical, cationic, anionic, and transition metal coordination processes. As illustrated in Eq. (11.3), the pendent styrene units serve – 0.6 (a)
– 0.8 –1.0
Heat flow (W/g)
–1.2
(b)
–1.4 –1.6 (c)
–1.8 –2.0 –2.2 –2.4 –2.6 0
50
100 Temperature (°C)
150
200
Figure 11.10 DSC curve comparisons (top) among three PP-g-PCL copolymers containing (a) 17, (b) 33, and (c) 57 wt% PCL. (Redrawn from Macromolecules 1994, 27, 1313. Copyright 1994 Am. Chem. Soc.)
Synthesis of Functional Polyolefins
217
as comonomers in the graft reactions:111 Polyolefin
Polyolefin
Monomers initiator
ⱍ
CHCH2
ⱍ
CHCH2
ⱍ
CHCH2
ⱍ
CHCH2
11:3 This process resembles graft-through polymerization, and the resulting graft copolymer has a polyole®n backbone and several polymer side chains that are bonded to the polyole®n at the middle of the polymer chain. The major concern of this graft reaction is the potential of crosslinking involving multiple polyole®n-bonded styrene units. Usually, the reaction has to be carried out in speci®c reaction conditions to obtain completely soluble graft copolymers with desirable compositions. To eliminate the concerns about crosslinking reactions, it is important to design a graft reaction with the pendent styrene units serving as the initiation or termination sites. In other words, the side chain polymer grafts to the backbone at its chain end, and the graft polymer basically has a similar molecular structure as those discussed in the previous sections. The following two graft reactions, involving a pendent styrene unit as the anionic initiating site and metallocene terminating site, respectively, are used to illustrate the general idea. 5.1
Anionic Reactions
The process begins with a metallation reaction of DVB-containing copolymer with alkyllithium (such as n-BuLi) to form a polyole®n containing pendent benzylic anions. By limiting the alkyllithium added to the reaction to the amount required to react with all of the DVB units in the copolymer, the metallation reaction between the pendent styrene unit and the alkyllithium is quantitative. In other words, no puri®cation is needed before adding anion-polymerizable monomers to continue the living anionic graft-from polymerization process. It is very interesting to note that the anionic polymerization of various monomers, such as methyl methacrylate, can take place at room temperature without causing any detectable side reactions, which may be associated with the stable benzylic anion initiator. After achieving the desired composition of the graft copolymer, the graft reaction can be terminated by adding a proton source, such as methanol or isopropanol. Thus, by using this easily controllable living graft-from reaction technique, a variety of graft copolymer compositions with well-de®ned side chain segments, including random and block copolymers, have been produced that are all completely soluble in organic solvents. One example is the preparation of a graft copolymer having an ethylene/1-octene/ DVB elastic backbone and several polystyrene side chains. The metallation reaction was carried out by mixing (with stirring) ethylene/1-octene/DVB terpolymer in
218
Functionalization of Polyolefins
cyclohexane solution with n-BuLi/TMEDA at ambient temperature for 1 h. To determine the ef®ciency, a small portion of metallated polymer solution was reacted with Me3SiCl at room temperature. The silylated polymer was isolated by precipitation in MeOH. Repeated washing with methanol was performed before drying the resulting polymer under vacuum. Figure 11.11 (bottom) shows the 1H NMR spectra of a silylated ethylene/1-octene/DVB terpolymer. All vinyl peaks have completely disappeared and the strong peak at 0.05 ppm corresponds to the methyl proton next to Si. Both the metallation and silylation ef®ciencies were almost 100%. The major portion of the metallated polymer solution was contacted with the styrene monomers. The anionic graft reaction was then carried out at ambient temperature for a designated reaction time; isopropanol
9
8
7
6
5 ppm
4
3
2
1
0
Figure 11.11 1H NMR spectrum of a silylated ethylene/1-octene/ DVB terpolymer (bottom) and a graft copolymer with ethylene/1octene/DVB backbone and polystyrene side chains (top). Both polymers were derived from the same starting ethylene/1-octene/DVB terpolymer having a DVB content of about 4 mol%.
Synthesis of Functional Polyolefins
219
was then added to terminate the reaction. The precipitated polymer was ®ltered and then subjected to fractionation. Figure 11.11 (top) shows the 1H NMR spectra of a graft copolymer containing more than 50 mol% polystyrene side chains. In addition to the chemical shifts for ethylene/1-octene, the new peaks ( 6.4±7.3 ppm) are due to aromatic protons in the PS side chains. Overall, the graft-from reactions were very effective ± more than 80% styrene monomer conversion within 1 h. The graft content increased proportionally with increasing monomer concentration and reaction time. Since the graft-from reaction involves a living anionic polymerization, it is reasonable to assume that each benzylic lithium produces one polymer side chain and the side chains have a similar molecular weight. The graft density, de®ned as the number of grafted side chains per 1000 carbons in the polyole®n backbone, is the same as the density of the benzylic anions. The side chain length is basically proportional to the reaction time and monomer concentration. 5.2
Metallocene Reactions
In metallocene graft reaction cases, the pendent styrene units in the -ole®n/DVB copolymer serve as chain transfer agents. As discussed in Chapter 9, the combination of styrene and hydrogen is a very effective chain transfer agent during metallocene polymerizations. Similar chemistry can be directly applied to the graft reaction of an ole®n/DVB copolymer containing pendent styrene units. The graft reaction mechanism is illustrated in Eq. (11.4.):
EP
EP
EP
(II)
(I)
ⱍ ⱍ CHCH2 HCZr+A– ⱍ + CH2 CH3 ⱍ PP CHCH2Zr+A– PP (III)
H2
ⱍ CH2 ⱍ CH2
+ Z+rH A– (V)
PP (IV)
11:4:
The metallocene polymerization of propylene was carried out in the presence of EP/DVB and hydrogen. During the polymerization of propylene (with a 1,2insertion manner) the propagation Zr±C site (I) reacts with the pendent styrene unit (with a 2,1-insertion manner) in the EP/DVB copolymer to form a graft copolymer (III) having several styrene-terminated polypropylene side chains. The catalytic Zr±C site in the graft copolymer (III) becomes inactive to propylene302 due to the combination of steric hindrance between the active site (Zr±C) and the incoming monomer (propylene with 1,2-insertion), and the formation of a complex between the adjacent phenyl group and the Zr ion. On the other hand, with the presence of hydrogen, the dormant Zr±C site (III) can react with hydrogen to form EP-g-PP (IV)
220
Functionalization of Polyolefins
Table 11.7
Summary of EP-g-PP graft copolymers
Reaction conditions C3/EP-DVB (psi/g) 20/2.6 20/2.6 20/2.6 20/2.6
1
H2 (psi) 0 2 6 12
Yield Cat. activity (g) (kg PP/(molZr.h)) 2.5 4.9 6.56 7.79
0 1864 3184 4128
Graft composition
Graft copolymer
EP-DVB (wt%)
PP (wt%)
Tg( C)
Tm( C)
100 32.3 30.5 28.5
0 67.7 69.5 71.5
±48 ±47 ±47 ±46
Ð 152 152 152
1
EP-DVB: E: 52.1 mol%; P: 43.7 mol%; and DVB: 4.2 mol%. Mw 68 600 and PDI 2.
and regenerate a Z r ±HA species (V) that is capable of reinitiating the polymerization of propylene and continuing the polymerization/graft reaction cycle. This process not only produces a desirable EP-g-PP graft polymer, but also maintains high metallocene catalyst activity. The PP's molecular weight is basically proportional to the [propylene]/[styrene] ratio. Table 11.7 summarizes the results of EP-g-PP graft copolymers using a racMe2Si[2-Me-4-Ph(Ind)]2ZrCl2/MAO catalyst with the presence of propylene and EP-DVB and hydrogen. This systematic study was conducted to evaluate the effect of hydrogen on the catalyst activity and graft copolymer composition. The presence of hydrogen is clearly a key factor in the success of this graft reaction. With an appropriate [propylene]/[hydrogen] ratio, the graft reaction can be as ef®cient as the homopolymerization of propylene itself. The graft length and density can be controlled by the combination of DVB units in the EP-DVB copolymer and the propylene concentration. Some very desirable EP-g-PP graft copolymers containing continuous PP domains and discrete EP domains have been prepared with highimpact TPO properties. 6
COMPATIBILIZATION OF POLYOLEFIN BLENDS
Polyole®n graft copolymers are very effective compatibilizers for improving interactions between their corresponding polyole®ns and other materials, including polymers and substrates. Three PE and PP blends will be used to demonstrate their advantages. 6.1
PP/PMMA Blends374
Optical microscopy was used to evaluate the graft copolymers' ability to act as phase compatibilizers for the blends of i-PP and PMMA homopolymers seen in Fig. 11.12. Polymer solutions were prepared in BHT-inhibited o-xylene at 135 C. The polymer ®lms were then solution cast onto glass microscope slides. After evaporating the xylene under N2 purge, the ®lms were covered with a slide cover. The polymer ®lms were then melted in a hot stage at 180 C for 15 min, and allowed to cool quiescently
Synthesis of Functional Polyolefins
221
Figure 11.12 Optical micrographs of (top) homopolymer blend with i-PP/PMMA 70/30 and (bottom) two homopolymers with PP-g-PMMA containing 30% MMA, i-PP/PP-gPMMA/PMMA 70/10/30. (Reproduced with permission from Macromolecules 1993, 26, 3467. Copyright 1993 Am. Chem. Soc.)
in the hot stage to room temperature for 20 min. The magni®cation on the microscope was 100 times. In blend A (70/30 wt% mixture of i-PP and PMMA homopolymers that were blended in solution) two distinct phases are visible ± the crystalline PP phase and an
222
Functionalization of Polyolefins
amorphous PMMA phase. Within the PP domain, the spherulite size varies greatly with a few extremely large crystallites and predominantly small spherulites. Blend B is a sample of blend A to which 10 wt% of PP-g-PMMA has been added in solution. The added graft copolymer was also 30 wt% PMMA so as not to change the overall composition. The graft copolymer behaves as a polymeric emulsi®er, and increases the interfacial interaction between the PP amorphous region and the PMMA to reduce the domain sizes. The most noticeable change in the micrographs is the disappearance of the visibly distinct PMMA domains. The large phase separated PMMA domains are now dispersed into the interspherulite regions, and cannot be detected with the optical microscope. The mode of nucleation within the polypropylene crystalline phase has changed, as evidenced by the now relatively homogeneous spherulite size. 6.2
PP/PC Blends375
It would be extremely advantageous if inexpensive, commodity PP could be effectively compatibilized in blends with the more costly engineering resin polycarbonate. Bisphenol-A polycarbonate (PC) has excellent high- and low-temperature physical properties even up to 140 C. Using PP as the matrix material with a compatibilized polycarbonate dispersed phase could greatly improve the mechanical properties, creating a toughened plastic. Since poly(-caprolactone) (PCL) and polycarbonate (PC) form a miscible blend, the graft copolymer PP-g-PCL should behave as an emulsi®er for PP and PC blends. Blends of polypropylene and polycarbonate (Mn 24 800 g/mol and PD 2.2) were mixed in a chlorobenzene solution. Optical microscope samples were solution cast directly from the chlorobenzene solution onto a glass slide, heated in a hot stage for 10 min at 200 C and allowed to cool quiescently. A 70/30 blend of PP/PC was observed by polarized optical microscopy. Figure 11.13 (top) shows gross phase separation of the spherulitic PP and the amorphous PC phases. The PC phases vary widely in both size and shape due to the lack of interaction with the PP matrix. A blend containing 70/30/10 of PP/PC/PP-gPCL, where the graft contains 57 wt% PCL, was used to evaluate the effect of the polymeric compatibilizer on the blend's morphology. Figure 11.13 (bottom) shows the micrograph of the compatibilized blend. Only small distorted spherulites are observed, and only a few very small distinct PC phases can be found. The PP-g-PCL is clearly proven to be an effective compatibilizer for PP and PC blends. It should also be noted that the ®lms of the compatibilized blends formed in the melt press were optically clear. This is unlike pure i-PP, which forms hazy, translucent ®lms. The lack of large spherulites in both the blend and the graft must minimize scattering. 6.3
PE/PS Blends376
It is interesting to study the compatibility of PE-g-PS copolymer in HDPE and PS blends. Both a polarized optical microscope and a scanning electron microscope
Synthesis of Functional Polyolefins
223
Figure 11.13 Optical micrographs of (top) two homopolymer blends with i-PP/PC 70/30 and (bottom) two homopolymers with PP-g-PCL, i-PP/PC/PP-g-PCL 70/30/10 (100). (Reproduced with permission from Macromolecules 1994, 27, 1313. Copyright 1994 Am. Chem. Soc.)
224
Functionalization of Polyolefins
Figure 11.14 Scanning electron micrographs of the cross-section of two polymer blends: (top) two homopolymers with PE/PS 50/50 (1000) and (bottom) two homopolymers and a PE-g-PS copolymer blend with PE/PE-g-PS/PS 45/10/45 (2000). (Reproduced with permission from Macromolecules 1997, 30,1272. Copyright 1997 Am. Chem. Soc.)
Synthesis of Functional Polyolefins
225
were used to examine the surfaces and bulk morphologies, respectively. Of the two blends comprising an overall 50/50 weight ratio of PE and PS, one is a simple mixture of 50/50 between HDPE and PS and the other is a 45/45/10 weight ratio of HDPE, PS, and PE-g-PS (containing 50 mol% PS). Similar to Figs 11.12 and 11.13, the polarized optical micrographs of the two blends are very different. A gross phase separation, with spherulitic PE and amorphous PS phases, is shown in the simple PE/ PS blend. The PS phases vary widely in both size and shape due to the lack of interaction with the PE matrix. On the other hand, a continuous crystalline phase results in the compatibilized blend. Basically, the large phase-separated PS domains are now dispersed into the interspherulite regions and cannot be detected by the optical microscope. The graft copolymer behaves as a polymeric emulsi®er, and increases the interfacial interaction between the PE crystalline and the PS amorphous regions to reduce the domain sizes. Figure 11.14 shows scanning electron micrographs, operating with secondary electron imaging, revealing the surface topography of the cold-fractured ®lm edges. The ®lms were cryofractured in liquid N2 to obtain an undistorted view representative of the bulk material. In the homopolymer blend, the polymers are grossly phase separated, as can be seen in Fig. 11.14 (top). The PS component exhibits nonuniform, poorly dispersed domains and voids at the fracture surface. This ``ball and socket'' topography is indicative of poor interfacial adhesion between the PE and PS domains, and represents PS domains that are pulled out of the PE matrix. Such pull-out indicates that limited stress transfer takes place between the phases during fracture. A similar blend containing graft copolymer shows a totally different morphology in Fig. 11.14 (bottom). The material exhibits ¯at mesa-like regions similar to pure PE. No distinct PS phases are observable, indicating that fracture occurred through both phases or that the PS phase domains are too small to be observed. The PE-g-PS is clearly proven to be an effective compatibilizer in PE/PS blends.
7
SUMMARY
A versatile method has been developed to prepare functional polyole®n graft copolymers having relatively well-de®ned molecular structures. The chemistry is based on reactive polyole®n copolymers containing comonomer units, such as borane monomers, p-methylstyrene, and divinylbenzene. In general, the reactive comonomers can be very effectively incorporated into polyole®ns, such as PE, PP, and EP, with narrow molecular weight and composition distributions by using metallocene technology. In turn, the pendent reactive groups, located along the polyole®n backbone, are very effective in engaging the subsequent graft polymerizations. In borane cases, upon oxygen oxidation reaction, the borane groups were quantitatively transformed to living free radicals that then initiated the living free radical polymerization of functional monomers. The whole process
226
Functionalization of Polyolefins
can effectively take place at ambient temperature. On the other hand, the p-methylstyrene side groups can be effectively metallated to form ``stable'' benzylic anions, which then initiate living anionic polymerization of many dienes, styrene, and functional monomers. In addition to the two processes of transforming reactive groups to living free radical and anionic initiators for graft-from polymerization, it is also possible to use reactive groups as comonomers or chain transfer agents in the subsequent polymerization reaction of constructing side chains. Divinylbenzene is an example that shows good reactivity in many subsequent polymerization mechanisms. This graft technique bene®ts greatly from the simplicity of its reaction process. Many functional polyole®n graft copolymers have been prepared and reported. Most of them showed good interfacial activities and improved morphology and adhesion between microdomain structures. Improved mechanical properties of polyole®n blends and composites have also been observed.
12 New Maleic-anhydride-modi®ed and Long-chain-branched Polyole®ns
1
INTRODUCTION
By far, maleic-anhydride-modi®ed polyole®ns are the most important class of functionalized polyole®ns in current commercial applications. Due to the unique combination of the low cost of the MA reagent and functionalization process, the high activity of succinic anhydride moiety, and good processibility, they are the popular choice of material for improving the compatibility, adhesion, and paintability of polyole®ns. Among them, maleic-anhydride-modi®ed PP (PP-MA) is the most investigated polymer, and one which has found applications in many commercial products ± glass-®ber-reinforced PP,378 anticorrosive coatings for metal pipes and containers,379 metal±plastic laminates for structural use,380 multilayer sheets of paper for chemical and food packaging,381 and polymer blends,382±384 to name a handful. As discussed in Chapter 5, the current commercial MA-modi®ed polyole®n products are prepared by chemical modi®cation of the pre-formed polymers, such as PE and PP homopolymers, under free radical conditions.385,386 This postpolymerization process is usually accompanied by many undesirable side reactions,186 such as -scission, chain transfer, coupling, and uncontrollable free radical addition reactions. The resulting MA-modi®ed polyole®ns, such as PE-MA and PPMA, have very complicated molecular structures. In general, the functionalization reactions compromise with many important physical properties, including mechanical properties (due to chain scission) and processibility (due to crosslinking). It is clear that there is both scienti®c and industrial interest in developing a facile method to prepare MA-modi®ed polyole®ns with well-de®ned molecular structures. As discussed in Chapters 6±9, the combination of metallocene and reactive comonomers, i.e. borane monomers, p-methylstyrene, and dienes, produces many new polyole®n copolymers containing reactive groups. The reactive groups in the polyole®n provide the selective sites for many functionalization reactions. It is very interesting to study their applicability in the maleic anhydride modi®cation of
228
Functionalization of Polyolefins
polyole®ns, especially the formation of MA-modi®ed polymers with controllable molecular structure (polymer molecular weight, MA concentration and location, and good processibility). Although some experimental results have been discussed in previous chapters, it is very interesting to summarize and compare all experimental results from different approaches ± especially the perspectives of the new reactive copolymer route. Section 12.2 will compare the reaction mechanisms and products, especially the MA-grafted polyole®ns. Section 12.3 will show new, well-de®ned MA-terminated polyole®ns with good control of polymer molecular weight and composition. The well-de®ned molecular structures allow the detailed study of the relationship between the molecular structure of MA-modi®ed polymer and the compatibility of polyole®n blends (Section 12.4). The new maleic-anhydride-modi®ed polymers are also used to prepare polyole®n superstructures, such as long-chain-branched polyole®ns with controlled branch density and length (Section 12.5).
2
COMPARISON OF FREE RADICAL GRAFTING PROCESSES
The key step of incorporating maleic anhydride moiety to polyole®ns involves the formation of free radicals in the polyole®n chain. As illustrated in Eq. (12.1), the current commercial process is based on the hydrogen abstraction of PE and PP homopolymers using an external alkoxyl radical:
R9O*
PE
(CH29CH2)
PP
CH3 ⱍ (CH29CH)
PE
* (CH29CH)
PP
CH3 ⱍ (CH29C) *
R ⱍ (CH29CH)x9(CH29CH)y ⱍ
R ⱍ (CH29CH)x9(CH29CH)y ⱍ
ⱍ CH3
ⱍ CH2 *
R ⱍ (CH29CH)x9(CH29CH)y ⱍ (CH2)n ⱍ B
O2
R ⱍ (CH29CH)x9(CH29CH)y ⱍ (CH2)n ⱍ O * * O ⱍ B
12:1
Maleic-anhydride-modified and Long-chain-branched Polyolefins
229
Both reactions can happen randomly in any monomer units along the polymer chains. The resulting secondary and tertiary radicals are directly located on the PE and PP backbones, respectively, and both are subject to undesirable side reactions, such as the crosslinking and chain scission discussed in Chapter 5. On the other hand, the reactive polyole®n copolymers containing reactive sites, i.e. p-methylstyrene (p-MS) or borane groups, provide the selective sites for forming stable free radicals. The resulting free radicals are located in the side chains. Such a process minimizes any disturbance (physical and chemical) in the crystalline polymer backbone (in PE and PP cases) and side reactions. In fact, the reaction can effectively take place in heterogeneous conditions by suspending polymer powders in a swellable solvent, since all the reactive comonomer units are located in the amorphous phases. As discussed in Chapter 7, Section 5.2, the alkoxyl radical selectively reacts with p-methylstyrene units to form a ``stable'' benzylic radical that then initiates the subsequent grafting reaction with maleic anhydride. Many high-molecular-weight MA-modi®ed PE and PP polymers have been prepared with a controllable quantity of MA units and good (melt and solution) processibility. The location and concentration of the incorporated MA units are basically predetermined by the p-MS units. On the other hand, borane groups located along the PE and PP chains can be quantitatively auto-oxidized by oxygen at ambient temperature to form stable alkoxyl radicals that are capable of initiating living free radical polymerization. As discussed in Chapter 6, Section 5, with the coexistence of maleic anhydride and styrene, the graft-from reaction takes place to incorporate multiple MA units (as part of a styrene/maleic anhydride alternating copolymer) at each borane site. In summary, a series of new MA-modi®ed polyole®ns (illustrated below), including PE, PP, and EP, with high concentration of grafted MA units have been prepared by this new reactive copolymer approach.
Polyolefin O:
O
:O
Polyolefin-g-MA O:
O
Polyolefin-g-SMA
Polyolefin SMA
:O
SMA
SMA: Styrene/maleic anhydride alternating copolymer
It is interesting to note the combination of narrow molecular weight and composition distributions of the starting reactive copolymers and the selective MA modi®cation at the reactive group sites. The resulting MA-grafted polyole®n has relatively well-de®ned molecular structure with known backbone molecular weight and MA concentration, as well as narrow molecular weight and composition distributions.
230 3
Functionalization of Polyolefins
NEW MALEIC-ANHYDRIDE-TERMINATED POLYOLEFINS120,350
The borane approach was extended to the preparation of MA-modi®ed polyole®ns with a terminal MA and multiple MA groups. In other words, the starting polymer containing a terminal borane group was used in the free radical graft-from reaction, and the reaction affords both on MA-terminated polyole®n and polyole®n-bpoly(styrene-alt-maleic anhydride) with well-controlled molecular structures:
: :
Polyolefin
R O ⱍ CH29CH9O9 O O
Polyolefin
CH29CH9O
SMA
Equation (12.2) shows the reaction steps involved in the preparation of MAterminated PP (PP-t-MA) and polypropylene-b-poly(styrene-alt-maleic anhydride) (PP-b-SMA) copolymer:
PP
CH3 ⱍ CH9CH29B
(12-I)
O2
CH3 ⱍ CH9CH29O* *O–B
PP MA
(12-III)
12:2
S/MA :
CH3 O ⱍ CH9CH29O9 O :
PP
(12-II)
PP
CH3 ⱍ CH9CH29O
SMA (12-IV)
O
As discussed in Chapter 9, the borane-terminated PP (PP-t-B) (12-I) can be prepared by chain transfer to the B±H group during the metallocene polymerization or hydroboration reaction of the chain-end unsaturated polyole®n. The borane group at the chain end was selectively oxidized by oxygen to form a polymeric radical (12-II) that is associated with a ``stable'' borinate radical.251,252 The polymeric radical reacts in situ with maleic anhydride to produce maleicanhydride-terminated PP (PP-t-MA) (12-III) with a single MA unit. In the presence of styrene, the polymeric radical initiates a ``stable'' copolymerization of styrene and maleic anhydride with an alternating manner. The resulting PP-b-SMA diblock copolymer (12-IV) contains both PP and alternating styrene-maleic anhydride (SMA) segments.
Maleic-anhydride-modified and Long-chain-branched Polyolefins
231
Figure 12.1 compares the 1H NMR spectra of the initial chain end unsaturated PP (u-PP), the corresponding PP-2-t-MA, and the reaction product of PP-2-t-MA and 3-aminopropyltrimethoxysilane. In Fig. 12.1(a), the u-PP sample shows three multiple peaks around 0.95, 1.35, and 1.65 ppm corresponding to CH3, CH2, and CH protons, which are accompanied by two singlets at 4.7 and 4.8 ppm corresponding to two vinylidene protons at the chain end. As shown in Fig. 12.1(b), both ole®nic proton peaks are completely beyond the limits of NMR sensitivity after hydroboration and MA graft-from reactions. However, there is no chemical shift corresponding to the methylene and methine protons of the terminal MA units,
(a)
(b)
(c)
9
8
7
6
5
4
3
2
1
0
ppm
Figure 12.1 1H NMR spectra of (a) starting u-PP-2, (b) the corresponding PP-2-tMA, and (c) the reaction product of PP-2-t-MA and 3-aminopropyltrimethoxysilane. (Redrawn from Macromolecules 1999, 32, 2525. Copyright 1999 Am. Chem. Soc.)
232
Functionalization of Polyolefins
which may be due to the long relaxation time of the MA group in the hydrocarbon polymer. It is interesting to note that the MA end group in PP-t-MA can be revealed by further imidization reactions of succinic anhydride groups with a 3-aminopropyltrimethoxysilane reagent. As shown in Fig. 12.1(c), two new peaks appear at 3.48 and 3.72 ppm corresponding to Si(OCH3)3 and N±CH2. The integrated intensity ratio of the chemical shifts at 3.48 ppm and the chemical shift between 0.9 and 1.9 ppm, and the number of protons both chemical shifts represent, indicate that more than 80% of PP chains contain a terminal MA group. Apparently, both hydroboration and oxidation reactions are not inhibited by the insolubility of polypropylene, and the MA end group in PP-t-MA is very reactive with the primary amino group. It is interesting to note that the same chemical reaction is also involved in the PP/polyamide reactive blends, which will be discussed later. Figure 12.2 compares IR spectra of the initial u-PP and three corresponding PP-b-SMA copolymers that were prepared with the same reagents and different
(d)
8 7
(c)
Absorbance
6 5 4 3 (b) 2 1 0 4000
(a) 3500
3000
2500 2000 Wavenumber
1500
1000
500
Figure 12.2 The comparison of IR spectra, (a) the starting u-PP-2, (b) PP-2-bSMA-1, (c) PP-2-b-SMA-2, and (d) PP-2-b-SMA-3. (Redrawn from Macromolecules 1999, 32, 2525. Copyright 1999 Am. Chem. Soc.)
Maleic-anhydride-modified and Long-chain-branched Polyolefins
233
reaction times. After the SMA graft-from reaction, several new absorption peaks were observed at 1860 and 1780 cm 1 corresponding to two CO vibrational stretching modes in succinic anhydride, at 900±950 cm 1, corresponding to the C±H deformation in succinic anhydride, and at 700, 750, and 580 cm 1 corresponding to styrene C±H deformation. It is clear that a very high concentration of MA groups has been incorporated with the styrene units in the PP-b-SMA block copolymer, and that the molecular weight of the SMA segment is basically proportional to the reaction time. Table 12.1 summarizes the experimental results of four comparative reaction sets that include four different initial u-PP polymers with various molecular weights and melting points. In each set, the experimental results are compared under various reaction conditions. In general, the graft ef®ciency was very high, and the major portion (>80%) of u-PP was converted to PP-b-SMA diblock copolymer. The occurrence of some unreacted PP polymers (< 20%) may be due to their completely saturated molecular structure, obtained during metallocene polymerization and Table 12.1
Summary of maleic-anhydride-modified PP copolymers Reaction conditions1
Sample No. uPP-2 PP-2-t-MA PP-2-b-SMA-1 PP-2-b-SMA-2 PP-2-b-SMA-3 PP-2-b-SMA-4 PP-2-b-SMA-5 PP-2-b-SMA-62 uPP-3 PP-3-t-MA PP-3-b-SMA-1 uPP-4 PP-4-t-MA PP-4-b-SMA-1 PP-4-b-SMA-2 PP-4-b-SMA-3 PP-4-b-SMA-4 PP-4-b-SMA-5 PP-4-b-SMA-62 uPP-9 PP-9-t-MA PP-9-b-SMA-1 PP-9-b-SMA-2 1
Fractionation results
Styrene Temp./time WSMA WPP WDiblock (ml) ( C)/(h) (g) (g) (g) Ð 0 2 2 2 2 2 3 Ð 0 0.1 Ð 0 0.1 2 2 2 2 2 Ð 0 0.1 2
Ð/Ð 45/4 45/1 45/4 45/16 20/16 0/16 20/16 Ð/Ð 45/4 45/4 Ð/Ð 45/4 45/4 45/4 45/10 20/10 0/10 20/10 Ð/Ð 45/4 45/4 45/10
Ð Ð 0.124 0.350 0.548 0.137 0.076 0.070 Ð Ð Ð Ð Ð Ð 0.302 0.510 0.146 0.035 0.045 Ð Ð Ð 0.506
Ð Ð 1.053 0.113 0.106 0.109 0.108 0.112 Ð Ð Ð Ð Ð Ð 0.134 0.130 0.124 0.131 0.128 Ð Ð Ð 0.193
Ð Ð 1.053 1.426 2.083 1.622 1.485 1.432 Ð Ð Ð Ð Ð Ð 1.274 2.014 1.443 1.407 1.411 Ð Ð Ð 1.658
Product properties Mv MA Tm 104 (wt%) ( C) 3.8 3.8 3.9 Ð Ð Ð Ð Ð 1.4 1.4 1.6 5.3 5.4 5.4 Ð Ð Ð Ð Ð 16 16 16 Ð
0 0.3 5.1 16 26 20 19 18 0 0.8 1.5 0 0.2 1.1 9.6 25 16 14 14 0 0.06 1.0 20
138.7 138.2 138.0 138.9 140.3 139.1 138.6 138.2 155.2 155.6 155.4 145.4 146.8 147.2 146.5 146.1 146.3 146.5 146.7 162.3 162.6 163.4 162.6
Tg ( C) Ð Ð Ð 216 213 214 213 213 Ð Ð Ð Ð Ð Ð 213 212 213 213 213 Ð Ð Ð 216
1 g borane-terminated PP, 2 g maleic anhydride, 20 ml benzene. 0.01 g pentaerythritol tetrakis(3,5-di-tert-butyl-4-hydroxyhydrocinnamate) inhibitor added to the reaction solution. Reproduced with permission from Macromolecules 1999, 32, 2525. Copyright 1999 Am. Chem. Soc. 2
234
Functionalization of Polyolefins
thermal degradation reactions. This observation is quite consistent with the previous 1 H NMR results of the imidized PP-t-MA, which also indicated about 80% of the initial u-PP chains contained a vinylidene group. It is very interesting to note that the undesirable side reaction of producing the ungrafted SMA polymer can be minimized by lowering the reaction temperature (in the PP-2-b-SMA-5 and PP-4-bSMA-5 cases) or adding some free-radical inhibitors (in the PP-2-b-SMA-6 and PP-4-b-SMA-6 cases). Apparently, the active site responsible for the ungrafted SMA polymers is basically a traditional free radical initiation site that is sensitive to temperature and inhibitors, and the one responsible for the graft-from reaction is a ``stable'' initiation site (II) (shown in Eq. 12.2), that is insensitive to the reaction temperature and inhibitor. In detail, more than 80% of the polymer chains in all four PP-t-MA copolymers contain a terminal MA unit, and there is about 0.3, 0.8, 0.2, and 0.06 wt% MA in PP-2-t-MA, PP-3-t-MA, PP-4-t-MA, and PP-9-t-MA, respectively. With the addition of styrene, the incorporation of MA units dramatically increases, and the concentration is dependent on the styrene concentration (comparing samples PP-4b-SMA-1 and PP-4-b-SMA-2) and reaction time (comparing samples PP-4-b-SMA2 and PP-4-b-SMA-3). In the PP-4-b-SMA-3 sample, 25 wt% MA incorporation indicates about 50 wt% SMA, assuming that an alternating copolymer segment existed in this diblock copolymer. In other words, the molecular weight of the SMA segment is about 5.3 104 g/mol, similar to that of the PP segment. Similar results were observed in other reaction sets. Despite the initial very high molecular weight (1.63 105 g/mol) of the u-PP-9 polymer, and thus the extremely low concentration of reactive sites, the chain extension took place and the amount of incorporated MA units dramatically increased. Sample PP-9-b-SMA-2 contains a high concentration (20 wt%) of MA units with an estimated molecular weight of > 2.6 105 g/mol, assuming an alternating SMA segment in the diblock copolymer. With an increasing SMA concentration in the PP structure, the solubility of the polymer is gradually reduced. Only the samples containing MA units of <1.5 wt% are soluble in decalin for molecular weight determination. However, all the samples are melt processible at 260 C. It is very interesting to note that all measured PP-tMA and PP-b-SMA samples (soluble in decalin) show molecular weights similar to the initial u-PP, indicating that no detectable degradation or crosslinking reactions happened during the MA grafting reactions. This is a great departure from the current maleic-anhydride-modi®ed PP technology. Figure 12.3 compares three DSC curves of uPP-2, PP-2-b-SMA-1, PP-2-b-SMA2, and PP-2-b-SMA-3 in the ®rst reaction set discussed in Table 12.1. An almost identical melting peak near 140 C corresponding to Tm of the PP polymer was observed in all samples. A clear new Tg at about 216 C corresponding to the alternating SMA copolymer was shown in the PP-2-b-SMA-3 sample. Two thermal transitions almost identical to those of the two corresponding PP and SMA polymers indicate a phase separation between the PP and SMA domains. Both polymer segments in the diblock copolymer must have long consecutive (undisturbed) sequences to form separate domains.
Heat flow (mV)
Maleic-anhydride-modified and Long-chain-branched Polyolefins
80 75 70 65 60 55 50 45 40 35 30 25 20 15 10
235
(d) (c) (b) (a)
50
75
100
125 150 175 Temperature (°C)
200
225
250
Figure 12.3 Comparison of three DSC curves of (a) u-PP-2, (b) PP-2-bSMA-1, (c) PP-2-b-SMA-2, and (d) PP-2-b-SMA-3. (Redrawn from Macromolecules 1999, 32, 2525. Copyright 1999 Am. Chem. Soc.)
4
APPLICATIONS IN PP/POLYAMIDE REACTIVE BLENDS120
These PP-t-MA and PP-b-SMA copolymers with relatively well-de®ned structures were used as the compatibilizers in the reactive PP/polyamide blends containing PP (Mw 288 000 and Mn 43 300 g/mol) and nylon 11 (Mn 24 800 g/mol). The reactive mixing was carried out in a Brabender mixer at 220 C for 20 min. Ideally, the PP-t-MA and PP-b-SMA copolymers would react with nylon 11 in situ right at the interfaces between the PP and nylon 11 domains and produce a PP-b-nylon 11 block copolymer and a PP-g-nylon 11 graft copolymer with ``brush-like'' microstructure, respectively, as illustrated in Eq. (12.3): PP-t-MA
PP-b-SMA
PP
CH3 ⱍ CH9CH29O O:
O
PP
:O
CH3 ⱍ CH9CH29O
O: O
CH29CH :O
Nylon 11
PP
PP
Nylon
Diblock copolymer
Brush-like copolymer
Nylon side chains
x
12:3
236
Functionalization of Polyolefins
The polymer blend's morphology was examined by observing the surface topography of cold-fractured ®lm edges using scanning electron microscopy. Figure 12.4 shows scanning electron micrographs of two homopolymer blends, including a PP-rich blend (PP/nylon 11 75/25 wt%) and a nylon 11-rich blend (PP/nylon 11 25/75 wt%). In both homopolymer blends, the polymers are grossly phase separated. This ``ball and socket'' topography is indicative of poor interfacial adhesion between the PP and nylon domains, and represents minor component domains that are pulled out of the major component matrix. Such pull-out indicates that no stress transfer takes place between the phases during fracture. (a)
(b)
Figure 12.4 Scanning electron micrographs of two homopolymer blends, including (a) a PP-rich blend (PP/nylon 11 75/25 wt%) and (b) a nylon-rich blend (PP/nylon 11 25/75 wt%). (Reproduced with permission from Macromolecules 1999, 32, 2525. Copyright 1999 Am. Chem. Soc.)
Maleic-anhydride-modified and Long-chain-branched Polyolefins
237
The introduction of the PP-t-MA compatibilizer containing only one reactive MA unit at the chain end into the PP/nylon 11 blend is expected to in situ form a PP-bnylon 11 diblock copolymer (as discussed in Eq. 12.3) which serves as a compatibilizer at the PP/nylon interfaces. Figures 12.5 and 12.6 compare two sets of ternary PP/PP-t-MA/nylon 11 blends, including a PP-rich system (PP/PP-t-MA/ nylon 11 70/5/25 wt%) and a nylon-rich system (PP/PP-t-MA/nylon 11 20/5/75 wt%). With 5 wt % PP-t-MA copolymer, the compatibility of the PP/nylon 11 blends is generally improved. Unexpectedly, both systems show that the compatibility is inversely proportional to the PP molecular weight of the PP-t-MA copolymer. In Figs 12.5(a) and 12.6(a), the PP-3-t-MA copolymer having a low molecular weight (Mw 15 000 and Mn 6000 g/mol) and Tm ( ~ 155 C) shows a quite effective compatibility in both PP-rich and nylon 11-rich blends. The results imply effective cocrystallization between the high-molecular-weight PP (Tm ~ 163 C , Mw 288 000, and Mn 43 300) and the low-molecular-weight PP segment of PP-t-MA. A certain MA concentration is necessary to secure the PP-tMA and nylon 11 in situ reaction, and to form a suf®cient quantity of PP-b-nylon 11 diblock copolymer to cover all the interfaces. The side-by-side comparisons between Figs 12.5(b) and 12.6(b) and between Figs 12.5(c) and 12.6(c) show signi®cantly better compatibility of PP-t-MA in the PPrich blends than in the corresponding nylon 11-rich blends. The clear pulled-out ``ball and socket'' topography in Figs 12.6(b) and 12.6(c) is indicative of poor interfacial adhesion between the PP discrete phases and the nylon 11 matrix. Some of the PP-b-nylon 11 diblock copolymers formed in situ in the nylon-rich system may have the tendency to form separate domains within the nylon matrix, which may signi®cantly reduce the effectiveness of the PP-t-MA copolymer. The same tendency was very clearly observed in some PP-b-SMA cases containing multiple MA units, which will be discussed below. It is very interesting to study the PP-b-SMA copolymer in the PP/nylon blend. The copolymer contains multiple reactive MA units at the PP chain end, and will form in situ the compatibilizer of the PP-g-nylon graft copolymer with a ``brushlike'' structure (as shown in Eq. 12.3). Figures 12.7 and 12.8 compare scanning electron micrographs of both the PP-rich system (with PP/PP-b-SMA/nylon 11 70/5/25 wt%) and the nylon 11-rich system (with PP/PP-b-SMA/nylon 11 20/5/75 wt%), respectively. The completely opposite results seen in these two systems was very unexpected. The PP-b-SMA copolymer shows almost no appreciable bene®t for the nylon 11rich blends. On the other hand, PP-b-SMA shows very good compatibility with the PP-rich blends ± and the higher the molecular weight, the better the compatibility. It is logical to expect that increasing the MA units in the PP-b-SMA copolymer would increase the graft density of the nylon 11 side chains in the PP-g-nylon 11 compatibilizer. The multiple nylon 11 chains in the PP-g-nylon 11 may increase their hydrogen-bonding interactions with the nylon 11 homopolymer, particularly in the nylon-rich blend environment. The surge of hydrogen bonding energy may overwhelm the cocrystallization energy between the PP-g-nylon 11 and the PP.
238
Functionalization of Polyolefins
(a)
(a)
(b)
(b)
(c)
(c)
Figure 12.5 Comparison of three ternary PP/PP-t-MA/nylon 11 70/5/25 wt% blends: (a) PP/PP-3-t-MA/nylon 11; (b) PP/PP-t-MA/ nylon 11; and (c) PP/PP-9-t-MA/nylon 11. (1500.) (Reproduced with permission from Macromolecules 1999, 32, 2525. Copyright 1999 Am. Chem. Soc.)
Figure 12.6 Comparison of three ternary PP/PP-t-MA/nylon 11 20/5/75 wt% blends: (a) PP/PP-3-t-MA/nylon 11; (b) PP/PP-4-tMA/nylon 11; and (c) PP/PP-9-t-MA/nylon 11. (1500.) (Reproduced with permission from Macromolecules 1999, 32, 2525. Copyright 1999 Am. Chem. Soc.)
Such a thermodynamic imbalance at two interfaces, i.e. nylon/PP-g-nylon and PP-gnylon/PP, may increase the tendency of PP-g-nylon to form discrete domains by itself in the continuous nylon 11 matrix. Therefore, the compatibility is largely lost (as shown in Fig. 12.8).
Maleic-anhydride-modified and Long-chain-branched Polyolefins
239
(a)
(a)
(b)
(b)
(c)
(c)
Figure 12.7 Comparison of three ternary PP/PP-b-SMA/nylon 11 70/5/25 wt% blends, including (a) PP/PP-3-b-SMA-1/nylon 11; (b) PP/PP-4-b-SMA-1/nylon 11, and (c) PP/PP-9-b-SMA-1/nylon 11. ( 1500.) (Reproduced with permission from Macromolecules 1999, 32, 2525. Copyright 1999 Am. Chem. Soc.)
Figure 12.8 Comparison of three ternary PP/PP-b-SMA/nylon 11 20/5/75 wt% blends, including (a) PP/PP-3-b-SMA-1/nylon 11; (b) PP/PP-4-b-SMA-1/nylon 11, and (c) PP/PP-9-b-SMA-1/nylon 11. ( 1500.) (Reproduced with permission from Macromolecules 1999, 32, 2525. Copyright 1999 Am. Chem. Soc.)
On the other hand, in the PP-rich system (Fig. 12.7) the formed ``brush-like'' PPg-nylon 11 compatibilizer may mostly stay at the interfaces due to the balanced energy between the effective cocrystallization of the PP (continuous major phases) and the PP segment of PP-g-nylon 11, and the interactions of nylon 11 (discrete
240
Functionalization of Polyolefins
minor phases) and the multiple nylon 11 side chains of the PP-g-nylon 11. This phenomenon is particularly clear in the ternary blends involving PP-9-b-SMA with high PP molecular weight. Figure 12.8(c) shows the ``ball and socket'' topography in the nylon 11-rich blend. On the other hand, Figure 12.7(c) displays no distinct globules in the corresponding PP-rich blend. Rather, a ¯at, mesa-like fracture surface is observed similar to that seen on a fractured PP homopolymer. Both the PP and nylon phases must be mechanically tied. It is very interesting to compare Figs 12.7(c) and 12.5(c). Both have similar ternary PP-rich blends and the same high molecular weight of the PP segment in the PP-9-t-MA and PP-9-b-SMA-1 compatibilizer structures. The only difference is the number of MA units (1 vs ~10 MA units) located at the PP-chain end. Comparison of the two micrographs clearly shows the advantage of high MA content in a highmolecular-weight PP-9-b-SMA-1 compatibilizer that can form strong interactions with both PP continuous and nylon 11 discrete phases. The high overall cocrystallization energy (due to the high PP molecular weight) may balance out with the mutiple interactions with nylon 11. Therefore, two strong and balanced PP/PP-9-bSMA-1 and PP-9-b-SMA-1/nylon 11 interactions secure the PP and nylon 11 interfaces.
5
APPLICATIONS IN LONG-CHAIN-BRANCHED POLYOLEFINS378
One of the most distinctive developments in metallocene catalysis is the preparation of polyethylene with a long-chain branching molecular structure379 by using a catalyst with constrained ligand geometry. Although only a very low concentration of long-chain-branched polymer is formed, and there are dif®culties in determining the details of the molecular structure, the long-chain-branched polyethylene (LCBPE) that exists in the PE product shows many unique rheological properties.380±382 These include higher viscosity at lower shear rates and lower viscosity at higher shear rates. The shear thinning is very important for improving the processibility of metallocene-produced polyethylene with a narrow molecular weight distribution. In addition, LCBPE polymers also show signi®cantly higher melt strength, which is essential in blow molding processes.383,384 So far, there is no example of long-chain-branched polypropylene (LCBPP) prepared by in situ polymerization. The polypropylene (PP) products prepared by both Ziegler±Natta and metallocene catalysts that are currently available have a predominantly linear structure. Although linear PP polymers have many desirable physical properties, they show low melt strength. As a result, PP has been limited in some end-use fabrications ± for example, extrusion coating and blow molding. The processibility is also a major concern in metallocene-prepared PP with narrow molecular weight distribution. To date, many functional polyole®ns with various functional groups located either in the side chains or chain end have been prepared. It is possible to use these
Maleic-anhydride-modified and Long-chain-branched Polyolefins
241
functional polyole®ns as building blocks to construct supermolecular structures, including long-chain-branched polymers. In this section, a new method will be discussed in the preparation of long-chain-branched PP with a relatively wellde®ned molecular structure, i.e. known backbone molecular weight, graft length, and graft density. The chemistry involves a graft-onto reaction between a maleicanhydride-grafted PP (PP-g-MA) (I) and several amine-group-terminated PP (PP-t-NH2) (II) chains, as shown in Eq. (12.4):
ⱍ : O: O O
ⱍ : O: O O
PP
ⱍ : O: N O
ⱍ : O: N O
PP
(I) PP
+ n
12:4
PP
–H2O
NH2
PP (II)
(III)
The formed LCBPP (III) has an imide linkage that connects the PP backbone and each PP side chain. Since the molecular structures of both the backbone and side chains are predetermined before the graft-onto reaction, the extent of the reaction can be controlled by the MA concentration in the PP-g-MA and the ratio between PP-g-MA and PP-t-NH2. An effective imidization reaction will produce a longchain-branched PP polymer (III) with a relatively well-de®ned molecular structure. Maleic-anhydride-terminated and grafted PP polymers, i.e. PP-t-MA and PP-gMA, respectively, were synthesized from the corresponding borane-containing polymers that were discussed in the previous section and Chapter 6, Section 5, respectively. Amine-terminated PP (PP-t-NH2) polymers can be prepared by two methods, as shown in Eq. (12.5):
+ LiMe2BH2
PP
Me3SiCl
H2NOSO3H
PP NH2
Chain-end unsaturated PP
12:5 PP
O O + H2N9(CH2)29NH2 O
PP
O N9(CH2)29NH2 O
The ®rst method involved hydroboration of chain-end unsaturated PP with dimethylborane385 that was prepared in situ from lithium dimethylboronhydride, then amination using hydroxyamine-o-sulfonic acid. The second method is to react MA terminated PP with 10-fold excess ethylene diamine at 130 C for 6 h. The
242
Functionalization of Polyolefins
unreacted diamine was completely removed after the imidization reaction. The graft-onto reaction was carried out by intensive mixing between MA- and NH2modi®ed PP polymers in xylene solution at 130 C for 5 h. The ®nal product was precipitated in acetone, washed, and dried at 50 C in a vacuum oven for 24 h. To determine the effectiveness of the imidization reaction, a maleic-anhydrideterminated PP (PP-t-MA) having Mw ~ 52 000 g/mol and Mw/Mn 2.6 was reacted with a NH2-terminated PP (PP-t-NH2) having Mw ~ 12 000 g/mol and Mw/Mn 2.2 with 1/1 mole ratio of MA/NH2 units in a xylene solution at 130 C for 5 h. The molecular weight change in the resulting diblock polymer offers very useful information about this coupling reaction. Figure 12.9 compares the GPC curves of the starting PP-t-MA and the resulting diblock PP containing both PP blocks. The homogeneous increase in the polymer's molecular weight with no signi®cant broadening of the molecular weight distribution strongly suggests the formation of a PP-b-PP diblock polymer, and indicates an effective imidization coupling reaction. The same coupling reaction was applied to the graft-onto reaction between PP64kg-MA (Mw ~ 64 000 g/mol) and PP12k-t-NH2 (Mw ~ 12 000 g/mol) with 1/1 mole ratio of MA/NH2 to produce the LCBPP polymer (III) containing an average of one PP backbone and six PP side chains. Figure 12.10 compares the IR spectra of PP-g-MA, a LCBPP sample formed at 130 C for 5 h, and the further heat-treated LCBPP ®lm at 130 C in a vacuum oven for 24 h (sample PP64k-g-PP12k-2 in Table 12.2). In Fig. 12.10(a), the starting PP-g-MA shows two absorbency peaks at 1860 and 1780 cm 1 corresponding to the symmetric and asymmetric stretching modes of the two carbonyl groups in succinic anhydride. After reacting with PP(12k)-t-NH2, Fig. 12.10(b) shows the disappearance of both succinic anhydride absorbencies and (a)
mV
(b)
20
24
28
32 Minutes
36
40
44
Figure 12.9 GPC curves of (a) PP-t-MA containing a terminal succinic anhydride group and (b) PP-b-PP diblock polymer formed by imidization coupling reaction. (Solvent: trichlorobenzene; temp.: 135 C.) (Redrawn from Macromolecules 1999, 32, 8678. Copyright 1999 Am. Chem. Soc.)
Absorbance
Maleic-anhydride-modified and Long-chain-branched Polyolefins
243
(c)
(b)
(a) 2000
1950
1900
1850
1800 1750 Wavenumber
1700
1650
1600
1550
Figure 12.10 FTIR spectrum of (a) the starting PP-g-MA, (b) LCBPP formed by reacting PP-g-MA with PP-t-NH2 in xylene solution at 130 C for 5 h, and (c) LCBPP formed after further coupling reaction in a vacuum oven at 130 C for 24 h. (Redrawn from Macromolecules 1999, 32, 8678. Copyright 1999 Am. Chem. Soc.)
the appearance of three new absorbencies, including a weak peak at 1770 and two strong peaks at 1710 and 1630 cm 1 corresponding to the carbonyl vibration modes of imide, carboxylic acid, and amide groups, respectively. It appears that the succinic anhydride groups in PP-g-MA react with the NH2 groups in terminal PP to form a product containing some imide groups, but mainly amide and carboxylic acid groups. After further heat treatment, Fig. 12.10(c) shows a decrease in the sample's absorbency at 1630 cm 1 and an increase in two strong peaks at 1770 and 1730 cm 1, corresponding to the imide group. The cyclization reaction took place to form imide groups as the linkages between the PP backbone and PP side chains. Figure 12.11 compares GPC curves for the LCBPP polymer (PP64k-g-PP12k-2) and the corresponding physical mixture (PP64k/PP12k) with identical mole ratios of backbone PP64k and branch PP12k. As expected, the physical mixture shows a bimodal and broad molecular weight distribution. On the other hand, the LCBPP shows a single GPC peak with a relatively narrow molecular weight distribution. It is very interesting to note that the GPC peak of the branched PP64k-g-PP12k polymer (with an expected molecular weight of 138k) is located between two peaks of the much lower molecular weight backbone PP64k and branch PP12k polymers. The results clearly imply the formation of long-chain-branched PP polymers. The long-chain branching structure reduces the radius of gyration and overall hydrodynamic volume of the polymer. The reduction in the hydrodynamic volume of LCBPP polymers was further supported by rheological study. Table 12.2 summarizes the intrinsic viscosity results of long-chain-branched PP polymers with various graft lengths and densities. It is also very interesting to compare their results with the corresponding physical mixtures.
244
Functionalization of Polyolefins
(b)
mV
(a)
20
24
28
32 Minutes
36
40
44
Figure 12.11 GPC curve comparison between (a) a physical mixture of the PP backbone and side chains and (b) the corresponding LCBPP polymer. (Redrawn from Macromolecules 1999, 32, 8678. Copyright 1999 Am. Chem. Soc.)
Table 12.2 Summary of long-chain-branched PP polymers and their corresponding mixtures Samples
Backbone (wt%)
Branch (wt%)
Graft density1
Mw (103)
[] (dl/g)2
PP64k-g-MA PP52k-t-NH2 PP23k-t-NH2 PP12k-t-NH2 PP64k/PP12k-1 PP64k-g-PP12k-1 PP64k/PP12k-2 PP64k-g-PP12k-2 PP64k/PP23k PP64k-g-PP23k PP64k/PP52k PP64k-g-PP52k
Ð Ð Ð Ð 73 73 47 47 32 32 29 29
Ð Ð Ð Ð 27 27 53 53 68 68 71 71
Ð Ð Ð Ð Ð 2 Ð 6 Ð 6 Ð 3
64 52 23 12 Ð 136 Ð 88 Ð 202 Ð 220
0.68 0.62 0.32 0.24 0.62 0.42 0.48 0.37 0.45 0.31 0.66 0.39
hr2i1/2 (nm)3
g 04
27.5 24.8 15.2 11.1 Ð 26.1 Ð 28.8 Ð 31.1 Ð 34.4
Ð Ð Ð Ð Ð 0.48 Ð 0.31 Ð 0.20 Ð 0.21
1 Graft density is the average number of PP grafts in each PP backbone, assuming 100% coupling reaction. 2 Intrinsic viscosity is measured in decahydronaphthalene at 135 C. 3 2 1/2 hr i (M []/)1/3, where M is molecular weight, is Flory constant (2.1 1021 (dl/g) cm3). 4 0 g []b/[]l; []l KM, where K 1.05 10 4 dl/g, 0.80, M Mw(backbone) Mw (branches). (Reproduced with permission from Macromolecules 1999, 32, 8678. Copyright 1999 Am. Chem. Soc.)
Maleic-anhydride-modified and Long-chain-branched Polyolefins
245
All the long-chain-branched PP polymers show signi®cantly lower intrinsic viscosity than their corresponding physical mixtures with identical compositions. The intrinsic viscosity of the physical mixture is roughly equal to the sum of the weight fraction of the intrinsic viscosity of each of the components. It is very interesting to note that the intrinsic viscosity of the LCBPP polymers shown in Table 12.2 systematically decreases with increases in both graft density and graft length. Comparing the samples of PP-64k, PP64k-g-PP12k-1, and PP64k-gPP12k-2 having the same backbone and graft length but increasing graft density (0, 2, 6 grafts/backbone, respectively), the intrinsic viscosity decreases from 0.68, 0.42 to 0.37 dl/g. Comparing the samples of PP-64k, PP64k-g-PP12k-2, and PP64k-gPP23k, having the same polymer backbone and graft density (6 grafts/backbone) but the increase of graft length (0k, 12k, and 23k g/mol, respectively), the intrinsic viscosity decreases from 0.68 to 0.37 and 0.31 dl/g. Zimm and Kilb introduced a branching parameter, g 0 , which is de®ned as g 0 []b/[]l, where []b and []l are the intrinsic viscosity of branched and linear polymers with the same molecular weight. Thus, all LCBPP polymers produced have a g 0 <<1, and the g 0 value proportionally decreases with increases in both graft density and graft length.
6
SUMMARY
The reactive copolymer approach clearly offers a facile and clean method to prepare maleic-anhydride-modi®ed polyole®ns with well-de®ned molecular structure. The chemistry is tremendously bene®ted by metallocene technology that results in reactive copolymers with narrow molecular weight and composition distributions, as well as controlling the concentration and location of reactive comonomer incorporation. The reactive groups in the copolymer provide the selective sites for MA modi®cation reaction. Such a reaction minimizes the undesirable side reaction in the polyole®n backbone and preserves its important properties. It is very exciting to discuss the well-de®ned MA-terminated polyole®ns, especially PP with controllable PP molecular weight and MA units at the polymer chain end. The new PP-MA block copolymers provide detailed information about the effects of PP-MA structure to the compatibility of the reactive PP/nylon blends. It is very unexpected that the PP-b-SMA copolymer, with its high-molecular-weight PP segment and multiple MA units, shows excellent compatibility in PP-rich blends and almost no detectable bene®t to nylon-rich blends. The new MA-modi®ed polymers with various molecular structures also offer an excellent opportunity to prepare other segmental polymers. An example presented is the new long-chainbranched polypropylene (LCBPP), which has a relatively well-de®ned molecular structure, i.e. predetermined PP backbone and side chains, as well as graft density. The long-chain branching structure is strongly supported by the formation of imide linkages and the reduction of overall hydrodynamic volume.
This Page Intentionally Left Blank
References
1. 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.
W.J. Storck, Chemical Eng. News, 78(26), 9 (2000). C. Vasile and R.B. Seymour, Handbook of Polyole®ns, Marcel Dekker (1993). D.C. Prevorsek, Polym. Mater. Encyclop., 11, 8411 (1996). G. VerStrate, Encyclop. Polym. Sci. Eng., 6, 522 (1986). B.D. Perlson and C.C. Schababerle, Plastics Recycling; Products and Process, (R.J. Ehrig, ed.), Hanser Publishers, Ch. 4 (1992). S.L. Aggarwal and O.J. Sweeting, Chem. Rev., 57, 665 (1957). K. Ziegler, E. Holzkamp, H. Martin, and H. Breil, Angew. Chem., 67, 541 (1955). R.J. Beam, Handbook of Plastic Materials and Technology (I.I. Rubin, ed.), John Wiley Ch. 30 (1990). G. Natta, J. Polym. Sci., 16, 143 (1955). A. Andresen, H.G. Cordes, J. Herwig, W. Kaminsky, A. Merk, R. Mottweiler, J. Pein, H. Sinn, and H.J. Vollmer, Angew. Chem. Int. Ed. Engl., 15, 630 (1976). H. Sinn and W. Kaminsky, Adv. Organomet. Chem., 18, 99 (1980). F.R.W.P. Wild, L. Zsolnai, G. Huttner, and H.H. Brintzinger, J. Organomet. Chem., 232, 233 (1982). F.R.W.P. Wild, M. Wasiucionek, G. Huttner, and H.H. Brintzinger, J. Organomet. Chem., 288, 63 (1985). J.A. Ewen, J. Am. Chem. Soc., 106, 6355 (1984). W. Kaminsky, K. Kulper, H.H. Brintzinger, and F.R.W.P. Wild, Angew. Chem. Int. Ed. Engl., 24, 507 (1985). V.K. Gupta, S. Satish, and I.S. Bhardwaj, J. Macromol. Sci. Rev. Macromol. Chem. Phys., C34(3), 439 (1994). A.E. Hamielec and J.B.P. Soares, Prog. Polym. Sci., 21, 651 (1996). E. Albizzati, U. Giannini, G. Collina, L. Noristi, and L. Resconi, Polypropylene Handbook, (E.P. Moore, Jr., ed.), Hanser Publishers, Ch. 2 (1996). W. Kaminsky and M. Arndt, Adv. Polym. Sci., 127 144 (1997). M. Peuckert and W. Keim, Organometallics, 2, 594 (1983). V.M. Mohring and G. Fink, Angew. Chem. Int. Ed. Engl., 24(11), 1001 (1985). U. Klabunde and S.D. Ittel, J. Mol. Catal., 41, 123 (1987). M. Brookhart, A.F. Volpe, Jr., and D.M. Lincoln, J. Am. Chem. Soc., 112, 5634 (1990). L.K. Johnson, C.M. Killian, and M. Brookhart, J. Am. Chem. Soc., 117, 6414 (1995). L.K. Johnson, S. Mecking, and M. Brookhart, J. Am. Chem. Soc., 118, 267 (1996). J. Boor, Jr., Ziegler±Natta Catalysts and Polymerizations, Academic Press (1979). P. Cossee, Tetrahedron Lett., 17, 12 (1960). P. Cossee, J. Catal., 3, 80 (1964). R.H. Grubbs, J. Am. Chem. Soc., 107, 3377 (1985).
248 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 41. 42. 43. 44. 45. 46. 47. 48. 49. 50. 51. 52. 53. 54. 55. 56. 57. 58. 59. 60. 61. 62. 63. 64. 65. 66.
References R.F. Jordan, J. Chem. Edu., 65(4), 285 (1988). P.G. Gassmann and M.R. Callstorm, J. Am. Chem. Soc., 108, 1718 (1986). J.J. Eisch, S.I. Pombrick, and G.X. Zheng, Organometallics, 12, 3856 (1993). X. Yang, L. Stern, and T.J. Marks, J. Am. Chem. Soc., 116, 10015 (1994). J.J. Eshuis, Y.Y. Tan, A. Meetsma, and J.H. Teuben, Organometallics, 11, 362 (1992). G. Guerra, P. Longo, L. Cavallo, P. Corradini, and L. Resconi, J. Am. Chem. Soc., 119, 4394 (1997). J.A. Ewen, M.J. Elder, R.L. Jones, L. Haspeslagh, J.L. Atwood, S.G. Bott, and K. Robinson, Makromol. Chem., Macromol. Symp., 48/49, 253 (1991). J.A. Ewen, R.L. Jones, A. Razavi, and J. Ferrara, J. Am. Chem. Soc., 110, 6255 (1988). M. Farina, G. Disilvestro, and P. Sozzani, Macromolecules, 26, 946 (1993). N. Herfert and G. Fink, Makromol Chem., Macromol. Symp., 66, 157 (1993). G.H. Llinas, S.H. Dong, D.T. Mallin, M.D. Rausch, Y.G. Lin, H.H. Winter, and J.C.W. Chien, Macromolecules, 25, 1242 (1992). W.J. Gauthier and S. Collins, Macromol. Symp., 98, 223 (1995). A. Razavi, L. Nafpliotis, D. Vereecke, K. DenDauw, J.T. Atwood, and U. Thewald, Macromol. Symp., 89, 345 (1995). J.A. Ewen, Macromol. Symp., 89, 181 (1995). G.W. Coates and R.M. Waymouth, Science, 267, 217 (1995). W. Kaminsky, A. Ahler, and N. Moller-Linderdorf, Angew. Chem. Int. Ed. Engl., 28, 1216 (1989). J.C.W. Chien and B.P. Wang, J. Polym. Sci., Polym. Chem., 28, 15 (1990). U. Stehling, J. Diebold, R. Kirsten, W. Roll, H.H. Brintzinger, S. Jungling, R. Mulhaupt, and F. Langhauser, Organometallics, 13, 964 (1994). S. Jungling, R. Mulhaupt, U. Stehling, H.H. Brintzinger, D. Fischer, and F. Langhauser, Macromol. Symp., 97, 205 (1995). L. Resconi, A. Fait, F. Piemontesi, M. Colonnesi, H. Rychlicki, and R. Ziegler, Macromolecules, 28, 6667 (1995). L. Resconi, F. Piemontesi, G. Franciscono, L. Abis, and T. Fiorani, J. Am. Chem. Soc., 114, 1025 (1992). W. Spaleck, A. Antberg, J. Rohrmann, A. Winter, B. Bachmann, P. Kiprof, J. Behn, and W.A. Herrmann, Angew. Chem. Int. Ed. Eng., 31, 1347 (1992). R. Mulhaupt, T. Duschek, and B. Riecer, Makromol. Chem., Macromol. Symp., 48/ 49, 317 (1991). A. Grassi, A. Zambelli, L. Resconi, E. Albizzati, and R. Mazzocchi, Macromolecules, 21, 617 (1988). T. Tsutsui, A. Mizuno, and N. Kashiwa, Polymer, 30, 428 (1989). T. Shiono, H. Kurosawa, O. Ishida, and K. Soga, Macromolecules, 26, 2085 (1993). G. Natta, G. Mazzanti, P. Longi, and F. Bernardini, Cheimica Industria, 40, 813 (1958). G. Natta, G. Mazzanti, P. Longi, and F. Bernardini, J. Polym. Sci., 31, 181 (1958). N.S. Nametkin, A.V. Topchiev, S.G. Dourgarian, and I.M. Tolchinski, Vysokomol. Soedin., 1(11), 1739 (1959). U. Giannini, G. Bruckner, E. Pellino, and A. Cassata, Polym. Lett., 5, 527 (1967). A. Carbonaro, A. Greco, and I.W. Bassi, Eur. Polym. J., 4, 445 (1968). J. Heller, D.O. Tieszen, and J. Parkinson, J. Polym. Sci., A1, 125 (1963). E.A. Hopkins and M.L Miller, Polymer, 4, 75 (1963). K.J. Clark and T. Powell, Polymer, 6, 531 (1965). R. Backsai, J. Polym. Sci., A3, 2491 (1965). T.C. Chung, Macromolecules, 21, 865 (1988). M.D. Purgett and O. Vogl, J. Polym. Sci. A: Polym. Chem., 26, 677 (1988).
References 67. 68. 69. 70. 71. 72. 73. 74. 75. 76. 77. 78. 79. 80. 81. 82. 83. 84. 85. 86. 87. 88. 89. 90. 91. 92. 93. 94. 95. 96. 97. 98. 99. 100. 101. 102. 103. 104. 105.
249
M.D. Purgett and O. Vogl, J. Polym. Sci. A: Polym. Chem., 27, 2051 (1989). T.C. Chung, U.S. Patent 4,734,472 (1989). T.C. Chung, U.S. Patent 4,812,529 (1989). S. Ramakrishnan, E. Berluche, and T.C. Chung, Macromolecules, 23, 378 (1990). A.W. Langer and R.R. Haynes, U.S. Patent 3,755,279 (1973). M.R. Kesti, G.W. Coates, and R.M. Waymouth, J. Am. Chem. Soc., 114, 9679 (1992). C.E. Wilen, M. Auer, and J.H Nasman, J. Polym. Sci. A: Polym. Chem., 30, 1163 (1992). G. Bertolini, L. Lanzini, L. Marchese, A. Roggero, A. Lezzi, and S. Costanzi, J. Polym. Sci. A: Polym. Chem., 32, 961 (1994). M.J. Schneider, R. Schafer, and R. Mulhaupt, Polymer, 38, 2455 (1997). U.M. Stehling, K.M. Stein, M.R. Kesti, and R.M. Waymouth, Macromolecules, 31, 2019 (1998). K.J. Clark and W.G. City, U.S. Patent 3,949,277 (1970). D.N. Schulz, K. Kitano, T.J. Burkhardt, and A.W. Langer, U.S. Patent 4,518,757 (1984). P. Aaltonen and B. Lofgren, Macromolecules, 28, 5353 (1995). P. Aaltonen, G. Fink, B. Lofgren, and J. Seppala, Macromolecules, 29, 5255 (1996). S. Bruzaud, H. Cramail, L. Duvignac, and A. Def®eux, Macromol. Chem. Phys., 198, 291 (1997). C.E. Wilen and J.H. Nasman, Macromolecules, 27, 4051 (1994). M.M. Marques, S.G. Correia, J.R. Ascenso, A.F.G. Ribeiro, P.T. Gomes, A.R. Dias, P. Foster, M.D. Rausch, and J.C.W. Chien, J. Polym. Sci. A: Polym. Chem., 37, 2457 (1999). U. Klabunde, U.S. Patent 4,698,403 (1987). F.C. Francis and M. Brookhart, J. Am. Chem. Soc., 117, 1137 (1995). S. Mecking, L.K. Johnson, L. Wang, and M. Brookhart, J. Am. Chem. Soc., 120, 888 (1998). S.G. Correia, M.M. Marques, J.R. Ascenso, A.F.G. Ribeiro, P.T. Gomes, A.R. Dias, M. Blais, M.D. Rausch, and J.C.W. Chien, J. Polym. Sci. A: Polym. Chem., 37, 2471 (1999). T. Simonazzi, D. Nicola, M. Aglietto, and G. Ruggeri, Comprehensive Polymer Science, First Supplement (G. Allen, ed.), Pergamon Press, New York, 133 (1982). A. Guyot, Polym. Adv. Technol., 7, 61 (1994). G. Xu and S. Lin, J. M. S. Rev. Macromol. Chem. Phys., C34(4), 555 (1994). Y.H.R. Jois and J.B. Harrison, J. M. S. Rev. Macromol. Chem. Phys., C36(3), 433 (1996). R. Singh, Prog. Polym. Sci., 17, 251 (1992). J. Jagur-Grodzinski, Prog. Polym. Sci., 17, 361 (1992). M.K. Naqvi and M.S. Choudhary, J. M. S. Rev. Macromol. Chem. Phys., C36(3), 601 (1996). T.C. Chung, ChemTech, 27, 496 (1991). T.C. Chung, J. Inorg. Organomet. Polym., 1, 37 (1991). T.C. Chung, J. Mol. Catal., 76, 15 (1992). T.C. Chung, Polym. News, 18, 38 (1993). S. Ramakrishnan and T.C. Chung, Macromolecules, 23, 4519 (1990). T.C. Chung, Macromolecular Symp., 89, 151 (1995). T.C. Chung, Trends Polym. Sci., 3, 191 (1995). T.C. Chung, Polym. Mater. Encyclop., 4, 2681 (1996). T.C. Chung, Polymer, 32, 1336 (1991). T.C. Chung, S. Ramakrishnan, and M.W. Kim, Macromolecules, 24, 2675 (1991). T.C. Chung and M. Chasmawala, Macromolecules, 24, 3718 (1991).
250 106. 107. 108. 109. 110. 111. 112. 113. 114. 115. 116. 117. 118. 119. 120. 121. 122. 123. 124. 125. 126. 127. 128. 129. 130. 131. 132. 133. 134. 135. 136. 137. 138. 139. 140. 141. 142. 143. 144. 145. 146. 147. 148. 149.
References T.C. Chung and M. Chasmawala, Macromolecules, 25, 5137 (1992). T.C. Chung and H.L. Lu, U.S. Patent 5,543,484 (1996). T.C. Chung and H.L. Lu, U.S. Patent 5,866,659 (1999). T.C. Chung and H.L. Lu, U.S. Patent 6,015,862 (2000). T.C. Chung, J. Elastom. Plast., 31, 298 (1999). T.C. Chung and J.Y. Dong, U.S. Patent 6,096,849 (2000). T.C. Chung, Functional Polymers; Synthesis and Applications (R. Arshady, ed.), ACS Professional Reference Book, 35 (1996). T.C. Chung, Polym. Mater. Encyclop., 8, 6412 (1996). T.C. Chung, Reactive Modi®ers for Polymers (S. Al-Malaika, ed.), Chapman & Hall, London, 303 (1997). T.C. Chung, Metallocene-Based Polyole®ns (W. Kaminsky, ed.), John Wiley, Vol. 2, 293 (1999). H.L. Lu and T.C. Chung, Functional Polymers (O. Patil, D.N. Schulz, and B.M. Novak, eds), ACS Book, 704, 163 (1998). T.C. Chung, Ole®n Polymerization; Emerging Frontiers (P. Arjunan, J.E. McGrath, and T.L. Hanlon, eds), ACS Book, 749, 104 (1999). T.C. Chung, H.L. Lu, and W. Janvikul, Polymer, 38, 1495 (1997). T.C. Chung and H.L. Lu, J. Mol. Catal. A: Chemical, 115, 115 (1997). B. Lu and T.C. Chung, Macromolecules, 32, 2525 (1999). G. Xu and T.C. Chung, J. Am. Chem. Soc., 121, 6763 (1999). G. Xu and T.C. Chung, Macromolecules, 32, 8689 (1999). H.C. Brown, Organic Synthesis via Boranes, Wiley-Interscience, New York (1975). H.C. Brown, R. Liotta, and G.W. Kramer, J. Org. Chem., 43, 1058 (1978). H. Noth and B. Wrackmeyer, Nuclear Magnetic Resonance Spectroscopy of Boron Compounds, Springer-Verlag, Berlin (1978). H.C. Brown, C.G. Scouten, and K.K. Wang, J. Org. Chem., 44, 2591 (1979). B.M. Mikhailov, Y.N. Bubnov, and A.V. Tsyban, J. Organomet. Chem., 154, 113 (1978). T.C. Chung, H.L. Lu, and C.L. Li, Polym. Int., 37, 197 (1995). T.C. Chung and D. Rhubright, Macromolecules, 24, 970 (1991). T.C. Chung and D. Rhubright, Macromolecules, 26, 3019 (1993). D. Rhubright, Ph.D Thesis, Pennsylvania State University (1993). S. Murahashi, S. Nozakura, and M. Sumi, Bull. Chem. Soc. Japan, 32, 670 (1959). V.Ia. Bogomol'nyi, Polym. Sci. (USSR), 2, 6 (1961). N.S. Nametkin, A.V. Topchiev, S.G. Durgar'ian, and I.M. Tolchinskii, Polym. Sci. (USSR), 2, 133 (1961). P. Longi, F. Greco, and U. Rossi, Makromol. Chem., 116, 113 (1968). S.L. Kapur and A.B. Deshpande, J. Sci. Ind. Res., 31(5), 254 (1972). K.J. Clark and T. Powell, Polymer, 6, 521 (1965). E.J. Vandenberg, J. Polym. Sci. C, 1, 207 (1963). U. Giannini, G. Bruckner, E. Pellino, and A. Cassata, J. Polym. Sci. C, 22, 157 (1968). U. Giannini, G. Bruckner, E. Pellino, and A. Cassata, Polym. Lett., 5, 527 (1967). C. Carlini, J. Polym. Sci., Polym. Chem. Ed., 18, 799 (1980). H.V. Holler and E.A. Youngman, U.S. Patent 3,761,458 (1973). J. Boor, Macromol. Rev., 2, 115 (1967). K.J. Clark, U.S. Patent 3,492,277 (1970). K. Tanaka, T. Iwata, and J. Sasaki, Jap. Patents 73/37,755 and 73/37,756 (1973). J.W. Collette, R. Ro, Y. Shin, and F.M. Sonnenberg, U.S. Patent 4,017,669 (1977). T. Iwata and J.J. Sasaki, Jap. Patent 73/43,191 (1973). J.W. Collette, R. Ro, and F.M. Sonnenburg, U.S. Patent 3,901,860 (1975). S. Datta and E.N. Kresge, U.S. Patent 4,987,200 (1991).
References 150. 151. 152. 153. 154. 155. 156. 157. 158. 159. 160. 161. 162. 163. 164. 165. 166. 167. 168. 169. 170. 171. 172. 173. 174. 175. 176. 177. 178. 179. 180. 181. 182. 183. 184. 185. 186. 187. 188.
251
U.M. Stehling, K.M. Stein, D. Fischer, and R.M. Waymouth, Macromolecules, 32, 14 (1999). G. Xu and T.C. Chung. Macromolecules, 33, 5803 (2000). A. Grassi, P. Longo, A. Proro, and A. Zambelli, Macromolecules, 22, 104 (1989). A.J. Sivak and L.A. Cullo, U.S. Patent 5,373,061 (1994). C.E. Wilen, M. Auer, and J.H. Nasman, Polymer, 33, 5049 (1992). R. Goretzki and G. Fink, Macromol. Rapid Commun., 19, 511 (1998). A. Tsuchida, C. Bolln, F.G. Sernetz, H. Frey, and R. Mulhaupt, Macromolecules, 30, 2818 (1997). L.M. Landoll and D.S. Breslow, J. Polym. Sci. A, 27, 2189 (1989). W.P. Baker, Jr. J. Polym. Sci., 42, 578 (1960). H. Higashi, K. Watabe, and S. Namikawa, J. Polym. Sci. B-5, 5, 1125 (1967). N. Yamazaki and S. Kambara, J. Polym. Sci. C-22, 22, 75 (1968). C.G. Overberger and G. Khattab, J. Polym. Sci. A-1, 7, 217 (1969). Y. Susuki and M. Saito, J. Polym. Sci. A-1, 9, 3639 (1971). R.N. Haszeldine, T.G. Hyde, and P.J. Tait, Polymer, 14, 215 (1973). R. Backsai, J. Polym. Sci. A-3, 3, 2491 (1965). G.A. Olah, Friedel±Crafts and Related Reactions, Interscience, New York (1963). A. Proto and D. Senatore, Macromol. Chem. Phys., 200, 1961 (1999). S.G. Correia, M.M. Marques, J.R. Ascenso, A.F.G. Ribeiro, P.T. Gomes, A.R. Dias, M. Blais, M.D. Rausch, and J.C.W. Chien, J. Polym. Sci. A: Polym. Chem., 37, 2471 (1999). E. Garagnai, R. Marzola, and A. Moro, Mater. Plast. Elastomer., 5, 298 (1982). A.F. Johnson and G.D. Simms, Composites, 17, 321 (1986). N. Subramanian and T.J. Nelson, Polym. Plast. Techn. Eng., 32, 635 (1993). R.J. Ashley, Adhesion, 12, 239 (1988). D. Jean-Marc, U.S. Patent 5,244,971 (1993). J.M. Felix and P. Gatenholm, J. Appl. Polym. Sci., 42, 609 (1991). G.E. Myers, et al. J. Polym. Mater., 15, 21 (1991). B. Majumdar, H. Keskkula, and D.R. Paul, Polymer, 35, 1386 (1994). G.H. Hu, J.J. Flat, and M. Lambla, Reactive Modi®ers for Polymers (S. Al-Malaika, ed.), Blackie Academic & Professional, London, p.1 (1997). A.H. Hogt, J. Meijer, and J. Jelenic, Reactive Modi®ers for Polymers (S. Al-Malaika, ed.), Blackie Academic & Professional, London, p.84 (1997). B. Ranby, Z.M. Gao, A. Hult, and P.Y. Zhang, Chemical Reactions on Polymers (J.L. Benham and J.F. Kinstle, eds), ACS Symposium Series 364, p.168 (1988). R.C. Kowalski, Reactive Extrusion: Principles and Practice (M. Xanthos, ed.), Hanser, Munich, p.7 (1992). T. Yagi, A.E. Pavlath, and A.G. Pittman, J. Appl. Polym. Sci., 27, 4019 (1982). D.E.J. Bergbreiter, Prog. Polym. Sci., 19, 529 (1994). J.C. Brosse, G. Legeay, and F. Poncin-Epaillard, Functional Polymers: Synthesis and Applications (R. Arshady, ed.), ACS, Washington, DC, p. 201 (1996). N.G. Gaylord, M. Mehta, and R. Mehta, J. Appl. Polym. Sci., 33, 2549 (1987). W. Heinen, C.H. Rosenmoller, C.B. Wenzel, H.J.M. de Groot, and J. Lugtenburg, Macromolecules, 29, 1151 (1996). N.G. Gaylord, H. Ender, L. Davis, Jr., and A. Takahashi, Modi®cation of Polymers (C.E. Carraher and M. Tsuda, eds), ACS Symposium Series 121, p. 469 (1980). G. Ruggeri, M. Aglietto, A. Petragnani, and F. Giadelli, Eur. Polym. J., 19, 863 (1983). S.H. Ryu, C.G. Gogos, and M. Xanthos, Adv. Polym. Technol., 11, 121 (1991/1992). K.E. Russell and E.C. Kelusky, J. Polym. Sci. A: Polym. Chem., 26, 2273 (1988).
252 189. 190. 191. 192. 193. 194. 195. 196. 197. 198. 199. 200. 201. 202. 203. 204. 205. 206. 207. 208. 209. 210. 211. 212. 213. 214. 215. 216. 217. 218. 219. 220. 221. 222. 223. 224. 225. 226. 227. 228.
References B.C. Trivedi and B.M. Culbertson, Maleic Anhydride, Plenum Press, New York (1982). A.R. Bunsell, Fibre Reinforcement of Composite Materials, Elsevier (1988). R.C. Constable, ANTEC, 127 (1994). H.G. Karian, ANTEC, 1665 (1995). J.B. Herater and E.M. Lacey, Mod. Plast., 41, 123 (1964). E.A. Flexman, Jr., Polym. Eng. Sci., 19, 564 (1979). B.E. Epstein, U.S. Patent 4,174,358 (1979). F. Ide, K. Kamada, and A. Hasegawa, Kobunshi Kagakun, 25, 107 (1968). A. Hogt, ANTEC, 1478 (1988). M. Xanthos. Reactive Extrusion: Principles and Practice (M. Xanthos, ed.), Hanser, Munich, 33 (1992). H.G. Fritz and B. Stoehrer, Int. Polym. Proc., 1, 31 (1986). K. Ebner and J.L. White, Int. Polym. Proc., 9, 233 (1994). G.H. Hu, J.J. Flat, and M. Lambla, Makromol. Chem., Macromol. Symp., 75, 137 (1993). C. Walling, E.R. Briggs et al., J. Am. Chem. Soc., 70, 1537 (1948). K. Dodgson and J.R. Ebdon, Eur. Polym. J., 13, 791 (1979). Q. Lin, M. Talukder, and C.U. Pittman, J. Polym. Sci. A: Polym. Chem., 33, 2375 (1995). S.D. Bruah and N.C. Laskar, J. Appl. Polym. Sci., 60, 649 (1996). V. Verney, E. Koerper, and A. Michel, Makromol. Chem., Macromol. Symp., 25, 187 (1989). S. Porejko, W. Gabara, and J. Kulesza, J. Polym. Sci. A-1, 5, 1563 (1967). N.G. Gaylord and M. Mehta, J. Polym. Sci.: Polym. Lett., 20, 481 (1982). G. DeVito, N. Lanzetta, M. Maglio, M. Malinconico, P. Musto, and R. Palumbo, J. Polym. Sci.: Polym. Chem., 22, 1335 (1984). N.G. Gaylord and M. Mehta, J. Appl. Polym. Sci., 33, 2549 (1987). S.W. Caywood, Jr., U.S. Patent 3,884,882 (1975). D.E. Bergbreiter, B. Srinivas, G.F. Xu, G. Aguilar, B.C. Ponder, H.N. Gray, and A. Bandella, Polymer Surfaces and Interfaces: Characterization, Modi®cation and Application (K.L. Mittal and K.W. Lee, eds), Utrecht, Netherlands, 3 (1997). A. Chapiro, J. Polym. Sci., 48, 109 (1960). G.G. Odian, T. Acker, and M. Sobel, J. Appl. Polym. Sci., 7, 245 (1963). K. Hayakawa and K. Kawase, J. Polym. Sci., A-1, 5 (1967). T.O. Neil, J. Polym. Sci. A, 10, 569 (1972). T. Kuroda, Y. Matushima, Y. Nakayama, and I. Toshima, U.S. Patent 3,494,480 (1970). R. Rangarajan, M. Vicic, and S. Lee, Polymer, 30, 933 (1989). R.H. Hansen, J.V. Pascale, T. De Benedicitis, and P.M. Rentzepis, J. Polym. Sci., A-3, 2205 (1965). M. Strobel, S. Corn, C.S. Lyons, and G.A. Korba, J. Polym. Sci.: Polym. Chem., 23, 1125 (1985). F. Poncin-Epaillard, B. Chevet, and J.C. Brosse, Eur. Polym. J., 26, 333 (1990). K. Inagaki, S. Tasaka, and K. Hibi, J. Polym. Sci., Polym. Chem., 30, 1425 (1992). P. Blais, D.J. Carlsson, G.W. Csullog, and D.M. Wiles, J. Colloid Interface Sci., 47, 636 (1974). K. Kato, J. Appl. Polym. Sci., 21, 2735 (1977). K.W. Lee and T.J. MaCarthy, Macromolecules, 21, 309 (1988). A. Deut and W. Berlin, Kl. Chem. Geol. Biol., 1, 427 (1963). K. Sadakata, K. Itoh, K. Kimura, T. Kishina, and Y. Nakamura, Jap. Patent 6,816,144. K. Nakatsuke, F. Ide, and K. Itoh, Jap. Patent 6,816,392.
References 229. 230. 231. 232. 233. 234. 235. 236. 237. 238. 239. 240. 241. 242. 243. 244. 245. 246. 247. 248. 249. 250. 251. 252. 253. 254. 255. 256. 257. 258. 259. 260. 261. 262. 263. 264. 265. 266. 267. 268. 269. 270. 271. 272.
253
M. Ichikama and R. Hirose, Jap. Patent 6,901,442. W. Gabara and S. Porejko, J. Polym. Sci., A-1, 5, 1547 (1967). M. Miura and S. Kamamatsu, Kobunshi Kaguku, 19, 175 (1962). K. Kawase and K. Hayakawa, Radiat. Res., 30, 116 (1967). A. Chapiro, J. Polym. Sci., 34, 439 (1959). C. Pugh and Y.L. Hsieh, J. Appl. Polym. Sci., 29, 3641 (1984). B.N. Mishra, D.S. Sood, and I.K. Mehta, J. Polym. Sci., Polym. Chem. Ed., 23, 1749 (1985). A.K. Mukherji and B.D. Gupta, J. Appl. Polym. Sci., 30, 2643 (1985). A.K. Mukherji and B.D. Gupta, J. Appl. Polym. Sci., 30, 3479 (1985). L. Odor, J. Polym. Sci. Pt C, 22, 477 (1968). S. Tazuke and H. Kimura, Makromol. Chem., 179, 2603 (1978). H. Kahayashi and I. Sakura, Jap. Kokai Tokyo Koho, 56, 671 (1979). D.C. Schram, G.M. Kroesen, and J. Beulens, J. Polym. Mater. Sci. Eng., 62, 25 (1990). F. Poncin-Epaillard, J.C. Brosse, and G. Legeay, Makromol. Chem., 189, 2293 (1988). N. Inagaki, S. Tasaka, and Y. Horikawa, Polym. Bull., 26, 283 (1991). F. Vigo, C. Uliana, and M. Traverso, Eur. Polym. J., 27, 779 (1991). T. Hirotsu and A. Arita, J. Appl. Polym. Sci., 42, 3255 (1991). T. Yamaguchi, S. Nakao, and S. Kimura, Macromolecules, 24, 5522 (1991). T.C. Chung, H.L. Lu, and C.L. Li, Macromolecules, 27, 7533 (1994). T.C. Chung and D. Rhubright, J. Polym. Sci. A: Polym. Chem., 31, 2759 (1993). S. Ramakrishnan, E. Berluche, and T.C. Chung, Macromolecules, 23, 378 (1990). T.C. Chung, W. Janvikul, R. Bernard, and G.J. Jiang, Macromolecules, 27, 26 (1994). T.C. Chung, W. Janvikul, and H.L. Lu, J. Am. Chem. Soc., 118, 705 (1996). T.C. Chung and Janvikul, J. Organomet. Chem., 581, 176 (1999). J. Furukawa and T. Tsuruta, J. Polym. Sci., 28, 227 (1958). N.L. Zutty and F.J. Welch, J. Polym. Sci., 43, 445 (1960). J.W.L. Fordham and C.L. Sturm, J. Polym. Sci., 33, 503 (1958). H.C. Brown and M.M. Midland, Chem. Commun., 699 (1971). S.B. Mirviss, J. Org. Chem., 32, 1713 (1967). A.G. Davies, K.U. Ingold, B.P. Roberts, and R.J. Tudor, Chem. Soc. B, 698 (1971). M.K. Georges, R.P.N. Veregin, P.M. Kazmaier, and G.K. Hamer, Macromolecules, 26, 2987 (1993). B. Lu, Ph. D. Thesis, Pennsylvania State University (1999). W. Chinsirikul, T.C. Chung, and I.R. Harrison, J. Thermoplast. Compos. Mater., 6, 18 (1993). S.H. Lee, C.L. Li, and T.C. Chung, Polymer, 35, 2980 (1994). T.C. Chung and S.H. Lee, J. Appl. Polym. Sci., 64, 567 (1997). T.C. Chung, U.S. Patent 5,837,365 (1998). T.C. Chung, U.S. Patent 5,976,434 (1999). M. Mulder, Basic Principles of Membrane Technology, Kluwer Academic Publishers, London (1990). S. Hiroyuki, M. Masahiro, T. Ryozo, T. Toshikazu, T. Kazumi, and F. Fumiaki, U.S. Patent 5,480,953 (1996). K. Matsuda and M. Kohno, U.S. Patent 4,409,339 (1983). Y.J. Wang, C.H. Chen, M.L. Yeh, G.H. Hsiue, and B.C. Yu, J. Membrane Sci., 53, 275 (1990). M. Toshio, H. Okihiko, O. Makoto, and S. Yukio, U.S. Patent 5,186,835 (1993). T.C. Chung, The Polymeric Materials Encyclopedia, 10, 8093 (1996). T.C. Chung and A. Kumar, Polym. Bull., 28, 123 (1992).
254 273. 274. 275. 276. 277. 278. 279. 280. 281. 282. 283. 284. 285. 286. 287. 288. 289. 290. 291. 292. 293. 294. 295. 296. 297. 298. 299. 300. 301. 302. 303. 304. 305. 306. 307. 308. 309. 310. 311. 312.
References T.C. Chung, D. Rhubright, and A. Kumar, Polym. Bull., 30, 385 (1993). T.C. Chung, F.J. Chen, J.E. Stanat, and A. Kumar, U.S. Patent 5,409,873 (1995). T.C. Chung, F.J. Chen, J.E. Stanat, and A. Kumar, U.S. Patent 5,414,177 (1995). T.C. Chung, F.J. Chen, J.E. Stanat, and A. Kumar, U.S. Patent 5,563,313 (1996). T.C. Chung, F.J. Chen, J.E. Stanat, and A. Kumar, U.S. Patent 5,571,885 (1996). T.C. Chung, F.J. Chen, J.E. Stanat, and A. Kumar, U.S. Patent 5,691,431 (1997). T.C. Chung, F.J. Chen, J.E. Stanat, and A. Kumar, U.S. Patent 5,770,539 (1998). T.C. Chung, R. Ding, and R.L. Shubkin, U.S. Patent 5,945,574 (1999). J.M.J. Frechet, Crown Ethers and Phase Transfer Catalystsin Polymer Science, Plenum Press, New York (1984). K.W. Powers, H.C. Wang, T.C. Chung, A.J. Jias, and J.A. Olkusz, U.S. Patent 5,162,445 (1992). S. Mohanraj and W. Ford, Macromolecules, 19, 2470 (1986). R.G. Jones and Y. Matsubayashi, Polymer, 33, 1069 (1992). D. Pini, R. Settambolo, A. Raffaelli, and P. Salvadori, Macromolecules, 20, 58 (1987). I. Piirma and J.R. Lenzotti, Brit. Polym. J., 21, 45 (1989). Y. Nagasaki and T. Tsuruta, Makromol. Chem., Rapid Commun., 7, 437 (1986). F. Bonaccorsi, A. Lezzi, A. Prevedello, L. Lanzini, and A. Roggero, Polym. Int., 30, 93 (1993). A. Onopchenko, J.G.D. Schulz, and R.J. Seekircher, J. Org. Chem., 37, 1414 (1972). L.P. Ferrari and H.D. Stover, Macromolecules, 24, 6340 (1991). T.C. Chung, H.L. Lu, and R.D. Ding, Macromolecules, 30, 1272 (1997). F. Bonaccorsi, A. Lezzi, A. Prevedello, L. Lanzini, and A. Roggero, Polym. Int., 30, 93 (1993). W. Kaminsky, K. Kulper, and H. Brintzinger, Angew. Chem., Int. Ed. Engl., 24, 507 (1985). J.A. Ewen, J. Am. Chem. Soc., 106, 6355 (1984). H.W. Turner, U.S. Patent 4,752,597 (1988). J.M. Canich, U.S. Patent 5,026,798 (1991). S.Y. Lai, J.R. Wilson, G.W. Knight, J.C. Stevens, and P.W. Chum, U.S. Patent 5,272,236 (1993). J.C. Stevens, F.J. Timmers, J.R. Wilson, G.F. Schmidt, P.N. Nickias, R.K. Rosen, G.W. Knight, and S.Y. Lai, Eur. Pat. Appl. 416,815 A2 (1991). F.G. Sernetz, R. Mulhaupt, and R.M. Waymouth, Macromol. Chem. Phys., 197, 1071 (1996). T.C. Chung and H.L. Lu, J. Polym. Sci. A: Polym. Chem., 35, 575 (1997). T.C. Chung and H.L. Lu, J. Polym. Sci. A: Polym. Chem., 36, 1017 (1998). H.L. Lu, S. Hong, and T.C. Chung, J. Polym. Sci. A: Polym. Chem., 37, 2795 (1999). A. Zambelli, C. Pellecchia, L. Oliva, P. Longo, and A. Crassi, Makromol. Chem., 192, 223 (1991). A. Zambelli, C. Pellecchia, and A. Proto, Macromol. Symp., 89, 373 (1995). P. Longo, A. Proto, and A. Zambelli, Macromol. Chem. Phys., 196, 3015 (1995). H.L. Lu, S. Hong, and T.C. Chung, Macromolecules, 31, 2028 (1998). B. Lu and T.C. Chung, J. Polym. Sci. A: Polym. Chem., 38, 1337 (2000). W. Heinen, C.H. Rosenmoller, C.B. Wenzel, H.J.M. de Groot, and J. Lugtenburg, Macromolecules, 29, 1151 (1996). R. Rengarjan, V.R. Oarameswaran, S. Lee, and P.L. Rinaldi, Polymer, 31, 1703 (1990). B. Rojas, J.G. Fatou, M.C. Martinez, and O. Laguna, Eur. Polym. J., 33, 725 (1997). K.E. Russell and E.C. Kelusky, J. Polym. Sci., A: Polym. Chem., 33, 555 (1995). K.W. Powers, H.C. Wang, T.C. Chung, A.J. Jias, and J.A. Olkusz, U.S. Patent 5,548,029 (1996).
References 313. 314. 315. 316. 317. 318. 319. 320. 321. 322. 323. 324. 325. 326. 327. 328. 329. 330. 331. 332. 333. 334. 335. 336. 337. 338. 339. 340. 341. 342. 343. 344. 345. 346. 347. 348. 349. 350. 351. 352. 353. 354. 355. 356.
255
M. Camps, Rev. Macromol. Chem. Phys., C22(3), 343 (1982/1983). L.P. Ferrari and H.D. Stover, Macromolecules, 24, 6340 (1991). C. Cozewith and G. Ver Strate, Macromolecules, 14, 482 (1971). J.C. Stevens, Stud. Surf. Sci. Catal., 89, 277 (1994). S. Kitagawa and I. Okada, U.S. Patent 4,366,296 (1982). S. Kitagawa and I. Okada, Polym. Bull., 10, 109 (1983). J. Lal and M.L. Senyek, U.S. Patent 4,551,503 (1985). R. Mulhaupt, W. Ovenall, and S.D.J. Ittel, J. Polym. Sci. A: Polym. Chem., 26, 2487 (1988). D.J. Lohse, S. Datta, and E. Kresge, Macromolecules, 24, 561 (1991). S. Machida and N. Tani, Jap. Patents 05-194665 and 05-194666 (1993). S. Marathe and S. Sivaram, Macromolecules, 27, 1083 (1994). B.P. Etherton, J.J. McAlpin, and E.N. Kresge, U.S. Patent 5,504,171 (1996). M. Yokoyama, Eur. Pat. Appl., 88310305.3 (1988). S. Machida, H. Shikuma, T. Tazaki, T. Tatsumi, and S. Kurokawa, U.S. Patent 5,608,009 (1997). H.C. Brown, K.W. Kim, M. Srebnik, and B. Singaram, Tetrahedron, 43, 4071 (1987). B.C. Trivedi and B.M. Culbertson, Maleic Anhydride, Plenum Press, New York (1982). Y. Doi, S. Ueki, and K. Soga, Macromolecules, 12, 814 (1979). Y. Doi, S. Ueki, and K. Soga, Makromol. Chem., 180, 1359 (1979). Y. Doi and T. Keii, Adv. Polym. Sci., 73/74, 201 (1986). Y. Doi, S. Suzuki, and K. Soga, Macromolecules, 19, 2896 (1986). X. Yasuda, M. Furo, and H. Yamamoto, Macromolecules, 25, 5115 (1992). C.M. Killian, D.J. Tempel, L.K. Johnson, and M. Brookhard, J. Am. Chem. Soc., 118, 11664 (1996). J.D. Scollard and D.H. McConville, J. Am. Chem. Soc., 118, 10008 (1996). R. Baumann, W.M. Davis, and R.R. Schrock, J. Am. Chem. Soc., 119, 3830 (1997). H. Hagihara, T. Shiono, and T. Ikeda, Macromolecules, 31, 3188 (1998). M. Brookhart, J.M. DeSimone, B.E. Grant, and M.J. Tanner, Macromolecules, 28, 5378 (1995). D.R. Bur®eld, Polymer, 25, 1817 (1984). T. Shiono, K. Yoshida, and K. Soga, Makromol. Chem., Rapid Commun., 11, 2085 (1990). T. Shiono, H. Kurosawa, and K. Soga, Makromol. Chem., 192, 2751 (1992). T. Shiono and K. Soga, Macromolecules, 25, 3356 (1992). T. Shiono, H. Kurosawa, and K. Soga, Macromolecules, 27, 2635 (1994). T. Shiono, H. Kurosawa, and K. Soga, Macromolecules, 28, 437 (1995). Y. Doi, G. Hizal, and K. Soga, Makromol. Chem., 188, 1273 (1987). P.F. Fu and T.J. Marks, J. Am. Chem. Soc., 117, 10747 (1995). K. Koo, P.F. Fu, and T.J. Marks, Macromolecules, 32, 981 (1999). K. Koo and T.J. Marks, J. Am. Chem. Soc., 120, 4019 (1998). K. Koo and T.J. Marks, J. Am. Chem. Soc., 121, 8791 (1999). T.C. Chung and J.Y. Dong, J. Am. Chem. Soc., 123, 4871 (2001). B. Lu and T.C. Chung, Macromolecules, 31, 5943 (1998). D. Suwanda, R. Lew, and S.T. Balke, J. Appl. Polym. Sci., 35, 1019 (1988). T. Tsutsui, A. Mizuno, and N. Kashiwa, Polymer, 30, 428 (1989). K.E. Russell and E.C. Kelusky, J. Polym. Sci. A: Polym. Chem., 26, 2273 (1988). M. Sclavons, V. Carlier, B. DeRoover, P. Franquinet, J. Devaux, and R. Legras, J. Appl. Polym. Sci., 62, 1205 (1996). S.L. Aggarwal, Block Copolymers, Plenum Press, New York (1970).
256 357. 358. 359. 360. 361. 362. 363. 364. 365. 366. 367. 368. 369. 370. 371. 372. 373. 374. 375. 376. 377. 378. 379. 380. 381. 382. 383. 384. 385.
References G. Riess, J. Periard, and A. Bandereet, Colloidal and Morphological Behavior of Block and Graft Copolymers, Plenum Press, New York (1971). M. Szwarc, Adv. Polym. Sci., 47, 1 (1982). R.N. Young, R.P. Quirk, and L. Fetters, J. Adv. Polym. Sci., 56, 1 (1984). M. Miyamoto, M. Sawamoto, and T. Higashimura, Macromolecules, 18, 123 (1985). J.P. Kennedy, U.S. Patent 4,946,899 (1990). W. Risse and R.H. Grubbs, Macromolecules, 22, 1558 (1989). R.R. Schrock, K.B. Yap, D.C. Yang, H. Sitzmann, L.R. Sita, and G.C. Bazan, Macromolecules, 22, 3191 (1989). T.C. Chung, W. Janvikul, R. Bernard, and G.J. Jiang, Macromolecules, 27, 26 (1993). T.C. Chung, W. Janvikul, R. Bernard, R. Hu, C.L. Li, S.L. Liu, and G.J. Jiang, Polymer, 36, 3565 (1995). R.E. Cohen, J. Polym. Sci. A: Polym. Chem., 24, 2457 (1986). M.A. Drzewinski, Macromol. Symp., 91, 107 (1995). Y. Doi, S. Ueki, and T. Keii, Coordination Polymerization (C.C. Price and E.J. Vandenberg, eds), Plenum, New York, 249 (1983). G. Natta, E. Beati, and F. Severine, J. Polymer Sci., 34, 548 (1959). B. Ranby and F. Guo, Polym. Preprints, 31, 446 (1990). T.C. Chung, G.J. Jiang, and D. Rhubright, U.S. Patent 5,286,800 (1994). T.C. Chung, G.J. Jiang, and D. Rhubright, U.S. Patent 5,401,805 (1995). T.C. Chung and G.J. Jiang, Macromolecules, 25, 4816 (1992). T.C. Chung, D. Rhubright, and G.J. Jiang, Macromolecules, 26, 3467 (1993). T.C. Chung and D. Rhubright, Macromolecules, 27, 1313 (1994). T.C. Chung, H.L. Lu, and R.D. Ding, Macromolecules, 30, 1272 (1997). H.L. Lu and T.C. Chung, J. Polym. Sci., Polym. Chem. Ed., 37, 4176 (1999). B. Lu and T.C. Chung, Macromolecules, 32, 8678 (1999). M.S. Jacovic, D. Pollock, and R.S. Porter, J. Appl. Polym. Sci., 23, 517 (1979). J.F. Vega and A. Santamaria, A. Munoz-Escalona and P. Lafuente, Macromolecules, 31, 3639 (1998). D. Yan, W.J. Wang, and S. Zhu, Polymer, 40, 1737 (1999). W.W. Graessley, Acc. Chem. Res., 10, 332 (1977). S.K. Goyal, Plast. Eng., 51, 25 (1995). B.J. Scheve, J.W. May®eld, and A.J. DeNicola, Jr. U.S. Patent 5,731,362 (1998). H.C. Brown, K. Kim, M. Srebnik, and B. Singaram, Tetrahedron, 43, 4071 (1987).
Index
Note: Figures and Tables are indicated by italic page numbers
A
acrylonitrile-grafted polypropylene ®lm 76 adhesion studies, PP/Al and PP/glass laminates 96±8 aldehyde-group-containing polyethyleneco-p-methylstyrene copolymers 133 aldehyde-group-containing polystyrene copolymers 52±3 4-(!-alkenyl)-2,6-di-t-butylphenol, copolymerization with propylene 54 alkoxyl radicals 66, 92, 228 alkyl-9-BBN, selective auto-oxidation of 91±3 alkyl halides, dehydrohalogenation of 57±8 allylsilane 37 copolymerization with -ole®ns 37 homopolymerization of 36 aluminum/polypropylene laminates, adhesion studies 96 amide-group-containing PE copolymers 123 1 H NMR spectra 123 amide groups, on surface-modi®ed polyole®ns 77 amino group, protection of 42, 45 amino-group-containing monomers, polymerization of 41±50 amino-group-containing polyethylene 148 1 H NMR spectra 147, 148 amino-group-containing polystyrenes 44±52
amino groups, on surface-modi®ed polyole®ns 77 amino-group-terminated polypropylene (PP-t-NH2) reaction with PP-g-MA 241, 242 synthesis of 174, 241 3-aminopropyltrimethoxysilane, reaction with maleic-anhydride-terminated polypropylene 232 anhydride group, as coupling agent 69 anionic reactions with divinylbenzene-containing copolymers 217±19 living anionic polymerization 189±94 applications, functionalized polyole®ns 7, 61, 95±103 aryl halides, copolymerization with ethylene 58 1,2-azobisisobutyronitrile (AIBN), as free radical initiator 66, 128
B
-hydride elimination 6, 59 -hydrogen transfer reaction 6 `ball-and-socket' topography 224, 225, 236, 237, 238 11 B NMR spectra borane-containing monomers 24, 26 poly(propylene-co-5-hexenyl-9-BBN) copolymers 31 BBN see 9-borabicyclononane benzoyl peroxide (BPO), as free radical initiator 69, 72, 75, 126, 128
258
Index
benzylic-alcohol-containing PE copolymers, 1H NMR spectra 122, 123 bibliography 247±56 bis(!-alkenyl)zinc compounds 157±8 1,3-bis(t-butylperoxyisopropyl)benzene, as free radical initiator 70 bisphenol-A polycarbonate 222 4-[N,N-bis(trimethylsilyl)aminoethyl] styrene, polymerization of 45, 46 4-[N,N-bis(trimethylsilyl)aminomethyl] styrene, polymerization of 45, 46 4-[N,N-bis(trimethylsilyl)amino] styrene polymerization of 45, 46, 47±8 with various catalyst/cocatalyst systems 47±8, 47 N,N-bis(trimethylsilyl)-1-amino-10undecene, copolymerization with ethylene 44 blends 7, 61, 220±5 compatibilization of 220±5, 235±40 PE/PS 222, 224±5 PP/PC 222 PP/PMMA 220±2 PP/polyamide 235±40 block copolymers 16±17, 179±95 9-borabicyclononane (9-BBN) as chain transfer agent 160, 162 reaction with dienes 23 9-borabicyclononane-containing monomers copolymerization with -ole®ns 27±35 homopolymerization of 25±7 preparation of 23±4 9-borabicyclononane-terminated polyethylene 160 9-borabicyclononane-terminated poly(ethylene-co-1-octene) 163±6 9-borabicyclononane-terminated poly(ethylene-co-styrene) 163±6 9-borabicyclononane-terminated polystyrene 161±2 borane cocatalysts 44, 45, 47, 48 borane-containing monomers 15, 23±4 copolymerization with -ole®ns 27±35 1-butene 33 by continuous reaction 33 ethylene 28±30 1-octene 33±5 propylene 30±3 homopolymerization of 25±7 deactivation modes 26
preparation of 23±4 solubility 81 stability to transition metal catalysts 24, 27 see also 5-hexenyl-9-borabicyclononane borane-containing polyole®ns hydroxylation of 32, 33, 82, 84, 85, 86, 89±90, 123, 146 oxidation of 146, 198 preparation of 27±35, 83, 86, 123, 145 reactions 146 borane-containing reactive polyole®ns 15, 81±104 borane conversion reactions 82 borane groups hydroxylation reactions 82, 83±90 stability to transition metal catalysts 24, 27, 81 versatility 82 borane reactive sites, living radical graft-from reaction using 198±202 borane-terminated polyethylene 160±1, 182 borane-terminated polyole®ns 16, 158±66 preparation of 160, 161±2, 172±4, 182 borane-terminated polypropylene 174, 185, 230 borane-terminated polystyrene 161±2, 186 DSC curve 189 glass transition temperature 188 melting temperature 188 borinate radicals compared with alkoxyl radicals 182 compared with nitroxide radicals 92±3 boron, position in Periodic Table 21, 22 brominated PE-p-MS copolymers DSC curves 130±1, 131 GPC curves 131±2, 131 1 H NMR spectra 130 8-bromo-1-octene, copolymerization with -ole®ns 58 brush-like molecular structures PP-g-nylon 11 graft copolymer 235 PP-OH copolymer 86, 87, 95 -scission reaction 14, 68 bulk functionalization 66±74 1,3-butadiene 136 1-butene, copolymerization with, borane-containing monomer 33
Index p-(3-butenyl)styrene, terpolymerization with ethylene and propylene 140 butyl arylate, in free radical polymerization 185 6-t-butyl-2-(1,1-dimethylhelpt-6-enyl)-4methylphenol, copolymerization with propylene 55 4-t-butyldimethylsiloxy-1,6-heptadiene, polymerization of 54 n-butyl methacrylate, in free radical polymerization 186
C 13
C NMR spectra aldehyde-group-containing polystyrene 52, 53 amino-group-containing polystyrenes 48±9 (DEPT-135), hydroxy-terminated poly(ethylene-co-styrene) 166 maleic-anhydride-modi®ed polypropylene 70 poly(styrene-co-[1,3-dimethyl-2(4-vinylphenyl]-imidazolidine) copolymer 51±2 -caprolactone (-CL), ring-opening reaction with hydroxylated polypropylene 214 carboxylated PE-p-MS copolymers DSC curves 124, 125, 132 1 H NMR spectra 121±2, 122 carboxylic acid groups effect on properties of PE-p-MS 133 protection of 56, 57 on surface-modi®ed polyole®ns 75, 77 !-carboxyl--ole®ns, terpolymerization with ethylene and propylene 57, 60 catalyst deactivation/poisoning 11 causes 11, 39 prevention of 12±13, 39±41, 42 catalyst-driven processes, historical aspects 2±4 catalysts immobilized 102±3 see also Lewis acid...; metallocene...; Ziegler±Natta catalysts chain-end unsaturated polymers chemical modi®cation of 172±5 synthesis of 172 see also polypropylene, chain-end unsaturated (u-PP)
259
chain extension reactions 185, 187±8, 192 chain scission reactions 14, 67, 68 chain transfer agents 157 chain transfer reactions 157±71, 182 borane-terminated polyole®ns 158±66, 182 control of 10±11 silyl-terminated polyole®ns 158 styrene-derivative-terminated polyole®ns 16±71 zinc-alkyl-terminated polyole®ns 157±8 charge transfer complex, between maleic anhydride and styrene 70 chlorinated PE-p-MS copolymers DSC curves 130±1, 131 GPC curves 131±2, 131 1 H NMR spectra 130 !-chloro--ole®ns, copolymerization with -ole®ns 58 5-chloropent-1-ene, failure to polymerize 58 p-chlorostyrene-terminated polypropylene 170±1 11-chloroundec-1-ene 58 cocrystallization, PP/PP-OH 87, 88, 102 cold plasma treatment 76 comb-like molecular structure, polyalcohols 84 composites 7, 61 contact angles polyethylene surfaces 77 polypropylene surfaces 77 PP/PP-OH membranes 99, 101 controlled rheology resins 14 corona-treated polypropylene ®lm 76 Cossee's mechanism 4±5 coupling reactions, for diblock copolymers 194±5 crosslinking reactions in post-polymerization 14, 67±8, 72, 73 in unsaturated polyole®ns 149, 150 cyclopentadienyl ligands see metallocene catalysts
D
diallylsilanes, polymerization of 37 diblock copolymers 15, 16, 175, 179±95 from living transition metal coordination polymerization 180±1
260
Index
(2,5-di-t-butylperoxy)-2,5-dimethylhexane, as free radical initiator 70 dicumyl peroxide, as free radical initiator 69, 73, 127 dienes co-/ter-polymerization with -ole®ns 137±45 types 136±7 differential scanning calorimetry see DSC curves diimine-containing catalysts 3±4, 59, 60 5-(N,N-diisopropylamino)-1-penetene copolymerization with ole®ns 44 homopolymerization of 44 4-(dimethylaminomethyl)styrene, homopolymerization of 45, 46 dimethyldiallylsilane copolymerization with propylene 37 polymerization of 37 1,3-dimethyl-2-(4-vinylphenyl) imidazolidine, copolymerization with styrene 50±1 diphenylphosphine-containing PE copolymers 123 1 H NMR spectra 123 direct polymerization approach 9±10, 11±13, 19±61 divinylbenzene (DVB) copolymerization with ethylene 141 in crosslinking reactions 149, 153 graft reactions using 216±20 incorporation into polyole®ns 15, 140±5 terpolymerization with ethylene and propylene 142±5 see also ethylene±propylene±divinylbenzene (EP-DVB) terpolymers DSC curves amino-group-containing polystyrenes 50 borane-terminated polystyrene 189 chain-end unsaturated polypropylene 235 EO-p-MS terpolymers 118 EP-p-MS terpolymers 116 hydroxylated polyethylene 148 polyethylene 109 poly(ethylene-co-1,4-hexadiene) 139, 148 poly(ethylene-co-p-methylstyrene) copolymers 109, 124, 131, 132, 208
compared with brominated and chlorinated copolymers 130±1, 131 compared with functionalized copolymers 124, 125 compared with polyethylene 109 polyethylene-g-polystyrene graft copolymers 208 poly(1-octene-co-1-hexen-6-ol) copolymers 36 poly(1-octene) and PO-OH 90 polypropylene and PP-OH 88 PP-g-PCL graft copolymers 216 PP-b-SMA diblock copolymers 234, 235 PP-g-SMA copolymers 95
E
early transition metal catalysts 2, 11, 23, 40 borane-containing monomers polymerized by 24±35 see also metallocene...; Ziegler±Natta catalysts emulsi®ers, graft polymers as 221, 222, 223, 224, 225 ene reaction EP-MA prepared using 74 maleation using 74, 152 EPDM see ethylene±propylene±diene (EPDM) terpolymers epoxidation, of unsaturated polyole®ns 153 ESR spectra, O2-oxidized PP-9-BBN 92 ester-group-containing monomers, polymerization of 56±7 ester groups, on surface-modi®ed polyole®ns 77 ether groups, on surface-modi®ed polyole®ns 77 ethylene copolymerization with amino-compounds 42±3, 44 borane-containing monomer 28±30 p-chlorostyrene 58 11-chloroundec-1-ene 58 divinylbenzene 141 1,4-hexadiene 137±9 methyl acrylate 60 p-methylstyrene 107±10 1-octene, with 9-BBN chain transfer agent 163±6
Index styrene, with 9-BBN chain transfer agent 163±6 terpolymerization with propylene and p-(3-butenyl)styrene 140 and !-chloro--ole®ns 58 and 11-chloroundec-1-ene 58 and divinylbenzene 142±5 and hydroxy compounds 54±5 and p-methylstyrene 115±16 and norbornene amino compound 43±4 and TMA-protected carboxylic acids 57, 60 and TMA-protected hydroxyole®ns 55, 60 see also polyethylene ethylene±p-chlorostyrene copolymer 58 ethylene±1-octene copolymers, 9-BBNterminated 163±6 ethylene±1-octene±divinylbenzene (EO-DVB) terpolymers, maleic anhydride modi®ed 151 ethylene±1-octene±p-methylstyrene (EO-p-MS) terpolymers 116±19 ethylene±propylene (EP) copolymers 1 acid-functionalized 57 amino-group-containing 44 attack by free radicals 14, 67 glass transition temperatures 115, 116, 142 graft copolymers (EP-g-PMMA) 201±2 glass transition temperatures 201 1 H NMR spectra 202 hydroxylated (EP-OH) 89±90 maleic-anhydride-modi®ed 73±4, 152±3 ethylene±propylene±diene (EPDM) terpolymers graft-from reaction with PMMA 201 hydroboration of 83, 89, 148 preparation of 135, 142 reaction with maleic anhydride 74, 152 ethylene±propylene±divinylbenzene (EP-DVB) terpolymers 142±5 crosslinking of 149, 150 DSC curves 143±4, 144 glass transition temperatures 142, 143, 144 1 H NMR spectra 142, 143 preparation of 142
261
ethylene±propylene±p-methylstyrene (EP-p-MS) terpolymers 115±16 DSC curves 116 glass transition temperatures 115 molecular weights 115 ethylene±styrene copolymers 9-BBN-terminated 163±6 uses 7 ethyl methacrylate, in free radical polymerization 185
F
fabrication methods 1 ®ltration membrane applications 99 ¯uorine-functionalization of polymer surface 77 Fox±Flory equation 211 free radical functionalization 66±8 disadvantages 10, 14 effect on ®nal polymer structure 68 free radical maleic anhydride grafting reaction 125±8, 228±9 for ethylene±propylene copolymers 73±4 for p-methylstyrene-containing polyole®ns 126±8 for polyethylenes 72±3, 126±7 for polypropylenes 69±71, 127 side reactions 125, 229 free radical polymerization, by oxidation adducts of trialkylboranes 91 FTIR spectra EP/DVB terpolymers after UV irradiation 149, 150 long-chain-branched poly propylene 243 functional groups effect on catalysts 12 protection of 12, 39, 40±1 functional-group-terminated polyethylenes 156 functionalized polyole®ns, applications 7, 61, 95±103 future prospects 6±7
G
gel permeation chromatography detectors 35 see also GPC curves GPC curves borane-containing polymers 25, 30 borane-terminated polyethylene 183
262
Index
GPC curves (cont.) chain-end unsaturated polypropylene 173 effect of trialkylboranes on ole®n polymerization 24 EP-DVB terpolymer 143, 144 esteri®ed poly(1-octene-co-5-hexenyl9-BBN) copolymers 34, 35 ethylene±propylene copolymer 144 long-chain-branched polypropylene 243, 244 p-methylstyrene-terminated polypropylene 169, 193 PE-b-PMMA diblock copolymer 183 poly(ethylene-co-5-hexenyl-9-BBN) copolymers 30 poly(ethylene-co-p-methylstyrene) 124, 131 compared with brominated and chlorinated copolymers 131±2, 131 compared with trimethylsilylcontaining copolymers 123, 124 polypropylene 169, 186 polypropylene diblock copolymers 186, 193, 242 PP-t-MA copolymer 242 PP-t-p-MS copolymer 169, 193 PP-b-PMMA diblock copolymer 186 PP-b-PP diblock copolymer 242 PP-b-PS diblock copolymer 193 graft copolymers 16±17, 197±226 as compatibilizers 220±5 polyethylene-based 204±9 poly(ethylene±propylene)-based 201±2, 219±20 polypropylene-based 199±201, 209±16, 219±20 graft density 206±7 values for polyethylene graft copolymers 206 graft-from reactions 65, 67 via borane reactive sites 198±202 initiators for 91, 93 via p-methylstyrene reactive sites 203±13 by ring-opening polymerization 213±16 styrene and maleic anhydride 94 graft length 207 values for polyethylene graft copolymers 206
graft polymerization initiation of 197 reactive groups in 198 graft reactions, via divinylbenzene sites 216±20
H 1
H NMR spectra aldehyde-group-containing polystyrene 52, 53 amide-group-containing PE-p-MS copolymers 123 benzylic-alcohol-containing PE-p-MS copolymers 122, 123 brominated PE-p-MS copolymers 129±30, 130 carboxylated PE-p-MS copolymers 121±2, 122 chain-end unsaturated polypropylene 231 chlorinated PE-p-MS copolymers 129±30, 130 diphenylphosphine-containing PE-p-MS copolymers 123 EPDM rubber 202 EP-DVB terpolymer 142, 143 EP-g-PMMA graft copolymer 202 halogenated PE-p-MS copolymers 128±30, 130 hydroxylated PE-p-MS copolymers 122, 123 hydroxy-terminated polyethylene 161 maleic-anhydride-grafted polyole®ns 127, 129, 231 maleic-anhydride-terminated polypropylene (PP-t-MA) 231 p-methylstyrene-terminated polypropylene 170, 191 PE-p-MS-g-MA copolymers 127, 129 PE-NH2 147 PE-1-octene-DVB-g-PS graft copolymer 218 PE-OH 147 PE-OSi(CH3)3 147 PE-b-PMMA diblock copolymers 184 PE-g-PMS graft copolymers 205±6, 205 PE-g-PS graft copolymers 204 polyethylene graft copolymers 204, 205 poly(ethylene-co-1,4-hexadiene) 147
Index polypropylene diblock copolymers 187, 193 polypropylene graft copolymers 210, 212, 215 polystyrene diblock copolymers 188 PP-t-MA 231 PP-t-p-MS 170, 191 PP-p-MS-g-MA copolymers 127, 129 PP-b-PBA diblock copolymer 187 PP-g-PB graft copolymer 211±12, 212 PP-g-PCL graft copolymers 215 PP-b-PEMA diblock copolymer 187 PP-g-PMMA graft copolymer 212, 213 PP-g-PMS graft copolymer 209, 210 PP-b-PS diblock copolymer 187, 193 PP-g-PS graft copolymer 209, 210 s-PS-g-PBMA graft copolymer 188 s-PS-g-PMMA graft copolymer 188 silane-terminated polystyrene 162, 163, 188 silylated ethylene/1-octene/ divinylbenzene terpolymer 218 trimethylsilyl-containing PE-p-MS copolymers 119±21 halogenation, of p-methylstyrenecontaining polyole®ns 128±32 halogen-containing monomers, polymerization of 57±8 !-halo--ole®ns, dehydrohalogenation of 57±8 1,4-hexadiene, copolymerization with ethylene 137±9 1-hexene amino-compound, copolymerization with propylene 43 copolymerization with allylsilane 37 8-bromo-1-octene 58 5-(N,N-diisopropylamino)-1penetene 44 5-hexenyl-9-borabicyclononane (H-BBN) copolymerization with 1-butene 33, 34 ethylene 29, 30 1-octene 33±4 propylene 31±2, 34 polymerization of 25±7 historical aspects 2±4 hydroboration of chain-end unsaturated polypropylene 172±4
263
of unsaturated polyole®ns 23, 83, 89, 145±8 hydrogen abstraction in bulk free radical grafting reaction 65, 69 in surface functionalization 75 susceptibility to 14, 67 hydrophilic PP/PP-OH membranes 98±102 contact angle(s) with water 99, 101 pore sizes 99, 101 preparation of 99 scanning electron micrographs 100 selectivity (rejection molecular weight) 101±2, 101 tensile strength 99 water permeability 101 hydroxy group, protection of 54 hydroxy-group-containing monomers, polymerization of 54±5 hydroxylated polyethylene 85±6 DSC curves 148 1 H NMR spectra 122, 147 hydroxylated polymers 32±3, 83±90 hydroxylated polypropylene 32±3, 86±9 applications 95±103 DSC curves 88 IR spectrum 200 ring-opening reaction with -caprolactone 214 thermal stability 87±8, 89 !-hydroxy--ole®ns, terpolymerization with ethylene and propylene 55, 60 hydroxy-terminated polyethylene 161 1 H NMR spectrum 184 hydroxy-terminated poly(ethylene-costyrene) 166 13 C (DEPT-135) NMR spectrum 166 hydroxy-terminated polystyrene 162 1-hydroxy-10-undecene, terpolymerization with ethylene and propylene 55
I
imide groups, on surface-modi®ed polyole®ns 77 imidization reaction, for preparation of long-chain-branched polypropylene 241 immobilized catalysts, on polyole®ns 102±3
264
Index
insertion reactions by propylene monomers 112, 113 by styrene monomers 112 interfacial agents, functionalized polyole®ns as 7, 61, 68, 97, 179 IR spectra borane-containing polymers 27 chain-end unsaturated polypropylene 172, 173, 232 ethylene±1-octene±divinylbenzene (EO-DVB) terpolymer 151 hydroxylated polypropylene (PP-OH) 200 poly(1-octene) compared with PO-OH 90 PP-g-PMMA graft copolymers 200 PP-b-SMA diblock copolymers 232±3, 232 PP-g-SMA graft copolymers 94 isotactic polymer structure 2, 3
K
Kelen±TuÈdos method 32, 33, 110 keto groups, on surface-modi®ed polyethylene 75
L
laminates 74, 87 PP/PP-OH 87, 88 lanthanocene-based catalysts 180 late transition metal catalysts 3±4, 12±13, 58±60 compared with metallocene-based catalysts 60 Lewis acid catalysts 102±3 Lewis acids 21 complexes formed with alkyl halides 57 Lewis bases 21 lithiated p-methylstyrene-terminated polypropylene 190 lithiated poly(ethylene-co-pmethylstyrene) copolymers 119±25 preparation of 119 reactions 120, 203 living anionic graft-from reaction, via p-methylstyrene reactive sites 203±13 living anionic polymerization for poly(propylene-b-styrene) 192 transformation from Ziegler±Natta polymerization 189±93
transformation to Ziegler±Natta polymerization 193±4 living radical graft-from reaction, via borane reactive sites 198±202 living radical polymerization, transformation from metallocene polymerization 181±9 living transition metal coordination polymerization 155±6 limitations 156, 179, 181 in preparation of diblock copolymers 180±1 long-chain-branched polyethylene (LCBPE) 240 long-chain-branched polyole®ns 240±5 long-chain-branched polypropylene (LCBPP) 240±5 FTIR spectra 243 GPC curves 243, 244 graft density 244 intrinsic viscosity 244 molecular weight 244 radius of gyration 244 synthesis of 241, 242
M
maleation reactions in post-polymerization grafting 69, 72, 73, 227 of unsaturated polyole®ns 150±3 maleic anhydride, as functional monomer 69 maleic-anhydride-modi®ed ethylene±1octene±divinylbenzene terpolymers 151 maleic-anhydride-modi®ed ethylene± propylene copolymers (EP-MA) 73±4 maleic-anhydride-modi®ed polyethylene (PE-g-MA) bulk modi®cation 72±3 surface modi®cation 75 maleic-anhydride-modi®ed polyole®ns 93±5, 227±40 maleic-anhydride-modi®ed polypropylene (PP-g-MA) 69±71, 93±4, 151±2 applications 227 FTIR spectrum 243 graft-onto reaction with PP-t-NH2 241, 242 melting temperature 93, 152 molecular weight 93, 152
Index maleic-anhydride-terminated polyole®ns 230±4 maleic-anhydride-terminated polypropylene (PP-t-MA) 174, 175, 230 amine-terminated polypropylene prepared from 241 as compatibilizer in PP/nylon 11 blends 235, 237, 238 GPC curve 242 1 H NMR spectra 231 melting temperature 233 molecular weight 233 reactions with 3-aminopropyltrimethoxysilane 232 with nylon 195, 235, 237 MAO cocatalysts see methylaluminoxane... melt free radical grafting 65, 68±74 ethylene±propylene copolymer with maleic anhydride 73 polyethylene with maleic anhydride 72±3 polypropylene with maleic anhydride 69±71 metallocene-based catalysts 3, 5±6, 135 classi®cation 6 compared with late transition metal catalysts 60 see also titanocene; zirconocene metallocene polymerization for amino-group-containing monomers 44 for borane-containing monomers 24±30, 83 for borane-terminated polyethylene 160 in diene/ole®n co-/terpolymerization 138, 140, 141, 142, 144 ester-group-containing monomers 57 for graft copolymers 219±20 living Ziegler±Natta polymerization 180±1 for long-chain-branched polyole®ns 240±5 in p-methylstyrene/ole®n co-/ terpolymerization 107±8, 110, 111±13, 115, 116 reactive polyole®ns formed by 14, 107±8, 110, 111±13, 115, 116 termination reactions 6
265
transformation to living free radical polymerization 181±9 metals/non-metals separation line 21, 22 methacrylate-grafted polypropylene ®bers 76 methyl acrylate, copolymerization with ethylene 60 methylaluminoxane-based (MAO) cocatalysts 3 in amino-compound/ole®n copolymerization 44 in amino-compound/styrene copolymerization 50 in borane/ole®n copolymerization 28, 29, 30 in !-carboxyl--ole®n/-ole®n copolymerization 57, 60 in !-chloro--ole®n/-ole®n copolymerization 58 in diene/ole®n co-/terpolymerization 138, 140, 142, 144 in homopolymerization of 4-[N,N-bis(trimethylsilyl)amino]styrene 47, 48 in hydroxy-compound/ole®n copolymerization 54, 55 methyl methacrylate, in free radical polymerization 92, 185, 186 4-methyl-1-pentene, copolymerization with, 5-(N,N-diisopropylamino)1-penetene 44 methylphenyldiallylsilane copolymerization with propylene 37 polymerization of 37 p-methylstyrene (p-MS) 15, 105 advantages 105 copolymerization with ethylene 107±10 reactivity ratios 110±11 copolymerization with propylene 111±14 elastomers with 114±19 p-methylstyrene-containing polyole®ns functionalization of 119±33 graft copolymers prepared from 203 halogenation of 128±32 lithiation of 119±25, 205 and maleic anhydride grafting reactions 125±8, 229 oxidation of 132±3 preparation of 107±19, 203
266
Index
p-methylstyrene-terminated polypropylene 167±70 GPC curves 168±9, 169, 193 1 H NMR spectrum 169, 170, 190, 191 molecular weight 168±9, 168, 170, 190 p-methylstyrene reactive sites, living anionic graft-from reaction using 126, 203±13, 229
N
Natta, Giulio 2, 9 nitrile groups, on surface-modi®ed polyole®ns 77 nitrogen-containing monomers, polymerization of 41±53 nitroxide radicals 92±3 NMR spectra see 11B...; 13C...; 1H NMR spectra norbornene amino-compounds copolymerization with ethylene 42±3 terpolymerization with ethylene and propylene 43±4
O
1-octene copolymerization with, boranecontaining monomer 33±5, 89 homopolymerization of, effect of trialkylborane 24 7-octenyl-9-borabicyclononane (O-BBN), polymerization of 25±6, 27 organoboranes, as chain transfer agents 158 oxidation of poly(ethylene-co-p-methylstyrene) copolymers 132±3 surface functionalization using 75
P
peel strength polypropylene/aluminum laminates 96 polypropylene/glass laminates 97 4-penten-1-ol 54 4-pentenyl-9-borabicyclononane (P-BBN), polymerization of 25±6 per¯uoroborate cocatalysts 44, 45, 47, 159 Periodic Table 21, 22 peroxides
hal¯ives at grafting temperature 69 thermal decomposition of 66 peroxylboranes 92, 182, 198 phenols, copolymerization with, propylene 55 photosensitizers, in UV-induced surface grafting reactions 76 plasma processes, surface modi®cation by 76±7 polyalcohols 84±5 polyamide/polypropylene blends 235±40 poly(4-aminoethylstyrene) 13 C NMR spectra 48±9 DSC curve 50 melting temperature 50 poly(4-aminomethylstyrene) 13 C NMR spectrum 48±9 DSC curve 50 melting temperature 50 poly(4-aminostyrene) 13 C NMR spectra 48±9 DSC curve 50 melting temperature 50 poly(4-[N,Nbis(trimethylsilyl)aminoethyl]styrene) converted to poly(4-aminoethylstyrene) 48 glass transition temperature 46 intrinsic viscosity 46 melting temperature 46 molecular weight 47 poly(4-[N,N-bis(trimethylsilyl)aminomethyl]styrene) converted to poly(4-aminomethylstyrene) 48 glass transition temperature 46 intrinsic viscosity 46 melting temperature 46 poly(4-[N,N-bis(trimethylsilyl)amino]styrene) converted to poly(4-aminostyrene) 48 glass transition temperature 46, 47 intrinsic viscosity 46, 47 melting temperature 46, 47 poly(1-butene), borane-containing, preparation of 33, 83, 86 polycarbonate (PC), in PP/PC blends 222 poly[4-(dimethylaminomethyl)styrene] glass transition temperature 46 melting temperature 46
Index polyethylene attack by free radicals 14, 67 borane-containing, preparation of 28±30, 83 borane-terminated (PE-t-B) 160±1, 182 GPC curve 183 molecular weights 160, 161 crystallinity 86, 138 DSC curve 109, 139 functional-group-terminated 156 fusion enthalpy 86 GPC curve 109, 139 high-density (HDPE) DSC curve 139 GPC curve 139 and maleic anhydride grafting 127 melting temperature 1, 126 molecular weight 126 production of 2, 40 silane-containing 36 hydrophilicity (contact angle with water) 77 hydroxylated (PE-OH) 85±6 1 H NMR spectrum 147 with p-(1-hydroxypropyl)styrene units 123 hydroxy-terminated (PE-t-OH) 161 1 H NMR spectrum 184 linear low-density (LLDPE) molecular structure 85 production of 44 uses 7 long-chain-branched (LCBPE) 240 low-density (LDPE) ®rst produced 2 maleic-anhydride-modi®ed 126, 127 melting temperature 126 molecular weight 126 silane-containing 36 maleic anhydride grafted 72, 75 surface modi®cation of 74, 75, 76±7 thermal properties 86 vinyl-grafted ®lms 76 polyethylene diblock copolymers 180±1 transformation from metallocene to living free radical polymerization 182±4 poly(ethylene-co-1,4-hexadiene) 137±9 crystallinity 138 DSC curves 139, 148
267
GPC study 139, 148 H NMR spectrum 147 hydroboration of 145, 146 hydroxylation of 148 melting temperature 138 molecular weight 138 poly(ethylene-co-1-hexen-6-ol) copolymers crystallinity 85, 86 fusion enthalpies 86 melting temperature 85, 86 thermal properties 86 poly(ethylene-co-5-hexenyl-9-BBN) copolymers, GPC curves 30 poly(ethylene-co-p-methylstyrene) copolymers (PE-p-MS) 107±11 amidation of 123 brominated copolymers 128±32 carboxylated copolymers 121±2, 132±3 chlorinated copolymers 128±32 crystallinity 110 DSC curves 109, 124, 131, 132, 208 fusion enthalpies 133 glass transition temperatures 109, 110, 117 GPC curves 109, 123, 124, 131 graft copolymers prepared from 204, 206 1 H NMR spectrum 121 halogenation of 128±32 hydroxylated copolymers 123 lithiated preparation of 119 reactions 120, 203, 208 lithiation of 119±25 maleic-anhydride-modi®ed 127 1 H NMR spectrum 127, 129 melting temperature 126, 127, 208 molecular weight 126 melting temperature 110, 126, 133 oxidation of 132±3 silylation of 119±20 poly(ethylene-co-1-octene) copolymers 9-BBN-terminated 163±6 molecular weight 164, 165 glass transition temperature 117 molecular weight 117 poly(ethylene-ter-1-octene-ter-divinyl benzene) terpolymers 151 graft copolymers with polystyrene 217±19 1
268
Index
poly(ethylene-ter-1-octene-ter-divinyl benzene) terpolymers (cont.) IR spectrum 151 maleic anhydride modi®ed 151 poly(ethylene-ter-1-octene-ter-p-methyl styrene) terpolymers 116±19 DSC curves 118±19, 118 glass transition temperatures 117 molecular weights 117, 118 polyethylene-g-polyacrylonitrile graft copolymers (PE-g-PAN) 208±9 polyethylene-b-poly(methyl-methacrylate) diblock copolymers (PE-b-PMMA) GPC curves 183 1 H NMR spectra 184 molecular weight 183 preparation of 182±3 polyethylene-g-poly(methyl methacrylate) graft copolymers (PE-g-PMMA) 208±9 polyethylene-g-poly(p-methylstyrene) graft copolymers (PE-g-PMS) 205±6 fusion enthalpy 206 glass transition temperature 206 graft density 206 graft length 206 1 H NMR spectra 205±6, 205 melting temperature 206 polyethylene-b-polypropylene diblock copolymers (PE-b-PP) 194 polyethylene/polystyrene blends 222, 224±5 polyethylene-g-polystyrene graft copolymers (PE-g-PS) 204±8 DSC curves 208 as emulsi®ers in PE/PS blends 224, 225 fusion enthalpy 206 glass transition temperature 206, 208 graft density 206 graft length 206 1 H NMR spectra 204 melting temperature 206, 208 poly(ethylene-co-propylene) copolymers glass transition temperatures 115, 116 graft copolymers 201, 219±20 molecular weights 115 see also ethylene±propylene (EP) copolymers poly(ethylene-ter-propylene-ter-divinyl benzene) terpolymers 142±5
DSC curves 143±4, 144 glass transition temperatures 142, 143, 144 1 H NMR spectrum 142, 143 poly(ethylene-ter-propylene-ter1,4-hexadiene) terpolymer and EP-g-PMMA graft copolymer 201 1 H NMR spectra 202 poly(ethylene-ter-propylene-ter-pmethylstyrene) terpolymers 115±16 DSC curves 116 glass transition temperatures 115 molecular weights 115 poly(ethylene-co-propylene)-g-poly (methyl methacrylate) graft copolymers (EP-g-PMMA) 201±2 1 H NMR spectra 202 poly(ethylene-co-propylene)-g-poly (methyl methacrylate) graft copolymers (EP-g-PMMA) 201 poly(ethylene-co-propylene)-g-polypropylene graft copolymers (EP-g-PP) 219±20 glass transition temperature 220 melting temperature 220 poly(ethylene-ter-propylene-ter-p(3-butenyl)styrene) terpolymer 140 poly(ethylene-co-styrene) copolymers (PE-S) 9-BBN-terminated 163±6 molecular weight 164, 165 hydroxy-terminated 166 13 C (DEPT-135) NMR spectrum 166 maleic-anhydride-modi®ed 126, 127 melting temperature 126 molecular weight 126 melting temperature 126 molecular weight 126 poly(ethylene-co-5-vinyl-2-norbornene) 139±40 intrinsic viscosity 140 poly(1-hexen-6-ol) 84±5 DSC curve 36, 90 IR spectrum 90 thermal stability 88, 89 poly(5-hexenyl-9-BBN) 26±7, 84 poly(5-hydroxy-1-pentene) 54 poly(11-hydroxy-1-undecene) 54 polyisobutylene (PIB), formation of 103
Index poly(5-N-isopropylaminopentene) 42 poly(methyl methacrylate) (PMMA) in EP-g-PMMA graft copolymer 201 in PE-b-PMMA diblock copolymer 182±3 in PP-b-PMMA diblock copolymer 180, 185 in PP-g-PMMA graft copolymer 199±201 in PS-b-PMMA diblock copolymer 187±8 poly(methyl methacrylate)/polypropylene blends 220±2 poly(p-methylstyrene) (PMS), in graft copolymers 205±6 poly(1-octene) borane-containing preparation of 33±4, 83, 86 properties 34 DSC curve 90 glass transition temperature 34, 36 hydroxylated (PO-OH) 35, 89, 90 IR spectrum 90 poly(1-octene-co-1-hexen-6-ol) copolymers 35, 89 DSC curves 36, 90 poly(1-octene-co-5-hexenyl-9-BBN) copolymers esteri®ed, GPC curves 34, 35 glass transition temperatures 34 molecular weights 34 poly(1-octen-6-ol) 84±5 X-ray diffraction pattern 84 poly(7-octenyl-9-BBN) 26±7, 84 IR spectrum 27 polypropylene acrylonitrile-grafted ®lm 75±6 amine-terminated (PP-t-NH2) intrinsic viscosity 244 molecular weight 244 radius of gyration 244 reaction with PP-g-MA 241, 242 synthesis of 174, 241 atactic (a-PP), silyl-terminated 158, 159 attack by free radicals 14, 67 borane-containing PP-9-BBN 91, 93, 94 preparation of 30±3, 83, 86 selective auto-oxidation of 91±2 borane-terminated (PP-t-B) 174, 185, 230
269
oxidation of 174±5, 230 reactions 174±5 as catalyst support material 102±3 chain-end unsaturated (u-PP) DSC curve 235 GPC curves 173 1 H NMR spectra 231, 231 hydroboration of 172±4, 241 IR spectra 173, 232 melting temperature 233, 235 molecular weight 172, 233 synthesis of 172 p-chlorostyrene-terminated (PP-t-p-Cl-St) 170±1 molecular weight 171 co-crystallization with PP-OH 87, 88, 102 controlled rheology resins 14 DSC curves 88 GPC curves 169, 186 graft copolymers 199±201 hydrophilicity 77 hydrophilic PP/PP-OH membranes 98±102 hydroxylated (PP-OH) 32±3, 86±9 applications 95±103 DSC curves 88 IR spectrum 200 thermal stability 87±8, 89 intrinsic viscosity 69 isotactic (i-PP) 2, 3 blends with PMMA 220±2 production of 3, 40 silane-containing 36 thermal degradation of 172 thermal stability 88, 89 long-chain-branched (LCBPP) 240±5 FTIR spectra 243 GPC curves 243, 244 graft density 244 intrinsic viscosity 244, 245 molecular weight 244 radius of gyration 244 synthesis of 241, 242 maleic-anhydride-modi®ed (PP-g-MA) 69±71, 93±4, 151±2 applications 227 graft-onto reaction with PP-t-NH2 241, 242 intrinsic viscosity 244 melting temperature 93, 128 molecular weight 93, 128, 244
270
Index
polypropylene (cont.) radius of gyration 244 maleic-anhydride-terminated (PP-t-MA) as compatibilizer in PP/nylon 11 blends 235, 237, 238 coupling with nylon 195 GPC curve 242 1 H NMR spectra 231±2, 231 melting temperature 233 molecular weight 233 reaction with 3-aminopropyltrimethoxysilane 232 melting temperature 128 methacrylate-grafted ®bers 76 p-methylstyrene-terminated (PP-t-p-MS) 167±70 GPC curves 168, 169, 193 1 H NMR spectrum 169, 170, 190, 191 molecular weight 168±9, 168, 170, 190 silated product 190, 191 molecular weight 128, 168 surface modi®cation of 74, 75±7, 87 syndiotactic (s-PP) 155 vinyl-grafted ®bers/®lms 76 zinc-alkyl-terminated 157 see also hydrophilic PP/PP-OH membranes polypropylene/aluminum laminates, adhesion studies 96 polypropylene diblock copolymers 185±6 transformation from metallocene to living free radical polymerization 185±6 polypropylene/glass laminates, adhesion studies 96±8 poly(propylene-co-1,4-hexadiene), hydroboration of 145, 146 poly(propylene-co-5-hexenyl-9-BBN) copolymers 31±2 11 B NMR spectra 31 melting temperature 93 molecular weight 93 oxidation of 32, 33, 92 poly(propylene-co-p-methylstyrene) copolymers (PP-p-MS) 111±14 fusion enthalpies 114 lithiated 209, 213 maleic-anhydride-modi®ed 127
1
H NMR spectrum 127, 129 melting temperature 128 molecular weight 128 melting temperature 112, 114, 128 metallocene copolymerization 111±13 molecular weight 114, 128 Ziegler±Natta copolymerization 113±14 polypropylene-b-nylon 11 diblock copolymers 235, 237 polypropylene-g-nylon 11 graft copolymers 235, 237 polypropylene-g-polyacrylonitrile graft copolymers (PP-g-PAN) 213 glass transition temperature 213 melting temperature 213 polypropylene/polyamide blends 235±40 polypropylene-g-polybutadiene graft copolymers (PP-g-PB) 211±12 1 H NMR spectra 211±12, 212 polypropylene-b-poly(butyl acrylate) diblock copolymers (PP-b-PBA) 185 1 H NMR spectra 187 polypropylene-g-polycaprolactone (PP-g-PCL) graft copolymers 214±16 DSC curves 216 as emulsi®er in PP/PC blends 222, 223 1 H NMR spectra 215 melting temperature 216 polypropylene/polycarbonate (PP/PC) blends 222, 223 polypropylene-b-poly(ethyl methacrylate) diblock copolymers (PP-b-PEMA) 185 1 H NMR spectra 187 polypropylene/poly(methyl methacrylate) blends 220±2 polypropylene-b-poly(methyl methacrylate) diblock copolymers (PP-b-PMMA) 180, 185 GPC curve 186 polypropylene-g-poly(methyl methacrylate) graft copolymers (PP-g-PMMA) 199±201, 213 as emulsi®er in PP/PMMA blends 221, 222 glass transition temperature 213 1 H NMR spectra 212 IR spectra 200, 201 melting temperature 213
Index polypropylene-g-poly(p-methylstyrene graft copolymers (PP-g-PMS) 209±11 fusion enthalpy 210 glass transition temperature 210 graft density 210 graft length 210 1 H NMR spectra 210 melting temperature 210 polypropylene-b-polypropylene diblock copolymers (PP-b-PP), GPC curves 242 polypropylene-b-polystyrene diblock copolymers (PP-b-PS) 185 GPC curve 193 1 H NMR spectra 187, 193 molecular weight 192 polypropylene-g-polystyrene graft copolymers (PP-g-PS) 209±11 DSC curves 211, 211 fusion enthalpy 210 glass transition temperature 210, 211 graft density 210 graft length 210 1 H NMR spectra 210 melting temperature 210 polypropylene-b-poly(styrene-alt-maleic anhydride) diblock copolymers (PP-b-SMA) 175, 230, 233±4 as compatibilizer in PP/polyamide blends 237±8, 239, 240 DSC curves 234, 235 glass transition temperature 233, 234 IR spectra 232±3, 232 melting temperature 233 molecular weight 233 synthesis of 233±4 polypropylene-g-poly(styrene-alt-maleic anhydride) graft copolymers (PP-g-SMA) 70±1, 94±5 glass transition temperature 93, 95 IR spectra 94±5, 94 melting temperature 93 polystyrene, syndiotactic (s-PS) 7 aldehyde-group-containing 52±3 amino-group-containing 44±52 borane-terminated (s-PS-t-B) 161±2, 186 DSC curve 189 glass transition temperature 188 intrinsic viscosity 162 melting temperature 162, 188
271
molecular weight(s) 162, 189 oxidation of 162 in diblock copolymers 186±9, 190, 192 glass transition temperature 46 hydroxy-terminated (s-PS-t-OH) 162 intrinsic viscosity 46 melting temperature 46, 50, 53 silane-terminated (s-PS-t-OSi(CH3)3) 162±3 1 H NMR spectra 163, 188 transformation from metallocene to living free radical polymerization 186±8 poly(styrene-co-[1,3-dimethyl-2-(4vinylphenyl]imidazolidine) copolymers 50±2 imidazolidine converted to aldehyde groups 52±3 poly(styrene-alt-maleic anhydride) (SMA alternating copolymer) 70, 93 in block polymer with polypropylene 173, 230 grafted onto polyole®ns 70±1, 94±5, 229 polystyrene-b-poly(n-butyl methacrylate) diblock copolymers (s-PS-b-PBMA) 187±8 1 H NMR spectra 186, 188 polystyrene-b-polyn-butyl methacrylate) diblock copolymers (s-PS-b-PBMA) molecular weight 189 reaction conditions 189 polystyrene/polyethylene blends 222, 224±5 polystyrene-b-poly(methyl methacrylate) diblock copolymers (s-PS-b-PMMA) 187±8 DSC curve 189 glass transition temperature 188 1 H NMR spectra 186, 188 melting temperature 188 molecular weight 189 reaction conditions 189 poly(vinyl alcohol), thermal instability 84±5 post-polymerization approach 10, 13±14, 63±78 advantages 65 disadvantages 65 side reactions 10, 14, 67±8, 78, 227
272
Index
propylene copolymerization with borane-containing monomers 30±3 8-bromo-1-octene 58 hydroxy-containing monomers 54 p-methylstyrene 111±14 silane-containing monomers 37 2,2,6,6-tetramethylpiperidinecontaining monomers 42±3 polymerization of with alkenylzinc compounds 157±8 with metallocene catalysts 172, 219±20 with p-methylstyrene and hydrogen chain transfer agents 167±8, 219 terpolymerization with ethylene and amino-containing norbornene 43±4 and p-(3-butenyl)styrene 140 and !-chloro--ole®ns 58 and divinylbenzene 142±5 and hydroxy compounds 54±5 and p-methylstyrene 115±16 and TMA-protected carboxylic acids 57, 60 and TMA-protected hydroxyole®ns 55, 60 see also polypropylene protection of functional groups 12, 39, 40±1 by electronic/steric protection 12, 40, 42, 45, 54 by spacer groups 41, 42, 55, 58 by steric groups 12, 40, 45, 54, 60
R
radiation-induced free radical grafting reaction 75±6 reactive extrusion 70, 81 reactive groups, requirements 14±15 reactive polyole®n approach 10±11, 14±16, 79±172 reactive polyole®ns borane-containing 15, 81±104 as `intermediates' in functionalization reactions 14 p-methylstyrene-containing 15, 105±33 requirements 14±15 with unsaturated groups 135±53
reactivity ratios 5-hexenyl-9-BBN and ole®ns 32, 33 p-methylstyrene anad ole®ns 110±11 references listed 247±56 re¯ection IR spectra polypropylene/glass laminates 97±8 silica glass 98 ring-opening graft-from reaction 213±16 ring-opening polymerization 213 RI (refractive index) detectors, in GPC 35
S
silane-containing monomers, polymerization of 36±7 silane-containing polyole®ns 21, 23, 36, 38 microstructures 23 silane-terminated polystyrene 162±3 1 H NMR spectra 163, 188 silazanic monomers, polymerization of 42 silica glass, re¯ection IR spectra 98 silicon, position in Periodic Table 21, 22 silyl-terminated polyole®ns 158 solvents, effect on Ziegler±Natta polymerization 11±12 spacer groups, protection of functional groups by 41, 42, 55, 58 steric groups, protection of functional groups by 12, 40, 45, 54, 60 steric interactions, in metallocene copolymerization of polypropylene and p-methylstyrene 113 styrene as comonomer with maleic anhydride 70 copolymerization with 1,3-dimethyl-2(4-vinylphenyl)imidazolidine 50±1 homopolymerization of 45, 46 styrene-derivative-terminated polyole®ns 166±71 styrene/maleic-anhydride alternating copolymer 70, 93, 229 in block polymer with polypropylene 173, 230 grafted onto polyole®ns 70±1, 94±5, 229 styrene monomers, amino-groupcontaining, polymerization of 44±50
Index surface modi®cation 66, 74±7 by chemical processes 75, 87 by corona/plasma processes 76±7 of hydrophilic PP membranes 98, 99 in PP/Al and PP/glass laminates 96 by radiation/photo processes 75±6 syndiotactic polymer structure 3
T
tacticity 2, 3 borane-containing polymers 27 terminal functional groups, polyole®ns with 15±16, 155±75 termination reactions 6, 155±6 2,2,6,6-tetramethylpiperidine-containing monomers, copolymerization with ole®ns 42±3 2,2,6,6-tetramethyl-1-piperidinyloxy (TEMPO) radical 92±3 TGA curves, polypropylene/PP-OH/ poly(1-hexen-6-ol) 89 titanocene-based catalysts 3, 5 in amino-compound/styrene copolymerization 50±3 amino-group-containing styrene monomers polymerized using 44±50 comparison of catalyst/cocatalyst systems 47 in diene/ole®n co-/terpolymerization 141, 144 in p-methylstyrene/ole®ns terpolymerization 115, 116 transition metal coordination polymerization 2±6 living polymerization 155±6 termination reactions 6 see also early...; late transition metal catalysts trialkylboranes amination of 148, 174 reactions 174 stability with transition metal catalysts 24 trimethylallylsilane 37 polymerization of 36 trimethylaluminum (TMA) carboxylic acid groups protected by 57, 60 hydroxy groups protected by 55, 60 trimethylsiloxy-containing polyethylene, 1 H NMR spectrum 147
273
4-trimethylsiloxy-1,6-heptadiene, homopolymerization of 54 trimethylsilyl-containing PE-p-MS copolymers 119 DSC curves 124, 125 GPC curves 123, 124 1 H NMR spectra 119±20, 121 trimethylsilyl-protected amino groups 42, 45 conversion to primary amino groups 48 trimethylsilyl-protected hydroxy groups 54 tris(2,2 0 ,2 00 -nona¯uorobiphenyl)borane, as cocatalyst in homopolymerization of 4-[N,N-bis(trimethylsilyl)amino] styrene 47, 48
U
10-undecen-1-oic acid, protection of carboxylic acid group 57 10-undecen-1-ol, protection of hydroxy group 54 unsaturated groups, reactive polyole®ns with 135±53 unsaturated polyole®ns crosslinking of 149±50 functionalization of 145±53 by epoxidation 153 by hydroboration 145±8 by maleation 150±3 uses (of polyole®ns) 1 UV detectors, in GPC 35 UV light, surface modi®cation using 76
V
vinyl arylate, in free radical polymerization 185 vinyl chloride, copolymerization with -ole®ns 58 vinylethylsilane 37 polymerization of 36±7 vinyl-modi®ed PE/PP ®bers/®lms 76 5-vinyl-2-norbornene (VNB), copolymerization with ethylene 139±40 4-vinylpyridine (4-VP), polymerization of 43 vinyltrimethylsilane 37 polymerization of 36±7
274
Index
W
wettability of polyole®n surfaces, effect of plasma treatment 76±7
Z
Ziegler, Karl 2 Ziegler±Natta catalysts 4±6 ®rst developed 2 metallocene-based 3, 5±6 polymerization mechanism 4±5 Ziegler±Natta polymerization and amino-group-containing monomers 41±4 and borane-containing monomers 24±35, 83, 86 chain transfer reactions 157 dienes/ole®ns co-/terpolymerization 138, 140, 141 and hydroxy-group-containing monomers 54±5 living polymerization 155 p-methylstyrene/propylene copolymerization 113±14
and silane-containing monomers 36±7 transformation from living anionic polymerization 193±4 transformation to living anionic polymerization 189±93 Ziegler polymerization 2 Zimm±Kilb branching parameter 245 zinc-alkyl-terminated polyole®ns 157±8 zirconocene-based catalysts 3, 5 in amino-compound/ole®n copolymerization 44 in borane/ole®n copolymerization 28±30 in !-chloro--ole®ns/-ole®ns copolymerization 58 in dienes/ole®ns co-/terpolymerization 138, 140, 141, 142, 145 in hydroxy-compound/ole®n copolymerization 54±5 in p-methylstyrene/ole®n copolymerization 107±8, 107, 111±12