nlreader.dll?bookid=12873&filena...

页码，1/1Document

2004-4-27http://www.netlibrary.com/nlreader/nlreader.dll?bookid=12873&filena...

Anionic Polymerization



SecLover

Pencil

PLASTICS ENGINEERING

Founding Editor

Donald E. Hudgin

Professor Clemson University

Clemson, South Carolina

1. Plastics Waste: Recovery of Economic Value, Jacob Leidner

2. Polyester Molding Compounds, Robert Burns

3. Carbon Black-Polymer Composites: The Physics of Electrically Conducting Composites, edited by Enid Keil Sichel

4. The Strength and Stiffness of Polymers, edited by Anagnostis EZ achariades and Roger S. Porter

5. Selecting Thermoplastics for Engineering Applications, Charles P. MacDermott

6. Engineering with Rigid PVC: Processability and Applications, edited by I. Luis Gomez

7. Computer-Aided Design of Polymers and Composites, D. H. Kaelble

8. Engineering Thermoplastics: Properties and Applications, edited by James M. Margolis

9. Structural Foam: A Purchasing and Design Guide, Bruce C. Wendle

10. Plastics in Architecture: A Guide to Acrylic and Polycarbonate, Ralph Montella

11. Metal-Filled Polymers: Properties and Applications, edited by Swapan K. Bhattacharya

12. Plastics Technology Handbook, Manas Chanda and Salil K. Roy

13. Reaction Injection Molding Machinery and Processes, F. Melvin Sweeney

14. Practical Thermoforming: Principles and Applications, John Florian

15. Injection and Compression Molding Fundamentals, edited by Avraam I. Isayev

16. Polymer Mixing and Extrusion Technology, Nicholas P. Cheremisinoff

17. High Modulus Polymers: Approaches to Design and Development, edited by Anagnostis E. Zachariades and Roger S. Porter

18. Corrosion-Resistant Plastic Composites in Chemical Plant Design, John H. Mallinson



19. Handbook of Elastomers: New Developments and Technology, edited by Anil K. Bhowmick and Howard L. Stephens

20. Rubber Compounding: Principles, Materials, and Techniques, Fred W. Barlow

21. Thermoplastic Polymer Additives: Theory and Practice, edited by John T. Lutz, Jr.

22. Emulsion Polymer Technology, Robert D. Athey, Jr.

23. Mixing in Polymer Processing, edited by Chris Rauwendaal

24. Handbook of Polymer Synthesis, Parts A and B, edited by Hans R. Kricheldorf

25. Computational Modeling of Polymers, edited by Jozef Bicerano

26. Plastics Technology Handbook: Second Edition, Revised and Expanded, Manas Chanda and Salil K. Roy

27. Prediction of Polymer Properties, Jozef Bicerano

28. Ferroelectric Polymers: Chemistry, Physics, and Applications, edited by Hari Singh Nalwa

29. Degradable Polymers, Recycling, and Plastics Waste Management, edited by Ann-Christine Albertsson and Samuel J. Huang

30. Polymer Toughening, edited by Charles B. Arends

31. Handbook of Applied Polymer Processing Technology, edited by Nicholas P. Cheremisinoff and Paul N. Cheremisinoff

32. Diffusion in Polymers, edited by P. Neogi

33. Polymer Devolatilization, edited by Ramon J. Albalak

34. Anionic Polymerization: Principles and Practical Applications, Henry L. Hsieh and Roderic P. Quirk

35. Cationic Polymerization: Mechanisms, Synthesis, and Applications, edited by Krzystof Matyjaszewski

Additional Volumes in Preparation

Macromolecular Design of Polymeric Materials, edited by Koichi Hatada, Tatsuki Kitayama, and Otto Vogl

Polyimides: Fundamentals and Applications, edited by Malay K. Ghosh and K. L. Mittal

Thermoplastics Melt Rheology and Processing, A. V. Shenoy and D. R. Saini

Practical Thermoforming: Second Edition, Revised and Expanded, John Florian



Page i

Anionic PolymerizationPrinciples And Practical Applications

Henry L. Hsieh Phillips Petroleum Company

Bartlesville, Oklahoma

Roderic P. Quirk The Maurice Morton Institute of Polymer Science

The University of Akron Akron, Ohio

MARCEL DEKKER, INC. NEW YORK • BASEL



Page ii

Library of Congress Cataloging-in-Publication Data

Hsieh, Henry L. Anionic polymerization: principles and practical applications/ Henry L. Hsieh, Roderic P. Quirk. p. cm. Includes index. ISBN-8247-9523-7 1. Addition polymerization. 2. Carbanions. I. Quirk, Roderic P. II. Title.QD281.P6H75 1996 547'.28—dc20 95–53987

CIP

The publisher offers discounts on this book when ordered in bulk quantities. Formore information, write to Special Sales/Professional Marketing at the address below.

This book is printed on acid-free paper.

Copyright © 1996 by Marcel Dekker, Inc. All Rights Reserved.

Neither this book nor any part may be reproduced or transmitted in any form or byany means, electronic or mechanical, including photocopying, microfilming, and recording, or by any information storage and retrieval system, without permission in writing from the publisher.

Marcel Dekker, Inc. 270 Madison Avenue, New York, New York 10016

Current printing (last digit):10 9 8 7 6 5 4 3

PRINTED IN THE UNITED STATES OF AMERICA



Page iii

To Diana and Donna for their patience and encouragement



Page v

PREFACE

The goal of polymer synthesis is to prepare a variety of polymers with control of the major variables that affect polymer properties. Chain reaction polymerizations, which proceed in the absence of chain termination and chain transfer reactions (so-called living polymerizations), provide the maximum degree of control for the synthesis of polymers with predictable, well-defined structures. Within the scope of monomers that can be polymerized anionically, living anionic polymerization, especially alkyllithium-initiated polymerization of styrene and diene monomers, provides the most versatile and elegant methodology for the synthesis of polymers with well-defined structures. For those monomer/initiator/solvent systems that proceed in the absence of termination and chain transfer, polymers can be prepared with control of molecular weight, molecular weight distribution, copolymer composition and microstructure, tacticity, chain-end and in-chain functional groups, architecture, and morphology.

The goal of this monograph is to provide the reader with the necessary background on both the fundamental and practical aspects of anionic polymerization to understand the scope and limitations of anionic polymerization of vinyl and diene monomers for controlled polymer synthesis and to read the relevant literature with critical insight. The process of choosing material for discussion has been selective rather than encyclopedic. For example, the major emphasis is on the organolithium-initiated polymerization of



Page vi

styrene and diene monomers, because these systems have proved to be the most important and useful for controlled polymer synthesis in academic research as well as in commercial applications. The controlled polymerization of methyl methacrylates and ring-opening block polymerization are also described.

No attempt has been made to provide a comprehensive review of all aspects of carbanion chemistry and anionic polymerization: the criterion used for literature selection has been the perceived value with regard to illustrating important concepts or generalities. Similarly, the examples of commercial products in Part V were selected based on the accessibility of reliable technical information for these products and on their usefulness to illustrate the general structure-property relationships of commercial materials. We take responsibility for any errors of omission and commission. To those authors whose work may have been slighted, we can only apologize.

Many aspects of anionic polymerization of vinyl and diene monomers are the subjects of current controversy. We have endeavored to rise above these controversies by providing relevant experimental data and attendant ambiguities without engaging in or giving credence to the polemics that sometimes accompany the data in the literature.

We wish to acknowledge the generosity of Phillips Petroleum Company for allowing one of us (H. L. H.) to publish a wealth of scientific information on anionic polymerization and technologies for commercial applications of anionically prepared polymers and for providing general support and encouragement during the preparation of this work. Henry Hsieh would also like to thank Teri Baldwin of Bartlesville, Oklahoma for her invaluable secretarial assistance and to acknowledge the hospitality of the faculty of the Department of Polymer Science, the University of Akron, during his stay as a Visiting Professor in 1993 to complete this manuscript.

Roderic Quirk would like to acknowledge the hospitality of Dr. Paul F. Rempp and his colleagues, especially Emile Franta and Yves Gnanou, at the Institut Charles Sadron, C.N.R.S., Strasbourg, France, during his stay as a Visiting Professor in 1991, during which time most of the initial writing was accomplished for Chapters 1–14. Dr. Quirk would also like to acknowledge the careful reading and helpful suggestions by Dr. Lewis J. Fetters (Exxon) and the following scientists at Bridgestone/Firestone: Dr. Tom Antkowiak, Dr. Pete Bethea, Dr. David Lawson, Dr. Tom Lynch, and Dr. James Oziomek. A special debt of gratitude is owed to his graduate students (the “anionic groupies”) for their considerable efforts in proofreading and authenticating references: Gilda Lizarraga, Dr. Jianxin Kuang, Jungsoo Kim, Qizhuo Zhuo, Huimin Yang, Yuhsin Tsai, Taejun Yoo, Shane Porzio, Li-Mei Lu, Jinsong Li, Sung Hoon Jang, Hong Dixon, and Kwansoo Han.

HENRY L. HSIEH RODERIC P. QUIRK



Page vii

CONTENTS

Preface v

I Structure and Bonding in Carbanionic Compounds

1 Structures of Carbanions and Organometallic Compounds

3

2 Stabilities of Carbanionic Species

33

3 Ion Pairs, Free Ions, and Stereochemistry in Carbanionic Chemistry

47

II Introduction to Anionic and Living Polymerization

4 Living Polymerizations: Definitions, Consequences, and Criteria

71

5 General Aspects of Anionic Polymerization

93

III Kinetics and Mechanism in Anionic Polymerization

6 Initiation Reactions in Anionic Polymerization: Kinetics of Addition of Organolithium Compounds to Vinyl Monomers

131

7 Propagation Reactions: Kinetics and Mechanism for Styrenes and Dienes in Hydrocarbon Solvents with Lithium as Counterion

155



Administrator

Underline

Administrator

Highlight

Page viii

8 Termination and Chain Transfer Reactions

173

9 Stereochemistry of Polymerization

197

10 Copolymerization

237

IV Anionic Synthesis of Polymers with Well-Defined Structures

11 Functionalized Polymers and Macromonomers

261

12 Block Copolymers

307

13 Star Polymers

333

14 Graft Copolymers

369

V Commercial Applications of Anionically Prepared Polymers

15 Commercial Applications of Anionically Polymerized Products

395

16 Polydiene Rubbers

421

17 Styrene-Diene Rubbers

447

18 Styrenic Thermoplastic Elastomers

475

19 Applications of Styrenic Thermoplastic Rubbers in Plastics Modifications, Adhesives, and Footwears

533

20 Clear Impact-Resistant Polystyrene

587

21 Nonfunctional Liquid Polybutadienes

607



Administrator

Underline

Administrator

Highlight

YFP

Highlight

I STRUCTURE AND BONDING IN CARBANIONIC COMPOUNDS



1 Structures of Carbanions and Organometallic Compounds

I. General Considerations

The chemistry of anionic polymerization of vinyl monomers is based on the fundamentals of carbanion chemistry. In order to understand the kinetics and mechanisms of anionic polymerization, it is necessary to consider the fundamental aspects of carbanion structure-stability-reactivity relationships. In this chapter the theoretical and experimental information relating to the structures of carbanions and organometallic compounds will be summarized.

A carbanion is defined as the conjugate base of a carbon acid: the species derived from an organic molecule by heterolytic fission of a carbon—hydrogen bond [1–3]. Experimental information regarding the structure and stability of simple unsolvated, free (i.e., without encumbering counterions) carbanions is quite limited. Structural information for carbanions in solution and in the solid state generally refers to species strongly complexed and interacting with counterions and sometimes solvent molecules (i.e., organometallic compounds in which the organic groups are formally attached to metal atoms by metal-carbon bonds). The term “carbanionoid” has been suggested to describe polar organometallic compounds [4]. X-ray crystallographic studies have provided the most detailed structural information available for a variety of carbanionic (generally organo-metallic) species [5–8]. It is generally assumed that these structures can be used as models for the corresponding species in solution. However, there is the possible



YFP

Highlight

YFP

Highlight

YFP

Underline

complication that crystal lattice interactions in the solid state can perturb the structure compared with those in solution or in the gas phase. For example, although ethyllithium exists predominantly as a hexameric species in hydrocarbon solution [9], the tetramer is observed in the solid state [10,11].

II. Alkyl Carbanions: Theoretical Results

A simple sp3-hybridized carbanion is isoelectronic with ammonia and on the basis of simple hybridization theory a pyramidal, sp3-hybridized structure would be predicted for the methyl anion [2]. In general, the available theoretical and experimental structural information is consistent with a pyramidal structure for simple alkyl carbanions [8,12]. However, the inversion barrier (Figure 1.1) for the methyl anion is calculated to be in the range of only 0–1.9 kcal/mole, which is consistent with estimates based on the vibronic structure of the photoelectron spectrum for CH3- in the gas phase

[13]. The corresponding inversion barrier in ammonia is 6 kcal/mole [14]. Schleyer and co-workers [14] have calculated

that the inversion barrier is only 1.5 kcal/mole and that the HCH bond angle is 109.0°. The small inversion barrier

represents the energy difference between the pyramidal and the planar forms, since the transition state for inversion

would correspond to the planar structure (see Figure 1.1). Furthermore, the electron affinity of the methyl anion with

respect to loss of an electron has been estimated to be only 1.8 ± 0.7 kcal/mole [13,15], which has been interpreted by

Schade and Schleyer [8] to indicate that the geometry of the methyl anion should be similar to the planar methyl radical.

In contrast to the methyl anion, the ethyl anion has not been observed experimentally and is predicted to be unstable (by 3.8–7.2 kcal/mole) with respect to electron loss [2,16]. It is predicted to be less stable than the methyl anion because of the destabilizing electronic effect of the methyl group (see Chapter 2).

III. Bonding in Organoalkali Compounds

The unique characteristics of alkyllithium compounds as initiators and lithium as a counterion in anionic polymerization relative to other organoalkali compounds [17–19] dictate a detailed consideration of their structure and bonding. From a

Figure 1.1 Structure of methyl carbanion and the pyramidal inversion process.



theoretical perspective, the approach of a counterion would be expected to perturb the electron density of a carbanion such that charge density would tend to be localized in the region between the cation and anion (see Structure 1.1) [8]. Organo-

lithium compounds are unique among the organic derivatives of the alkali metals since they generally exhibit properties characteristic of both covalent and ionic compounds [4,20]. In general, simple alkyllithium compounds exist as aggregated species in the solid state, in solution, and even inthe gas phase [6,21–24]. Thus, the persistence of tetrameric structures in the gas phase has been deduced by the presence of R3Li4+ ions using electron-impact mass spectrometry [25–27]. A notable exception is

gaseous bis(trimethylsilyl)methyllithium, which is monomeric at 140°C [28]. Although other alkali metal compounds

are also aggregated in solution and in the solid state, they have negligible vapor pressure and, unlike many alkyllithium

compounds, they are generally insoluble in hydrocarbon solvents [20]. Methyl derivatives of potassium, rubidium, and

cesium are ionic compounds with quite different structures and properties compared to the corresponding alkyllithium

compounds [5,8]. Lithium is unique among the alkali metals in having the highest electron affinity, electronegativity,

and ionization potential, coupled with the smallest covalent and ionic radius in the group, and the availability of

relatively low-lying unoccupied p orbitals for bonding [6,29].

The exact nature of the bonding in alkyllithium compounds is a subject of considerable interest and controversy [4,8,12]. A variety of theoretical and experimental evidence has been presented regarding the relative amount of ionic and covalent character of the carbon-lithium bond. Part of the problem is associated with semantics, since “An unambiguous and generally acceptable definition of covalent and ionic is hard to achieve” [12]. On the basis of calculations based on an electrostatic model that minimizes coulombic energy, for example, Streitwieser [30] concluded that the x-ray structure of the methyllithium tetramer is consistent with a completely ionic model consisting of a tetrahedron of four carbon-negative point charges with an interpenetrating tetrahedron of four lithium-positive point charges. Furthermore, Streitwieser et al. [31,32] have performed ab initio self-consistent field molecular orbital (SCFMO) calculations for monomeric methyllithium from which it was concluded that the carbon-lithium bond can be de-



scribed as a charge transfer complex with essentially no shared electron covalent character, since the degree of electron transfer was calculated to be 0.8e. The results from recent, more elaborate ab initio calculations with near Hartree-Fock basis sets plus configuration interaction indicated that the degree of charge separation is in the range of 0.55–0.60 electrons for monomeric methyllithium [33]. The atomic charges on lithium were calculated to be +0.23, +0.25, and +0.15 electronic units and the atomic charges on carbon were calculated to be -0.27, -0.31, and -0.23 electronic units for the dimer, tetramer, and hexamer, respectively [34]. These calculations offer a possible rationalization for the increased reactivity of less aggregated alkyllithiums (e.g., in initiation of anionic polymerization; see Chapters 4 and 6): the less aggregated species, in general, have a somewhat higher degree of ionic character in the carbon-lithium bond. The enthalpies of dissociation of the tetramer to dimers and unaggregated species were calculated to be 36.4 and 100.4 kcal/mol, respectively [34]. However, another recent ab initio calculation study at the restricted Hartree-Fock level concluded that monomeric methyllithium is 88% ionic and that the bonding in oligomers is largely electrostatic and more ionic than in the monomer [35]. The enthalpies of dissociation of the tetramer to dimers and to unaggregated species were calculated to be 34.3 and 122.9 kcal/mol, respectively [35]. Thus, sophisticated ab initio calculations by different groups provide quite different interpretations with respect to the nature of carbon-lithium bonding [12].

Perhaps the most important information regarding the nature of carbon-lithium bonding comes from direct experimental evidence. A number of nuclear magnetic resonance (NMR) spectroscopic investigations of alkyllithium compounds in solution have shown the presence of both 13C-7Li and 13C-6Li scalar coupling interactions between the alpha carbon and lithium using both lithium (6Li and 7Li) and carbon (13C) NMR data [36–43]. Schleyer and co-workers [42,43] have concluded that these coupling constants depend primarily on the state of aggregation (n) as shown in Equations 1.1 and 1.2. In general, however, n

J(13C,6Li) = 1/n(17 ± 2)Hz (1.1)

J(13C,7Li) = 1/n(45 ± 5)Hz (1.2)

represents the number of lithium atoms directly bound to a given 13C in terms of the NMR time scale [40]. For example, in a static tetrameric aggregate (slow intra-aggregate exchange limit), a carbon interacts with only the three lithium atoms in that tetrahedral face; however, in the fast intra-aggregate exchange limit, the carbon effectively interacts with all four lithium atoms in the aggregate [40].

Since ionic bonding should provide only relatively small coupling constants, the partially covalent nature of the carbon-lithium bond is indicated by these observations as well as INDO-type molecular orbital calculations of these coupling constants for various methyllithium aggregates [44]. However, Streitwieser



and co-workers [31] have proposed that 13C-7Li coupling constants can be explained by spin polarization and that they are not necessarily indicative of covalent bonding, in accord with their premise of ionic carbon — lithium bonding [30]. In NMR spectroscopy, this spin polarization mechanism of transmittal of spin information between nuclei is termed through-space coupling [45]. Although transmittal by a purely through-space mechanism is possible, it would be “unusual” [45]. Given the fact that structural models for aggregates in the solid state provide evidence for lithium-carbon bond distances within their respective covalent radii (see next section), an ad hoc through-space rationalization for the 13C-7Li coupling constants appears to be unnecessary. It is significant to note that there is no evidence for scalar 6Li-7Li coupling in these compounds, indicating the absence of significant bonding interaction between the lithium atoms [46]. This indicates that the electron density is primarily concentrated in regions to maximize Li-C-Li bonding interactions.

Schade and Schleyer [8] have concluded that the very high dipole moment (6D) [47] of monomeric methyllithium indicates a high degree of ionic character of the carbon-lithium bond; however, this dipole moment was estimated from the experimentally observed force constant via a two-point curve of bond length vs. force constant. Perhaps it is not surprising that such a dipole moment has been interpreted in terms of both a lithium cation-stabilized methyl anion [35] and a 40% covalent bond [33] via ab initio molecular orbital calculations.

Thus, there is no satisfactory, generally accepted description of the bonding in simple alkyllithium compounds [12], despite conclusions to the contrary [4,5]. Several salient aspects of carbon-lithium bonding should be emphasized, however:

1. Lithium is unique among the alkali metals, in terms of size, charge density, and atomic orbitals available for bonding. If significant covalent bonding interaction exists for any organoalkali compounds, it would be expected for lithium.

2. There is a discontinuity in properties and chemistry between organolithium compounds and the other organoalkali compounds. For example, poly-lithiation of many hydrocarbons occurs [48,49], and this argues against completely ionic descriptions of C-Li bonding [12]. Thus, a variety of procedures have been reported for the preparation of dilithiomethane [50], and perlithiated propyne can be prepared by the reaction of propyne with excess butyllithium [51].

3. Theoreticians using state-of-the-art, ab initio methods cannot agree on the nature of bonding in organolithium compounds [4,12,30–35]; thus, the complexity and ambiguity of the situation should be apparent.

4. The unique oligomeric structures (described in terms of electron-deficient bonding with bridging organic groups) and properties (such as hydrocarbon solubility and volatility) of organolithium compounds suggest that carbon-lithium bonding is not simply an ionic interaction, as with other organoalkali



compounds. However, Schade and Schleyer [8] have stated that this is an erroneous conclusion and these observations can be interpreted in terms of a small, inner cluster of lithium ions on the inside and nonpolar hydrocarbon moieties on the outside of the aggregates. This concept is illustrated by the reported hydrocarbon solubility of potassium salts of 2,3-dimethyl-3-pentanol and 3-ethyl-3-pentanol [52].

5. The unique nature of carbon-lithium bonding is directly responsible for the ability of only lithium among the alkali metals to polymerize 1,3-dienes to high 1,4-microstructure polydienes stereospecifically [17-19,53,54] (see Chapter 9).

Thus, the true nature of carbon-lithium bonding is perhaps best summarized by the conclusions that “Clearly, the C,Li bond cannot be as covalent as a C,H bond” [40] at one extreme, and that the C,Li bond is more covalent than the C,Met bonds for the heavier alkali metals, at the other.

IV. Structures in the Solid State

The x-ray crystal structures of a variety of simple alkyl organoalkali compounds have been determined [5–8]. In the solid state, methyllithium [55–57], ethyllithium [10,11], and t-butyllithium [58] are arranged into tetrameric aggregates that can be described as interpenetrating tetrahedra of alkyl groups and lithium atoms (see Figure 1.2). An alternative view that provides some insight into the bonding arrangement in these compounds is that the alkyl groups are located above each of the triangular Li3 faces of the tetrahedron formed by Li4. This picture is consistent with the proposal that these are electron-deficient compounds, in which the number of formal bonds (bonding atom — atom contacts) exceeds the number of available valence electron pairs [59,60]. The x-ray crystal structure of 3-lithio-1-methoxybutane is analogous to the tetrameric structures of methyllithium and ethyllithium, with each triangular Li3 face occupied by the C-3 carbon, except that each oxygen atom coordinates to lithium [61]. An analogous structure with nitrogen coordination to lithium has been found for 3-dimethylamino-1-lithiopropane [62]. When two heteroatoms are present in the same alkyllithium molecule, dimeric coordination is observed. Thus, 3-lithio-1,5-dimethoxypentane [63] is reported to exist as a dimer and 1,1-bis[(dimethylamino)methyl]-2-propyllithium [64] (see Figure 1.3) exhibits dimeric units by x-ray analysis. In contrast to the structures of these alkyllithiums, the crystal structures of other organoalkali compounds, such as the potassium, rubidium, and cesium analogs, are clearly ionic, with hexagonal structures corresponding to the nickel arsenide class [5,6,8]; however, methyl sodium exhibits a tetrameric structure analogous to methyllithium [65].

Cyclohexyllithium cocrystallizes with two molecules of benzene to form a hexameric structure in the solid state with lithium occupying the apices of a



Figure 1.2 Models for alkyllithium compounds based on x-ray crystal structures.



Figure 1.3 Model of the x-ray structure of 1,1-bis

[(dimethylamino)methyl]-2-propyllithium.

distorted octahedron, six cyclohexyl groups positioned above six of the eight octahedral Li3 faces, and a benzene molecule lying above each of the two vacant faces located on opposite sides of the octahedron as shown in Figure 1.2 [66]. The proposed bonding arrangement in the hexamer is analogous to the structure of the tetramers, with electron-deficient bonding of the alkyl groups to the Li3 faces and sp2-hybridized lithium. 1-Lithiomethyl-2,2,3,3-tetramethylcyclopropane crystallizes to form a similar hexameric structure; however, the two transoid faces without hydrocarbon groups remain unoccupied [67]. The x-ray crystal structure of n-butyllithium, which is a liquid at room temperature, has been examined at -90°C; this compound also exhibits the hexameric aggregate structure (distorted octahedral) with two empty and opposite triangular faces [58]. A hexameric structure for trimethylsilylmethyllithium has also been reported [68]. Although bis(trimethylsilyl)methyllithium is monomeric in the gas phase, in the solid state a polymeric structure is formed with only one methylene carbon bridging between two lithium atoms [28]. In contrast, the more hindered tris(trimethylsilyl)methyllithium crystallizes to form a dimeric structure [69].

Cocrystallization of alkyllithium compounds with chelating Lewis bases generally results in solid-state aggregates with lower degrees of aggregation. Thus, a dimeric structure has been determined for the N,N,N',N'-tetramethylethylenediamine (TMEDA) complex of 1-lithiobicyclo[1,1,0]butane [70]. The adduct of t-butyllithium with diethyl ether exists as a dimer consisting of a four-membered ring with two bridging alpha carbons of the t-butyl groups [58]. The steric demands of the t-butyl groups are probably responsible for the presence of only one solvating base unit and the absence of the normal tetrahedral coordination around lithium. A monomeric structure has been found for the pentamethyldiethylenetriamine (PMDETA) adduct of bis(trimethylsilyl)methyllithium [71].

The x-ray crystal structures of alkyllithium compounds can provide indirect information regarding the nature of carbon—lithium bonding. With the proviso that structures in the solid state may be distorted by crystal lattice interactions, the carbon—lithium bonding distances in alkyllithium aggregates in the solid state can



be compared with expected bonding distances for hypothetical covalent bonds. A compilation of carbon-lithium bond lengths in unsolvated alkyllithium compounds determined by x-ray crystallography is listed in Table 1.1. Analogous data for complexes with Lewis bases and intermolecular heteroatoms are listed in Table 1.2. Also listed in Table 1.1 is the expected carbon-lithium covalent bond distance (2.11 Å) based on the sum of the covalent radii of lithium (1.34Å) [72] and carbon (0.772Å) [73a].

The shortest lithium-carbon distances in n-butyllithium, t-butyllithium, methyllithium, ethyllithium, and cyclohexyllithium in the solid state are 2.14 [58], 2.15 [58], 2.19 [46], 2.31 [8], and 2.18Å [52], respectively (see Table 1.1). The

Table 1.1 Bond Lengths in Unsolvated Alkyllithium Compounds

Compound

Degree of Aggregation

C-Li (ºA)

Li-Li (ºA)

Reference

Model 2.11a 2.68b 72,73a

Bis (trimethylsilyl)methyllithium 1c 2.03 — 28

nd 2.14, 2.18, 2.21, 2.22, 2.27

28

Tris (trimethylsilyl)methyllithium 2 2.29 2.35 69

2.30

Methyllithium 4 2.31 2.68 57

Ethyllithium 4 2.19 2.42 10,11

2.25 2.60

2.47 2.63

t-Butyllithium 4e 2.15–2.37 2.38–2.43 58

Cyclohexyllithium 6f 2.18 2.397 66

2.30 2.968

n-Butyllithium 6d 2.14 2.41–2.44 58

2.18 2.90–2.97

3-(Lithiomethyl)-1,1,2,2- Tetramethylcyclopropane

6 2.12 2.16

2.46 2.98

67

2.30

Trimethylsilylmethyllithium 6 2.20 2.45 68

2.28 3.18

aSum of the covalent radii for carbon and lithium.



bMetallic radius for lithium.

cGas phase.

dCrystal structure for polymeric aggregate.

eData obtained at -90°C.

fAdduct with two benzene molecules cocrystallized with six cyclohexyllithiums.



Table 1.2 Bond Lengths in Lewis-Base Solvated and Heteroatom-Coordinated Alkyllithium Compounds

Compound

Degree of Aggregation

C-Li (ºA)

Li-Li (ºA)

Reference

Bis(trimethylsilyl)methyllithium (PMDETA)

1 2.13 71

Benzyllithium (THF) (TMEDA) 1 2.21 74

α-(Trimethylsilyl)benzyllithium (TMEDA)

1 2.13 75

α-(Phenylthio)benzyllithium (THF) 1 2.21 75

Triphenylmethyllithium (TMEDA) 1 2.23 76

Bicyclo[1,1,0]butan-1-yllithium (TMEDA)

2 2.23 2.74 70

1,1-Bis(dimethylamino)-methyl)-2- propyllithium

2 2.19 2.40 64

2.21

(t-BuLi)(Et2O) 2a 2.19 77

2.19

2-Lithio-2-methyl-1,3-dithiane(TMEDA)

2 2.19 78

Cyclopropyllithium (LiBr) 2 2.20 2.56–3.21 79

2.26

2.30

Methyllithium(TMEDA) 4 2.23 2.56–2.57 80

2.27

2.28

3-Dimethylamino-1-lithiopropane 4 2.24 2.469 62

2.27 2.539

2.28 2.467

3-Lithio-1-methoxybutane 4 2.29 2.51 61

2.36 2.47

(n-BuLi)( t-BuOLi) 4 2.15 77

2.22



Benzyllithium (triethylenediamine) n 2.21 81

Benzyllithium (Et2O) n 2.19–2.23 82

aData obtained at -90°C.



C-Li bond distance for monomeric bis(trimethylsilyl)methyllithium in the gas phase is only 2.03Å [25]. Thus, these observed bond distances are suggestively close to the sum of the covalent radii (2.11Å), which provides some support for significant orbital overlap and covalent bonding in alkyllithium compounds. However, this conclusion is mitigated by the data in Table 1.2, which indicate that similar bond distances (2.13–2.20Å) are observed for Lewis base and heteroatom complexes of analogous alkyllithium compounds. For the base complex, it would be expected that the carbon-lithium bond would be more polarized: more ionic than the corresponding unsolvated aggregates. Despite this observation, the carbon-lithium bond distances observed by x-ray crystallography in the solid state are within distances expected to result in significant orbital overlap, (i.e., covalent bonding interactions).

V. Structures in Hydrocarbon Solution

Alkyllithium compounds are generally soluble in hydrocarbon solution, unlike other organoalkali compounds [4,17,20,23]. This solubility is a consequence of the aggregation behavior of these compounds. Thus, a variety of colligative property studies have established that simple alkyllithium compounds are associated into dimers, tetramers, and hexamers in hydrocarbon solvents [4,21–23]. Structures analogous to those observed by x-ray crystallography have been proposed for the corresponding associated species in solution. The degrees of aggregation of organolithium compounds at room temperature have been determined by colligative property measurements and representative results are summarized in Table 1.3. It is noteworthy that methyllithium, vinyllithium, allyllithium, and phenyllithium are all insoluble in hydrocarbon solvents. The degree of association of organolithium compounds depends on the structure of the organic moiety, the solvent, the concentration, and the temperature. In general, simple, unhindered, straight-chain alkyllithium compounds such as n-butyllithium and ethyllithium are associated into hexamers in hydrocarbon solution. However, NMR evidence for higher degrees of association of primary organolithiums at lower temperatures has been presented [58]. The average degree of association is decreased by increasing steric bulk of the alkyl group; alkyllithium compounds with branching at either the α- or β-carbon tend to associate into tetramers. Both isopropyllithium and trimethylsilylmethyllithium exhibit concentration-dependent degrees of association in hydrocarbon solvents; a shift from tetrameric to hexameric association occurs as the concentration is increased [84]. In contrast, isopropyllithium is tetrameric in benzene [84]. It has been reported that 2-methylbutyllithium, although hexameric in pentane at 0.89 m (30°C), has a lower association number (3.2) at lower concentrations (0.048 m) and a higher association number (7.6) at lower temperatures (-12°C) [85]. The more hindered sec-butyllithium [86] and t-butyllithium [87] are tetrameric in hydrocarbon solution, and menthyllithium



Table 1.3 Aggregation States of Organolithium Compounds in Hydrocarbon Solution

Compound Solvent na Methodb Reference

C-2H5Li Benzene 6 F 90–92

Benzene 4.5–6.0 F 93

6.1 F 9,84

Cyclohexane 6.0 F 9,84,94

n-C4H9Li Benzene 6.3 I 95

6.0 F 9,84

Cyclohexane 6.2 I 95

n-C5H11Li Benzene 6.0 V 96

n-C8H17Li Benzene 6.0 V 96

CH2=CH-CH2CH2Li Cyclopentane 6.1 V 97

i-C3H7Li Cyclohexane 4.0 (<0.03 m)

F 84

>4 (>0.03 m) F 84

Benzene 4.0 (<0.1 m) F 84

>4 (>0.1 m) F 84

(CH3)3SiCH2Li Cyclohexane 6 F 84

Benzene 4 (<0.06 m) F 84

>4 (>0.06 m) F 84

CH3CH2CH(CH3)CH2Li Pentane 6 (0.89M) V 85

7.6 (-12°C) V 85

3.2 (0.048M) V 85

(CH3)2N(CH2)3Li Benzene 4 F 98

(CH3CH2)2N(CH2)3Li Benzene 4 F 98

CH3OCH2CH2(CH3)CHLi Benzene 4 I 61

sec-C4H9Li Cyclohexane 4 F 86

Benzene 4 F 86

t-C4H9Li Benzene 4 B 87



[88] is dimeric in hydrocarbon solution. Benzyllithium [9] and the analogous poly(styryl)lithium [89] have been reported to be dimeric in benzene solution. The general conclusions from these data are that the average degree of aggregation of organolithium compounds in hydrocarbon solvents is quite dependent on the structure of the organic moiety, the solvent, the concentration, and the temperature. Thus, the degree of association of alkyllithiums tends to decrease with increasing steric bulk of the alkyl group, delocalization of charge, decreasing concentration, increasing temperature, and substitution of an aromatic solvent for an aliphatic solvent. These results should be kept in mind when considering the effects of initiator structure, counterion, solvent, temperature, and concentration on polymerization kinetics, diene stereochemistry, and reactions of polymeric organolithium compounds.

VI. Structures in Polar Solutions and in the Presence of Lewis Bases

Lewis bases exert dramatic effects on the rate, stereochemistry, and reaction pathways in organolithium chemistry [23]. A partial explanation for these observations can be deduced from the effects of Lewis bases on the degree of association of organolithium compounds, as shown in Table 1.4. In general, the average degree of association of organolithium compounds is decreased in the presence of Lewis bases and polar solvents. Thus, most simple alkyllithium compounds that are hexameric in hydrocarbon solution (Table 1.3) are associated into tetramers in the presence of ethers and amines at room temperature; vinyllithium, phenyllithium, and allyllithium compounds are the exceptions in exhibiting dimeric association behavior in ethers at low concentrations. In contrast to the temperature dependence of the association behavior in hydrocarbon media [85] (see Table 1.3), decreasing temperature favors lower degrees of association in polar media [42,102,103]. Thus, at -108°C in THF, t-butyllithium is dimeric [42], sec-butyllithium exists as a monomer-dimer equilibrium (88/12) [42], and n-butyllithium is a mixture of dimers and trimers (81/19) [103]. NMR evidence has been presented that supports the observations of lower degrees of association at low temperatures in polar solvents [42,102,104] and that shows that the addition of N,N,N',N',N"-pentamethyldiethylenetriamine converts all dimers to monomers [42].

The nature of the interaction of Lewis bases and polar solvents with simple and polymeric organolithium compounds is important since in the presence of such compounds the high 1,4 stereospecificity of the lithium counterion with dienes is lost; polydienes with high 1,2 or 3,4 microstructures (high vinyl structures) are obtained (see Chapter 9) [17–19,53,54,105,106]. An order of basicity of Lewis base donors (B) toward alkyllithiums has been determined by measuring the heats of this interaction (∆H, kcal/mole) at 25°C in hydrocarbon solvents at



Table 1.4 Degree of Association of Organolithium Compounds in Polar Solvents

Compound Solvent na Reference

CH3Li THF 4 113

Et2O 4 113

C2H5Li Et2O 4 114

n-C4H9Li Et2O 4 113

THF 2.4 (D/T = 81/19) b 103

2.8 (D/T = 61/39) b 103

sec-C4H9Li THF 1.1 (0.158m) 42

M/D (88/12) b

t-Butyllithium THF 1.1 (0.172m) b 42

Allyllithium Et2O 2.0 (0.1M) 114

>10(>1.5M) 114,115

THF 2.1b 116

2-Methyl-2-phenylpropyllithium Et2O 2 117

trans-1-Propenyllithium Et2O 4 114

2-Propenyllithium Et2O 2 (0.04M) 114

4 (0.84M) 114

C6H5Li THF 2 113

1.6 (M/D = 39/61)b 42

C6H5CH2Li THF 1 113

(CH3)2NCH2CH2CH2Li 1,4-dioxane 1 98

9-(2-Hexyl)fluorenyllithium THF 1.39–1.63 99

aDegree of association

b108°C.

M, monomeric species; D, dimer; T, trimer.

low ratios of [Lewis base] / [lithium] using high dilution, solution calorimetry (Eq. 1.3) [17,107–109]. The observed general basicity order is THF > 2-CH3THF >



mically than triethylamine [17,107–109]. It is significant to note that a similar basicity order has been observed for polymeric organolithium compounds [17,110–112]. The highest enthalpies of interaction were observed for interactions with TMEDA and the enthalpies decreased in the order TMEDA > diglyme > THF > 2,5Me2THF > dioxane > tetramethylethylenediphosphine > Et3P > Et2O Et3N.

VII. Structure and Bonding in Benzylic, Allylic, and Enolate Anions

A. Benzyl Carbanions

The benzyl carbanion represents some ambiguity in terms of its structure. On the basis of hybridization, placing the unshared pair of electrons in an orbital with more s character would be favored [118]. However, stabilization by delocalization of charge into the aromatic benzene ring is optimized for sp2 hybridization of the benzylic carbon, and this is the minimum energy structure obtained by molecular orbital calculations, as depicted in Figure 1.4 [2,119].

The bonding in benzyllithium is not precisely defined. The only x-ray crystallographic structure determinations are for complexes with Lewis bases. Thus, the quinuclidene complex of benzyllithium consists of infinite chains of benzyllithium ion pairs bridged by triethylenediamine groups coordinated to the lithium cation [81]. The lithium cation is positioned asymmetrically within bonding distance of the benzylic carbon and both C(1) and one of the ortho carbons on the benzene ring, C(2), as depicted in Figure 1.5 by an orbital description of the interaction of lithium with the HOMO of the benzyl anion [81]. In contrast, the x-ray crystal structure of the diethyl ether complex of benzyllithium consists of infinite chains of alternating benzyl and lithium ions with lithium also

Figure 1.4 π Orbital picture of the benzyl anion.



Figure 1.5 HMO of the benzyl anion and the lithium cation.

coordinated to Et2O [82]. The shortest lithium-carbon distances in these complexes are 2.21Å and 2.19Å for the TMEDA [81] and Et2O [82] complexes, respectively (see Table 1.2). In both of these benzyllithium complexes, the benzyl ligands are planar with lithium ions in approximately symmetrical positions above and below the benzylic carbon atoms [5].

In contrast to these results, the adduct of benzyllithium with both THF and TMEDA is monomeric in the solid state and exhibits significant pyramidalization of the benzylic carbon atom [74]. This has been attributed to the charge-localizing effect of the small lithium cation [120]. Pyramidal benzylic carbons have also been observed in the x-ray crystal structures of both the TMEDA complex of α-(trimethylsilyl)benzyllithium [75] and the THF complex of α-(phenylthio)-benzyllithium [75] (see Table 1.2).

13C NMR studies of the 13C chemical shifts and J(13C-H) coupling constants of benzyllithium indicate appreciable sp3 character at the benzylic carbon in benzene solution compared to THF [74,120–122]. An empirical correlation has been observed between the magnitude of the J(13C-H) coupling constant and the percentage of s-character in the carbon hybrid orbital formally used to bond to hydrogen (see Eq. 1.4 [123,124]). Assuming that this relationship is applicable to

J(13C-H) = 5x (% s) [Hz] (1.4)

benzyllithiums, an sp3-hybridized CH2- group (tetrahedral) would be expected to exhibit a J(13C-H) = 125Hz compared to J(13C-H) = 167Hz for an sp2-hybridized CH2- group (planar, trigonal) [74,120]. The observed coupling constants of 134 Hz and 131Hz for benzyllithium [125,126] and the THF/TMEDA complex of benzyllithium [74], respectively, have been interpreted in terms of a pyramidal benzylic CH2- group [74]. The observation that J(13C-H) = 151Hz for the penta-



methyldiethylenetriamine (PMDTA) adduct of benzylpotassium in THF [120] is consistent with both a planar, trigonal structure for the potassium derivative and structural differences between the lithium and potassium compounds.

1H NMR studies of ring hydrogen chemical shifts indicate that there is more delocalization of charge in THF (0.6 electron) compared to benzene solution (0.3 electron) [2,122,127]. Ultraviolet-visible spectroscopic studies of benzyllithium indicate quite different absorbances in benzene (λmax = 292 nm) and in THF (λmax = 330 nm) [127]. The maximum absorbance peak is shifted to 362 nm for the free benzyl carbanion in THF [128,129]. It should be noted that benzyllithium is dimeric in benzene [9] and unassociated in THF [113] (see Tables 1.3, 1.4). These results suggest that there is considerably more interaction between lithium and the benzylic carbon in hydrocarbon solution resulting in less charge delocalization into the benzene ring, than in THF solutions or compared to other alkali metal counterions. However, this conclusion is attenuated by the fact that n-butyllithium exhibits quite different absorbances in hexane (λmax = 210 nm) and benzene (λmax = 278–282 nm) compared with Et2O (λmax = 240 nm) and THF (λmax = 270 nm) [130–132].

Oliver and co-workers [133] have reviewed 7Li NMR chemical shift data and have concluded that the bonding in benzyllithium is mainly ionic, based on chemical shift calculations and the relative magnitudes of the various shielding terms.

Poly(styryl)lithium (PSLi) is reported to be dimeric in benzene [89,134] and cyclohexane [135] solutions; in the presence of sufficient amounts of THF, conversion to the unassociated form occurs [89,134] (see Table 1.5). The bonding situation is even less clear for poly(styryl)lithium versus benzyllithium since the ultraviolet-visible absorption spectrum of PSLi exhibits λmax = 334 nm in benzene and λmax = 330 nm in THF [136]. It should be noted that the free benzyl carbanion exhibits λmax = 362 nm in THF [128,129]. This suggests that lithium causes considerable perturbation of the charge distribution even in THF.

B. Allylic Carbanions

The simple allylic anion presents a classic example of a highly delocalized and therefore presumably planar, sp2-hybridized π system (see Figures 1.6, 1.7) [2]. The isolated allyl anion is calculated to have a planar structure [2]. The structures of many organolithium compounds with p-delocalized carbanions interacting with lithium cations have often been interpreted in terms of multicenter covalent interactions between the unoccupied atomic orbitals of lithium and the appropriate occupied molecular orbitals of the carbanion. The highest occupied molecular orbital of allyllithium [6,7], as shown in Figure 1.8, consists of the highest occupied allyl molecular orbital and a vacant p orbital on lithium. The geometry of



Table 1.5 Degree of Association of Polymeric Organolithium Compounds

Pli Solvent na Reference

Poly(styryl)lithium Benzene 2 89

Cyclohexane 2 134,135

n-Hexane 137

Poly(isoprenyl)lithium n-Hexane 2 89,134

Benzene 2 89,134

Cyclohexane 2 134,138

Heptane 2, 3 139

Cyclohexane 2.8–4.1 135

Poly(butadienyl)lithium n-Hexane 2 89

2 100

Benzene 2.1b(0.0495m) 140

3.7b(0.33m) 140

2 134

2 and >100 141

Cyclohexane 2 138

Cyclohexane 3.9–4.3 135

aAverage degree of association.

bNeopentylallyllithium.

allyllithium predicted from ab initio calculations is shown in Figure 1.9; the central hydrogen is bent toward and the two inner hydrogens at C-1 and C-3 are bent away from the lithium [142,143]. Although this geometry could be duplicated when the valence orbitals on lithium were omitted, it was concluded that multicenter bonding of the carbon atoms with lithium does contribute to the total energy [6,7,144].

X-ray crystal structure data for allyllithium have been obtained for the TMEDA complex [145] and the pentamethyldiethylenetriamine complex [146].



Figure 1.6 Allyl bonding molecular orbital.



Figure 1.7 Allyl anion resonance structures.

In the TMEDA complex, the terminal C atoms of the allyl groups are linked to Li atoms forming polymeric chains as shown schematically in Figure 1.10[145]. In contrast, the pentamethyldiethylenetriamine complex is monomeric and exhibits asymmetrical bonding of the allyl group to lithium (see Figure 1.11); the two different lithium-terminal carbon bond distances are 2.25Å and 2.72Å[146].

It is enlightening to note that in the light of recent variable temperature 13C NMR data for deuterated derivatives of allyllithium indicating that allyllithium forms an asymmetrical dimeric structure in THF[116], a lowest energy asymmetrical dimer structure was readily calculated[116]. The corresponding sodium and potassium compounds were concluded to have symmetrical structures using the same probe. Allyllithium was previously reported to be predominantly monomeric in THF with some dimers present (n = 1.3)[147]. Allyllithium is highly aggregated in Et2O (>10 units per aggregate)[114,115].

The 1H NMR spectrum of allyllithium in THF changes from an AB4 to an AA'BB'X type spectrum when the temperature is lowered to -87°C; other organoalkali allyl compounds exhibit AA'BB'X patterns at room temperature[6,115,148,149]. It has been concluded that the chemical shifts correspond to an essentially delocalized ionic species[149].

The barrier to rotation about a C-C bond has been utilized as a measure of the allylic resonance stabilization energy of the planar structure[2]. In fact, it has been

Figure 1.8 Highest occupied allyllithium molecular orbital.



Figure 1.9 Ab initio geometry of allyllithium.

concluded that the presence of hindered rotation can be used to rule out covalent structures for allyllithium (see Figure 1.12). In THF, the barriers to rotation for the organometallic species CH2CHCH2M are 10.7, 16.7, and 18 kcal/mole for M = Li, K, and Cs, respectively[150]. It would be expected that the lithium—carbon bonding interaction (covalent bonding or charge localization), because of partial covalent character in the transition state, reduces the rotational barrier appreciably[6]. However, the unexpected fact that allyllithium forms an unsymmetrical dimeric structure in THF requires that these simplified conclusions based on isolated molecules be modified[116,143].

The most sophisticated theoretical calculations give values for the barrier to rotation within the range of 19–24 kcal/mole for the free allyl anion[143,151,152]. From photoattachment spectroscopy, the electron affinity of the allyl radical

Figure 1.10 Schematic representation of the

x-ray crystal structure of the allyllithium TMEDA complex.



Figure 1.11 Schematic representation of the crystal structure

of the PMDETA complex of allyllithium.

(24.5 kcal/mol) and the resonance stabilization energy (14.5 kcal/mol) have been reported [153].

3-Neopentylallyllithium has been prepared in hydrocarbon solutions and has been shown to aggregate predominantly into dimers in benzene [100,154]. Cis and trans isomers do not interconvert on the 1H NMR time scale. The large upfield shift of the γ protons (δ = 4.637 and 4.495 ppm for trans and cis, respectively) indicates some degree of delocalization of charge from the α position (more delocalization occurs in THF and Et2O, in which δ = 4.1 and 3.6 ppm, respectively). The 1H NMR studies of a propagating poly(butadienyl)lithium chain end have been interpreted in terms of virtually 100% 1,4 chain-end structure, the lithium being σ-bonded to the α-carbon [55].

The ultraviolet (UV)-visible absorption spectra for allyl anions in THF are sensitive to the counterion [148]. The UV absorption maxima are observed at 318, 346, and 344 nm for the lithium, potassium, and cesium counterions, respectively. These results are analogous to the effects of the addition of polar solvents or counterion changes observed for benzyllithium (i.e., bathochromic shifts in UV absorption maxima), and are consistent with formation of more charge delocalized

Figure 1.12 Classical σ-bonded

structure of allyllithium.



allylic ions with larger alkali metal counterions. 3-Neopentylallyllithium exhibits a UV absorption maximum at 270 nm in n-pentane [100].

C. Enolate Anions

A simple enolate anion can be represented in terms of two contributing resonance structures, which show that the negative charge can be delocalized onto both carbon and oxygen (see Figure 1.13). Given the higher electronegativity of oxygen [73b], it would be expected that the oxyanion form (A) would contribute more than the carbanion form (B) to the stabilization of the anion. The available x-ray data for the crystal structures of enolate anions are consistent with this expectation. For example, the structure of the THF complex of LiCH2C(O)C(CH3)3 derived from 3,3-dimethyl-2-butanone consists of a tetrameric array of lithium atoms coordinated to a tetrameric array of the oxygen atoms of the enolate anion [156]. In addition, each lithium atom is coordinated to the oxygen of a complexed THF molecule. This structure is directly analogous to the interpenetrating tetrahedra of alkyl groups and lithium atoms (see Figure 1.2) observed for alkyllithium compounds such as methyllithium and ethyllithium. The observed carbon(1)- carbon(2) bond length of 1.34Å is consistent with essentially double bond character as shown in Figure 1.14. The corresponding solvent-free compound has an analogous hexameric structure [157,158]. The hexameric structure consists of two hexameric units with approximate S6 symmetry and opposing Li and O atoms on each face. The C1—C2 bond length in the hexamer is 1.330Å. The Li-O bond distances are 1.94–1.99Å and 1.85–1.95Å in the tetramer and hexamer, respectively.

An interesting aspect of the structure of the unsolvated hexameric aggregate is that the bonding distance between C-1 and Li-2 is only 2.53Å, which is within distances for significant π-bonding interaction. The analogous sodium derivative aggregates as a tetramer in the solid state [157].

A measure of the resonance stabilization of CH2CHO- comes from the barrier to rotation about the CC bond [2]. The most recent calculation yielded a value of 40 kcal/mole [151]. It is noteworthy that the corresponding calculated

Figure 1.13 Acetaldehyde enolate anion resonance structures.



Figure 1.14 Bonding arrangement in lithium enolate.

barriers for the allyl anion are 22 [151] and 19 kcal/mole [152]. Recent ab initio calculations indicate that the charge distribution and bond lengths for alkali metal acetaldehyde enolates are relatively insenstive to the counterion, in contrast to the corresponding carbanion derivatives [159]. The aggregation states of lithioiso-butyrophenone are reported to be tetrameric in dioxolane and predominantly dimeric in dimethoxyethane [160].

D. Ester Enolate Anions

The structures of lithium ester enolates have also been determined by x-ray crystallography. It should be noted that x-ray analysis of ester enolates must be carried out at low temperatures because of their instability, even in the crystalline state, with respect to decomposition to ketenes and alkoxides [161]. Although two stereoisomeric forms are possible, the crystal structure determination for the lithium ester enolate of t-butyl propionate complexed with TMEDA indicated only the (Z)-configuration (see Eq. 1.5) [162]. This ester enolate was associated

(1.5)

into dimers with Li2O2 four-membered rings. The lithium—oxygen distances were in the range of 1.90–1.95Å, which are similar to those observed for the lithium enolates of ketones [156–158]. Also analogous to the lithium enolates, the observed C2–C3 (a in Eq. 1.5) bond distance of 1.35Å is consistent with essentially double-bond character for this bond and concentration of the negative charge on the oxygen atom used to bond to lithium.

In contrast, the lithium ester enolate of methyl 3,3-dimethylbutanoate (see Eq. 1.6) complexes with THF to form tetrameric aggregates in the solid state [162]. However, only the (Z)-configuration is observed for this ester enolate also.



(1.6)

The structure of the tetramer consists of interpenetrating tetrahedra of lithium and oxygen as described for the lithium enolate of 3,3-dimethyl-2-butanone [156]. The average lithium—oxygen distance was 1.96Å and the formal double bond (a in Eq. 1.6) has the expected bond distance of 1.34Å.

The aggregation states of α-lithio esters of carboxylic acids are reported to be more than hexameric in benzene [(CH3)2CLiCO2C(CH3)3] and to vary from 2.3 [(CH3)2CLiCO2C(CH3)3] to 3.5 [(CH3)2CLiCOCH3 and (CH3)2CLiCO2CH2CH3] in tetrahydrofuran [163]. It is also important to note that lithium t-butoxide is associated into predominantly hexamers in benzene and also in diethyl ether; in THF and even pyridine the predominant degree of association is tetrameric [164].

VIII. Summary

Theoretical calculations for anions are difficult and many of the results obtained with less than the most sophisticated ab initio methods give results that are not reliable [2]. Theory predicts that free carbanions are relatively unstable species that readily lose an electron to form the corresponding radical. They have very low barriers to pyramidal inversion also.

Calculations for organometallic species seem to give consistent results with regard to structure. However, the interpretation of theoretical calculations with respect to the amount of covalent interaction for organolithium compounds is controversial.

Lithium perturbs the electron distribution for carbanions and tends to concentrate charge in the region between carbon and lithium.

Lithium—carbon bond distances are within what would be expected to be bonding distances. Therefore, contributions from covalent bonding interactions would be expected.

References

1. D. J. Cram, Fundamentals of Carbanion Chemistry, Academic Press, New York, 1965.

2. R. H. Nobes, D. Poppinger, W.-K. Li, and L. Radom, in Comprehensive Carbanion Chemistry, Part C, E. Buncel and T. Durst, Eds., Elsevier, Chapt. 1, 1987, p. 1.

3. C. Lambert and P. von Rague Schleyer, Methoden Org. Chem. (Houben-Weyl), 4th ed., 1952-, Bd. E 19d, 1993, p. 1.



4. C. Lambert and P. von Rague Schleyer, Angew. Chem. Int. Ed. Engl., 33, 1129 (1994).

5. E. Weiss, Angew. Chem. Int. Ed. Engl., 32, 1501 (1993).

6. J. L. Warden, “Alkali Metals,” in Comprehensive Organometallic Chemistry; The Synthesis, Reactions and Structures of Organometallic Compounds, G. Wilkinson, F. Gordon, A. Stone and E. W. Abel, Eds., Pergamon Press, Oxford, 1982, Vol. 1, p 43.

7. W. N. Setzer and P. von Rague Schleyer, Adv. Organomet. Chem., 24, 353 (1985).

8. C. Schade and P. von Rague Schleyer, Adv. Organomet. Chem., 27, 169 (1987).

9. T. L. Brown, Accounts Chem. Res., 1, 23 (1968).

10. H. Dietrich, Acta Crystallogr., 16, 681 (1963).

11. H. Dietrich, J. Organomet. Chem., 205, 291 (1981).

12. W. J. Hehre, L. Radom, P. v. R. Schleyer and J. A. Pople, Ab Initio Molecular Orbital Theory, Wiley, New York, 1986, p. 401.

13. G. B. Ellison, P. C. Engleking and W. C. Lineberger, J. Am. Chem. Soc., 100, 2556 (1978).

14. G. W. Spitznagel, T. Clark, J. Chandrasekhar, and P. v. R. Schleyer, J. Comput. Chem., 3, 363 (1982).

15. B. M. Smirnov, Negative Ions, McGraw-Hill, New York, 1982.

16. P. v. R. Schleyer, G. W. Spitznagel and J. Chandrasekhar, Tetrahedron Lett., 27, 4411 (1986).

17. R. N. Young, R. P. Quirk, and L. J. Fetters, Adv. Polym. Sci., 56, 1 (1984).

18. M. Morton, Anionic Polymerization: Principles and Practice, Academic Press, New York, 1983.

19. S. Bywater, in Encyclopedia of Polymer Science and Engineering, 2nd edn., J. I. Kroschwitz, Ed., Wiley-Interscience, New York, 1985, Vol. 2, p. 1.

20. G. E. Coates, M. L. H. Green, and K. Wade, Organometallic Compounds, Vol. 1, The Main Group Elements, Methuen, London, 3rd ed., 1967.

21. T. L. Brown, Adv. Organomet. Chem., 3, 365 (1966).

22. T. L. Brown, Pure Appl. Chem., 23, 447 (1970).

23. B. L. Wakefield, The Chemistry of Organolithium Compounds, Pergamon Press, Oxford, 1974.

24. M. E. O'Neill and K. Wade, in Comprehensive Organometallic Chemistry; The Synthesis, Reactions and Structures of Organometallic Compounds, G. Wilkinson, F. G. A. Stone and E. W. Abel, Eds., Pergamon Press, Oxford, 1982, Vol. 1, p. 1.

25. J. W. Chinn, Jr., and R. J. Lagow, Organometallics, 3, 75 (1984).

26. W. McLean, P. T. Murray, and R. C. Jarnagin, J. Chem. Phys., 69, 2715 (1978).

27. D. Plavsic, D. Srzic, and L. Klasinc, J. Phys. Chem., 90, 2075 (1986).



33. G. D. Graham, D. S. Marynick, and W. N. Lipscomb, J. Am. Chem. Soc., 102, 4572 (1980).

34. G. D. Graham, S. Richtsmeier, and D. A. Dixon, J. Am. Chem. Soc., 102, 5759 (1980).

35. E. Kaufmann, K. Raghavachari, A. E. Reed, and P. v. R. Schleyer, Organometallics, 7, 1597 (1988).

36. G. Fraenkel, M. Henrichs, J. M. Hewitt, B. M. Su, and M. J. Geckle, J. Am. Chem. Soc., 102, 3345 (1980).

37. D. Seebach, R. Hässig, and J. Gabriel, Helv. Chim. Acta, 66, 308 (1983).

38. G. Fraenkel, A. Chow, and W. R. Winchester, J. Am. Chem. Soc., 112, 6190 (1990).

39. B. Lindman and S. Forsen, in NMR and the Periodic Table, R. K. Harris and B. E. Mann, Eds., Academic Press, New York, 1978, p. 166.

40. W. Bauer and P. v. R. Schleyer, in Advances in Carbanion Chemistry, V. Snieckus, Ed., JAI Press, London, 1992, Vol. 1, p. 89.

41. R. D. Thomas, in Isotopes in the Physical and Biomedical Science, E. Buncel and J. R. Jones, Eds., Elsevier, Amsterdam, 1991, Vol. 2, p. 367.

42. W. Bauer, W. R. Winchester, and P. von Rague Schleyer, Organometallics, 6, 2371 (1987).

43. W. Bauer, M. Feigel, G. Müller, and P. von R. Schleyer, J. Am. Chem. Soc., 110, 6033 (1988).

44. T. Clark, J. Chandrasekhar, and P. von R. Schleyer, J. Chem. Soc. Chem. Commun., 672 (1980).

45. H. Gunther, D. Moskau, and D. Schmalz, Angew. Chem. Int. Ed. Engl., 26, 1212 (1987).

46. T. L. Brown, L. M. Seitz, and B. Y. Kimura, J. Am. Chem. Soc., 90, 3245 (1968).

47. L. Andrews and T.G. Carver, J. Phys. Chem., 72, 1743 (1968).

48. J. W. Chinn, Jr., J. A. Gurak, and R. J. Lagow, in Lithium: Current Applications in Science, Medicine and Technology, R. O. Bach, Ed., John Wiley, New York, 1985, p. 291.

49. P. v. R. Schleyer, Pure Appl. Chem., 56, 151 (1984).

50. A. Maercker, M. Theis, A. J. Kos, and P. v. R. Schleyer, Angew. Chem., Int. Ed. Engl., 22, 733 (1983).

51. R. West, P. A. Carney, and I. C. Mineo, J. Am. Chem., Soc., 87, 3788 (1965).

52. S. D. Smith, A. Ashraf, and S. J. Clarson, Polym. Prep., Am. Chem. Soc., Div. Polym. Chem., 35(2), 466 (1994).

53. E. Schoenberg, H. A. Marsh, S. J. Walters, and W. M. Saltman, Rubber Chem. Tech., 52, 526 (1979).

54. D. H. Richards, Chem. Soc. Rev., 6, 235 (1977).

55. E. Weiss and E. A. C. Lucken, J. Organomet. Chem., 2, 197 (1964).

56. E. Weiss and G. Hencken, J. Organomet. Chem., 21, 265 (1970).



61. G. W. Klumpp, P. J. A. Geurink, A. L. Spek, and J. M. Duisenberg, J. Chem. Soc. Chem. Commun., 814 (1983).

62. K. S. Lee, P. G. Williard, and J. W. Suggs, J. Organomet. Chem., 299, 311 (1986).

63. M. Vos, F. J. J. de Kanter, M. Schakel, N. J. R. van Eikemea Hommes, and G. W. Klumpp, J. Am. Chem. Soc., 109, 2187 (1987).

64. W. Moene, M. Vos, F. J. J. de Kanter and G. W. Klumpp, J. Am. Chem. Soc., 111, 3463 (1989).

65. E. Weiss, G. Sauermann, and G. Thirase, Chem. Ber., 116, 74 (1983).

66. R. Zerger, W. Rhine, and G. D. Stucky, J. Am. Chem. Soc., 96, 6048 (1974).

67. A. Maercker, M. Bsata, W. Buchmeier, and B. Engelen, Chem. Ber., 117, 2547 (1984).

68. B. Tecle, A. F. M. Maqsudur Rahman, and J. P. Oliver, J. Organomet. Chem., 317, 267 (1986).

69. W. Hiller, M. Layh, and W. Uhl, Angew. Chem. Int. Ed. Eng., 30, 324 (1991).

70. R. P. Zerger and G. D. Stucky, J. Chem. Soc., Chem. Commun., 44 (1973).

71. M. F. Lappert, L. M. Engelhardt, C. L. Raston, and A. H. White, J. Chem. Soc., Chem. Commun., 1323 (1982).

72. L. Sutton, Ed., Table of Interatomic Distances and Configurations in Molecules and Ions, Spec. Publ. No. 11, Chemical Society, London, 1958; No. 18, 1965.

73. L. Pauling, The Nature of The Chemical Bond, 3rd ed., Cornell University Press, Ithaca, New York, 1960, (a) p. 225; (b) p. 93.

74. W. Zarges, M. Marsch, K. Harms, and G. Boche, Chem. Ber., 122, 2303 (1989).

75. W. Zarges, M. Marsch, K. Harms, W. Koch, G. Frenking and G. Boche, Chem. Ber., 124, 543 (1991).

76. J. J. Brooks and G. D. Stucky, J. Am. Chem. Soc., 94, 7333 (1972).

77. M. Marsch, K. Harms, L. Lochmann, and G. Boche, Angew. Chem. Int. Ed. Eng., 29, 308 (1990).

78. R. Amstutz, D. Seebach, P. Seiler, B. Schweizer, and J. D. Dunitz, Angew. Chem. Int. Ed. Eng., 19, 53 (1980).

79. H. Schmidbaur, A. Schier, and U. Schubert, Chem. Ber., 116, 1938 (1983).

80. H. Köster, D. Thoennes, and E. Weiss, J. Organomet. Chem., 160, 1 (1978).

81. S. P. Patterman, I. L. Karle, and G. D. Stucky, J. Am. Chem. Soc., 92, 1150 (1970).

82. M. A. Beno, H. Hope, M. M. Olmstead, and P. P. Power, Organometallics, 4, 2117 (1985).

83. G. Fraenkel, M. Henrichs, J. M. Hewitt, B. M. Su, and M. J. Geckle, J. Am. Chem. Soc., 102, 3345 (1980).

84. H. L. Lewis and T. L. Brown, J. Am. Chem. Soc., 92, 4664 (1970).



91. T. L. Brown and M. T. Rogers, J. Am. Chem. Soc., 79, 1859 (1957).

92. T. L. Brown, R. L. Gerteis, D. A. Baffus, and J.A. Ladd, J. Am. Chem. Soc. 86, 2135 (1964).

93. T. L. Brown, D. W. Dickerhoff, and D. A. Baffus, J. Am. Chem. Soc., 84, 1371 (1962).

94. T. L. Brown, J. A. Ladd, and G. N. Newman, J. Organomet. Chem., 3, 1 (1965).

95. D. Margerison and J. P. Newport, Trans. Faraday Soc., 59, 2058 (1963).

96. D. Margerison and J. D. Pont, Trans. Faraday Soc., 67, 353 (1971).

97. J. B. Smart, R. Hogan, P. A. Scherr, M. T. Emerson, and J. P. Oliver, J. Organomet. Chem., 64, 1 (1974).

98. K.-H. Thiele, E. Langguth, and G. E. Muller, Z. Anorg. Allg. Chem., 462, 152 (1980).

99. M. M. Exner, R. Waack, and E. C. Steiner, J. Am. Chem. Soc., 95, 7009 (1973).

100. W. H. Glaze, J. E. Hanicak, M. L. Moore, and J. Chaudhuri, J. Organomet. Chem., 44, 39 (1972).

101. G. W. Klumpp, H. Luitjes, M. Schakel, F. J. J. de Kanter, R. F. Schmitz, and N. J. R. van Eikema Hommes, Angew. Chem. Int. Ed. Eng., 31, 633 (1992).

102. G. Fraenkel, A. Chow, and W. R. Winchester, J. Am. Chem. Soc., 112, 6190 (1990).

103. W. Bauer and D. Seebach, Helv. Chim. Acta, 67, 1972 (1984).

104. J. F. McGarrity and G. A. Ogle, J. Am. Chem. Soc., 107, 1805 (1984).

105. S. Bywater, in Comprehensive Polymer Science, Chain Polymerization I, G. C. Eastmond, A. Ledwith, S. Russo, and P. Sigwalt, Eds., Pergamon Press, Elmsford, NY, 1989, Vol. 3, p. 433.

106. T. A. Antkowiak, A. E. Oberster, A. F. Halasa, and D. P. Tate, J. Polym. Sci., Part A-1, 10, 1319 (1972).

107. R. P. Quirk, D. E. Kester, and R. D. Delaney, J. Organomet. Chem., 59, 45 (1973).

108. R. P. Quirk and D. E. Kester, J. Organomet. Chem., 72, C23 (1974).

109. R. P. Quirk and D. E. Kester, J. Organomet. Chem., 127, 111 (1977).

110. R. P. Quirk and D. McFay, Makromol. Chem., Rapid Commun., 1, 71 (1980).

111. R. P. Quirk, in Anionic Polymerization: Kinetics, Mechanisms, and Synthesis, J. E. McGrath, Ed., ACS Symposium Series 166, Am. Chem. Soc. Washington, D.C., 1981, p. 117.

112. R. P. Quirk and D. McFay, J. Polym. Sci. Polym. Chem. Ed., 19, 1445 (1981).

113. P. West and R. Waack, J. Am. Chem. Soc., 89, 4395 (1967).

114. G. R. Brubaker and P. Beak, J. Organomet. Chem., 136, 147 (1977).

115. P. West, J. I. Purmort, and S. V. McKinley, J. Am. Chem. Soc., 90, 797 (1968).

116. W. R. Winchester, W. Bauer, and P. v. R. Schleyer, J. Chem. Soc. Chem. Commun., 177



121. R. Waack, L. D. McKeever, and M. A. Doran, Chem. Commun., 117 (1969).

122. L. D. McKeever, in Ions and Ion-Pairs in Organic Reactions, M. Szwarc, Ed., Wiley, London, 1972, Vol. 1, p. 263.

123. N. Muller and D. E. Pritchard, J. Chem. Phys., 31, 768, 1471 (1959).

124. F. W. Wehrli, A. P. Marchand, and S. Wehrli, Interpretation of Carbon-13 NMR Spectra, 2nd Ed., Wiley, New York, 1983, p. 66.

125. J. P. C. M. van Dongen, H. W. D. van Dijkman, and M. J. A. de Bie, Recl. Trav. Chim. Pays-Bas, 93, 29(1974).

126. K. Takahashi, Y. Kondo, and R. Asami, Org. Magnet. Reson., 6, 580 (1974).

127. S. Bywater and D. J. Worsfold, J. Organomet. Chem., 33, 273 (1971).

128. B. Bockrath and L. M. Dorfman, J. Am. Chem. Soc., 97, 3307 (1975).

129. L. M. Dorfman, R. J. Sujdak, and B. Bockrath, Acc. Chem. Res., 9, 352 (1976).

130. J. B. Smart, R. Hogan, P. A. Scherr, L. Ferrier, and J. P. Oliver, J. Am. Chem. Soc., 94, 8371 (1972).

131. J. R. Urwin and P. J. Reed, J. Organomet. Chem., 15, 1 (1968).

132. T. Holm. Acta Chem. Scand., Ser. B., 32, 162 (1978).

133. P. A. Scherr, R. J. Hogan, and J. P. Oliver, J. Am. Chem. Soc., 96, 6055 (1974).

134. M. Morton, L. J. Fetters, R. A. Pett, and J. F. Meier, Macromolecules, 3, 327 (1970).

135. D. J. Worsfold and S. Bywater, Macromolecules, 5, 393 (1972).

136. S. Bywater, A. F. Johnson, and D. J. Worsfold, Can. J. Chem., 42, 1255 (1964).

137. M. M. Al-Jarrah and R. N. Young, Polymer, 21, 119 (1980).

138. L. J. Fetters and M. Morton, Macromolecules, 7, 552 (1974).

139. A. Hemandez, J. Semel, H. C. Broecker, H. G. Zachmann, and H. Sinn, Makromol Chem., Rapid Commun., 1, 75 (1980).

140. H. L. Makowski and M. Lynn, J. Macromol. Chem., 1, 443 (1966).

141. L. J. Fetters, N. P. Balsara, J. S. Huang, H. S. Jeon, K. Almdal, and M. Y. Lin, Macromolecules, 28, 4996 (1995).

142. T. Clark, E. D. Jemmis, P. von R. Schleyer, J. S. Binkley, and J. A. Pople, J. Organomet. Chem., 150, 1 (1978).

143. N. J. R. van Eikema Hommes, M. Bühl, P. von R. Schleyer, and Y.-D. Wu, J. Organomet. Chem., 409, 307 (1991).

144. T. Clark, C. Rohde, and P. von R. Schleyer, Organometallics, 2, 1344 (1983).



155. M. Morton, R. D. Sanderson, R. Sakata, and L. A. Falvo, Macromolecules, 6, 186 (1973).

156. R. Amstutz, W. B. Schweizer, D. Seebach, and J. D. Dunitz, Helv. Chim. Acta, 64, 2617 (1981).

157. P. J. Williard and G. B. Carpenter, J. Am. Chem. Soc., 107, 3345 (1985).

158. P. J. Williard and G. B. Carpenter, J. Am. Chem. Soc., 108, 462 (1986).

159. C. Lambert, Y.-D. Wu, and P. von Rague Schleyer, J. Chem. Soc., Chem. Commun., 255 (1993).

160. L. M. Jackaman and B. C. Lange, J. Am. Chem. Soc., 103, 4494 (1981).

161. D. Seebach, Angew. Chem. Int. Ed. Engl., 27, 1624 (1988).

162. D. Seebach, R. Amstutz, T. Laube, W. Bernd Schweizer, and J. D. Dunitz, J. Am. Chem. Soc., 107, 5403 (1985).

163. V. Halaska and L. Lochmann, Coll. Czech. Chem. Commun., 38, 1780 (1973).

164. V. Halaska, L. Lochmann, and D. Lim, Coll. Czech. Chem. Commun., 33, 3245 (1968).



2 Stabilities of Carbanionic Species

I. Introduction

A carbanion can be defined as the species formed by removal of a proton from a carbon acid (i.e., by Eq. 2.1 [1,2]). The species on the left side of Eq. 2.1 is the

(2.1)

conjugate acid of the carbanion and the acidity constant for this acid (Keq) can be measured or estimated by a variety of methods [1,3–12]. These acidity constants are generally used as a guide to carbanion stability, making the assumption that the stronger carbon acid is associated with the formation of a more stable carbanionic species, and vice versa. In this chapter systematic studies of carbon acidity and carbanion stability will be discussed.

II. Acidity of Carbon Acids in the Gas Phase

Experimental information regarding the structure and stability for simple unsolvated, free (i.e., without encumbering counterion) carbanions is quite limited. The stabilities of various carbanionic species have been obtained in the gas phase by



high-pressure mass spectrometry, flowing after-glow, and ion cyclotron resonance spectrometry [7,8,11,12]. The gas phase acidity of a neutral acid, A-H, can be described in terms of the standard free energy change (∆G, kcal/mole) for the deprotonation reaction in the gas phase (Eq. 2.2). The results of gas phase acidity measurements for a variety of carbon and other acids are shown in Table 2.1.

(2.2)

It is important to note the following significant differences between acidities in the gas phase and in solution [7,12–14]:

1. Toluene is more acidic than water in the gas phase, but 1020 times less acidic in solution [15].

2. Fluorene is almost 105 times less acidic than cyclopentadiene in solution [10], but is the stronger acid in the gas phase.

3. The gas-phase acidity order t-BuOH > i-PrOH > EtOH > MeOH is reversed in solution [13,16–19].

Table 2.1 Comparison of Gas Phase Stabilities of Anions Derived from Carbon, Oxygen, and Nitrogen Acids

Compound ∆Gacid (kcal/mole) a Compound ∆Gacid (kcal/mole) a

CH3CH3412 CH3CN 364

CH4409 (CH3CH2)3COH 364

CH3CH2CH2CH3407 CH3CO2CH3

362

(CH3)3CH 406 CH3COCH3362

CH2=CH2404 CH3CHO 360

∆ -H 403 C6H5NH2359

NH3397 (C6H5)2CH2

359

C6H6396 CH3SO2CH3

359

(CH3)2NH 389 (C6H5)3CH 353

H2O 384 CH3NO2352

(i-Pr) 2NH 383 Cyclopentadiene 348

CH3OH 373 Fluorene 344

C6H5CH3372 CH2(CO2Et) 2 342

CH3CH2OH 371 C6H5OH 342

(CH3)2CHOH 369 CH3CO2H 342



4. The gas phase basicity of simple amines increases with alkyl substitution as observed in solution [13]. This result and item 3 above show that in the gas phase larger alkyl groups can stabilize a charge of either sign, by either withdrawing or donating electrons as needed, presumably via a charge-induced polarization of the electron distribution [7,19].

The principal conclusion from the data in Table 2.1 is that the acidity orders in the gas phase are often quite different from those observed in solution [2,3,6–19]. The gas-phase data refer to the properties of isolated molecules, whereas the results in solution include the interactions with the solvent and counterions. Thus, stability orders observed in solution are not necessarily based on simple intrinsic structural effects, but are composite effects that include the differential effects of solvent on all species involved, as well as the intrinsic acidity of the solute. These solvation effects include the differential solvation energy of the neutral conjugate acids, differential solvation energy of the ions, as well as differential solvation of the ions and ion-pairing effects. Unfortunately from the standpoint of understanding structure-stability relationships, all of these effects will be involved in the measured values of solution-phase acidities. These solvation energies can be as large as 50–90 kcal/mol [6,7,13]. Since these solvation energies are much larger than intrinsic acidity differences for most pairs of compounds, the relative acidities in solution often reflect the differential solvation energies more than intrinsic structural effects [6,7,13,14].

The dominant role of solvent on acidity in solution is supported by the effects of solvation on linear free energy relationships [6,13,14]. The importance of solvent in affecting structure—stability—reactivity relationships can be illustrated by comparison of the Hammett rho (ρ) values for ionization of meta- and para- substituted benzoic acids in water, dimethylsulfoxide (DMSO), and the gas phase, for which the ρ values are 1.0, 2.5, and ˜10, respectively [6]. It was concluded that the polar and resonance effects of the ring substituents are small by comparison with the strong, specific solvation interactions with the carboxylic acid group in solution [6].

The lore of physical organic chemistry is intimately related to perceived understanding of structure—stability—reactivity relationships in terms of fundamental effects such as polarization, resonance, inductive, steric, and hybridization [19]. Therefore, it is important to try to determine which of these “effects” [7,13,14,18–20] is fundamentally correct (i.e., observed in both the gas phase and in solution).

One important conclusion from the gas phase data is that there is a general increase in acidity with increasing molecular size; it has been concluded that polarization is the dominant effect responsible for this general trend [7,19]. In solution, smaller ions tend to be more strongly solvated and thus the solution acidity order may often favor the smaller, more highly solvated anion; this may



explain some of the discrepancies between acidity orders observed in solution compared to the gas phase. Thus, the ability of larger, more polarizable alkyl groups to increase acidity in the gas phase for carbon acids, alcohols and the conjugate acids of amines can be rationalized [7,13,14,19]. In solution the substitution of larger alkyl groups would tend to decrease solvation energies and thus lower acidities, as observed [10,14].

Delocalization effects would be expected to result in analogous reversals in acidity orders between solution and gas phase results because of the inverse correlation between charge delocalization and solvation [7,13,14]. This provides an explanation for the observation that fluorene is approximately 105 times less acidic than cyclopentadiene in solution [10], whereas fluorene is the stronger acid in the gas phase [7,14] (see Table 2.1).

Consistent with predictions based on resonance and stabilization by delocalization of charge, it is observed that introduction of an unsaturated group adjacent to a deprotonation site tends to increase the gas phase acidity analogous to the effects observed in solution. Thus, propene is a much stronger acid than propane, toluene is much stronger than methylcyclohexane, and cyclopentadiene is much stronger than cyclopentane [8]. Although gas phase acidity data relating to the effects of polar substituents on acidity are limited, the following general order of the decreasing ability of substituents to acidify carbon-hydrogen bonds can be deduced:

-NO2 > RSO2- > RCO- > -CN RSO- > C6H5- > CH2=CH-

Another tenet of physical organic chemistry is the relationship between hybridization and acidity [1]. In general, it is more energetically favorable to place electron density in an orbital with more s character, since a p orbital has a node at the positively charged nucleus. In accord with these concepts, a linear correlation is observed between the pKa of carbon acids in solution and the percentage of s-character of the carbon orbital used in bonding to the ionizing hydrogen that produces the acidity order [1]:

acetylene (sp) > ethylene (sp2) > cyclopropane (sp2.28) > ethane (sp3)

In the gas phase, acetylene (sp) is a much stronger acid than both benzene (sp2) and methane (sp3); the gas phase increased acidity increments (δ∆) are 0 < 8 9 < 37 kcal/mole for CH3CH3 < CH2 = CH2 ∆-H < HC CH, respectively [7,8,11].

III. Acidity of Carbon Acids in Solution

A. Water Solution

In principle, the relative stabilities of carbanions (R-) can be evaluated by measuring the acid dissociation constants (Ka) of the corresponding conjugate



acids (RH) as shown in Eq. 2.3, where S is a solvent molecule and H+ (S) is the

(2.3)

conjugate acid of the solvent. For reasonably strong acids (i.e., pKa = 0–12), the aqueous standard state provides a useful measure of acid dissociation constants. However, for many hydrocarbon acids, reference to the aqueous standard state can only be made indirectly because these acids have estimated acid dissociation constants (pKas) larger than 20 (i.e., Ka < 10-20). A variety of kinetic and thermodynamic methods, including acidity function techniques [21], have been used to determine the relative acidity order of hydrocarbon acids or the relative stabilities of the corresponding carbanions [1,3–6,10]. A summary of representative hydrocarbon acidity data referenced to the aqueous standard state is listed in Table 2.2. General reviews [3–6,10] and the references listed in Table 2.2 should be consulted for more detailed, critical insight into the reliability and limitations of these results.

In order to interpret this acidity data in terms of the effects of substituents on acidity, it is necessary to assume that the acidities of these compounds primarily reflect the ability of the respective substituents to stabilize the negative charge in the anion (i.e., to assume that differential solvent interactions in the corresponding conjugate acids and in the carbanions can be ignored). Although water has a high dielectric constant (ε = 78.5) [22] and thus ion-pairing effects can be neglected, water is a hydroxylic solvent capable of interacting strongly with solutes and ions by specific interactions such as hydrogen bonding. To the extent that such differential, specific hydrogen-bonding occurs, the results will not simply reflect the intrinsic electronic effects of the substituents on acidity. These differential solvent effects would be expected to occur primarily for small ions with localized charges as discussed previously. Subject to these limitations, these data can be used to generate an approximate order of the decreasing ability of substitutents to acidify carbon-hydrogen bonds in aqueous solution [3]:

-NO2 > RCO- > RSO2- > -CO2R > -CN > -H > -R (alkyl)

his order is similar to the order derived from acidities in the gas phase, except for the reversal of the RCO- and RSO2- groups.

B. Dimethylsulfoxide Solution

The most extensive investigations of the effect of structure on acidity for carbon acids have been carried out in dimethylsulfoxide using a carbon acid indicator method [6,10,15,23–26]. An absolute scale of acidities was established by anchoring the scale with potentiometric measurements for stronger acids (pKa 7–12) [24]. The effect of structure on acidity was defined by the ionic acidity differences (i.e., relative acidities), using comparative equilibria of the type shown in Eq. 2.4,

(2.4)



Table 2.2 Acidity of Hydrocarbon Acids Referenced to the Aqueous Standard State

Carbon AcidpKa Reference

Pentacyanocyclopentadiene -11 28

CH(CN) 3 -5.1 29

CH(NO2)3 0.2 30

CH2(NO2)2 3.6 30

CH(COCH3)3 5.8 31,32

CH2(COCH3)2 9.0 31,32

CH3NO2 10.2 33

CH2(CN)2 11.2 34

CH2(SO2CH2CH3)2 12.5 35,36

CH2(CO2CH2CH3)2 13.3 34

C6H5COCH3 16 37

18.6 38

Cyclopentadiene 16.0 39

(CH3)3SiCOCH3 16.4 40

CH3CHO 16.7 41

Cyclopentanone 17 35, 36

CH3COCH3 20 42

19.6 30

19.3 43,44

CH3SO2CH3 23 34

CH3CN 25 34

31.5a 23

CH3CO2Et 27–28a 23

CH3CON(CH3)2 31–32a 23

a Estimated by extrapolation.



where InH represents an indicator of known acidity whose conjugate base concentration, [In-K+], can be readily measured spectrophotometrically. It was concluded that, because of the high dielectric constant of DMSO (ε = 48.9) [27], the use of dilute solutions of potassium salts of carbanions (ca. 0.01 M), and other experimental evidence such as conductance measurements [41], these results involve dissociated ions and ion pairing is generally avoided [6,45–47]. Ion-pairing effects are observed for strongly chelating anions such as those formed from β-diketones such as CH3COCH2COCH3 [46]. Ion pairing was reported to stabilize the carbanion relative to the delocalized indicator anion, leading to an apparent increase in acidity [10]. As an example of the dramatic effects of ion pairing on apparent acidity measurements, the equilibrium constant for the equilibrium shown in Eq. 2.5 (M+ is the counterion) changes from a value of 0.007 for



(2.5)

K+ to 4 for Li+ [48,49]. A proposed criterion for detecting ion pairing involved the addition of a standard solution of potassium iodide (5–10 equivalents of common ion) in DMSO and monitoring any spectral changes that would indicate variations in the concentration of the indicator anion, [In-] [46,47]. A drop in absorbance greater than that due to dilution was interpreted in terms of a shift to the right for the equilibrium shown in Eq 2.6, where R- is the carbanion whose

(2.6)

acidity constant is being measured and R-,K+ is the corresponding ion pair. Another useful criterion to detect ion-pairing effects was the effect on the absorbance of the indicator anion from the addition of a cryptand or crown ether (e.g., 18-crown-6 ether), which complexes specifically with potassium [46,47]; in the absence of ion pairing, no change in absorbance will be observed.

The acidity constants (pKa ± 0.1) listed in Table 2.3 refer to dilute solutions in DMSO as the standard state. For the methane series of carbon acids, the order of the ability of substitutents to acidify adjacent carbon-hydrogen bonds in dimethylsulfoxide solution decreases in the order:

This order is similar to the effects observed in the gas phase, except that the sulfonyl group is less acidifying than the carbonyl and similar to the cyano group in DMSO vs. the gas phase. The order -NO2 > RCO- > -CN appears to be universal since it is observed in the gas phase, in aqueous solution, and in DMSO.

The use of DMSO as solvent for acidity measurements for weak carbon acids is limited by the leveling effect of the solvent [52]; the pKa of DMSO is estimated to be 35.1 [16]. Therefore, accurate determination of acidities is difficult for carbon acids within 4–5 pKa units of this value. It was concluded that the upper pKa limit was 31–32 for acidity measurements in DMSO [16]. Carbon acids with pKa values in the range of 32–35 were estimated by an extrapolation technique [50]. The acidifying effects (∆pKa) of a number of substituents were measured for a variety of acidic carbon acids with two or more acidifying substituents; these increments were then used to estimate the acidity of the corresponding mono-substituted compounds [50].

In Table 2.3 the acidity scale results in DMSO are also compared with results obtained for corresponding ion-paired species (see Chapter 3 for discussion of ion pair chemistry) in cyclohexylamine, a solvent of much lower dielectric constant



Table 2.3 Comparison of the Acidities of Carbon Acids in DMSO at 25°C [pK(DMSO)] and Ion Pair Acidities in Cyclohexylamine (Cesium Cyclohexylamide) [pK(CsCHA)]

Compound pK(DMSO) Reference pK(CsCHA) [3]

9-Cyanofluorene 8.3 24 —

9-Carboxymethylfluorene

10.3 24 —

CH2(CN)211.1 24 —

Nitromethane 17.2 24 —

9-Phenylfluorene 17.9 24 (18.49)b

Cyclopentadiene 18.0 26 16.25

Carbazole 19.9 26 —

Fluorene 22.6 24 23.04

Pyrrole 23.05 26 —

Acetophenone 24.7 24 —

Acetone 26.5 24 —

Phenylacetylene 28.7 47 23.24 (LiCHA)c

(CH3)3CC CH — 25.52 (LiCHA)c

C6H5CH2SOCH329.0 24 —

C6H5SO2CH329.0 24 27.2

(C6H5)2PCH2P(C6H5)2— 31.1

Triphenylmethane 30.6 24 31.45

C6H5CH2SC 6H530.8 25 —

CH3CO2Et 30–31a 23 —

CH3SO2CH331.1 24 —

CH3CN 31.3 24 —

Diphenylmethane 32.2 10 33.38

Benzofuran — 36.84

Benzothiophene — 37.05

Thiophene — 38.2

CH3CON(CH3)2 34–35a 23 —



(ε = 6) [24]. It should be noted that these ion pair “pKa” values are all relative to an arbitrary standard, often the pKa of 9-phenylfluorene, which was assumed to have a value of 18.5 as determined by acidity function measurements.

C. Cyclohexylamine and Tetrahydrofuran Solutions

These relative ion pair acidities in cyclohexylamine listed in Table 2.3 refer to the type of equilibrium shown in Eq. 2.7, where the species R-M+ and R'-M+

(2.7)

represent the ion-paired species that may or may not be aggregated in the solution. In spite of the assumptions required in these methods and the approximate nature of the results, there is surprisingly good agreement regarding the relative acidities (∆pKa) of hydrocarbon acids that form delocalized anions, as shown in Table 2.3. In fact, these two data sets for delocalized ions exhibit a linear correlation with each other over a range of 15 pKa units (r = 0.998) [3]. The pK(DMSO) data also exhibit a linear correlation with the gas phase data over a range of 20 pKa units (R2 = 0.998) [14]. It has been noted that the agreement breaks down when comparing phenylacetylene, which forms a localized anion, with fluorenes, in which the negative charge is highly delocalized [6]; in cyclohexylamine ∆pKa between phenylacetylene and fluorene is approximately zero, while in DMSO the ∆pKa is calculated to be 6 (see Table 2.3).

A revised cesium ion pair acidity scale relative to fluorene (assigned pKa = 22.9) [10] has been developed in tetrahydrofuran (ε = 7.39) [22] by Streitwieser and co-workers [53]. The acidity measurements in THF were linearly related to the previous data in cyclohexylamine (± 0.18 pKa units) [3]. An extrapolated pKa value of 40.9 was estimated for toluene using this scale. This scale has been used to estimate the acidity of a variety of other types of carbon acids including α,ω-diphenylpolyenes [54]. For example, the ion pair and free ion pKa values of 1,3-diphenylpropene were estimated to be 27.85 and 26.17, respectively [54]. The acid-strengthening effect of trimethylsilyl groups has also been investigated [55]. Although the acidity of 9-trimethylsilyfluorene (pKa = 21.6) is only enhanced slightly compared to fluorene (pKa = 22.9), the acidity of benzyltrimethylsilane (pKa = 37.5) is 3.4 pKa units higher than the estimated pKa of toluene (40.9). The acidity of tris(trimethylsilyl)methane (pKa = 36.8) was used to obtain an extrapolated value of 47 for the pKa of methane [55].

With respect to the use of these ion-pair acidity scales as guides to the acidities of carbon acids and the stabilities of carbanions, their potential limitations interms of their sensitivity to specific solvent and counterion interactions vis à vis intrinsic electronic effects should be noted. Furthermore, the consequences of ion-pairing on stability measurements have been demonstrated, especially for carbanions with localized charge [14]. Recognizing these limitations,



one can use these results to make predictions and to correlate the effects of structure on the relative stabilities and reactivities of carbanionic species in these types of solutions, since ion pairs are generally involved in “real world” chemistry (see Chapter 3).

D. Electrochemical Determinations in Dipolar Aprotic Solutions

It is not possible to measure directly the acid dissociation constants for very weak hydrocarbon acids such as alkanes. An electrochemical method has been developed for estimating the acidity of weak carbon acids using the thermodynamic sequence shown in Eq. 2.8 [56–58]. Hydrocarbon acids were compared to tri

(2.8)

phenylmethane as the reference standard. The energy of the first step corresponds to the standard bond dissociation energy (SBDE) of the hydrocarbon in the gas phase. The second step corresponds to the reversible potential for electrochemical reduction of the radical to the anion. The third step is common to all substrates and is not actually considered. For isobutane, the difference in pKa was estimated from the difference in standard bond dissociation energies coupled with the difference in oxidation potentials of the conjugate bases, which were determined by reversible polarographic reductions of the corresponding alkyl iodides that takes place in two steps as shown in Eq. 2.9. The polarography of t-butyl iodide was carried out

(2.9)

in dipolar aprotic solvents such as acetonitrile and 1,2-dimethoxyethane in the presence of a supporting electrolyte such as a tetraalkylammonium perchlorate. It is important to note that some of the data exhibited a strong dependence on solvent and supporting electrolyte (∆pKa = 5 for isobutane). To estimate the acidities of other hydrocarbon acids, reversible electrochemical oxidation potentials of the corresponding organolithium compounds were determined in THF/hexamethyl-phosphoramide with 0.2 M LiClO4 [58]. Using SBDE and pKa values for tri-phenylmethane of 75 kcal/mol and 31.5, respectively, pKa values for isobutane (70.7), propene (47.1–48.0), toluene (44.4–45.2), and cyclopentadiene (22.2–23.4) have been reported [56–58]. It should be noted that the values for propene, toluene and cyclopentadiene refer to ion-pair acidities [58]. The pKa of methane (58 ± 5) was estimated by extrapolation methods [58]. Bordwell [10] has noted that the SBDE of triphenylmethane may be in error by as much as 6 kcal/mol. In spite of these limitations, this method represents the only approximate thermo-dynamic data for hydrocarbons in the high pKa region (>50). It should be noted that a pKa value of 56 has been estimated for methane by extrapolation of acidity data for other more acidic hydrocarbons (see Table 2.3) [10,15,26].



IV. Summary

pKa scales for carbon acids can provide fundamental information on the relationships between structure and reactivity. They provide quantitative information on the effects of structure on the stability of carbanions. Studies of the effect of solvent on pKa can provide insight into the factors that affect anion and ion solvation. The most reliable data for deducing intrinsic acidities and the effects of structure on acidity are provided by gas-phase studies. In solution, the effects of differential solvation of carbon acids and anions (generally with their attendant counterions) dominate acidity measurements for localized anions. Acidities in solution for acids that form delocalized anions correlate well with gas-phase data. The ability of certain substituents to acidify carbon acids in the gas phase and in solution (H2O and DMSO) are in the order -NO2 > RCO- > -CN. The orders for other substituents depend on the medium.

Carbanion stability information deduced from acidity measurements for the corresponding hydrocarbon conjugate acids can be used to make predictions regarding the relative reactivity of monomers in order to choose appropriate initiators for various monomers, to predict and understand the required order of addition of monomers for block copolymer formation, to understand the copolymerization behavior of various monomers, and to predict chain transfer activity from Bronsted acids to propagating anionic chain ends.

It is assumed that useful insight about kinetics and reactivity can be deduced from thermodynamic data. This assumption may or may not be valid depending on the nature of the transition state relative to the structure and energy of the carbanion[59,60]. However, there is no necessary direct relationship between thermodynamic stability and kinetic reactivity. Kinetics is concerned with free energies of activation and defining the structure and energy of the transition states and intermediates, as well as the reaction coordinates. Thermodynamics deals with the differences in free energy between reactants and products, which may or may not be related in structure and energy to the reactants, intermediates, and transition states whose energies are germane to kinetics.

References


2. W.-K. Li, R. H. Nobes, D. Poppinger, and L. Radom, in Comprehensive Carbanion Chemistry, Part C, E. Buncel and T. Durst, Eds., Elsevier, Chapt. 1, 1987, p. 1.

3. A. Streitwieser, Jr., E. Juaristi, and L. L. Nebenzahl, in Comprehensive Carbanion Chemistry, Part A, E. Buncel and T. Durst, Eds., Elsevier, Chapt. 7, 1980, p. 323.

4. J. R. Jones, The Ionization of Carbon Acids, Academic Press, New York, 1973.

5. O. A. Reutov, I. P. Beletskaya and K. P. Butin, CH-Acids, Pergamon, New York, 1978.

6. F. G. Bordwell, Pure Appl. Chem., 49, 963 (1977).



7. M. J. Pellerite and J. I. Brauman, in Comprehensive Carbanion Chemistry, Part A, E. Buncel and T. Durst, Eds., Elsevier, Chapt. 2, 1980, p. 55.

8. S. G. Lias, J. E. Bartmess, J. F. Liebman, J. L. Holmes, R. D. Levin, and W. G. Mallard, J. Phys. Chem. Ref. Data, 17, Suppl. 1, 1988.

9. W. K. McEwen, J. Am. Chem. Soc., 58, 1124 (1936).

10. F. G. Bordwell, Acc. Chem. Res., 21, 456 (1988).

11. R. R. Squires, Acc. Chem. Res., 25, 461 (1992).

12. C. Lambert and P. von R. Schleyer, Methoden Org. Chem. (Houben-Weyl) 4th Ed. 1952–, Bd. E19d, “Carbanionen”, Georg Thieme Verlag, Stuttgart, 1993, p. 1.

13. R. W. Taft, Prog. Phys. Org. Chem., 14, 247 (1987).

14. R. W. Taft and F. G. Bordwell, Acc. Chem. Res., 21, 463 (1988).

15. D. Algrim, J. E. Bares, J. C. Branca, and F. G. Bordwell, J. Org. Chem., 43, 5024 (1978).

16. W. N. Olmstead, Z. Margolin, and F. G. Bordwell, J. Org. Chem., 45, 3295 (1980).

17. D. T. Grimm and J. E. Bartmess, J. Am. Chem. Soc., 114, 1227 (1992).

18. R. W. Taft, I. A. Koppel, R. D. Topsom, and F. Anvia, J. Am. Chem. Soc., 112, 2047 (1990).

19. J. I. Brauman and L. K. Blair, J. Am. Chem. Soc., 92, 5986 (1970).

20. E. M. Arnett, J. Chem. Ed., 45, 793 (1968).

21. C. H. Rochester, Acidity Functions, Academic Press, New York, 1970.

22. A. A. Maryott and E. R. Smith, Nat. Bur. Std. Circ., 514 (1951).

23. F. G. Bordwell and H. E. Fried, J. Org. Chem., 46, 4327 (1981).

24. W. S. Matthews, J. E. Bares, J. E. Bartmess, F. G. Bordwell, F. J. Cornforth, G. E. Drucker, Z. Margolin, R. J. McCallum, G. J. McCollum, and N. R. Vanier, J. Am. Chem. Soc., 97, 7006 (1975).

25. F. G. Bordwell, X. Zhang, and J.-P. Cheng, J. Org. Chem., 56, 3216 (1991).

26. F. G. Bordwell, G. E. Drucker, and H. E. Fried, J. Org. Chem., 46, 632 (1981).

27. J. F. Coetzee and C. D. Ritchie, Eds., Solute—Solvent Interactions, Marcel Dekker, New York, 1969.

28. O. W. Webster, J. Am. Chem. Soc., 88, 3046 (1966).

29. R. H. Boyd, J. Phys. Chem., 67, 737 (1963).

30. V. K. Slovetskii, S. A. Shevelev, A. A. Fainzil'Berg, and S. S. Novikov, Zhur. Vses. Khim. Obshch. Mendeleeva, 6, 599, 707 (1961); cited in ref. 3.

31. G. Schwarzenbach and E. Felder, Helv. Chim. Acta, 27, 1701 (1944).

32. G. Schwarzenbach and K. Lutz, Helv. Chim. Acta, 23, 1162 (1940).



42. R. P. Bell, Trans. Faraday Soc., 39, 253 (1943).

43. Y. Chiang, A. J. Kresge, and N. P. Schepp, J. Am. Chem. Soc., 111, 3977 (1989).

44. Y. Chiang, A. J. Kresge, Y. S. Tang, and J. Wirz, J. Am. Chem. Soc., 106, 460 (1984).

45. J. H. Exner and E.C. Steiner, J. Am. Chem. Soc., 96, 1782 (1974).

46. W. N. Olmstead and F. G. Bordwell, J. Org. Chem., 45, 3299 (1980).

47. F. G. Bordwell, D. Algrim, and H. E. Fried, J. Chem. Soc. Perkin Trans. II, 726 (1979).

48. W. T. Ford, J. Am. Chem. Soc., 92, 2857 (1970).

49. E. C. Steiner and J. M. Gilbert, J. Am. Chem. Soc., 85, 3054 (1963).

50. F. G. Bordwell and D. J. Algrim, J. Am. Chem. Soc., 110, 2964 (1988).

51. C. H. Langford and R. L. Burwell, J. Am. Chem. Soc., 82, 1503 (1960).

52. E. J. King, Acid—Base Equilibria, Pergamon Press, New York, 1965, p. 299.

53. A. Streitwieser, J. C. Ciula, J. A. Krom, and G. Thiele, J. Org. Chem., 56, 1074 (1991).

54. G. Thiele and A. Streitwieser, J. Am. Chem. Soc., 116, 446 (1994).

55. A. Streitwieser, L. Xie, P. Wang, and S. M. Bachrach, J. Org. Chem., 58, 1778 (1993).

56. R. Breslow and J. L. Grant, J. Am. Chem. Soc., 99, 7745 (1977).

57. R. Breslow and R. Goodin, J. Am. Chem. Soc., 98, 6076 (1976).

58. B. Jaun, J. Schwarz, R. Breslow, J. Am. Chem. Soc., 102, 5741 (1980).

59. G. S. Hammond, J. Am. Chem. Soc., 77, 334 (1955).

60. T. H. Lowry and K. S. Richardson, Mechanism and Theory in Organic Chemistry, 3rd ed., Harper & Row, New York, 1987, p. 212.



3 Ion Pairs, Free Ions, and Stereochemistry in Carbanionic Chemistry

I. Ion Pairs and Free Ions

A. Introduction

Carbanions, free radicals, and carbocations are reaction intermediates. An intermediate is defined as a transient chemical species, with a lifetime appreciably longer than a molecular vibration (corresponding to a local potential energy minimum or depth greater than RT), that is formed (directly or indirectly) from the reactants and reacts further to give (either directly or indirectly) the products of a chemical reaction [1]. Unlike most other reaction intermediates, a wide variety of carbanions can be prepared, isolated and characterized. This chapter will review the nature of carbanions in solution, especially their interaction with counterions, and their stereochemistry.

B. Winstein Spectrum of Ionic Species: Ion Pairs

Oppositely charged ions in solution interact with each other with an attractive force according to Coulomb's law, (Eq. 3.1):

(3.1)



where zi represents the charge on a given ion, e is the unit electric charge, r is the interionic distance, and D is the dielectric constant of the medium [2]. An ion pair can be operationally defined in terms of a critical interionic distance (rc) at which the electrostatic interaction energy (Ec, see Eq. 3.1) is equal to the average kinetic energy per ion pair (3/2RT = 0.98 kcal/mol at 25°C) [2,3]. If the interionic separation is larger than rc, then the ions are categorized as free ions (see Eq. 3.2).

(3.2)

It is apparent from Eq. 3.1 that the interionic distance at which an ion pair is formed (rc) will depend on the dielectric constant of the medium. Thus, counterions are considered to be paired by this definition [3] when they are within distances of approximately 3.6Å in water (D = 78.5) [4] or as far away as 120Å in benzene (D = 2.3) [5] or dioxane (D = 2.2) [5]. For a 1M solution of a salt, the average distance between ions is 9.4Å; at this distance the interaction energy (Eq. 3.1) corresponds to 10.1 kcal/mol in acetic acid (D = 6.1) [5] but only 0.79 kcal/mol in water [2].

It has been proposed that ion pairs can exist in at least two forms [6–8]. The contact or tight ion pair corresponds to two ions in direct contact (i.e., without any intervening solvent) and is labeled R-,M+ in Figure 3.1. As the two ions separate, a second energy minimum is envisioned that corresponds to the intervention of a layer of solvent around the ion(s); conversely, as two ions approach one another, it can be envisioned that removal of the last layer of solvent to form the contact ion pair (desolvation) requires considerable energy [8]. Thus, the ionic species formed at larger distances with a sheath of solvent is called the loose or solvent-separated ion pair, labeled as R-/M+ in Figure 3.1. Szwarc [9] has noted that loose or

Figure 3.1 Potential energy curve for an ion pair as a function

of interionic distance.



solvent-separated ion pairs will exist only when at least one of the ions possesses a tight solvation shell; conversely, if the interaction of both ions with the solvent is weak, only tight or contact ion pairs will exist. The equilibrium between contact and solvent-separated ion pairs will depend on the structure of the carbanion, counterion, solvent, and temperature.

For delocalized carbanionic systems (such as the allylic and benzylic species) involved as propagating species in anionic polymerization, the presence and role of ion pairs has been carefully and clearly elucidated [10]. Thus, in addition to the aggregated (1) and unassociated (2) species that can exist in hydrocarbon solution (see Chapter 1), in polar solvents it is necessary to consider the intervention of free ions (5), the contact (3) and solvent-separated (4) ion-paired carbanion species as shown in Scheme 3.1 where Mt represents a metallic counterion such

Scheme 3.1

as an alkali metal cation. In principle, each of these types of intermediates can participate as reactive propagating species in anionic polymerization under certain experimental conditions. Thus, the kinetics of propagation can be complicated by the participation of more than one type of these carbanionic intermediates as active propagating centers because each carbanionic species would be expected to react with monomer with its own unique rate constant as shown in Scheme 3.2, where [M] represents the concentration of monomer.

Scheme 3.2

C. Experimental Evidence for Carbanion Ion Pairs

One of the unique aspects of carbanion chemistry is that these intermediates can be generated in a variety of media and exhibit stabilities and lifetimes that permit their direct observation and characterization. Although the Winstein spectrum of ion-paired species [8] was proposed on the basis of the kinetics and stereochemistry of nucleophilic substitution reactions proceeding via carbocation intermediates, the first direct spectral observations of chemically distinguishable forms of ion pairs in solution were reported for carbanion intermediates [11–13].



The most extensive studies have been carried out for salts of the fluorenyl anion, FL-, whose basic structure is shown below. Spectrophotometric studies of

alkali metal salts of fluorenyl carbanions in solvents of low dielectric constant show dramatic changes with temperature as shown in Figure 3.2. For example, the ultraviolet (UV)-visible spectrum of fluorenyl sodium in THF exhibits an absorption peak at 355 nm; on cooling, the intensity of this band decreases and a new absorption band at 373 nm appears and grows in intensity. All of these changes were reversible on warming. It was found that the addition of highly ionized common ion salts such as sodium tetraphenylboron did not affect these spectral changes [14]. This result indicates that the equilibria do not involve free ions

Figure 3.2 UV-visible spectra for fluorenyl sodium in THF as a function

of temperature. (From Ref. 14; reprinted by permission of the American Chemical Society.)



whose concentration would decrease in the presence of common ion (see Eq. 3.2), but this result is consistent with the interconversion of different types of ion pairs whose concentrations are not affected by the presence of common ions. Further evidence for the absence of significant free ion contributions was obtained by conductance measurements that indicated a negligible concentration of free ions at the concentrations used for the spectrophotometric studies (ca. 10-3 M) [15]. No dependence of these spectral changes on concentration was observed; this result was interpreted to mean that these spectral changes were not due to the effects of ion pair aggregation that should be concentration dependent. Thus, the spectral changes were interpreted in terms of a temperature-dependent equilibrium between two types of ion pairs, the contact (or tight) ion pairs [Na+,FL-] and the solvent-separated (or loose) ion pairs [Na+//FL-] as shown in Eq. 3.3.

(3.3)

A summary of the spectral data and fraction of solvent-separated ion pairs for a variety of alkali and alkaline earth salts of fluorene is shown in Table 3.1.

The first obvious conclusion from the data in Table 3.1 is that ion pair

Table 3.1. UV-Visible Spectral Data and Fractions of Solvent-Separated Ion Pairs for Fluorenyl Salts at Room Temperature

Cation (Solvent)

λmax(nm)

ε a Contact Solvent-Separated Fssb

Li(THP) c 5.6 349 373 0.30

Na(THP) 356 <0.01

K (THP) 362 <0.01

Cs(THP) 364 <0.01

Li(THF) 7.4 349 373 0.80

Na(THF) 356 373 0.04

Li (DME) 7.2 373 1.0

Na (DME) 356 373 0.8

K(DME) 362 0.1

Cs(DME) 364 <0.01

Ba(THF) 7.4 347 <0.05

Sr(THF) 343 0.07

aSolvent dielectric constant.bMole fraction of solvent-separated ion pairs.cTHP, tetrahydropyran; THF, tetrahydrofuran; DME, dimethoxyethane.

Source: Ref. 13.



equilibria depend on counterion and solvent. Since the fluorenyl carbanion is a highly delocalized ion, it is assumed that this species is not specifically solvated (see Chapter 2). In general, the conversion from contact to solvent-separated ion pairs involves charge separation and cation solvation. Thus, this interconversion is favored by increasing the polarity and cation-solvating power of the solvent in the order DME > THF > THP, regardless of the counterion. The actual complexity of the solvent-cation interaction has been indicated by the fact that the fraction of solvent-separated ion pairs decreases in the order 0.8, 0.25, 0.02 upon changing the solvent from THF to 2-methyltetrahydrofuran to 2,5-dimethyltetrahydrofuran; thus, steric effects have a dramatic effect on cation solvation. For the interpretation of counterion effects, two factors can be considered. First, the interaction between the counterion and the carbanion would be expected to decrease with increasing size of the cation. In these polar media, this factor is not significant since the formation of solvent-separated ion pairs is favored by decreasing the size of the cation. The second important factor is the solvation of the cation. The interaction of the solvent with the cation would be expected to decrease with increasing size of the cation; this seems to be the important effect of counterion variation for ion pair equilibria in polar media. Thus, lithium which has the smallest size, forms the largest fraction of solvent-separated ion pairs for any given solvent system. Another interesting aspect of the data in Table 3.1 is the fact that although the absorption maximum attributed to the contact ion pair is quite sensitive to the specific cation, the peak corresponding to the solvent-separated ion pair does not depend on the cation or the solvent. This is consistent with the hypothesis that no specific anion solvation is involved for the delocalized fluorenyl carbanion. Furthermore, it is reported that the absorption maximum for the free ion is very similar to that of the solvent-separated ion pair. The absorption maximum corresponding to the contact ion pair undergoes a bathochromic shift to longer wavelengths with increasing radius of the cation for the alkali metal cations. This is consistent with perturbation of the charge distribution in the carbanion by the presence of the counterion (i.e., electronic charge would be expected to be concentrated in the vicinity of the counterion) (see Chapter 1). This effect would be expected to be larger for the small lithium cation and to decrease with increasing cation size.

Crown ethers are cyclic oligomers of ethylene oxide and their derivatives [16]. The structures of several common crown ethers are shown below. The uniqueness of crown ethers is that they are very strong cation-coordinating species. The binding strength varies with the effective size of the cavity within the crown ether and also with the size of the cation. When dicyclohexyl-18-crown-6 is added to a solution of fluorenylsodium in THF, the UV-visible absorption maximum changes from 355 nm to 372 nm [17]. This has been interpreted in terms of the complete conversion of the contact ion pair to the solvent-separated ion pair



for which the formation constant is estimated to be >107. In contrast, addition of 4'-methylbenzo-15-crown-5 to a solution of fluorenylsodium in THF causes a shift in the UV-visible absorption maximum to two new peaks at 359 nm and at 370 nm [18]. The smaller shoulder at 370 nm was interpreted in terms of formation of a small amount of solvent-separated ion pair, but the main peak at 359 nm was attributed to a contact ion pair with external solvation of the sodium cation by the added crown ether.

The effect of temperature on the equilibrium between solvent-separated and contact ion pairs should be considered. In general, the enthalpy of dissociation to form the solvent-separated ion pair is exothermic. However, the conversion of the contact to the solvent-separated ion pair will involve specific solvation of the cation. This solvation process would be expected to have a negative entropy; however, the contribution of this unfavorable entropy to the equilibrium will



decrease with decreasing temperature in accordance with the Gibbs free energy equation (Eq. 3.4). Another factor that will favor solvation and formation of

∆G = ∆H - T ∆S (3.4)

solvent-separated ion pairs is the fact that most solvents exhibit a negative temperature coefficient for the dependence of dielectric constant on temperature; thus, the dielectric constant of the medium will increase with decreasing temperature that will favor formation of solvent-separated ion pairs [9,13].

The ion pair dissociation constants, Ki (see Eq. 3.3), and thermodynamic parameters for formation of solvent-separated from contact ion pairs are shown in Table 3.2. These data show that the formation of solvent-separated ion pairs is an exothermic process with a large, unfavorable, negative entropy. The dependence of the ion pair dissociation constant on counterion is also large and in the expected order (i.e., largest for lithium and decreasing with increasing cation size).

The final factor that should be considered is the effect of carbanion structure. The primary effect of changing the structure of the carbanion will be to alter the interaction between the carbanion and the counterion. This interaction, analogous to the interaction of ions with solvent, will increase with decreasing size of the ions. Thus, it would be expected, and it is observed, that the fraction of solvent-separated ion pairs increases with increasing anion size. The fraction of solvent-separated ion pairs in DME at 25°C is only 0.5 for indenyllithium but greater than 0.99 for fluorenyllithium; similarly, in THF at 25°C, this fraction is 0.8 for fluorenyllithium but >0.99 for 2,3-benzofluorenyllithium [13].

D. Free Ions

The concentration of free ions in solvents such as ethers used in anionic polymerization is quite small as deduced from conductivity measurements [15,19]. However, the importance of free ions as kinetically active species cannot be ignored because they exhibit exceptionally high reactivity toward monomers [10]. The contribution of free ions can be dramatically reduced by the addition of common

Table 3.2. Dissociation Constants and Thermodynamic Parameters for Ion Pair Dissociation (Eq. 3.3) in THF at 20°C

Fluorenyl Salt 102 Ki ∆Hi (kcal/mol) ∆Sio (e.u.)

LiFL 290 -7.0 -22

NaFL 6 -7.6 -33

CsFL <1

Source: Ref. 13.



ions that effectively shift the equilibrium in Eq. 3.2 toward the formation of ion pairs.

The ionization constants, Kd, and thermodynamic parameters for formation of free ions from ion pairs are shown in Table 3.3. Since both contact and solvent-separated ions may be present in solution, it is important to consider which ion-pair species is involved in the equilibrium in order to understand these results (see Table 3.1). The potassium and cesium salts exist primarily in the contact ion pair form in both solvents. The sodium salt changes from contact ion pairs to solvent-separated ion pairs upon cooling to -60°C; in DME it is essentially all solvent separated below -10°C. The lithium salt is 80% and essentially 100% solvent separated in THF and DME, respectively.

It is interesting to compare these data with the corresponding data for the ion pair dissociation process listed in Table 3.2. In general, these two processes exhibit the same dependence on counterion, exothermicity, and negative entropy. This is expected from consideration of the Winstein spectrum for carbanions (Scheme 3.1) since the formation of the solvent-separated ion pair is a process that involves dissociation and charge separation analogous to the dissociation process to form free ions. Furthermore, both of these equilibria should respond to temperature in

Table 3.3. Dissociation Constants and Thermodynamic Parameters for Formation of Free Ions (Eq. 3.2)

Fluorenyl Salt

Solvent

Temperature (°C)

107Kd

(M)∆Hd

(kcal/mol)∆Sd° o

(e.u.)

LiFL THF 20 34 -3.5 -37

-70 155

DME 15 62 -2.9 -33

-65 214

NaFL THF 20 7.5 -8.3 -56

-70 480

DME 20 64 -3.9 -37

-70 370

KFI THF 20 1.9 -4.7 -47

-70 33

DME 20 29 -5.9 -45

-70 314

CsFL THF 20 0.15 -2.9 -46

-70 0.55

DME 15 2.5 -4.6 -45

-65 30

Source: Ref. 19.



the same manner by increasing the fraction of dissociated species as the temperature is lowered, because these dissociation processes are exothermic with negative entropy terms.

II. Stereochemistry in Carbanion Chemistry

A. Introduction

Two points can be gleaned from the theoretical studies of the structure of carbanions and their isoelectronic relationship to amines (see Chapter 1): (a) simple alkyl carbanions would be expected to exist in a pyramidal configuration and thus should be capable of existing in two enantiomeric forms (see Scheme 3.3); and (b) relatively low barriers are predicted for pyramidal inversion of simple carbanions and the pyramidal inversion process interconverts the enantiomeric forms of the carbanions as shown in Scheme 3.3.

Scheme 3.3

It would be expected that ion pairing and/or some degree of covalent character in the carbon-metal bonds of organometallic compounds would tend to slow down the inversion process (see Scheme 3.4). The configurational inversion process illustrated in Scheme 3.4 requires that the counterion (e.g., lithium) moves

Scheme 3.4



from one face to the other. To the extent that a charge-separated species is formed as a transition state or intermediate in this isomerization, this would increase the barrier relative to the simple inversion process illustrated in Scheme 3.3. Possible mechanisms for inversion could involve migration of lithium from one face of the carbon skeleton to the other via a dissociated type of ion pair intermediate (see Scheme 3.5), or by formation of a less dissociated ion pair in which rotation of

Scheme 3.5

Scheme 3.6

the carbanion occurs (see Scheme 3.6). It should be noted that these schemes neglect the known propensity of organolithium compounds to exist predominantly as aggregated species in solution (see Chapter 1). However, in general, the information available to characterize the stereoisomerism process is insufficient to deduce whether it is the associated or unassociated species that is involved; hence, simplifying, unassociated structures are schematically used herein.

B. Configurational Stability of Carbanions

General Aspects

Information about the structure and immediate environment of reaction intermediates has often been deduced through use of stereochemical studies [20]. The stereochemical course of substitution reactions of chiral carbanions was first



deduced from studies of reactions of 2-methyl-1-phenyl-1-butanone [21–23]. These detailed studies showed that for chiral 1 the rates of hydrogen-deuterium

exchange (kex) were equal to the rates of racemization (krac) in mixtures of dioxane, deuterium oxide, and sodium deuteroxide; in addition, the rate of acetate-catalyzed bromination (kBr) was found to be equal to the rate of racemization (krac). Furthermore, a primary deuterium isotope effect (kH/kD = 6.1) has been reported for the bromination of methyl cyclohexyl ketone in methanol [24]. These results have been interpreted in terms of formation of a symmetrical carbanion (enolate anion) intermediate that could be captured from either face with equal probability as shown in Scheme 3.7. It is assumed that the rate-determining step

Scheme 3.7

(krds) is removal of the proton to form the enolate intermediate; thus, all subsequent steps do not affect the observed rates and the same rate constants are observed for racemization, bromination, and hydrogen-deuterium exchange. Protonation could also occur on oxygen to form the symmetrical enol intermediate that would also result in the same kinetic and stereochemical outcome.

Of course, it should be recognized that these results can equally well be interpreted in terms of a chiral carbanion intermediate if it is assumed that the rate of



configurational interconversion (see Scheme 3.3) is rapid relative to protonation (deuteration). With regard to the lifetime of chiral carbanionic intermediates in hydroxylic media, it has been reported that a chiral carbanionic intermediate can be trapped in 9N aqueous potassium hydroxide solution as shown in Scheme 3.8

Scheme 3.8

[25]. These results do not require the intervention of a chiral carbanion; the collapse of an unsymmetrically solvated carbanion (e.g., hydrogen-bonded to the carboxylic acid) prior to formation of a symmetrically solvated species would also explain the observations.

The first study describing the stereochemistry of formation and substitution reactions for chiral organometallic compounds was the reaction of optically active 2-octyl iodide with sec-butyllithium in 94% petroleum ether/6% diethyl ether at -70°C followed by carboxylation to give 2-methyloctanoic acid with 20% overall retention of configuration [26]. If the reaction product was allowed to warm to 0°C, a racemic product was obtained. If it is assumed that both of these reactions (lithium—iodide exchange and carbonation) proceed with retention of configuration, these results can be interpreted in terms of formation of a chiral organolithium compound able to retain its configuration at low temperatures.

The treatment of optically active di-sec-butylmercury with racemic 2-octyllithium in pentane at temperatures below 0°C followed by carboxylation to give 2-methylbutanoic acid (see Scheme 3.9) proceeded with 13–83% net retention of configuration depending on the reaction time before carbonation [27]. In the presence of 6% diethyl ether, only the racemic acid was obtained under conditions in which 55% retention was observed in hydrocarbon solution. The effect of ether on the configurational stability of this organolithium compound is consistent with racemization occurring either via a less aggregated or a more dissociated form of the organolithium compound (e.g., via a dimer vs. the tetrameric aggregate) or via



Scheme 3.9

a solvent-separated ion pair versus the contact ion pair. The lithium—bromide exchange and subsequent carbonation reactions of cis- and trans-2-methylcyclopropyl bromide have likewise been reported to proceed stereospecifically to produce the corresponding isomers with overall retention of configuration [28]. Analogous studies with optically active 2,2-diphenyl-1-methylcyclopropyl bromide confirmed the configurational stability of cyclopropyllithium species and showed that these results were not dependent on reaction time or the presence of ether additives [20,29]. The stereochemical stability of cyclopropyllithiums is consistent with expected high amounts of s-orbital character (J13C-H = 162 Hz, 32% s character, sp2.28) in the CH bond [30]; in general, a linear correlation is observed between acidity (pKa) and the amount of s character in the carbon orbital used in the CH bond [20,31,32].

The stereochemical integrity of alkyllithium compounds can also be affected by intramolecular interactions with coordinating groups. It has been reported that lithium—tin exchange of the chiral tributyltin derivative, 2, followed by alkylation with dimethylsulfate produced the corresponding methyl derivative with retention of configuration even in tetrahydrofuran as shown in Scheme 3.10 [33].

Scheme 3.10



The rates and energetics of configurational inversion in alkyllithium compounds have been investigated by 1H nuclear magnetic resonance (NMR) spectroscopy [34,35]. The 1H NMR spectrum for neohexyllithium (3,3-dimethylbutyl-lithium) in diethyl ether changes from an AA'BB' pattern at - 18°C to an A2B2 pattern upon heating [34]. It was concluded that the observed spectral changes, resulting from averaging of the vicinal coupling constants, were only consistent with inversion of configuration at the alpha carbon as represented below (Eq. 3.5). There was no effect of dilution on the inversion rate, suggesting a unimolecular

(3.5)

process for inversion. Since organolithium compounds are aggregated in diethyl ether (see Chapter 1), it is presumed that inversion takes place in an aggregated species (not shown). The calculated energy barrier for inversion was 15 ± 2 kcal/mol. Analogous studies of 2-methylbutyllithium in pentane determined that the free energy of activation for inversion was 15.8 ± 1 kcal/mol compared to a free energy of activation of 13.3 ± 1.5 kcal/mol for the interaggregate exchange processes [35]. The degree of aggregation of 2-methylbutyllithium in pentane was determined to be hexameric (0.89M) [35]. The low barrier for configurational inversion even in hydrocarbon media suggests that this process occurs within an aggregated species without formation of dissociated, ion-paired species (see Chapter 1). A schematic representation of the geometry of the transition state for an inversion process occurring in a dimer is shown below [36,37]. Interaction of a lithium atom from the back side would generate a symmetrical, planar transition state structure at the inverting carbon as shown below in Structure 3.2.



Configurations of Benzyl Carbanions

In general, benzylic carbanions exist as planar, charge-delocalized species [38]. An apparent exception to this generalization is 7-phenylnorbornyllithium [39]. In contrast to 7-phenylnorbornylpotassium, which exhibits only two aliphatic 13C NMR peaks, the C2 resonance for 7-phenylnorbornyllithium splits into two peaks at low temperature corresponding to separate resonances for C2 and C6 (see structures below). It was concluded that this reversible process was consistent with configurational inversion of a pyramidal carbanion structure (see structures below in Scheme 3.11). Other nonpyramidal structures such as the asymmetrical planar ion pair structure (see below) were ruled out on the basis that the large chemical shifts observed for the bridgehead carbons in the lithium versus the

Scheme 3.11

potassium derivative were only consistent with a fundamental change in structure. It is important to note that the J13C-H coupling constant for benzyllithium changes from 116 Hz in benzene to 132 Hz in THF [40]. Since a general correlation has been developed between the J13C-H coupling constants and the amount of s character in the carbon orbitals used to bond to hydrogen [41,42], these results suggest that there is increased sp3 character in the α-carbon in benzyllithium in benzene compared to THF [40]. Another interesting facet of the structure of benzyllithiums is the fact that the barrier to rotation about the aryl-methylene bond causes the ortho and meta protons and carbons to exhibit splitting due to chemical nonequivalence [43–46]. For example, the adduct of t-butyllithium and α-methylstyrene (see structures below, Eq. 3.6) exhibits separate resonances for each of the ortho and meta protons in the 1H NMR spectrum [43]. From the temperature dependence of the spectra and the coalescence temperature, the barrier to rotation



(3.6)

about the aryl-methylene bond was estimated to be 14 kcal/mole. A lower barrier to rotation (11.9 kcal/mole) was observed for the corresponding adduct with styrene [43]. Analogous effects are observed for the corresponding carbon resonances [44–46].

Configurations of Allyl Carbanions

The delocalized structure of allylic anions suggests that substituted allylic anions should exist in stereoisomeric forms with barriers to interconversion that reflect the resonance energy of the anion as discussed previously in Chapter 1. The adduct of t-butyllithium and butadiene (i.e., neopentylallyllithium) exists as a mixture of cis (Z) and trans (E) isomers in toluene (1/3 ratio) [47] and in benzene (1/2.6 ratio) [48]. These isomers did not equilibrate at temperatures up to 70°C. Based on 13C NMR analysis it has been proposed that neopentylallyllithium exists as a mixture of covalent and ionic delocalized forms in fast dynamic equilibrium (see covalent and ionic forms in Scheme 3.12) [49]. It should be noted that these

Scheme 3.12



species are aggregated in hydrocarbon solution (see Chapter 1) and the preferred configuration may be influenced by the structure of the aggregate [47]. The adduct of isoprene and t-butyllithium displays a similar preference for the trans chain end structure (T/C = 3/1) [50].

In the presence of Lewis base additives and in polar media, the cis (Z) isomer is favored (cis/trans > 2/1) for the t-butyllithium adducts of both butadiene [48,51] and isoprene [52]. With less sterically demanding alkyl groups, the preference for the Z configuration is enhanced in polar media [38].

Achiral Carbanions in Chiral Environments

Extensive studies by Cram and co-workers [20] have shown that the stereochemistry of exchange reactions for carbon acids that form achiral, planar carbanions can proceed via complete retention, racemization, complete inversion, or even via inversion without exchange (isoracemization) depending upon the carbon acid, base, acidity of the proton donor, solvent polarity, and temperature. A general description of the stereochemical course of the exchange reaction is shown in Scheme 3.13. The various stereochemical pathways for D/H exchange reactions

Scheme 3.13

Table 3.4. Stereochemical Pathways for D/H Exchange Reactions

Stereochemistry Characteristics Limiting kexkα

Retention kα = 0

Racemization kex = kα 1

Inversion kinv = kex = 1/2kα 0.5

Isoracemization (Isoinversion) kex = 0 0



are shown in Table 3.4, where kα is the rate constant for racemization, kinv is the rate constant for inversion, and kex is the rate constant for isotopic exchange. The most studied carbon acid systems with regard to the stereochemical course of exchange reactions under a variety of conditions are various substituted fluorene systems, whose structures are shown below. Some representative data for ex-

change reactions are shown in Table 3.5. Thus, the stereochemical course of exchange reactions of carbanions does not by itself unambiguously define the structure of the carbanion (planar or pyramidal). In general, chiral reactions can occur if the neighboring environment of the carbanion is chiral, even when the carbanion is planar and achiral. Furthermore, racemization will occur for either a planar or pyramidal carbanion when the reaction conditions, especially increased solvent polarity, promote dissociation to form more dissociated species such as solvent-separated ion pairs.

III. Summary

The structures of organometallic compounds in solution vary with the counterion, the solvent, and the temperature. As the polarity of the solvent increases, a shift to more ionic structures has been observed. The spectrum of active species in carbanion reactions can range from organolithium aggregates in hydrocarbon media to free ions in polar solvents. Experimental evidence suggests that there are two types of ion pairs: contact and solvent separated ion pairs. Solvent-separated ion pairs are favored by cation solvation, polar media, and low temperatures. In

Table 3.5. Relative Rates of Exchange vs. Racemization for Base-Catalyzed Hydrogen Isotope Exchange Reactions

Carbon Acid Solvent Base Kex/kα

3.3 THF NH3148

3.3 DMSO NH3 1.0

3.3 CH3OH Pr3N 0.65

3.3 THF Et3N 0.23

3.4 THF, 1.5M t-BuOH, 10-4M Pr3NHI Pr3N 0.1



general, carbanionic species do not maintain their configurational integrity. Chiral alkyllithium compounds can maintain their configuration at low temperatures in nonpolar media. The stereochemical course of carbanion reactions can vary from retention, racemization, inversion, and isoinversion depending on the structure of the carbanion, the counterion, the solvent, and the temperature. Substrates that generate achiral carbanionic intermediates can give chiral products when the reaction conditions favor or provide a chiral environment.

References

1. Glossary of Terms Used in Physical Organic Chemistry, Pure Appl. Chem., 51, 1725 (1979).

2. E. M. Kosower, An Introduction to Physical Organic Chemistry, Wiley, New York, 1968, p. 344.

3. J. E. Gordon, The Organic Chemistry of Electrolyte Solutions, Wiley, New York, 1975.

4. A. A. Maryott and E. R. Smith, Nat. Bur. Std. Circ., 514 (1951).

5. Solute-Solvent Interactions, J. F. Coetzee and C. D. Ritchie, Eds., Marcel Dekker, New York, 1969, p. 52.

6. E. Grunwald, Anal. Chem., 26, 1696 (1954).

7. H. Sadek and R. M. Fuoss, J. Am. Chem. Soc., 76, 5897, 5905 (1954).

8. S. Winstein, E. Clippinger, A. H. Fainberg, and G. C. Robinson, J. Am. Chem. Soc., 76, 2597 (1954).

9. M. Szwarc, in Ions and Ion Pairs in Organic Reactions, M. Szwarc, Ed., Wiley-Interscience, New York, Vol. 1, 1972, p. 15.

10. M. Szwarc, Adv. Polym. Sci., 49, 1 (1983).

11. T. E. Hogen-Esch and J. Smid, J. Am. Chem. Soc., 87, 669 (1965).

12. J. Smid, in Ions and Ion Pairs in Organic Reactions, M. Szwarc, Ed., Wiley-Interscience, New York, Vol. 1, 1972, p. 85.

13. T. E. Hogen-Esch, Adv. Phys. Org. Chem., 15, 153 (1977).



16. C. J. Pedersen and H. K. Frensdorff, Angew. Chem. Int. Ed. Engl., 11, 16 (1972).

17. K. H. Wong, G. Konizer, and J. Smid, J. Am. Chem. Soc., 92, 666 (1970).

18. U. Takaki, T. E. Hogen-Esch, and J. Smid, J. Am. Chem. Soc., 93, 6760 (1971).

19. T. Ellingsen and J. Smid, J. Phys. Chem., 73, 2712 (1969).


21. C. L. Wilson, J. Chem. Soc., 1550 (1936).

22. S. K. Hsu and C. L. Wilson, J. Chem. Soc., 623 (1936).



29. H. M. Walborsky, F. J. Impastato, and A. E. Young, J. Am. Chem. Soc., 86, 3283 (1964).

30. W. A. Bernett, J. Chem. Ed., 44, 17 (1967).

31. M. J. Pellerite and J. I. Brauman, in Comprehensive Carbanion Chemistry, Part A, E. Buncel and T. Durst, Eds., Elsevier, 1980, Chapt. 2, p. 55.

32. R. W. Taft and F. G. Bordwell, Acc. Chem. Res., 21, 463 (1988).

33. W. C. Still and C. Sreekumar, J. Am. Chem. Soc., 102, 1201 (1980).

34. M. Witanowski and J. D. Roberts, J. Am. Chem. Soc., 88, 737 (1966).

35. G. Fraenkel, W. E. Beckenbaugh, and P. P. Yang, J. Am. Chem. Soc., 98, 6878 (1976).

36. T. Clark, P. von Rague Schleyer, and J. A. Pople, J. Chem. Soc., Chem. Commun., 137 (1978).

37. E. Kaufmann, K. Raghavachari, A. E. Reed, and P. von Rague Schleyer, Organometallics, 7, 1597 (1988).

38. J. L. Wardell, in Comprehensive Organometallic Chemistry; The Synthesis, Reactions and Structures of Organometallic Compounds, G. Wilkinson, F. G. A. Stone, and E. W. Abel, Eds., Pergamon Press, Oxford, 1982, Vol. 1, p. 43.

39. P. R. Peoples and J. B. Grutzner, J. Am. Chem. Soc., 102, 4709 (1980).

40. L. D. McKeever, in Ions and Ion Pairs in Organic Reactions, Vol. 1, M. Szwarc, Ed., Wiley-Interscience, New York, 1972, p. 263.

41. D. M. Grant and W. M. Lichman, J. Am. Chem. Soc., 87, 3994 (1965).

42. R. Waack, M. A. Doran, E. B. Baker, and G. A. Olah, J. Am. Chem. Soc., 88, 1272 (1966).

43. S. Brownstein and D. J. Worsfold, Can. J. Chem., 50, 1246 (1972).

44. K. Matsuzaki, M. Henmi, T. Kanai, and T. Iwamoto, Makromol. Chem., Rapid Commun., 3, 83 (1982).

45. G. Fraenkel, M. J. Geckle, A. Kaylo, and D. W. Estes, J. Organomet. Chem., 797, 249 (1980).

46. G. Fraenkel and M. J. Geckle. J. Am. Chem. Soc., 102, 2869 (1980).

47. W. H. Glaze, J. E. Hanicak, M. L. Moore, and J. Chaudhur, J. Organomet. Chem., 44, 39 (1972).

48. S. Bywater, D. J. Worsfold, and G. Hollingsworth, Macromolecules, 5, 389 (1972).

49. G. Fraenkel, A. F. Halasa, V. Mochel, R. Stumpe, and D. Tate, J. Org. Chem., 50, 4563 (1985).


51. R. T. McDonald, S. Bywater, and P. Black, Macromolecules, 20, 1196 (1987).

52. C. J. Dyball, D. J. Worsfold, and S. Bywater, Macromolecules, 72, 819 (1979).



II INTRODUCTION TO ANIONIC ANDLIVING POLYMERIZATION



4 Living Polymerizations: Definitions, Consequences, and Criteria

I. Introduction

In principle, living polymerizations provide the most versatile methodologies for the preparation of macromolecules with well-defined structures and low degrees of compositional heterogeneity [1–6]. Using these methodologies it is possible to synthesize macromolecular compounds with control of a wide range of compositional and structural parameters including molecular weight, molecular weight distribution, copolymer composition and microstructure, stereochemistry, branching, and chain-end functionality [1–6]. Since the insightful description of the “living” nature of anionic polymerizations of styrene and diene monomers by Szwarc and co-workers in 1956 [7,8], a variety of other mechanistic types of living polymerizations have been developed including cationic [6,9–11], radical [12,13], Ziegler-Natta [14,15], ring-opening metathesis [5,16–18], and group-transfer polymerization [4,19–21]. In addition, qualifying descriptions such as the “degree of livingness,” quasi-living [22], pseudo-living [12,23], immortal [24], truly living [4,11], and livingness enhancement [25] have been advanced. In this chapter the meaning of living polymerization will be presented, the experimental consequences of living polymerization will be outlined, and a critical appraisal of the various criteria used to deduce whether a given system is living or not will be discussed [26].


2004-4-22http://www.netlibrary.com/nlReader/nlReader.dll?bookid=12873&filena...

II. Definitions

A living polymerization is a chain polymerization that proceeds in the absence of the kinetic steps of termination or chain transfer [7,8,27]. A recent review states that “The term living polymers is often used to describe systems in which active centers remain after complete polymerization, so that a new batch of monomer subsequently added will add to the existing chains and increase their degree of polymerization” [11]. The limitations of this alternative description will be discussed in connection with criteria for living polymerizations.

It is worthwhile for clarity to reconsider the definitions for all of the terms in the above definition [28]. A chain reaction is a reaction in which one or more reactive reaction intermediates are continuously regenerated, usually through a repetitive cycle of elementary steps (the propagation steps). In chain reaction polymerization, reactive intermediates of the same type, generated in successive steps or cycles of steps, differ in molar mass. Thus, living polymerizations belong to the mechanistic class of chain reaction (chain-growth) polymerizations as compared to step reaction (step-growth) or condensation polymerizations.

A reaction intermediate is a transient chemical species, with a lifetime appreciably longer than a molecular vibration (corresponding to a local potential energy minimum or depth greater than RT), that is formed (directly or indirectly) from the reactants and reacts further to give (either directly or indirectly) the products of a chemical reaction (e.g., free radicals, carbanions, and carbocations are reactive intermediates).

Initiation is a reaction or process generating free radicals (or some other reactive reaction intermediates) that then participate in a chain reaction; initiation is illustrated in Eqs. 4.1 and 4.2, where I is an initiator precursor, I* is the initiating species, M is a monomer molecule, and the symbol * is used to represent a radical, cationic or anionic center.

(4.1)

(4.2)

Propagation is the continuous regeneration of reactive intermediates, through a repetitive cycle of elementary steps, which differ in molar mass; propagation is illustrated in Eq. 4.3, where i is an index indicating the degree of polymerization of the growing polymer chain,

(4.3)

Termination is the step in a chain reaction in which the reactive reaction intermediates are destroyed or rendered inactive, thus ending the chain; termination is illustrated in Eq. 4.4, where P is a dead or inactive polymer with respect to chain growth.

(4.4)



Chain transfer (intermolecular) involves the transfer, by the reactive intermediate end of a growing chain polymer, of an atom (or group) to or from another molecule (or polymer); the intramolecular analog involves transfer to or from a site within the polymer chain. The growth of the polymer chain is thereby terminated but a new reactive intermediate, capable of chain propagation and polymerization, is simultaneously created. Chain transfer is illustrated in Eqs. 4.5 and 4.6 where A-X is a chain transfer agent and A* is a new reactive intermediate capable of continuing the chain growth reaction by reinitiating chain growth at a rate comparable to the normal chain propagation rate. The rate of reinitiation is the important feature that differentiates chain transfer from retardation and inhibition, in which the rate of reinitiation of chain growth after transfer is slower than the rate of propagation [29].

(4.5)

(4.6)

It should be noted that reversible chain transfer and/or chain termination reactions occur for certain polymerizations that show characteristics in common with living polymerizations [6,24]. The question of whether these systems should be classified as living will be deferred to a later section.

III. Controlled Syntheses Using Living Polymerizations

Polymerization reactions that proceed in the absence of chain transfer and chain termination provide versatile methodologies for the preparation of polymers with control of the major variables that affect polymer properties. Well-defined polymers can be prepared with low degrees of compositional heterogeneity. The control variables and the synthetic applications of living polymerizations are outlined in this section.

A. Molecular Weight

Molecular weight is one of the most important variables affecting polymer properties. High molecular weight is the key feature of polymers that differentiates them from low-molecular-weight compounds as delineated in the macromolecular hypothesis by Staudinger [30,31]. The molecular weight in a living polymerization is controlled by the stoichiometry of the reaction and the degree of conversion. For a monofunctional initiator under ideal conditions, one polymer chain is formed for each initiating molecule or species. Thus, at complete conversion the expected number average molecular weight can be calculated as shown in Eq. 4.7. When two initiator molecules form one difunctional initiating species, as is the case for alkali metal naphthalenide anion initiators [27,32], the predicted molecular weight



is twice as high per mole of initiator, as shown in Eq. 4.8. At intermediate degrees of conversion, the molecular weight is related to the grams of monomer consumed and Eq. 4.7 is replaced by Eq. 4.9. From Eq. 4.9 it is predicted that the number

Mn = g of monomer/moles of initiator (4.7)

Mn = g of monomer /(1/2) moles of initiator (4.8)

Mn = g of monomer consumed/moles of initiator (4.9)

average molecular weight (or Xn, the number average degree of polymerization) will be a linear function of conversion for a living polymerization. From a practical point of view, it is possible to prepare polymers with predictable molecular weights ranging from 3 g/mol to > 106 g/mol using living polymerizations [33].

B. Molecular Weight Distribution

In general, it is possible to prepare a polymer with a narrow molecular weight distribution (Poisson distribution) using a living polymerization when the rate of initiation is competitive with the rate of propagation [1]. This condition ensures that all of the chains grow for essentially the same period of time.

The relationship between the polydispersity and the degree of polymerization for a living polymerization is shown in Eq. 4.10 [30,34,35]; the second approxi-

Xw/Xn = 1 + [Xn/(Xn + 1)2] ≅ 1 + [1/Xn] (4.10)

mation is valid for high molecular weights. Thus, it is predicted that the molecular weight distribution will decrease with increasing molecular weight for a living polymerization system. From a practical perspective, it is easier to make highermolecular-weight polymers with narrow molecular weight distributions; the preparation of lower-molecular-weight polymers with narrow molecular weight distributions generally requires careful attention to experimental details. Broader molecular weight distributions are obtained using less active initiators [36], with mixtures of initiators [37], or with continuous addition of initiator as involved in a continuous flow, stirred tank reactor [38–42].

C. Block Copolymers

One of the unique aspects of living polymerizations is the fact that all of the chains retain their active centers when all of the monomer has been consumed. Consequently, if additional monomer is introduced into the system, the molecular weight will increase by an amount that can be calculated using Eq. 4.9. Furthermore, if a different monomer is added, a diblock copolymer will be formed (see Chapter 12) [43,44]. Sequential addition of monomer charges can generate diblocks, A—B, triblocks such as A—B—A, A—B—C, and even more complex multi



block structures. It is important to note that, in principle, each of the blocks in these polymers can be prepared with controlled molecular weight and narrow molecular weight distribution.

D. Chain-End Functionalized Polymers

Another consequence of the fact that all of the chains retain their active centers when the monomer has been consumed is the ability to effect controlled termination reactions. This provides a methodology for the synthesis of chain-end functionalized polymers with a variety of functional end groups as shown in Eqs. 4.11 and 4.12 for living anionic and cationic [6] polymerizations (see Chapter 11)

(4.11)

(4.12)

[45,46], respectively. Of course, the efficiency of these ω-functionalization reactions is often less than 100%.

An alternative methodology for the synthesis of functionalized polymers using living polymerization is the use of a functionalized initiating species. If a functional group (or a suitably protected functional group) is incorporated into the initiator, that functional group will be at the initiating end of every polymer molecule as shown in Eqs. 4.13–4.15, where X is the functional group in the initiating species X-I* and X-P is the α-functionalized polymer. In principle, this method can produce 100% functionalized polymers with controlled molecular weight and narrow molecular weight distribution because of the absence of chain termination and chain transfer in living polymerization.

(4.13)

(4.14)

(4.15)

E. Star-Branched Polymers

An extension of the concept of controlled termination reactions is the ability to prepare star-branched polymers by deliberate, controlled termination reactions with multifunctional linking reagents as shown in Eq. 4.16, where L is a linking

(4.16)

agent of functionality “n” (see Chapter 13) [47–49]. For example, termination of a living anionic polymerization with a tetrafunctional electrophile such as silicon tetrachloride will produce a four-armed star polymer as shown in Eq. 4.17. The

(4.17)



branched polymer will have a predictable, well-defined structure, since a sample of the base arm polymer can be obtained prior to the linking reaction. Analogous to the functionalization chemistry described in Eqs. 4.13–4.15, the use of a multifunctional initiator in a living polymerization will alsoproduce a branched polymer structure with the number of arms corresponding to the functionality of the initiator. This is illustrated in Scheme 4.1 for a trifunctional initiator produc-

Scheme 4.1

ing a three-armed, star-branched polymer. The control of the arm molecular weights and molecular weight distributions is less precise in this method than in the end-linking procedure (Eq. 4.16) because it is dependent on the efficiency and relative rate of initiation for all initiating sites in the multifunctional initiator molecule compared to the rate of propagation; thus, heterogeneity of arm molecular weights is more likely in this method than in the end-linking chemistry described in Eq. 4.16.

IV. Experimental Criteria for Living Polymerizations

A. Criteria

The following experimental criteria have been proposed and utilized as diagnostic characteristics for living polymerizations [26]. In the following sections each of these criteria will be independently and critically evaluated as experimental criteria for living polymerization.

1. Polymerization proceeds until all of the monomer has been consumed. Further addition of monomer results in continued polymerization.

2. The number average molecular weight, Mn (or Xn, the number average degree of polymerization), is a linear function of conversion.

3. The number of polymer molecules (and active centers) is a constant, which is sensibly independent of conversion.

4. The molecular weight can be controlled by the stoichiometry of the reaction.

5. Narrow-molecular-weight distribution polymers are produced.



6. Block copolymers can be prepared by sequential monomer addition.

7. Chain-end functionalized polymers can be prepared in quantitative yield.

8. Linearity of a kinetic plot of rate of propagation as a function of time as shown in Eq. 4.18.

(4.18)

9. Determination of linearity of a kinetic plot of the left side of Eq, 4.19 versus time, t [50].

(4.19)

Criterion 1

The criterion that the polymerization proceeds until all of the monomer is consumed and further addition of monomer results in continued polymerization was enunciated by Szwarc and colleagues [8] in their first paper describing the concept and consequences of living anionic polymerization. In fact, this criterion is the basis of the description of living polymers in the Encyclopedia of Polymer Science and Engineering [11]. Even though the complete consumption of monomer is not in itself a useful criterion, the ability to continue polymerization upon addition of additional monomer is an important characteristic of living polymerization. However, if a system is a living polymerization, all of the chains must retain their active centers during the time scale of the laboratory experiment; therefore, this criterion should state that all of the chains continue to grow when additional monomer has been added. The practical experimental verification of this criterion requires that the molecular weight and molecular weight distribution be determined before and after the second monomer addition using, for example, size exclusion chromatography (SEC). Furthermore, it is advisable to add sufficient additional monomer so that the final molecular weight distribution does not overlap with the molecular weight distribution of the initial base polymer. If the system is living, the molecular weights of all polymer chains will increase and the corresponding SEC elution volumes will be decreased; no residual base polymer should be observed. Both chain termination and chain transfer reactions will produce dead polymer chains that will not increase in molecular weight upon addition of more monomer. This is a useful and rigorous criterion if used correctly.

Criterion 2

The criterion that the molecular weight increases linearly with conversion has been used quite often recently to document that a polymerization is living. As discussed in the previous section, at complete conversion the expected number average molecular weight can be calculated as shown in Eq. 4.7. At intermediate



degrees of conversion, the molecular weight is related to the grams of monomer consumed and Eq. 4.7 is replaced by Eq. 4.9. From Eq. 4.9 it is predicted that the number average molecular weight (or Xn, the number average degree of polymerization) will be a linear function of conversion for a living polymerization. Unfortunately, this is not a rigorous criterion [26]. If termination is occurring, the number of chains will still be a constant throughout the polymerization and the relationships shown in Eqs. 4.7 and 4.9 will still apply. This limitation has been noted recently by Penczek and colleagues [50].

To illustrate the limitations of this criterion, it is useful to examine the data reported for the sec-butyllithium-initiated, incremental polymerization of styrene with deliberate 5% termination prior to the next incremental monomer addition [26]. The results are shown in Table 4.1 and in Figure 4.1. It is apparent from the SEC results that termination is occurring in these polymerizations as shown in Figure 4.1, curve D for the final polymer. Small peaks are observed for the incrementally terminated polystyrenes at Mn = 6 × 103 g/mol and 12 × 103 g/mol; the peak corresponding to Mn = 18 × 103 g/mol is not resolved.

When the molecular weight data corresponding to these experiments are plotted as a function of conversion (or g of monomer polymerized), a linear plot is obtained as shown in Figure 4.2. Thus, even in this system where 15% of the polymer chains have been terminated prior to the final incremental styrene monomer addition, the linear relationship between Mn and % conversion is still maintained as required by Eq. 4.9. It is apparent from these results that a linear plot of Mn (or Xn) versus % conversion is not a rigorous test of a living polymerization. This type of plot will detect chain transfer reactions but it is not sensitive to termination reactions. A linear plot will always be obtained even if chain termination is occurring, if there is no chain transfer.

Table 4.1. Effect of Methanol Termination (5% per Incremental Monomer Addition) on Mn, Mw, and Mw/Mn as a Function of Conversion for sec-Butyllithium-Initiated, Incremental Polymerization of Styrene

Molecular Weight Dataa

Sample

Percentage Conversion

Mn(calc) (g/mol)

Mn (peak) (SEC) (g/mol)

Mn (SEC)(g/mol)

Mw/Mn

(SEC)

A 25 5.5 × 103 5.6 × 103 5.4 × 103 1.02

B 58 13.6 × 103 13.5 × 103 13.2 × 103 1.03

C 80 19.4 × 103 19.0 × 103 18.0 × 103 1.04

D 100 24.8 × 103 25 × 103 23.1 × 103 1.07

aFor the total polymer obtained at the given percentage conversion.



Figure 4.1 SEC curves for sec-butyllithium-initiated polymerization

of styrene with incremental monomer addition and deliberate termination

(5% termination per incremental monomer addition) (see Table 4.1). A. After first increment: 5% termination,

95% active. B. After second increment: 10% termination, 90% active. C. After third increment: 15% termination,

85% active. D. Final polymer. (From Ref. 26; reprinted by permission of the Society of Chemical Industry,

London, UK.)

Criterion 3

The requirement that the number of polymer molecules be constant and independent of conversion for a living polymer system is subject to limitations similar to those discussed for Criterion 2. This criterion is especially sensitive to the occurrence of chain transfer, since this will increase the number of polymer molecules. However, it is not in itself a good diagnostic test for termination reactions, because termination reactions will not change the total number of molecules. Both Criteria 2 and 3 can be regarded as necessary but not sufficient criteria for a living



Figure 4.2 Plot of Mn vs. percentage conversion for incremental monomer

addition and deliberate termination (see Table 4.1). ——, Mn of total polymer; ————,

Mn value of main peak in SEC curve of total polymer. (From Ref. 26; reprinted by

permission of the Society of Chemical Industry, London, UK.)

polymerization. Therefore, they should only be used in conjunction with other, more definitive criteria.

Criterion 4

For a living polymerization system, the number average molecular weight should be a simple function of the degree of conversion of the monomer and the stoichiometry of the reaction as described in Eq. 4.9. Obviously, this criterion depends on the quantitative utilization of the initiator before all of the monomer has been consumed. Thus, it is sensitive to the presence of impurities, which would change the effective number of moles of initiator, and, consequently, the number of active chain ends and polymer molecules. In general, termination reactions will increase the observed molecular weight relative to the calculated one (for example, see Table 4.1 and Figure 4.2) and chain transfer reactions will decrease the molecular weight relative to the stoichiometric value. However, the utilization of the experimentally observed number average molecular weight as a criterion is limited by the error limits of the methods used to determine the number average molecular weight. As shown by the data in Table 4.1, the calculated and observed values of Mn are similar for the final polymer, even though increments of 5% of the polymer chains were terminated at 25%, 50%, and 75% conversion.



Criterion 5

Perhaps no other aspect of living polymerization has generated so much confusion as the molecular weight distribution[11,26,50]. The essential requirements for formation of a polymer with a Poisson molecular weight distribution were clearly enunciated by Flory[30,34] and Henderson and Szwarc[35] as follows:

1. The growth of each polymer molecule must proceed exclusively by consecutive addition of monomers to an active terminal group.

2. All of these active termini, one for each molecule, must be equally susceptible to reaction with monomer, and this condition must prevail throughout the polymerization.

3. All active centers must be introduced at the outset of the polymerization.

4. There must be no chain transfer or termination (or interchange).

5. Propagation must be irreversible (i.e., the rate of depropagation should be vanishingly small).

A narrow or monodisperse molecular weight distribution has been operationally defined as a polymer exhibiting a value of Mw/Mn < 1.1[51]. In general, it is possible to prepare a polymer with a narrow molecular weight distribution using a living polymerization when the rate of initiation is competitive with the rate of propagation (requirement 3). This condition ensures that all of the chains grow for essentially the same period of time. If the relative rate of initiation is slow, the number of monomer units added during the time interval starting when the first chain is initiated and ending when the last chain has been initiated will be a measure of the polydispersity of the sample. It is useful to review the work of Gold[3,52], who developed quantitative treatments for the relationships describing the dependence of molecular weight distribution on the ratio of propagation to initiation rates; some of the results of these calculations are illustrated in Table 4.2. It is apparent that one may obtain “narrow molecular weight distributions” even

Table 4.2. Influence of the Ratio of kp/ki on Molecular Weight Distributions

R(kp/ki) Xw/Xn for Xn = ca.100

10-1 1.008

0.5 1.01

10 1.019

102 1.242 (Xn = 60) not all initiator used

1.01 (Xn = 911)

106 1.25 (Xn = 5,000) not all initiator used (1%)

Source: Ref. 52.



for systems that exhibit quite divergent values of the ratio of propagation rate constant to the initiation rate constant.

Another more subtle requirement for obtaining narrow molecular weight distribution polymers is that all of the active chain termini must be equally susceptible to reaction with monomer (i.e., all living polymer chains grow with the same rate constant: requirement 2). Thus, if there is more than one type of active center and each type propagates with a different propagation rate constant, these species must be in rapid equilibrium so that all chains grow uniformly. Living polymerizations that involve more than one type of propagating active center are commonly encountered in ionic polymerizations where covalent species, contact ion pairs, solvent-separated ion pairs, and free ions can all be involved (see Chapter 3)[32]. In fact, the broadening of the molecular weight distribution caused by the contributions of different types of propagating species has been used to determine various equilibrium and rate constants involved in living anionic polymerizations in polar media[32,53,54].

Another necessary condition for obtaining a narrow molecular weight distribution in a living polymerization is that propagation must be irreversible (requirement 5). In this regard, it should be noted that living polymerizations with monomers that have accessible ceiling temperatures can produce polymers with molecular weight distributions that broaden with time. These polymerizations proceed in the absence of chain termination and chain transfer reactions, and they are clearly living polymerizations as noted by Henderson and Szwarc[35], contrary to classifications described in a recent review[11,50]. However, these living polymerizations can and often do produce polymers with broad molecular weight distributions.

Only one of the five requirements for formation of a polymer with a Poisson molecular weight distribution relates to the living nature of the polymerization (i.e., requirement 4). The implication is clear; even when the system is living, if one of the other requirements is not fulfilled then a narrow molecular weight distribution polymer will not be formed. For example, consider requirement(3). If all of the chains are initiated at approximately the same time, they will grow for the same length of time and therefore one will obtain a narrow molecular weight distribution polymer. This requires that the rate of initiation (Ri) be comparable to the rate of propagation (Rp). Perhaps not surprisingly, there are many living polymerizations that, although they proceed in the absence of chain transfer and chain termination, provide relatively broad molecular weight distributions (i.e., Mw/Mn > 1.1). This aspect of living anionic polymerization was clearly illustrated by Hsieh and McKinney[36], who showed that relatively broad molecular weight distribution polystyrenes and polydienes resulted from the use of less reactive initiators such as n-butyllithium and (in some cases) t-butyllithium compared to the very reactive sec-butyllithium. Trepka[37] has reported that broadened molecular weight distributions can also be obtained when a mixture of initiators is



used. Also it should be noted that living polymerizations with monomers such as α-methylstyrene that have accessible ceiling temperatures can produce polymers with broad molecular weight distributions, because the propagation reaction is reversible. These polymerizations proceed in the absence of chain termination and chain transfer reactions and are clearly living polymerizations [35]. Therefore, the inability to obtain a narrow molecular weight distribution does not necessarily indicate that the polymerization is not living.

It is also possible to obtain narrow molecular weight distributions in systems which are clearly not living. As shown by the data in Table 4.1, one can obtain a narrow molecular weight distribution even when significant amounts of termination are occurring. Ver Strate and co-workers [55] have reported that a homogeneous Ziegler-Natta polymerization system utilizing a tube-flow reactor and efficient, rapid initiation of chains at low temperatures produces polymers with relatively narrow molecular weight distribution (Mw/Mn < 1.1). The key point was that the chains were all initiated at approximately the same time and were terminated rapidly in the flow system before significant contributions from chain transfer and chain termination reactions occurred. In contrast to living systems, the molecular weight distribution broadened with increasing reaction time relative to initiation.

Another important aspect of molecular weight distributions is that even bimodal and trimodal molecular weight distribution polymers can exhibit values of Mw/Mn that appear to be narrow [26]. The sample calculations below for samples 1 and 2 illustrate that even relatively narrow molecular weight distributions of Mw/Mn = 1.12 and 1.21, in fact, can be very deceptive. The type of synthetic mixtures described in samples 1 and 2 should not be described as monodisperse.

Sample 1: 0.5 wt fraction, Mn = 50 × 103 g/mol; 0.5 wt fraction Mn = 100 × 103 g/mol

calculate: Mn = 67 × 103 g/mol; Mw = 75 × 103 g/mol; Mw/Mn = 1.125

Sample 2: 0.33 wt fraction Mn = 50 × 103 g/mol; 0.33 wt fraction Mn = 100 × 103 g/mol; 0.33 wt fraction Mn = 150 × 103 g/mol

calculate: Mn = 81.8 × 103 g/mol; Mw = 99 × 103 g/mol; Mw/Mn = 1.21

The general equation that describes the relationship between Mw/Mn and the weight fraction, w1 of the lower-molecular-weight component Mn(1), for a mixture of two monodisperse polymers, Mn(1) and Mn(2), is shown in Eq. 4.20; where n is the molecular weight ratio [Mn(2)/Mn(1)] and Mn(2) > Mn

(1). For a mixture of

[Mw/Mn] = -[n(n-2) + 1/n]w12 + [n(n-2) + 1/n]w1 + 1 (4.20)

two monodisperse polymers whose molecular weights differ by a factor of two (i.e., n = 2), this equation exhibits a maximum value (Mw/Mn = 1.126) when w1 = 0.5, regardless of molecular weight. When the mixture of two monodisperse



polymers has Mn values that differ by a factor of three (i.e., n = 3), the maximum polydispersity is 1.33 when w1 = 0.5.

To illustrate further the limitations of the use of Mw/Mn as a criterion of living polymerizations, samples 3 and 4 have been prepared by combining equimolar amounts of two and three monodisperse polystyrene standards, respectively, as described in Table 4.3. These samples have been analyzed by SEC and the SEC traces are shown in Figure 4.3. It is clear that samples exhibiting relatively narrow molecular weight distributions, 1.11 and 1.15, respectively, would not be described as monodisperse based on their SEC behavior as illustrated in Figure 4.3; these polymers are clearly bimodal and trimodal, respectively. For these reasons, it is important to reserve the term “narrow molecular weight distribution” to polymers with Mw/Mn values < 1.1 as advocated in the review by Fetters [51] and also to utilize SEC data generated with an appropriate high resolution column set.

Thus, operationally, narrow molecular weight distributions should not be used per se as a criterion for living polymerizations. It is neither a necessary nor sufficient condition to indicate whether a given polymerization system is living [11,50].

Criterion 6

In addition to the important preparative applications, the ability to prepare block copolymers by sequential monomer addition can be used as a diagnostic test for polymerizations that proceed in the absence of chain transfer and chain termination in the same manner as Criterion 1. For example, the use of this criterion has been illustrated for the preparation of a poly(styrene-b-butadiene) diblock copolymer (Mn = 11.3 × 103 g/mol; Mw/Mn = 1.02) whose SEC chromatogram is shown in Figure 4.4 [26]. The absence of a peak corresponding to the base polystyrene (Mn = 1.9 × 103 g/mol; Mw/Mn = 1.04) is an indication of the absence of termination or chain transfer reactions. Note: the new peak should be clearly

Table 4.3. Samples for SEC Analysis of Mixtures of Monodisperse Polymers

Mixture Molecular Weight Data

Sample

Mn of Component Polymers (g/mol)

Weight Fraction

Mn Mw

(g/mol)Mz

Mw/Mn

3 19,000 0.5

34,500 0.5 23,480 25,940 28,720 1.11

4 19,000 0.33

34,500 0.33

48,000 0.33 29,780 34,290 38,040 1.15



Figure 4.3 SEC curves of mixtures of monodisperse polymers (see Table 4.3). ———, Sample 3; ——, Sample 4.

(From Ref. 26; reprinted by permission of the Society of Chemical Industry, London, UK.)

separated from the original base peak for a valid test of this criterion. Thus, this criterion is sensitive to these reactions and, like Criterion 1, it is a useful as one of several criteria required for determination of the living nature of a polymerization.

Criterion 7

Controlled termination of living polymerization systems can, in principle, result in chain-end functionalized polymers. However, as a diagnostic test of living polymerizations it is limited. This conclusion is based on the fact that most functionalization reactions do not proceed quantitatively to produce functionalized polymers [45,46,56]. Furthermore, analytical methods generally have experimental errors associated with them that limit the practical application of functionality determinations as criteria of a living polymerization. It should be noted that the sensitivity of these methods decreases with increasing molecular weight.

Criterion 8

The kinetics of propagation for a living polymerization should follow relatively simple pseudo-first-order behavior as shown in Eq. 4.21. Thus, if the concentra-



Figure 4.4 SEC curves for polystyrene-block-polybutadiene

synthesis. ———, polystyrene base polymer; ——, polystyrene-block-polybutadiene. (From Ref. 26; reprinted by permission of the

Society of Chemical Industry, London, UK.)

(4.21)

tion of active propagating species is constant (i.e., if there is no chain termination) integration of Eq. 4.21 will provide Eq. 4.18. However, the presence of chain transfer will not affect the kinetics since the number of active propagating species,

(4.18)

[P*], is not affected. Hence, a plot of the left side of Eq. 4.18 (In [M]o/[M]) as a function of time should be linear. This is a very useful criterion that supplements Criterion 2, since Criterion 2 is sensitive to chain transfer but insensitive to chain termination, while Criterion 8 is sensitive to chain termination but insensitive to chain transfer.

Criterion 9

The use of this relationship has been advocated as a simple experimental criterion of living polymerizations [50]. This equation is generated by combining Eq. 4.22, which is similar to Eq. 4.9, with Eq. 4.23. The uniqueness of this approach is that Eq. 4.22 is a diagnostic test for chain transfer, while Eq. 4.23 is a diagnostic test for chain termination; thus, their combination provides a useful criterion for living



(4.22)

(4.23)

polymerizations. It is only necessary to determine the dependence of DPn on time to apply this criterion. If the plot is linear, both chain transfer and chain termination are absent.

B. Equilibrium Polymerizations

The subject of equilibrium polymerizations vis-à-vis living polymerizations deserves special attention because of the ambiguity of various criteria for living polymerization as they relate to these systems. As discussed under Criterion 5, if living polymerizations are carried out near the ceiling temperature for a reversible polymerization, the monomer will not be completely consumed and the molecular weight distribution will broaden with reaction time because of the reversibility of the polymerization. In spite of the fact that no termination and no chain transfer reactions are involved, part of Criterion 1, (i.e., complete consumption of monomer) will not be applicable. However, addition of more monomer will result in continued polymerization and an increase in molecular weight for all chains. Addition of a second monomer will also result in block copolymer formation (Criterion 6). This also means that Criterion 2 will be applicable, even though the polymerization will not necessarily proceed to complete conversion depending on the reaction temperature relative to the ceiling temperature; thus, a plot of Mn vs. conversion of monomer will be linear in accord with Eq. 4.9. An equilibrium polymerization will satisfy Criterion 3, because the number of polymer molecules, all with reactive chain ends, will remain constant. Because of incomplete consumption of monomer, the number average molecular weight calculated from the initial stoichiometry of the polymerization (Eq. 4.7) will not be applicable (Criterion 4); however, the Mn value will be in accord with Eq. 4.9, based on the total amount of monomer consumed. As discussed previously, Criterion 5, relating to the molecular weight distribution, will not generally be applicable to a reversible living polymerization because Mw/Mn will tend toward a value of 2 with increasing reaction time in spite of the absence of chain transfer or chain termination reactions. To the extent that any functionalization reaction will proceed quantitatively to produce end-functionalized polymer, an equilibrium polymerization will satisfy Criterion 7 as well as any other living polymerization system. The kinetics of reversible, living polymerization (Criterion 8) will follow pseudo-first-order kinetics except that the monomer concentration must be modified to take into account the equilibrium monomer concentration as shown in Eq. 4.24 [57]. Thus, consideration of the nature of an equilibrium polymerization



(4.24)

system vis-à-vis the criteria for living polymerizations helps to clarify the limitations of the criteria themselves. Once again through this exercise, it is apparent that no one criterion of living polymerization is satisfactory for defining a living polymerization unambiguously.

C. Time

There has always been a tacit assumption that the categorization of a given system as a living polymerization was based on results obtained on the laboratory time scale: no chain termination or chain transfer reactions occurred within the normal time required to complete the polymerization and carry out any deliberate chemical reactions with the active polymer chain ends [32]. It is a matter of judgment with respect to this practical laboratory time scale; however, the absence of chain termination and chain transfer during this time period is not subject to judgment. This means that a polymerization system that only exhibits transient living characteristics, such as the previously described Ziegler-type polymerization with propagation only faster than termination and transfer in a flow system [55], would not be classified as living. Matyjaszewski [58] h s proposed that very specific and practical criteria be used to define living polymerizations kinetically. Thus, a living polymerization is characterized by kp/kt > 104 mol-1L and kp/ktr > 104, such that less than 10% of the chains would be deactivated in a time period of t 1000 sec.

D. Reversible Termination

Reversible termination reactions have been described in several polymerization systems that exhibit characteristics of living polymerizations. If termination is reversible, as shown in Scheme 4.2, then the “termination event” is not such (i.e.,

Scheme 4.2

reactive reaction intermediates are not kinetically destroyed or rendered inactive) [28]. The net effect of reversible termination is to reduce the effective kinetic concentration of the reactive chain ends by an amount reflecting their equilibrium concentration. Rather than coining new names or acronyms for this type of system, it has been proposed that this type of behavior be described as living



polymerization with reversible termination [26]. In effect, these types of polymerizations belong to abroad class of living polymerizations in which the active, propagating species is in equilibrium with inactive chains. These systems should exhibit most of the characteristics of living polymerizations as described in Criteria 1, 2, 4–6, and 8. Criterion 3 should be modified to state that the “effective” number of polymer molecules (and active centers) is a constant and it may not be possible to prepare chain-end functionalized polymers quantitatively (Criterion 7) in a system involving reversible termination because of the chemistry of those chain ends that are in the inactive state [59,60].

E. Reversible Chain Transfer Systems

Reversible chain transfer reactions have been described in several polymerization systems that exhibit characteristics of living polymerizations. Analogous to reversible termination, if chain transfer is reversible, as shown in Scheme 4.3, the “chain transfer” event does not effectively generate a dead polymer molecule as encompassed by the usual definition of chain transfer [28]. The net result of reversible chain transfer is to increase the effective number of growing chains

Scheme 4.3

relative to the number expected based on the moles of initiator (Eq. 4.9). The concentration of growing chains would be equal to the number of moles of initiator plus the number of moles of chain transfer agent, and uniform chain growth would be expected if the rate of reversible chain transfer is fast compared to the rate of propagation. Rather than coining new names or acronyms for this system, it has been proposed that this type of behavior be described as living polymerization with reversible chain transfer [26]. These systems should exhibit most of the characteristics of living polymerizations as described in Criteria 1, 2, 5, 6, and 8. Criteria 3 and 4 should be modified to state that the number of polymer molecules is a constant and the molecular weight can be controlled by the stoichiometry of the reaction, if it is recognized that the total number of polymer molecules will include the number of moles of initiator plus the number of moles of chain transfer agent. It may not be possible to prepare chain-end functionalized polymers quantitatively (Criterion 7) in a system involving reversible chain



transfer because of the chemistry of those chain ends in the inactive state. It should be noted that this type of polymerization has been described in the literature as an “immortal polymerization” [24].

V. Summary

A living polymerization is a chain reaction polymerization that proceeds in the absence of the kinetic steps of termination or chain transfer. The terms living polymerization with reversible termination and living polymerization with reversible chain transfer should be used to describe polymerizations that proceed in the absence of the kinetic steps of irreversible chain transfer and irreversible chain termination, respectively. For other systems that have some of the characteristics of living polymerizations, that is how they should be described: as “possessing characteristics of living polymerizations.”

A variety of experimental criteria can be used to determine whether a given polymerization is a living polymerization. In general, no one criterion is satisfactory to determine whether a given polymerization is living or not; this is because chain transfer and chain termination reactions can have different experimentally observable consequences and various criteria have different sensitivities to these side reactions.

Living polymerizations provide methodologies for the synthesis of polymers with control of a wide range of compositional and structural parameters including molecular weight, molecular weight distribution, copolymer composition (i.e., block and random copolymers), stereochemistry, molecular architecture, and chain-end functionality.

References

1. P. Rempp, E. Franta, and J.-E. Herz, Adv. Polym. Sci., 86, 145 (1988).

2. S. Bywater, in Encyclopedia of Polymer Science and Engineering, second edition, Vol. 2, J. I. Kroschwitz, Ed., Wiley-interscience, New York, 1985, p. 1.


4. O. W. Webster, Science, 251, 887 (1991).

5. R. H. Grubbs and W. Tumas, Science, 243, 907 (1989).

6. J. P. Kennedy and B. Ivan, Designed Polymers by Carbocationic Macromolecular Engineering: Theory and Practice, Hanser Publishers, Munich, 1992.

7. M. Szwarc, Nature, 178, 1168 (1956).

8. M. Szwarc, M. Levy, and R. Milkovich, J. Am. Chem. Soc., 78, 2656 (1956).

9. T. Higashimura and M. Sawamoto, Adv. Polym. Sci., 62 (1984).

10. M. Sawamoto and T. Higashimura, in Encyclopedia of Polymer Science and Engi



Administrator

Highlight

Administrator

Underline

neering, J. I. Kroschwitz, Ed., Wiley-interscience, New York, 1989, Supplemental Volume, p. 399.

11. S. Penczek and P. Kubisa, in Encyclopedia of Polymer Science and Engineering, J. I. Kroschwitz, Ed., Wiley-Interscience, New York, 1989, Supplemental Volume, p. 380.

12. H. J. Harwood, in Encyclopedia of Polymer Science and Engineering, J. I. Kroschwitz, Ed., Wiley-Interscience, New York, 1989, Supplemental Volume, p. 429.

13. S. R. Turner and R. W. Blevins, Macromolecules, 23, 1856 (1990).

14. Y. Doi and T. Keii, Adv. Polym. Sci., 73/74, 201 (1986).

15. T. Keii, Y. Doi, and K. Soga, in Encyclopedia of Polymer Science and Engineering, J. I. Kroschwitz, Ed., Wiley-Interscience, New York, 1989, Supplemental Volume, p. 437.

16. R. Grubbs and B. M. Novak, in Encyclopedia of Polymer Science and Engineering, J. I. Kroschwitz, Ed., Wiley-Interscience, New York, 1989, Supplemental Volume, p. 420.

17. R. R. Schrock, Accounts Chem. Res., 23, 158 (1990).

18. R. Grubbs, W. Risse, and B. M. Novak, Adv. Polym. Sci., 102, 47 (1992).

19. O. W. Webster, in Encyclopedia of Polymer Science and Engineering, J. I. Kroschwitz, Ed., Wiley-Interscience, New York, 1987, Vol. 7, p. 580.

20. R. P. Quirk and G. Bidinger, Polym. Bull., 22, 63 (1989).

21. W. J. Brittain, Rubber Chem. Tech., 65, 580 (1992).

22. J. P. Kennedy and E. Marechal, Carbocationic Polymerization, Wiley-Interscience, New York, 1981, pp. 453–456.

23. E. J. Goethals, F.D'Haese, R. DeClercq, and D. Van Meirvenue, Polym. Prepr., Am. Chem. Soc. Div. Polym. Chem., 29(2) 61 (1988).

24. S. Inoue and T. Aida, in Encyclopedia of Polymer Science and Engineering, J. I. Kroschwitz, Ed., Wiley-Interscience, New York, 1987, Vol. 7, p. 412.

25. C. G. Cho, B. A. Feit, and O. W. Webster, Macromolecules, 23, 1918 (1990).

26. R. P. Quirk and B. Lee, Polym. Int., 27, 359 (1992).

27. M. Szwarc, Carbanions, Living Polymers and Electron Transfer Processes, Interscience, New York, 1968.

28. V. Gold, Ed., Pure Appl. Chem., 51, 1725 (1979).

29. G. Odian, Principles of Polymerization, 3rd ed., Wiley, New York, 1991, p. 259.

30. P. J. Flory, Principles of Polymer Chemistry, Cornell University Press, Ithaca, NY, 1953.

31. H. Staudinger, Chem. Ber., 53, 1073 (1920).




40. C. C. Chang, J. W. Miller, Jr., and G. R. Schorr, J. Appl. Polym. Sci., 39, 2395 (1990).

41. G. L. Frontini, G. E. Elicabe, D. A. Couso, and G. R. Meira, J. Appl. Polym. Sci., 31, 1019 (1986).

42. G. L. Frontini, G. E. Elicabe, and G. R. Meira, J. Appl. Polym. Sci., 33, 2165 (1987).

43. R. P. Quirk, Rubber Chem. Tech., 64, 450 (1991).

44. R. P. Quirk, D. J. Kinning, and L. J. Fetters, in Comprehensive Polymer Science, Vol. 7, S. L. Aggarwal, Ed., Pergamon Press, Elmsford, NY, 1989, p. 1.

45. R. P. Quirk, in Comprehensive Polymer Science, First Supplement, S. L. Agarwal and S. Russo, Eds., Pergamon Press, Oxford, UK, 1992.

46. R. P. Quirk, J. Yin, S.-H. Guo, X. W. Hu, G. J. Summers, J. Kim, L.-F. Zhu, J.-J. Ma, T. Takizawa, and T. Lynch, Rubber Chem. Tech., 64, 648 (1991).

47. B. J. Bauer and L. J. Fetters, Rubber Chem. Tech., 51, 406 (1978).

48. S. Bywater, Adv. Polym. Sci., 30, 89 (1979).

49. J. Roovers, in Encyclopedia of Polymer Science and Engineering, 2nd ed., J. Kroschwitz, Ed., Wiley, New York, 1985, Vol. 2, p. 478.

50. S. Penczek, P. Kubisa, and R. Szymanski, Makromol. Chem., Rapid Commun., 12, 77 (1991).

51. L. J. Fetters, in Encyclopedia of Polymer Science and Engineering, 2nd ed., J. I. Kroschwitz, Ed., Wiley-Interscience, New York, 1987, Vol. 10, p. 19.

52. L. Gold, J. Chem. Phys., 28, 91 (1958).

53. L. L. Bohm, M. Chmelir, G. Lohr, B. J. Schmitt, and G. V. Schulz, Adv. Polym. Sci., 9, 1 (1973).

54. L. L. Bohm, G. Lohr, and G. V. Schulz, Ber. Bunsen-Ges., 78, 1064 (1974).

55. G. VerStrate, C. Cozewith, and S. Ju, Macromolecules, 21, 3360 (1988).


57. P. Rempp and E. W. Merrill, Polymer Synthesis, 2nd ed., Huthig & Wepf Verlag, Heidelberg, 1991, p. 165.

58. K. Matyjaszewski, J. Polym. Sci. Part A: Polym. Chem., 31, 995 (1993).

59. B. Ivan and J. P. Kennedy, Macromolecules, 23, 2880 (1990).

60. K. Matyjaszewski and P. Sigwalt, Polym. Int., 35, 1 (1994).



5 General Aspects of Anionic Polymerization

I. Experimental Procedures

Detailed procedures for anionic polymerization are available in the literature [1–6]. In general, inert atmosphere [7] or high vacuum techniques [1,7] are required because of the reactivity of carbanions toward oxygen, moisture, and carbon dioxide [8]. The use of less than rigorous experimental procedures is often characterized by the appearance of dimer peaks resulting from air oxidation reactions [9–12].

II. Polymerizable Monomers

A. General Aspects

Monomers in two broad classifications are amenable to anionic polymerization [13]: vinyl, diene, and carbonyl-type monomers with difunctionality provided by one or more double bonds; and cyclic (e.g., heterocyclic) monomers with difunctionality provided by a ring that can open by reaction with nucleophiles.

The polymerizability of vinyl monomers cannot be deduced from the thermodynamics of polymerization [14,15]. In general, almost all vinyl monomers exhibit negative free energies of polymerization, that is, if a suitable pathway exists, the polymerization will proceed spontaneously to form the polymer from the monomer. The proviso that there exists a suitable pathway provides the major



Administrator

Underline

limitation on the polymerizability of monomers. The distinguishing factor for monomer polymerizability is the mechanism of polymerization. Thus, for anionic polymerizability it is generally considered that there must be substituents on the double bond that can stabilize the negative charge that develops in the transition state for the monomer addition step as shown in Eq. 5.1. These substituents must

(5.1)

also be stable to the reactive anionic chain ends; thus, relatively acidic, proton-donating groups (e.g. amino, hydroxyl, carboxyl, acetylene functional groups) or strongly electrophilic functional groups that react with bases and nucleophiles [8] must not be present or must be protected by conversion to a suitable derivative [16]. In general, substituents that stabilize negative charge by anionic charge delocalization are the substituents that render vinyl monomers polymerizable by an anionic mechanism. Such substituents include aromatic rings, double bonds, as well as carbonyl, ester, cyano, sulfoxide, sulfone, and nitro groups (see Chapter 2). The corresponding classes of polymerizable monomers are shown in Table 5.1. The general types of heterocyclic monomers that can be polymerized anionically are listed in Table 5.2 (see Chapter 24). The ability to polymerize the corresponding deuterated monomers should also be noted; deuterated polymers are useful for small-angle neutron-scattering investigations of chain conformations in solution and chain dynamics [17]. The range of monomers that can be polymerized anionically without the incursion of termination and transfer reactions includes styrenes, dienes, methacrylates, epoxides, episulfides, cyclic siloxanes and lactones [13,17–22].

Monomers with polar substituents such as carbonyl, cyano, and nitro groups often undergo side reactions with initiators and propagating anions; therefore, controlled polymerization to provide high-molecular-weight polymers is not always possible. Even though some termination reactions occur with certain monomers, especially those with polar functionalities, they can be used to prepare the last-formed block in a sequence. Under these circumstances, control of variables such as molecular weight and molecular weight distribution will be lost. The types of monomers that have been polymerized anionically, but that do not produce living, stable, carbanionic chain ends, include acrylonitriles, cyanoacrylates, propylene oxide, vinyl ketones, acrolein, vinyl sulfones, vinyl sulfoxides, vinyl silanes, halogenated monomers, ketenes, nitroalkenes, and isocyanates [18,19,21,22].

It is noteworthy and unexpected that the simplest vinyl monomer, ethylene, can be polymerized by an anionic mechanism [18,23–27]. Although the propagating primary carbanion in ethylene polymerization would not be expected to be



Administrator

Underline

Administrator

Underline

Administrator

Underline

Administrator

Highlight

Administrator

Highlight

Administrator

Highlight

Administrator

Highlight

Administrator

Highlight

Administrator

Highlight

Administrator

Highlight

Administrator

Underline

Administrator

Text Box

[化]丙烯醛

Administrator

Pencil

Administrator

Pencil

Administrator

Text Box

乙烯酮, 烯酮

Administrator

Pencil

stable relative to the substituted carbanions corresponding to the other monomers listed in Table 5.1, the conversion of a double bond to two single bonds provides the energetic driving force for this reaction.

An important aspect of monomer reactivity in anionic polymerization is the relationship among monomer reactivity, the stability of the corresponding propagating carbanionic species, and the appropriate initiating species. It is satisfying to note that there appears to be a general relationship between monomer reactivity in anionic polymerization and the stability of the anions formed by nucleophilic addition or ring opening as deduced from the pKa values for the conjugate acids of these anions. Thus, the monomers that form the least stable anions (i.e., have the largest values of pKa for the corresponding conjugate acids) are the least reactive monomers in anionic polymerization; in turn, these less reactive monomers require the use of the most reactive, organometallic initiators as shown in Table 5.3. In general, an appropriate initiator is an anionic species that has a reactivity similar to the propagating carbanionic species [17]. If the initiator is too reactive, side reactions are promoted. If the initiator is relatively unreactive, the initiation reaction may be slow or inefficient.

Similar considerations should be kept in mind for block copolymer synthesis with regard to the reactivity of the polymeric anionic initiator with a second block-forming monomer. Thus, in general, the more reactive propagating anions corresponding to a less reactive monomer group (higher carbanion conjugate acid pKa) can initiate polymerization of more reactive monomers in a group that forms more stable anions (lower carbanion conjugate acid pKa), but not vice versa [204]. These relationships are illustrated in Table 5.3, which lists basic monomer types, the pKa of the conjugate acid of the carbanionic species involved in propagation, and appropriate initiating species for each class of monomers.

B. Monomers with Protected Functional Groups

Functional groups with relatively acidic hydrogens would normally participate in termination or chain transfer reactions in anionic polymerization [17]. These groups can be protected, however, by conversion to suitable derivatives that are stable to the anionic polymerization conditions (initiator and propagating anion) and that can be removed readily after the polymerizations [16,205]. These protected functional groups are listed in section 3.B in Table 5.1. Thus hydroxyl, phenol, and amine functional groups can be protected by conversion to the corresponding silyl derivatives. Mild acid hydrolysis or reaction with fluoride ion in methanol can be used to regenerate the hydroxyl and amine functional groups [16,205]. Aldehyde and ketone functional groups can be protected by conversion to the corresponding imidazolidine, aminal, or acetal derivatives, respectively. The carboxyl functional groups can be protected by conversion to the oxazoline derivative or by use of the hindered t-butyl ester. Many of these protecting groups



Administrator

Underline

Administrator

Underline

Administrator

Underline

Administrator

Highlight

Administrator

Highlight

Administrator

Highlight

Administrator

Underline

Administrator

Underline

Administrator

Highlight

Administrator

Highlight

Administrator

Underline

Administrator

Underline

Administrator

Highlight

Administrator

Highlight

Administrator

Highlight

Administrator

Highlight

Administrator

Underline

Administrator

Underline

Administrator

Text Box

唑啉

Table 5.2 Anionically Polymerizable Heterocyclic

Monomers

Epoxides [170–173]

Cyclic sulfides [172,174]

Lactones and lactides [175–180]

Cyclic carbonates [181–184]

Lactams (by activated monomer polymerization) [185–187]

Cyclosiloxanes (D3 and D4) [188–192]

Cyclic phosphorous compounds [193]

Source:Refs. 166–169



Table 5.3 Relationships Among Monomer Reactivity, Carbanion Stability, and Suitable Initiatorsa

Monomer Type PKa (DMSO) PKa (H2O) Initiatorsb

Ethylene 56 Rli

Dienes and 44 NR2-, RLi,

Styrenes 43 RMtb, naphthalene radical anionsc, cumyl- K+, Mt,

Acrylonitrile 32 RMgX

Alkyl 30–31 (195) 27–28 (195) Fluorenyl-,

Methacrylates Ar2C-, ketyl radical anionsd

Vinyl ketones 26 19 (197)

Oxiranes 29–32 16–18 (198) RO-

Thiiranes 17 12–13 (199)

Nitroalkenes 17 10–14 (200)

Siloxanes 10–14 (201, 202) RO-, OH-

Lactones 12 4–5 (203) RCO2-

Cyanoacrylates 11 (196) HCO3-, H2O

Vinylidene cyanide 11 11 (196)

apKa values refer to the conjugate acid of the anionic propagating intermediate. PKa values in DMSO are from Ref. 194 unless noted in parentheses after the number. The references for pKa values in H2O are listed in parentheses after the number.

bMt refers generally to alkali metals (Li, Na, K, Rb, Cs).

cFor example, naphthalene radical anion (Li+, Na+, K+).

dAr2CO .

are themselves not stable to initiators or carbanionic propagating species and require the use of low temperatures for their controlled polymerization [16,205].

C. Styrene Monomers

Styrene monomers are important for living anionic polymerization and many commercial polymers. They can be polymerized in hydrocarbon or polar aprotic media. The ability to polymerize styrenes with both aromatic and side-chain substituents provides routes to polymers with either hydrophobic or hydrophilic character. In addition, the use of protecting groups for reactive functional groups



Administrator

Underline

temperature, and chain end concentration (see Chapter 9). Lithium is unique among the alkali metal counterions in producing high 1,4-polydienes in hydrocarbon media. High 1,4-polydienes and their copolymers generally have low glass transition temperatures and produce good elastomers. The glass transition temperature increases with increasing amounts of side-chain vinyl microstructure (see Chapter 9).

E. Vinylpyridines

Because of the reactivity of the pyridine ring towards nucleophilic attack [8], the selection of an appropriate initiator and reaction conditions is important to effect the successful polymerization of 2-vinylpyridine (see references cited in Table 5.1, section 3.D). Controlled polymerization of 2-vinylpyridine has been effected by addition of lithium chloride at -78°C in tetrahydrofuran (THF) [206]. In general, the anionic polymerization of 4-vinylpyridine is not carried out because the polymer precipitates from solution during the polymerization [207].

F. Alkyl Methacrylates

The proper choice of initiator and reaction conditions is essential for the controlled anionic polymerization of alkyl methacrylates (see Chapter 23). In general, a less reactive initiator such as 1,1-diphenylhexyllithium, formed by the addition of butyllithium to 1,1-diphenylethylene (Eq. 5.2) (see Chapter 6), is an effective

(5.2)

initiator [121,208–212]. More reactive initiators such as butyllithium react with the ester carbonyl group in competition with the Michael addition to the conjugated double bond; less than half of the initiator molecules initiate chain growth [213]. These polymerizations must be carried out at low temperatures (e.g. -78°C), although it has been reported that polymerizations can be carried out at higher temperatures in the presence of lithium chloride [214,215]. It is even possible to effect controlled anionic polymerization of t-butyl acrylate in the presence of lithium chloride [127–129] or lithium 2-(2-methoxyethoxy)ethoxide [216] in THF at low temperatures. In addition, the anionic polymerization of t-butyl methacrylate can be carried out at room temperature [217]. The solvent has a dramatic effect on the stereochemistry of the polymerization (see Chapter 23). Highly isotactic polymer can be formed in hydrocarbon solvents such as toluene, whereas highly syndiotactic polymer is formed in tetrahydrofuran [218].



Administrator

Underline

Administrator

Text Box

乙烯基吡啶

Administrator

Underline

III. Solvents and Lewis Base Additives

The range of useful solvents for anionic polymerization is limited by the high reactivity (basicity and nucleophilicity) of the initiators and propagating anionic chain ends. For styrene and diene monomers, the solvents of choice are alkanes and cycloalkanes, aromatic hydrocarbons, and ethers [1,18,111]; the use of alkenes has also been described, although some chain transfer can occur, especially at elevated temperatures and in the presence of Lewis bases [219]. Benzene and toluene provide enhanced rates of initiation and propagation relative to the alkanes (see Chapters 6,7); however, they can participate in chain transfer and metalation processes under certain conditions. Even benzene undergoes relatively rapid metalation by butyllithium in the presence of N,N,N',N'-tetramethylethylene-diamine (TMEDA) [8,220–223]; thus, polymerization or metalation-grafting reactions using this initiator system must be carried out in saturated hydrocarbon solvents (see Chapter 14) [224–228]. Toluene undergoes chain transfer reactions during styrene and diene polymerizations [229,230]. The importance of these chain transfer reactions increases with increasing temperature and in the presence of polar additives such as ethers or amines. A useful alternative aromatic solvent to benzene and toluene is t-butylbenzene, which has no acidic benzylic protons, a low freezing point (-58.1°C), and a higher boiling point (168.5°C) [231].

Ethers can react with organometallic initiators, as well as polystyryl and polydienyl anions, to decrease the concentration of active centers and to terminate chain growth [8,232]. The pseudo-first order termination reaction of the chain ends in polar solvents such as THF can have significant effects on the ability to prepare block copolymers efficiently [233]. The rate of reaction with ethers decreases in the order Li > Na > K [232]. For example, dilute solutions of poly(styryl)lithium in THF at room temperature decompose at the rate of a few percent each minute [232,234]. Alkyllithium initiators also react relatively rapidly with ethers as shown in Scheme 5.1 for the decomposition of sec-butyllithium in tetrahydrofuran [8,235–237]. The order of reactivity of organolithium compounds with ethers is tertiary RLi > secondary RLi > primary RLi > phenyllithium >

Scheme 5.1



Administrator

Underline

Administrator

Underline

Administrator

Highlight

Administrator

Highlight

Administrator

Underline

Administrator

Underline

methyllithium > benzyllithium [8,236,238–241]. An approximate order of reactivity of ethers towards alkyllithium compounds is dimethoxyethane, THF > tetrahydropyran > diethyl ether > diisopropyl ether [238–240,242]. For example, n-butyllithium is completely decomposed in THF at room temperature within 2 h [236]; the half-life of n-decyllithium in diethyl ether is 72 h [242]. The half-life of t-butyllithium in dimethoxyethane is only 11 min at -70°C [240].

Even tertiary amines can react with alkyllithium compounds; isopropyl-lithium in benzene-petroleum ether reacts with 2 equivalents of triethylamine at 15–19°C to form 4-methylpentanoic acid in 25% yield after carbonation [243]. The importance of these reactions can be minimized by working at lower temperatures (e.g., <0°C) [236,238,240]; it is also advisable to use only the minimum amounts of ethers and other Lewis bases required as additives.

For less reactive anionic chain ends such as those involved in propagation for heterocyclic monomers, a wider range of solvents can be utilized. For example, dipolar aprotic solvents such as dimethylsulfoxide (DMSO), N,N-dimethylformamide (DMF), and hexamethylphosphoramide (HMPA) can be used for polymerizations of epoxides. It has been reported that the addition of coordinating ligands such as crown ethers, cryptands, diamines [N,N,N',N'-tetramethylethylene-diamine, 1,2-bis(piperidino)ethane], and glymes (1,2-dimethoxyethane, etc.) is effective in changing the kinetics and/or the stereochemistry in anionic vinyl and ring-opening polymerization [13,189]. For example, at low temperatures (0°C) it has been reported that polybutadiene with high (>90%) 1,2-microstructure is formed in the presence of 1,2-bis(piperidino)ethane [244].

IV. Initiators

A. Alkali Metals

The direct use of alkali metals as initiators for anionic polymerization of diene monomers is primarily of historical interest. One of the earliest reports is the patent disclosure of Mathews and Strange [245] issued in 1911, which detailed the use of metallic sodium to polymerize isoprene and other dienes. Independently and simultaneously, Harries [246] described the use of sodium metal to polymerize butadiene, isoprene, and 2,3-dimethyl-1,3-butadiene. Tornqvist [247] has chronicled the ensuing debate regarding who had prepared the first synthetic rubber. Buna (butadiene and natrium) rubber was a commercial rubber produced in Germany; an analogous rubber was also produced in Russia. Interest in alkali metal-initiated polymerization of 1,3-dienes culminated in the discovery by Stavely and co-workers [248] at Firestone Tire and Rubber Company that polymerization of neat isoprene with lithium dispersion produced high cis-1,4-polyisoprene, similar in structure and properties to Hevea natural rubber.

The mechanism of the anionic polymerization of styrenes and 1,3-dienes



Administrator

Underline

Administrator

Underline

Administrator

Highlight

Administrator

Underline

Administrator

Text Box

穴状配体

Administrator

Pencil

Administrator

Pencil

Administrator

Text Box

丁纳橡胶

Administrator

Pencil

initiated by alkali metals has been described in detail by Szwarc [249]. Initiation is a heterogeneous process occurring on the surface of the metal (Mt) by transfer of an electron to adsorbed monomer as shown in simplified form in Scheme 5.2.

Scheme 5.2

The initially formed radical anions rapidly dimerize to form dianions [250,251]. Monomer addition to these dianions forms adsorbed oligomers that eventually desorb and continue growth in solution. Unlike homogeneous anionic initiation processes with organometallic compounds, this heterogeneous initiation reaction continues to generate new active chain ends during the course of the subsequent propagation reactions. As a consequence, there is little control of molecular weight and relatively broad molecular weight distributions have been reported for the soluble polymer obtained from these bulk polymerizations (Mw/Mn = 3–10) [252]. A polybutadiene polymer produced using sodium metal as initiator has been characterized as exhibiting a high degree of branching and gel content (45%) combined with inhomogeneity in composition and molecular weight distribution [248,252,253].

These reactions are useful for preparation of homogeneous difunctional initiators from α-methylstyrene in polar solvents such as tetrahydrofuran. Because of the low ceiling temperature of α-methylstyrene [254], dimers or tetramers can be formed depending on the alkali metal system, temperature, and concentration. The structures of the dimer and tetramer correspond to initial tail-to-tail addition to form the most stable dianion, as shown in Eq. 5.3 [255].

(5.3)

The unique role of lithium among the alkali metals is clear from the data in Tables 5.4 and 5.5, which list the microstructures reported for alkali metalinitiated polymerizations of isoprene and 1,3-butadiene, respectively. In general, elastomeric polydienes with high 1,4-microstructure can only be prepared using lithium as the counterion.

B. Radical Anions

Many aromatic hydrocarbons react with alkali metals in polar aprotic solvents to form stable solutions of the corresponding radical anions, as shown in Eq. 5.4.



Table 5.4 Microstructures of Alkali-Metal-Initiated Polyisoprenes

Polyisoprene Microstructure (%)

Alkali Metal cis-1,4 trans-1,4 3,4 1,2

Li 94 0 6 0

Na 0 43 51 6

K 0 52 40 8

Rb 5 47 39 8

Cs 4 51 37 8

Source: Ref. 256.

These solutions can be analyzed by ultraviolet (UV)-visible spectroscopy and stored for use [249,257]. Radical anions have occupied

(5.4)

a special position in the development of living anionic polymerization. The naphthalene radical anion initiator system (Eq. 5.5) was used by Szwarc and co-

(5.5)

workers [258] in their initial enunciation of the concepts of living anionic polymerization. The unpaired electron is added to the lowest unoccupied molecular

Table 5.5 Microstructures of Alkali-Metal-Initiated Polybutadienes

Polybutadiene Microstructure (%)

Alkali Metal cis-1,4 trans-1,4 1,2

Li 35 52 13

Na 10 25 65

K 15 40 45

Rb 7 31 62

Cs 6 35 59

Source: Ref. 256.



orbital of the aromatic hydrocarbon and a delocalized radical anion is formed. This oxidation-reduction reaction of the aromatic hydrocarbon with the metal is reversible; thus, sodium metal and naphthalene are reformed when THF is removed [259]. Representative resonance structures of the naphthalene radical anion are shown in Scheme 5.3.

Scheme 5.3

Several important aspects of the use of radical anions as homogeneous initiators for living anionic polymerization should be noted. Perhaps the most important consideration is that these radical anions can only be formed efficiently in polar aprotic solvents such as tetrahydrofuran (THF) and the glymes. For example, although sodium naphthalene formation is 95% complete in tetrahydrofuran, this radical anion is formed in less than 1% yield in diethyl ether [249]. For biphenyl, that has a lower electron affinity compared to naphthalene, only 20% of the corresponding radical anion is formed by sodium reduction in THF [249]. In general, these aromatic radical anion initiators can only be used in such dipolar aprotic solvents; however, the addition of lithium naphthalene in tetrahydrofuran to a benzene solution produces a finely divided suspension that reacted with styrene monomer to generate narrow molecular weight distribution polymers [260].

Aromatic radical anions such as sodium naphthalene react with monomers such as styrene by reversible electron transfer to form the corresponding monomer radical anions as shown in Scheme 5.4. Although the equilibrium between the

Scheme 5.4

radical anion of the monomer and the aromatic radical anion lies far to the left because of the low electron affinity of the monomer [258], this is an efficient initiation process because the resulting monomer radical anions rapidly undergo



dimerization reactions (step 2 in Scheme 5.4) with rate constants that approach diffusion control [249]. For example, the rate constants for dimerization of the radical anions of diphenylethylene (Eq. 5.6) are 1.2 × 108, 3.5 × 108, 1 × 109, and

(5.6)

3 × 109 L/mol sec for the Li+, Na+, K+ and Cs+ salts, respectively [249]. Based on these kinetic results, Szwarc [249] has concluded that the addition of monomer to the monomer radical anion is of little importance in the initiation process by electron transfer.

These reactions or the corresponding direct reduction reactions with alkali metals can be used to prepare oligomeric dianionic initiators from monomers such as α-methylstyrene, which have accessible ceiling temperatures (Tc = 61°C) [254] as shown in Scheme 5.5. It is interesting to note that while it is reported that the reduction of α-methylstyrene by sodium potassium alloy produces the dimeric

Scheme 5.5



dianionic initiators in THF [261], the reduction with sodium metal forms the tetrameric dianions as the main products [262]. These dianionic initiators are formed and used in polar solvents such as THF.

The consequence of the necessity to use polar solvents for aromatic radical anion initiators and the corresponding dianionic initiators (see Scheme 5.5) is that polydiene microstructure is high in 1,2- and 3,4- addition stuctures (i.e., the high 1,4-stereospecificity observed with lithium in hydrocarbon solvent is lost in polar solvents such as tetrahydrofuran; see Chapter 9). In addition, polar solvents accelerate the rate of propagation; this will tend to broaden the molecular weight distribution because of the disparity between the rate of propagation relative to the rate of initiation (see Chapter 4). Furthermore, in polar solvents, there is an equilibrium (Winstein spectrum, see Chapter 3) between contact ion pairs, solvent-separated ion pairs and free ions, each of which can propagate with different rate constants [249,263]. As a consequence of these equilibria, broader molecular weight distributions can result when the interchange between different propagating species is not fast relative to propagation [263–265]. A further complication with radical anion initiators and oligomeric α-methylstyrylsodium is that it has been reported that on aging they form side reaction products with reduced reactivity for polymerization [266]. The common practice of adding sodium tetraphenylborate common ion salt to suppress formation of free ions can also result in formation of inactive species [267]. Thus, not all anionic polymerization systems provide well-controlled, living polymerizations; it is always necessary to characterize the polymer products adequately [17].

Monomers that can be polymerized with aromatic radical anions include styrenes, dienes, epoxides, and cyclosiloxanes. If the aromatic radical anions are too stable, they will not efficiently initiate polymerization of less reactive monomers; thus, the radical anion of anthracene cannot initiate styrene polymerization [249].

C. Alkyllithium Compounds

A variety of simple alkyllithium compounds are readily available commercially in hydrocarbon solvents such as hexane and cyclohexane. Methyllithium and phenyllithium are available in diethyl ether solutions, since they are not soluble in hydrocarbon solvents. In general, commercially available solutions are used directly as anionic polymerization initiators; however, solutions of alkyllithium compounds frequently show turbidity associated with the formation of lithium alkoxides by oxidation reactions or hydroxides by reaction with moisture. Although these species contribute to the total basicity of the solution as determined by simple acid titration, they do not react with allylic and benzylic bromides or ethylene dibromide rapidly in ether solvents. This is the basis for the double titration method of determining the amount of carbon-bound lithium reagent in a



given sample [268,269]. Purification of many alkyllithium compounds can be effected by recrystallization (ethyllithium [270]), sublimation (ethyllithium [270], t-butyllithium [270], isopropyllithium [270]), or distillation (sec-butyllithium [271]). Since methyllithium [272] and phenyllithium [273] are crystalline solids insoluble in hydrocarbon solution, they can be precipitated in these solutions and then dissolved in an appropriate polar solvent. Unfortunately, n-butyllithium is noncrystalline and has too high a boiling point to be purified by distillation [8].

The important differences between the various alkyllithium compounds are their degrees of aggregation in solution (see Chapter 1) and their relative reactivity as initiators for anionic polymerization of styrene and diene monomers. The relative reactivities of alkyllithiums as polymerization initiators are intimately linked to their degree of association, as shown below, with the average degree of association in hydrocarbon solution, where known (see Chapter 1), indicated in parentheses after the alkyllithium [1,13,18,274,275]:

Styrene polymerization: menthyllithium (2) > sec-BuLi (4) > i-PrLi (4–6) >

i-BuLi > n-BuLi (6) > t-BuLi (4)

Diene polymerization: menthyllithium (2) > sec-BuLi (4) > i-PrLi (4–6) >

t-BuLi (4) > i-BuLi > n-BuLi (6)

It is clear that, in general, the less associated alkyllithiums are more reactive as initiators than the more highly associated species. The effect of solvent on initiator reactivity is also consistent with the importance of association phenomena. Aromatic solvents, that tend to decrease the average degree of association and promote dissociation processes of aggregates, are reported to lead to initiation rates 102-103 faster than in aliphatic solvents [271,274,275].

Alkyllithium initiators are primarily used as initiators for polymerizations of styrenes and dienes. They effect quantitative, living polymerization of styrenes and dienes; one unique aspect of lithium-based initiators in hydrocarbon solution is that elastomeric polydienes with high 1,4-microstructure are obtained (see Tables 5.4 and 5.5; Chapter 9). In general, these initiators are too reactive for alkyl methacrylates and vinylpyridines (see Table 5.3; Chapter 23). n-Butyllithium is used commercially to initiate anionic homopolymerization and copolymerization of butadiene, isoprene, and styrene with linear and branched structures. Because of the high degree of association (hexameric), n-butyllithium-initiated polymerizations are often effected at elevated temperatures (> 50°C) to increase the rate of initiation relative to propagation and thus to obtain polymers with narrower molecular weight distributions [276] (see Chapter 4). Hydrocarbon solutions of this initiator are quite stable at room temperature for extended periods of time; the rate of decomposition per month is 0.06% at 20°C [277].

sec-Butyllithium is the second most important organolithium initiator. It is used commercially to prepare styrene-diene block copolymers because it can ini-



Administrator

Underline

Administrator

Underline

Administrator

Highlight

Administrator

Underline

tiate styrene polymerization rapidly compared to propagation. Even polystyrene blocks with relatively low molecular weights (10,000–15,000 g/mole) can be prepared with stoichiometric control and narrow molecular weight distributions. Hydrocarbon solutions of sec-butyllithium are thermally less stable than n-butyllithium solutions; the rate of decomposition is 1.4% per month at 20°C [277].

D. Copolymerization Initiators

The copolymerization of styrene and dienes in hydrocarbon solution with alkyllithium initiators produces a tapered block copolymer structure because of the large differences in monomer reactivity ratios for styrene (rs < 0.1) and dienes (rd > 10) [13,18,111]. In order to obtain random copolymers of styrene and dienes, it is necessary to add either small amounts of a Lewis base such as tetrahydrofuran or an alkali metal alkoxide (Na,K,Rb,Cs). In contrast to Lewis bases, which promote formation of undesirable vinyl microstructure in diene polymerizations [278], the addition of small amounts of an alkali metal alkoxide such as potassium amyloxide ([ROK]/[Li] = 0.08) is sufficent to promote random copolymerization of styrene and diene without producing significant increases in the amount of vinyl microstructure [279,280].

E. Difunctional Initiators

Difunctional initiators are of considerable interest for the preparation of triblock copolymers, telechelic polymers, and macrocyclic polymers. Although triblock copolymers can be prepared with monofunctional initiatiors using a three-step, sequential monomer addition process, with difunctional initiators they can be formed in a more efficient two-step process [13,20,281]. Difunctional initiators also provide a methodology to prepare new triblock copolymers that cannot be prepared by the three-step, sequential monomer addition route because the chain ends formed from the first monomer are too stable to initiate the polymerization of the second monomer; for example, a difunctional initiator can be used for the direct synthesis of poly(ethylene oxide)-block-polystyrene-block-poly(ethylene oxide) [20]. Difunctional initiators provide direct, efficient methods for the formation of α,ω-difunctional polymers (i.e., telechelic polymers; see Chapter 11) [282], by termination reactions of the polymeric α,ω-dianions with electrophilic functionalization agents. In analogous fashion, termination of the α,ω-dianions with a difunctional, electrophilic coupling agent under high dilution conditions promotes intramolecular cyclization reactions to form macrocyclic polymers [283].

Aromatic radical anions, such as lithium naphthalene or sodium naphthalene, are efficient difunctional initiators (see Scheme 5.4) [249,257,284]. However, the need to use polar solvents for their formation and use limits their utility for diene polymerization since the unique ability of lithium to provide high 1,4-polydiene microstructure is lost in polar media [13,18,111,283,284]. As a consequence, a



Administrator

Underline

Administrator

Highlight

Administrator

Underline

Administrator

Underline

Administrator

Highlight

Administrator

Underline

major research challenge has been to discover a hydrocarbon-soluble, dilithium initiator that could initiate the polymerization of styrene and diene monomers to form monomodal α,ω-dianionic polymers at rates faster than or comparable to the rates of polymerization (i.e., to form narrow molecular weight distribution polymers) [20,22,285].

The methodology for preparation of hydrocarbon-soluble, dilithium initiators is generally based on the reaction of an aromatic divinyl precursor with two moles of butyllithium. Because of the tendency of organolithium chain ends in hydrocarbon solution to associate and form electron-deficient dimeric, tetrameric or hexameric aggregates (see Chapter 1) [8,111,286,287], however, attempts to prepare dilithium initiators in hydrocarbon media have generally resulted in the formation of insoluble, three-dimensionally associated species [13,285,288–292]. These precipitates are not effective initiators because of their heterogeneous initiation reactions with monomers that tend to result in broader molecular weight distributions (Mw/Mn > 1.1) [288,290,292]. Soluble analogs of these difunctional initiators have been prepared by addition of either small amounts of weakly basic additives such as triethylamine [293] or anisole [294], which have relatively minor effects on diene microstructure [295]. Another method to solubilize these initiators is to use a seeding technique, by which small amounts of diene monomer are added to form a hydrocarbon-soluble, oligomeric dilithium-initiating species [289,296].

The stoichiometric reaction of m-diisopropenylbenzene with two moles of sec-butyllithium in the presence of triethylamine (Scheme 5.6) has been reported to produce a useful, hydrocarbon-soluble dilithium initiator because of the low

Scheme 5.6



ceiling temperature of the monomer [297,298], which is analogous in structure to α-methylstyrene; however, other studies suggest that oligomerization occurs (see Scheme 5.7) with this divinyl compound and that functionalities higher than two result [299]. Higher functionalities result from oligomerization of the desired diadduct with additional divinyl precursor as shown in Scheme 5.7.

Scheme 5.7

In analogous fashion, the use of m-divinylbenzene as a dilithium precursor has been reported [300]. However, oligomerization occurs on treatment of divinylbenzene with butyllithium, resulting in initiators with functionalities greater than two [288]. From a commercial perspective this oligomerization and lack of precise functionality control is not necessarily a problem and useful multifunctional initiators have been prepared from the reaction of butylithium with varying amounts of divinylbenzene (commerical divinylbenzene contains 22% meta, 11% para and 66% o-, m-, and p-ethylvinylbenzene) [301], often in the presence of styrene or diene monomer to provide solubility (seeding technique). The reaction of pure m-divinylbenzene with sec-butyllithium in toluene at -49°C in the presence of triethylamine ([Et3N]/[Li] = 0.1) has been reported to produce the corresponding dilithium initiator in quantitative yield [302]. Polymerization of butadiene with this initiator in toluene at -78°C produced polybutadiene with low polydispersity (Mw/Mn = 1.06), high 1,4-microstructure (87%), and low Tg (-86°C).

Although a plethora of divinyl aromatic compounds has been investigated as



precursors for hydrocarbon-soluble dilithium initiators [288], one of the only systems demonstrated to produce a hydrocarbon-soluble dilithium initiators is based on 1,3-bis(1-phenylethenyl)benzene [281,303–312]. The addition reaction of sec-butyllithium with 1,3-bis(1-phenyl-ethenyl)benzene (Scheme 5.8) pro

Scheme 5.8

ceeds rapidly and efficiently to produce the corresponding dilithium species in toluene [313] or in cyclohexane [304]. This dilithium initiator is soluble in hydrocarbon media such as cyclohexane, benzene and toluene (even at -20°C) [306], but it is not soluble in n-hexane [308]. This hydrocarbon-soluble dilithium adduct also functions as an efficient difunctional initiator for the preparation of homopolymers and triblock copolymers with relatively narrow molecular weight distributions [303–305]. However, it is necessary to add a small amount of Lewis base [310] or two equivalents of lithium alkoxide (e.g., lithium sec-butoxide) [281] to produce narrow, monomodal molecular weight distributions. Added lithium sec-butoxide is the preferred additive, since high 1,4-polybutadienes are obtained [281].

F. Functionalized Initiators

The use of alkyllithium initiators that contain functional groups provides a versatile method for the preparation of end-functionalized polymers and macromonomers. For a living anionic polymerization, each functionalized initiator molecule will produce one macromolecule, with the functional group from the initiator residue at one chain end and the active carbanionic propagating species at the other chain end. Thus, in contrast to most functionalization procedures that involve postpolymerization termination reactions with electrophilic reagents [314], the use of a functionalized initiator retains the anionic chain end and the



ability to prepare block and star-branched polymers with the functional group at the initiating end. For example, dimethylaminopropyllithium can be prepared in hydrocarbon solution and has been used to prepare polystyrenes and polydienes with tertiary amine end group functionality [315,316]. However, many functional groups such as hydroxyl, carboxyl, phenol, and primary amine are not stable in the presence of reactive dienyllithium and styryllithium chain ends. Therefore, it is necessary to convert these functional groups into suitable derivatives, (i.e., protected groups), which should be stable to the carbanionic chain ends and can be removed readily after the polymerization is completed [16]. Examples of these protected functional initiators include the hydroxyl-protected initiator, 6-lithiohexyl acetaldehyde acetal [317] and a primary amine-protected initiator, 4-bis(tri-methylsilyl)aminophenyllithium [318]. New protected hydroxyl group initiators, 6-(t-butyldimethylsiloxy)hexyllithium [319,320] and 3-(t-butyldimethylsiloxy)-propyllithium [321] have recently been described. In one application, for example, they can be used to synthesize functionalized, star-branched polyisoprenes [321].

G. Cumyl Potassium

Cumyl potassium (pKa > 43 based on toluene; see Table 5.3) is another useful initiator for anionic polymerization of a variety of monomers, including styrenes, dienes, methacrylates, and epoxides. This carbanion is readily prepared from cumyl methyl ether, as shown in Eq. 5.7 [322]. It is necessary to remove the

(5.7)

potassium methoxide salt that precipitates from the solution; cooling to low temperature prior to filtration is recommended. The concentration of active initiating species can be determined by titration with standardized acid or by using this initiator with a known amount of styrene monomer and measuring the number average molecular weight of the polymer, if it is assumed that one initiator moiety produces one polystyrene macromolecule (see Chapter 4). This initiator is generally used at low temperatures in polar solvent such as THF, that limits the microstructure of polydienes to low 1,4-contents (high 1,2 microstructure). This initiator is useful for the synthesis of polystyrene-block-poly(ethylene oxide) diblock copolymers [323]; analogous polymerizations using alkyllithium initiators are complicated by the lack of reactivity of lithium alkoxides as initiators for ethylene oxide polymerization [324,325].



H. 1,1-Diphenylmethylcarbanions

The carbanions based on diphenylmethane (pKa = 32) (see Table 5.3) are useful initiators for vinyl and heterocyclic monomers, especially alkyl methacrylates at low temperatures [121,208–212]. Addition of lithium chloride [127,214,215], lithium tert-butoxide [326], or lithium 2-(2-methoxyethoxy)ethoxide [216] has been shown to narrow the molecular weight distribution and improve the stability of active centers for anionic polymerization of both alkyl methacrylates and tertbutyl acrylate. A surprise has been that these more stable carbanions can also efficiently initiate the polymerization of styrene and diene monomers [217]. Diphenylmethyllithium can be prepared by the metalation reaction of butyllithium with diphenylmethane; in addition, the adduct of butyllithium and 1,1-diphenylethylene is conveniently prepared in either hydrocarbon or polar solvents such as THF, as shown in Eq. 5.2. This reaction can also be utilized to prepare functionalized initiators by reacting butyllithium with a substituted 1,1-diphenylethylene derivative. For example, polymers end-functionalized with primary amine, tertiary amine, phenol, and bis(phenol) groups have been prepared in essentially quantitative yields by using the reaction of sec-butyllithium with the corresponding substituted (or protected) 1,1-diphenylethylene [314].

I. Fluorenyl Carbanions

Salts of fluorene (pKa = 22.6) (see Table 5.3) are more hindered and less reactive than many other organometallic initiators. These carbanions can be readily formed by reaction with alkali metal derivatives, as shown in Eq. 5.8 for 9-methylfluorene

(5.8)

[327]. Carbanion salts of 9-methylfluorene are preferable to fluorene, since the latter generate chain ends that retain reactive, acidic fluorenyl hydrogens that can participate in chain transfer reactions [328–330]. Fluorenyl salts are useful initiators for the polymerization of alkyl methacrylates, epoxide, and thiirane monomers.

J. Ester Enolate Initiators

In principle, ester enolate anions should represent the ideal initiators for anionic polymerization of alkyl methacrylates. Although general procedures have been developed for the preparation of a variety of alkali metal enolate salts, many of these compounds are unstable except at low temperatures [286,331,332]. Useful



initiating systems for acrylate polymerization have been prepared from complexes of ester enolates with alkali metal alkoxides [280,281].

V. Summary

A wide variety of nonpolar monomers, polar vinyl monomers (see Chapter 23) and heterocyclic monomers (see Chapter 24) can be polymerized in a controlled fashion by anionic polymerization. The most versatile, controlled polymerization methodology is the alkyllithium-initiated polymerization of styrene and diene monomers in hydrocarbon solution. Even within this rather limited scope of structural variation, a wide variety of substituted monomers and monomers with protected functional groups can be polymerized. The intelligent design and execution of controlled polymer synthesis require the careful matching of monomer, initiator reactivity and reaction conditions. Although nonpolar solvents are preferred, polar solvents have utility for certain monomers and the addition of Lewis bases can be used to promote initiation, propagation, and controlled termination reactions.

References

1. M. Morton and L. J. Fetters, Rubber Chem. Tech., 48, 359 (1975).

2. L. J. Fetters and M. Morton, in Macromolecular Syntheses, J. A. Moore, Ed., Wiley, New York, 1977, Collective Vol. 1, p. 463.

3. J. C. Falk, M. A. Benedetto, J. Van Fleet, and L. Ciaglia, Macromol. Synth., 8, 61 (1982).

4. L. C. Anand, A. Takahashi, and N. G. Gaylord, in Macromolecular Syntheses, Collective Volume I, J. A. Moore, Ed., Wiley, New York, 1977, p. 389.

5. D. Braun, H. Cherdron, and W. Kern, Practical Macromolecular Organic Chemistry, translated by K. J. Ivin, Harwood Academic Publishers, New York, 1984, p. 172.

6. R. W. Richards and J. L. Thomason, Polymer, 23, 1988 (1982).

7. D. F. Shriver and M. A. Drezdzon, The Manipulation of Air-Sensitive Compounds, 2nd ed., Wiley-Interscience, New York, 1986.

8. B. J. Wakefield, The Chemistry of Organolithium Compounds, Pergamon Press, New York, 1974.

9. L. J. Fetters and E. M. Firer, Polymer, 18, 306 (1977).

10. R. P. Quirk and W.-C. Chen., J. Polym. Sci. Polym. Chem. Ed., 22, 2993 (1984).

11. E. J. Panek and G. M. Whitesides, J. Am. Chem. Soc., 94, 8768 (1972).

12. J. M. Catala, G. Riess, and J. Brossas, Makromol. Chem., 178, 1249 (1977).

13. S. Bywater, in Encyclopedia of Polymer Science and Engineering, 2nd ed., Vol. 2, J. I. Kroschwitz, Ed., Wiley-Interscience, New York, 1985, p. 1.

14. H. Sawada, Thermodynamics of Polymerization, Marcel Dekker, New York, 1976.

15. G. Odian, Principles of Polymerization, 3rd ed., Wiley, New York, 1991, p. 199.

16. S. Nakahama and A. Hirao, Prog. Polym. Sci., 15, 299 (1990).





19. C. G. Overberger, J. E. Mulvaney, and A. M. Schiller, in Encyclopedia of Polymer Science and Technology, N. Bikales, Ed., Wiley, New York, Vol. 2, 1965, p. 95.

20. G. Riess and G. Hurtrez, in Encyclopedia of Polymer Science and Engineering, 2nd ed., J. I. Kroschwitz, Ed., Wiley, New York, 1985, Vol. 2, p. 324.

21. M. Fontanille, in Comprehensive Polymer Science, Vol. 3, Chain Polymerization I, G. C. Eastmond, A. Ledwith, S. Russo, and P. Sigwalt, Eds., Pergamon, New York, 1989, p. 365.

22. R. Jerome, R. Fayt, and T. Ouhadi, Prog. Polym. Sci., 10, 87 (1984).

23. W. E. Hanford, J. R. Roland, and H. S. Young, U.S. patent, 2,337,779 (1945).

24. A. W. Langer, J. Polym. Prepr. Am. Chem. Soc., Div. Polym. Chem., 7, 132 (1966).

25. M. Aldissi, F. Schue, K. Geckeler, and M. Abadie, Makromol. Chem., 181, 1425 (1980).

26. D. E. Bergbreiter, J. R. Blanton, R. Chandran, M. D. Hein, K.-J. Huang, D. R. Treadwell, and S. A. Walker, J. Polym. Sci., Polym. Chem., 27, 4205 (1989).

27. C. V. Uglea and I. I. Negulescu, Synthesis and Characterization of Oligomers, CRC Press, Boca Raton, FL, 1991, p. 215.

28. G. K. Rickle, J. Macromol. Sci. Chem., A23, 1287 (1986).

29. G. K. Rickle, J. Macromol. Sci. Chem., A24, 93 (1987).

30. R. Asami, J. Oku, M. Takeuchi, K. Nakamura, and M. Takaki, Polym. J., 20, 699 (1988).

31. J. W. Mays and N. Hadjichristidis, Polym. Bull, 22, 471 (1989).

32. R. P. Quirk, M. T. Sarkis, and D. J. Meier, in Advances in Elastomers and Rubber Elasticity, J. Lal and J. E. Mark, Eds., Plenum Press, New York, 1986, p. 143.

33. J. C. Chen and L. J. Fetters, Polym. Eng. Sci., 27, 1300 (1987).

34. D. A. Conlon, J. V. Crivello, J. L. Lee, and M. J. O'Brien, Macromolecules, 22, 509 (1989).

35. H. Yuki, Y. Okamoto, Y. Kuwae, and K. Hatada, J. Polym. Sci., A-1, 7, 1933 (1969).

36. J. Geerts, M. Van Beylen, and G. Smets, J. Polym. Sci., A-1, 7, 2859 (1969).

37. K. Se, M. Kijima, and T. Fujimoto, Polym. J., 20, 791 (1988).

38. M. Shima, E. Ogawa, and K. Konishi, Makromol. Chem., 777, 241 (1976).

39. E. Ogawa and M. Shima, Polym. J., 18, 443 (1986).

40. I. Konigsberg and J. Jagur-Grodzinski, J. Polym. Sci. Polym. Chem. Ed., 21, 2535 (1983).

41. I. Konigsberg and J. Jagur-Grodzinski, J. Polym. Sci. Polym. Chem. Ed., 21, 2649 (1983).



47. K. Kajiwara, H. Suzuki, H. Inagaki, M. Maeda, and T. Tsuruta, Makromol. Chem., 187, 2257 (1986).

48. Y. Nagasaki, Y. Tamura, and T. Tsuruta, Makromol. Chem., Rapid Commun., 9, 31 (1988).

49. H. Nakagawa, Y. Matsushita, and S. Tsuge, Polymer, 28, 1512 (1987).

50. R. M. Nasirova, L. S. Murav'eva, E. A. Mushina, and B. A. Krentsel, Russian Chem. Rev., 48, 692 (1979).

51. D. J. Worsfold, Macromolecules, 3, 514 (1970).

52. R. N. Young and L. J. Fetters, Macromolecules, 11, 899 (1978).

53. T. Ishizone, A. Hirao, and S. Nakahama, Macromolecules, 24, 625 (1991).

54. T. Ishizone, S. Wakabayashi, A. Hirao, and S. Nakahama, Macromolecules, 24, 5015 (1991).

55. S. J. Whicher and J. L. Brash J. Appl. Polym. Sci., 30, 2297 (1985).

56. S. J. Whicher and J. L. Brash, J. Polym. Sci. Polym. Chem. Ed., 19, 1995 (1981).

57. T. Kase, M. Imahori, T. Kazama, Y. Isono, and T. Fujimoto, Macromolecules, 24, 1714 (1991).


59. N. Yamazaki, S. Nakahama, A. Hirao, J. Goto, Y. Shiraishi, F. Martinez, and H. M. Phung, J. Macromol. Sci., Chem., 16, 1129 (1981).

60. A. Hirao, Y. Shiraishi, F. Martinez, H. M. Phung, S. Nakahama, and N. Yamazaki, Makromol. Chem., 184, 961 (1983).

61. T. Ishizone, A. Hirao, S. Nakahama, T. Kakuchi, K. Yokota, and K. Tsuda, Macromolecules, 24, 5230 (1991).

62. T. Ishizone, R. Kato, Y. Ishino, A. Hirao, and S. Nakahama, Macromolecules, 24, 1449 (1991).

63. A. Hirao, Y. Ishino, and S. Nakahama, Makromol. Chem., 187, 141 (1986).

64. A. Hirao, K. Takenaka, S. Packirisamy, K. Yamaguchi, and S. Nakahama, Makromol. Chem., 186, 1157 (1985).

65. A. Hirao and S. Nakahama, Macromolecules, 20, 2968 (1987).

66. A. Hirao, Y. Ishino, and S. Nakahama, Macromolecules, 21, 561 (1988).

67. K. Suzuki, K. Yamaguchi, A. Hirao, and S. Nakahama, Macromolecules, 22, 2607 (1989).

68. A. Hirao, K. Kato, and S. Nakahama, Macromolecules, 25, 535 (1992).

69. A. Hirao, T. Hatayama, T. Nagawa, M. Yamaguchi, K. Yamaguchi, and S. Nakahama, Macromolecules, 20, 242 (1987).

70. T. Taki, A. Hirao, and S. Nakahama, Macromolecules, 24, 1455 (1991).

71. A. Hirao, T. Hatayama, and S. Nakahama, Macromolecules, 20, 1505 (1987).



80. D. Ades, M. Fontanille, and J. Leonard, Can. J. Chem., 60, 564 (1982).

81. D. Ades, M. Fontanille, J. Leonard, and M. Thomas, Eur. Polym. J., 19, 305 (1983).

82. P. Chaumont, G. Beinert, J. Herz, and P. Rempp, Makromol. Chem., 180, 2061 (1979).

83. Y. Okamoto, H. Takano, and H. Yuki, Polym. J., 1, 403 (1970).

84. P. Chaumont, G. Beinert, and J. Herz, Eur. Polym. J., 18, 875 (1982).

85. P. Lutz, G. Beinert, E. Franta, and P. Rempp, Polym. Prepr. Am. Chem. Soc., Div. Polym. Chem., 20(2), 122 (1979).

86. P. Lutz, G. Beinert, and P. Rempp, Makromol. Chem., 183, 2787 (1982).

87. K. Saigo and F. Watanabe, J. Polym. Sci. Polym. Chem., 27, 2611 (1989).

88. H. Gisser and H. E. Mertwoy, Macromolecules, 2, 461 (1969).

89. A. Rembaum, J. Moacanin, and R. Haack, J. Macromol. Chem., 1, 657 (1966).

90. A. Douy, G. Jouan, and B. Gallot, Makromol. Chem., 177, 2945 (1976).

91. M. Stolka, J. F. Yanus, and J. M. Pearson, Macromolecules, 9, 710, 715, 719 (1976).

92. J. F. O'Malley, J. F. Yanus, and J. M. Pearson, Macromolecules, 5, 158 (1972).

93. Y. Morishima, Y. Itoh, T. Hashimoto, and S. Nozakura, J. Polym. Sci. Polym. Chem. Ed., 20, 2007 (1982).

94. T. Rhein, S. Boileau, and R. C. Schulz, Makromol. Chem., 188, 2151 (1987).

95. D. Engel and R.C. Schulz, Eur. Polym. J., 19, 967 (1983).

96. D. Engel and R. C. Schulz, Makromol. Chem., 182, 3279 (1981).

97. H. L. Hsieh and I. W. Wang, in Recent Advances in Anionic Polymerization, T. E. Hogen-Esch and J. Smid, Eds., Elsevier, New York, 1987, p. 411.

98. H. Yuki, Prog. Polym. Sci. Jpn., 3, 141 (1972).

99. A. R. Luxton, A. Quig, M.-J. Delvaux, and L. J. Fetters, Polymer, 19, 1320 (1978).

100. C. C. Meverden and T. E. Hogen-Esch, Makromol. Chem., Rapid Commun., 5, 749 (1984).

101. I. G. Krasnoselskaya and B. L. Erussalimsky, Makromol. Chem., Rapid Commun., 6, 191 (1985).

102. I. G. Krasnoselskaya and B. L. Erussalimsky, Acta Polym., 37, 72 (1986).

103. M. Müller and R. W. Lenz, Makromol. Chem., 190, 1153 (1989).

104. C. C. Meverden and T. E. Hogen-Esch, J. Polym. Sci. Polym. Chem. Ed., 23, 159 (1985).

105. A. H. Soum, C.-F. Tien, T. E. Hogen-Esch, N. B. D'Accorso, and M. Fontanille, Makromol. Chem. Rapid Commun., 4, 243 (1983).



113. X. Zhongde, J. Mays, C. Xuexin, N. Hadjichristidis, F. C. Schilling, H. E. Bair, D. S. Pearson, and L. J. Fetters, Macromolecules, 18, 2560 (1985).

114. D. Blondin, J. Regis, and J. Prud'homme, Macromolecules, 7, 187 (1974).

115. T. Suzuki, Y. Tsuji, Y. Takegami, and H. J. Harwood, Macromolecules, 12, 234 (1979).

116. K. Takenaka, T. Hattori, A. Hirao, and S. Nakahama, Macromolecules, 22, 1563 (1989).

117. Y.-X. Ding and W. P. Weber, Macromolecules, 21, 532, 2672 (1988).

118. X. F. Zhong and B. Francois, Makromol. Chem., 191, 2735, 2743 (1990).

119. C. Petzhold, R. Stadler, and H. Frauenrath, Makromol. Chem. Rapid Commun., 14, 33 (1993).

120. R. P. Quirk and T.-L. Huang, in New Monomers and Polymers, B. M. Culbertson and C. U. Pittman, Eds., Plenum, New York, 1984, p. 329.

121. G. D. Andrews and L. R. Melby, in New Monomers and Polymers, B. M. Culbertson and C. U. Pittman, Eds., Plenum, New York, 1984, p. 357.

122. A. H. E. Muller, in Recent Advances in Anionic Polymerization, T. E. Hogen-Esch and J. Smid, Eds., Elsevier, New York, 1987, p. 205.

123. K. Hatada, T. Kitayama, and K. Ute, Prog. Polym. Sci., 13, 189 (1988).

124. H. Yuki and K. Hatada, Adv. Polym. Sci., 31, 1 (1979).

125. K. Hatada, K. Ute, K. Tanaka, T. Kitayama, and Y. Okamoto, in Recent Advances in Anionic Polymerization, T. E. Hogen-Esch and J. Smid, Eds., Elsevier, New York, 1987, p. 195.

126. R. D. Allen, T. E. Long, and J. E. McGrath, Polym. Bull., 15, 127 (1986).

127. R. Fayt, R. Forte, C. Jacobs, R. Jerome, T. Ouhadi, Ph. Teyssie, and S. K. Varshney, Macromolecules, 20, 1442 (1987).

128. J. P. Hautekeer, S. K. Varshney, R. Fayt, C. Jacobs, R. Jerome, and Ph. Teyssie, Macromolecules, 23, 3893 (1990).

129. S. K. Varshney, C. Jacobs, J. P. Hautekeer, P. Bayard, R. Jerome, R. Fayt, and Ph. Teyssie, Macromolecules, 24, 4997 (1991).

130. H. Ozaki, A. Hirao, and S. Nakahama, Macromolecules, 25, 1391 (1992).

131. T. Kitayama, T. Shinozaki, T. Sakamoto, M. Yamamoto, and K. Hatada, Makromol. Chem., Suppl., 15, 167 (1989).

132. M. A. Buese and T. E. Hogen-Esch, Macromolecules, 17, 118 (1984).

133. M. A. Buese and T. E. Hogen-Esch, J. Am. Chem. Soc., 107, 4509 (1985).

134. A. Zilkha, B.-A. Feit, and M. Frankel, J. Polym. Sci., 49, 231 (1961).

135. A. Ottolenghi, S. Barzakay, and A. Zilkha, J. Polym. Sci., A, 1, 3643 (1963).

135. A. Tsukamoto, J. Polym. Sci., A, 3, 2767 (1965).



142. N. Yamashita, A. Tadokoro, E. Ozau, T. Maeshima, I. C. Baianu, and L.-S. Wei, J. Macromol. Sci. Chem., A24, 1223 (1987).

143. N. Yamashita, S. Morita, M. Nishino, and T. Maeshima, J. Polym. Sci. Polym. Chem., 27, 239 (1983).

144. C. G. Overberger and A. M. Schiller, J. Polym. Sci. C, 1, 325 (1963), 530 (1961).

145. A. R. Lyons and E. Catterall, Eur. Polym. J., 7, 839 (1971).

146. D. T. Dotcheva and C. B. Tsvetanov, Makromol. Chem., 186, 2103 (1985).

147. A. T. Blomquist, W. J. Tapp, and J. R. Johnson, J. Am. Chem. Soc., 67, 1519 (1945).

148. R. W. H. Berry, R. J. Mazza, and S. F. Sullivan, Makromol. Chem., 185, 559 (1984).

149. I. C. Eromosele, D. C. Pepper, and B. Ryan, Makromol. Chem., 190, 1613 (1989).

150. S. M. Aharoni, Polym. Bull., 9, 186 (1983).

151. W. Mormann, A. Schwabe, and R. Sikora, Makromol. Chem., 187, 133 (1986).

152. I. C. Eromosele and D. C. Pepper, J. Polym. Sci. A, 25, 3499 (1987).

153. V. Skarda, D. Ivkovich, and M. M. Labes, J. Polym. Sci. Polym. Chem. Ed., 23, 107 (1985).

154. T. Hagiwara, J. Mizota, H. Hamana, and T. Narita, Makromol. Chem. Rapid Commun., 6, 169 (1985).

155. T. Otsu, A. Tatsumi, and A. Matsumoto, J. Polym. Sci. Chem., 24, 113 (1986).

156. P. Hodge, E. Khoshdel, and A. A. Naim, Polym. Commun., 27, 322 (1986).

157. T. Hagiwara, M. Takeda, H. Hamana, and T. Narita, Makromol. Chem. Rapid Commun., 8, 167 (1987).

158. T. Hagiwara, T. Someno, H. Hamana, and T. Narita, J. Polym. Sci. Polym. Chem. Ed., 26, 1011 (1988).

159. T. Oishi, M. Fujimoto, N. Yoshimoto, and T. Kimura, Polym, J., 21, 655 (1989).

160. I. Cho and W.-T. Kim, J. Polym. Sci. Polym. Chem., 25, 2791 (1987).

161. I. Cho and K.-D. Ahn, J. Polym. Sci. Polym. Chem., 17, 3183 (1979).

162. R. P. Quirk and R. Bhatia, Rubber Chem. Tech., 62, 332 (1989).

163. Q. Zhou, G. Manuel, and W. P. Weber, Macromolecules, 23, 1583 (1990).

164. C. X. Liao and W. P. Weber, Macromolecules, 25, 1639 (1992).

165. T. Kitayama and H. K. Hall, Jr., Macromolecules, 20, 1451 (1987).

166. T. Tsuruta and Y. Kawakami, in Comprehensive Polymer Science, Vol. 3, Chain Reaction Polymerization I, G. C. Eastmond, A. Ledwith, S. Russo, and P. Sigwalt, Eds., Pergamon, New York, 1989, p. 31.



tion I, G. C. Eastmond, A. Ledwith, S. Russo, and P. Sigwalt, Eds., Pergamon, New York, 1989, p. 467.

173. F. E. Bailey, Jr., and J. V. Koleske, Alkylene Oxides and Their Polymers, Marcel Dekker, New York, 1991.

174. P. Sigwalt and N. Spassky, in Ring-Opening Polymerization, K. J. Ivin and T. Saegusa, Eds., Elsevier, New York, 1984, Vol. 2, p. 603.

175. D. B. Johns, R. W. Lenz, and A. Luecke, in Ring-Opening Polymerization, K. J. Ivin and T. Saegusa, Eds., Elsevier, New York, 1984, Vol. I, p. 461.

176. H. R. Kricheldorf, M. Berl, and N. Scharnagl, Macromolecules, 21, 286 (1988).

177. R. Jerome and Ph. Teyssie, in Comprehensive Polymer Science, Vol. 3, Chain Reaction Polymerization I, G. C. Eastmond, A. Ledwith, S. Russo, and P. Sigwalt, Eds., Pergamon, New York, 1989, p. 501.

178. Ph. Dubois, R. Jerome, and Ph. Teyssie, Makromol. Chem. Macromol. Symp., 42/43, 103 (1991).

179. Ph. Dubois, R. Jerome, and Ph. Teyssie, Macromolecules, 24, 977 (1991).

180. H. R. Kricheldorf, I. Kreiser-Saunders, and N. Scharnagl, Makromol. Chem. Macromol. Symp., 32, 285 (1990).

181. H. Hocker, H. Keul, S. Kuhling, and W. Hovestadt, Makromol. Chem. Macromol. Symp., 42/43, 145 (1991).

182. D. J. Brunelle and T. G. Shannon, Makromol. Chem. Macromol. Symp., 42/43, 155 (1991).

183. H. Keul, H. Hocker, E. Leitz, K.-H. Ott, and L. Morbitzer, Makromol. Chem., 189, 2303 (1988).

184. D. J. Brunelle, in Ring-Opening Polymerization, D. J. Brunelle, Ed., Hanser, New York, 1993, p. 309.

185. J. Sebenda, in Comprehensive Polymer Science, Vol. 3, Chain Reaction Polymerization I, G. C. Eastmond, A. Ledwith, S. Russo, and P. Sigwalt, Eds., Pergamon, New York, 1989, p. 511.

186. H. Sekiguchi, in Ring-Opening Polymerization, K. J. Ivin and T. Saegusa, Eds., Elsevier, New York, 1984, Vol. 2, p. 809.

187. J. Sebenda, Makromol. Chem. Macromol. Symp., 32, 105 (1990).

188. P. V. Wright, in Ring-Opening Polymerization, K. J. Ivin and T. Saegusa, Eds., Elsevier, New York, 1984, Vol. 2, p. 1055.

189. S. Boileau, in Ring-Opening Polymerization, J. E. McGrath, Ed., ACS Symposium Series 286, American Chemical Society, Washington, D.C., 1985, p. 23.

190. D. Fish, I. M. Khan, and J. Smid, Makromol. Chem. Macromol. Symp., 32, 241 (1990).

191. T. C. Kendrick, B. M. Parbhoo, and J. W. White, in Comprehensive Polymer Science, Vol. 4, Chain Reaction Polymerization II, G. C. Eastmond, A. Ledwith, S. Russo and P. Sigwalt, Eds., Pergamon, New York, 1989, p. 459.



195. F. G. Bordwell and D. J. Algrim, J. Am. Chem. Soc., 110, 2964 (1988).


197. Y. Chiang, A. J. Kresge, and Y. S. Tang, J. Am. Chem. Soc., 106, 460 (1984).

198. W. Reeve, C. M. Erikson, and P. F. Aluotto, Can. J. Chem., 57, 2747 (1979).

199. A. Ohno and S. Oae, in Organic Chemistry of Sulfur, S. Oae, Ed., Plenum Press, New York, 1977, p. 128.

200. R. G. Pearson and R. L. Dillon, J. Am. Chem. Soc., 75, 2439 (1953).

201. L. Allred, E. G. Rochow, and F. G. A. Stone, J. Inorg. Nucl. Chem., 2, 416 (1956).

202. H. Arm, K. Hochstrasser, and P. W. Schindler, Chimia, 28, 237 (1974).

203. G. Kortum, W. Vogel, and K. Andrussow, Dissociation Constants of Organic Acids in Aqueous Solution, Butterworths, London, 1961.

204. R. P. Quirk, D. J. Kinning, and L. J. Fetters, in Comprehensive Polymer Science, Vol. 7, Specialty Polymers & Polymer Processing, S. L. Agarwal, Ed., Pergamon, New York, 1989, p. 1.

205. A. Hirao and S. Nakahama, Prog. Polym. Sci., 17, 283 (1992).

206. J. W. Klein, J.-P. Lamps, Y. Gnanou, and P. Rempp, Polymer, 32, 2278 (1991).

207. S. K. Varshney, X. F. Zhong, and A. Eisenberg, Macromolecules, 26, 701 (1993).

208. D. Freyss, P. Rempp, and H. Benoit, Polym. Lett., 2, 217 (1964).

209. D. M. Wiles and S. Bywater, Polym. Lett., 2, 1175 (1964).

210. D. M. Wiles and S. Bywater, Trans. Faraday Soc., 61, 150 (1965).

211. R. P. Quirk and L. Zhu, Brit. Polym. J., 23, 47 (1990).

212. B. C. Anderson, G. D. Andrews, P. Arthur, Jr., H. W. Jacobson, L. R. Melby, A. J. Playtis, and W. H. Sharkey, Macomolecules, 14, 1599 (1981).

213. K. Hatada, T. Kitayama, K. Fujikawa, K. Ohta, and H. Yuki, in Anionic Polymerization, J. E. McGrath, Ed., ACS Symposium Series No. 166, 1981, p. 327.

214. S. K. Varshney, J. P. Hautekeer, R. Fayt, R. Jerome, and Ph. Teyssie, Macro-molecules, 23, 2618 (1990).

215. D. Kunkel, A. H. E. Müller, M. Janata, and L. Lochmann, Makromol. Chem., Macromol, Symp., 60, 315 (1992).

216. J.-S. Wang, R. Jerome, Ph. Bayard, M. Patin, and Ph. Teyssie, Macromolecules, 27, 4635 (1994).

217. A. H. E. Müller, Makromol. Chem., 182, 2863 (1981).




Ledwith, S. Russo, and P. Sigwalt, Eds., Pergamon Press, Oxford, UK, 1989, Vol. 6, p. 403.

227. J. Roovers and P. M. Toporowski, Polym. Commun., 27, 266 (1986).

228. M. M. F. Al-Jarra, J. K. H. Al-Kafifi, and R. L. Apikian, Brit. Polym. J., 18, 256 (1986).

229. A. L. Gatzke, J. Polym. Sci., A-1, 7, 2281 (1969).

230. L. S. Wang, J. C. Favier, and P. Sigwalt, Polym. Commun., 30, 248 (1989).

231. The Merck Index, Eleventh ed., Merck & Co., Rahway, NJ, 1989, p. 238, no. 1551.

232. M. D. Glasse, Prog. Polym. Sci., 9, 133 (1983).

233. A. Gourdenne and P. Sigwalt, Eur. Polym. J., 3, 481 (1967).

234. D. N. Bhattacharyya, C. L. Lee, J. Smid, and M. Szwarc, J. Phys. Chem., 69, 612 (1965).

235. J. M. Mallan and R. L. Bebb, Chem. Rev., 69, 693 (1969).

236. H. Gilman and B. J. Gaj, J. Org. Chem., 22, 1165 (1957).

237. A. Rembaun, S.-P. Siao, and N. Indictor, J. Polym. Sci., 56, S17 (1962).

238. H. Gilman, A. H. Haubein, and H. Hartzfeld, J. Org. Chem., 19, 1034 (1954).

239. D. Seyferth and H. M. Cohen, J. Organomet. Chem., 1, 15 (1963).

240. J. J. Fitt and H. W. Gschwend, J. Org. Chem., 49, 209 (1984).

241. H. Gilman and H. A. McNich, J. Org. Chem., 27, 1418 (1962).

242. H. Gilman and G. L. Schwebke, J. Organometal. Chem., 4, 483 (1965).

243. P. D. Bartlett, S. J. Tauber, and W. P. Weber, J. Am. Chem. Soc., 91, 6362 (1969).

244. A. F. Halasa, D. F. Lohr, and J. E. Hall, J. Polym. Sci. Polym. Chem. Ed., 19, 1357 (1981).

245. F. E. Mathews and E. H. Strange, British Patent 24,790 (1910).

246. C. H. Harries, Justus Liebigs Ann. Chem., 383, 184, 190 (1911); U.S. Patent 1,058,056 (April 8, 1913).

247. E. G. M. Tornqvist, in Polymer Chemistry of Synthetic Elastomers, Part I, J. P. Kennedy and E. G. M. Tornqvist, Eds., Interscience, 1968, p. 21.

248. F. W. Stavely and coworkers, Indust. Eng. Chem., 778 (1956).


250. H. C. Wang, G. Levin, and M. Szwarc, J. Am. Chem. Soc., 100, 3969 (1978).

251. H. C. Wang, G. Levin, and M. Szwarc, J. Phys. Chem., 83, 785 (1979).

252. E. W. Duck and J. M. Locke, in The Stereo Rubbers, W. M. Saltman, Ed., Wiley Interscience, New York, 1977, p. 139.



260. R. P. Quirk and P. Cheng, Macromolecules, 19, 1291 (1986).

261. C. L. Lee, J. Smid, and M. Szwarc, J. Phys. Chem., 66, 904 (1962).

262. D. H. Richards and R. L. Williams J. Polym. Sci., 11, 89 (1973).

263. L. L. Böhm, G. Löhr, and G. V. Schulz, Ber. Bunsen-Ges., 78, 1064 (1974).

264. L. J. Fetters, in Encyclopedia of Polymer Science and Engineering, J. I. Kroschwitz, Ed., Wiley, New York, 1987, Vol. 10, p. 19.

265. G. Meyerhof, Z. Elektrochem., 61, 1245 (1957); Z. Phys. Chem., 23, 100 (1960).

266. G. V. Schulz, L. L. Böhm, M. Chmelir, G. L-öhr, and V. B. J. Schmitt, IUPAC International Symposium on Macromolecular Chemistry, Budapest, 1969, pp. 223–249.

267. V. B. J. Schmitt and G. V. Schulz, Makromol. Chem., 121, 184 (1969).

268. H. Gilman and F. K. Cartledge, J. Organomet. Chem., 2, 447 (1964).

269. T. R. Crompton, Comprehensive Organometallic Analysis, Plenum, New York, 1987, p. 181.


271. S. Bywater and D. J. Worsfold, J. Organomet. Chem., 10, 1 (1967).

272. T. L. Brown and M. T. Rogers, J. Am. Chem. Soc., 79, 1859 (1957).

273. G. R. Brubaker and P. Beak, J. Organomet. Chem., 136, 147 (1977).

274. H. L. Hsieh and W. H. Glaze, Rubber Chem. Tech., 43, 22 (1970).

275. S. Bywater, in Comprehesive Chemical Kinetics, C. H. Bamford and C. F. H. Tipper, Eds., Elsevier, Amsterdam, 1976, p. 1.

276. H. L. Hsieh and O. F. McKinney, Polym. Lett., 4, 843 (1966).

277. R. J. Bauer, in Ullman's Encyclopedia of Industrial Chemistry, 5th ed., VCH Verlagsgesellschaft, Weinheim, Germany, 1990, Vol. A 15, p. 393.


279. C. F. Wofford and H. L. Hsieh, J. Polym. Sci., Part A-1, 7, 461 (1969).

280. H. L. Hsieh and C. F. Wofford, J. Polym. Sci., Part A-1, 7, 449 (1969).

281. R. P. Quirk and J.-J. Ma, Polym. Int., 24, 197 (1991).

282. E. J. Goethals, Ed., Telechelic Polymers: Synthesis and Applications, CRC Press, Boca Raton, FL, 1989.

283. P. Rempp, C. Strazielle, and P. Lutz, in Encyclopedia of Polymer Science and Engineering, 2nd ed., Vol. 9, J. I. Kroschwitz, Ed., Wiley-Interscience, New York, 1987, p. 183.

284. M. J. M. Abadie and L. Satibi, Eur. Polym. J., 23, 423 (1987).



289. P. Guyot, J. C. Favier, H. Uytterhoeven, M. Fontanille, and P. Sigwalt, Polymer, 22, 1724 (1981).

290. P. Guyot, J. C. Favier, M. Fontanille, and P. Sigwalt, Polymer, 23, 73 (1982).

291. I. Obriot, J. C. Favier, and P. Sigwalt, Polymer, 28, 2093 (1987).

292. A. Soum, M. Fontanille, and A. Aboudalle, Polymer, 29, 1528 (1988).

293. R. P. Foss, H. W. Jacobson, and W. H. Sharkey, Macromolecules, 10, 287 (1977).


295. S. Bywater, in Comprehensive Polymer Science, Vol. 3, Chain Polymerization I, G. C. Eastmond, A. Ledwith, S. Russo, and P. Sigwalt, Eds., Pergamon Press, Elmsford, New York, 1989, p. 433.

296. L. H. Tung, G. Y.-S. Lo, and D. E. Beyer, Macromolecules, 11, 616 (1978).

297. G. Beinert, P. Lutz, E. Franta, and P. Rempp, Makromol. Chem., 779, 551 (1978).

298. P. Lutz, E. Franta, and P. Rempp, Polymer, 23, 1953 (1982).

299. G. G. Cameron and G. M. Buchan, Polymer, 20, 1129 (1979).

300. P. Lutz, E. Franta, and P. Rempp, C. R. Acad. Sci. Ser. C, 283, 123 (1976).

301. B. J. Bauer and L. J. Fetters, Rubber Chem. Technol., 51, 406 (1978).

302. R. D. Sanderson, A. H. A. Roediger, and G. J. Summers, Polym. Int., 35, 263 (1994).

303. L. H. Tung and G. Y. Lo, in Advances in Elastomers and Rubber Elasticity, J. Lal and J. E. Mark, Eds., Plenum Press, New York, 1986, p. 129.

304. A. D. Broske, T. L. Huang, R. D. Allen, J. M. Hoover, and J. E. McGrath, in Recent Advances in Anionic Polymerization, T. E. Hogen-Esch, and J. Smid, Eds., Elsevier, New York, 1987, p. 363.

305. T. E. Long, A. D. Broske, D. J. Bradley, and J. E. McGrath, J. Polym. Sci. Polym. Chem., 27, 4001 (1989).

306. G. Schulz and H. Höcker, Angew. Chem. Int. Ed., 19, 219 (1980).

307. T. Bastelberger and H. Höcker, Angew. Makromol. Chem., 125, 53 (1984).

308. L. H. Tung and T. Y.-S. Lo, Macromolecules, 27, 2219 (1994).

309. C. J. Bredeweg, A. L. Gatzke, G. Y.-S. Lo, and L. H. Tung, Macromolecules, 27, 2225 (1994).

310. G. Y.-S. Lo, E. W. Otterbacher, A. L. Gatzke, and L. H. Tung, Macromolecules, 27, 2233 (1994).

311. G. Y. L. Lo, E. W. Otterbacher, R. G. Pews, and L. H. Tung, Macromolecules, 27, 2241 (1994).

312. A. L. Gatzke and D. P. Green, Macromolecules, 27, 2249 (1994).



320. N. Shepherd and M. J. Stewart, U.S. Patent 5,331,058 (1994).

321. D. L. Handlin, R. C. Bening, and C. L. Willis, U.S. Patent 5,376,745 (1994).

322. K. Ziegler and H. Dislich, Chem. Ber., 90, 1107 (1957).

323. F. Calderara, Z. Hruska, G. Hurtrez, T. Nugay, and G. Riess, Makromol. Chem., 194, 1411 (1993).

324. R. P. Quirk and J.-J. Ma, J. Polym. Sci. Part A Polym. Chem., 26, 2031 (1988).

325. Y. Wang, C. M. Kausch, M. Chun, R. P. Quirk, and W. L. Mattice, Macromolecules, 28, 904 (1995).

326. M. Janata, L. Lochmann, and A. H. E. Müller, Makromol. Chem., 194, 625 (1993).

327. R. P. Quirk and G. P. Bidinger, Polym. Bull., 22, 63 (1989).

328. V. Warzelhan, H. Höcker, and G. V. Schulz, Makromol. Chem., 179, 2221 (1978).

329. D. Lassalle, S. Boileau, and P. Sigwalt, Eur. Polym. J., 13, 599 (1977).

330. A. Deffieux, P. Sigwalt, and S. Boileau, Eur. Polym. J., 20, 77 (1984).

331. D. Seebach, Angew. Chem. Int. Ed., 27, 1624 (1988).

332. G. Boche, K. Harms, and M. Marsch, Makromol. Chem. Macromol. Symp., 67, 97 (1993).



III KINETICS AND MECHANISM IN ANIONIC POLYMERIZATION



6 Initiation Reactions in Anionic Polymerization: Kinetics of Addition of Organolithium Compounds to Vinyl Monomers

I. Model Reactions

The mechanism of initiation of anionic polymerization of vinyl monomers with alkyllithium compounds and other organometallic compounds is complicated by association and cross-association phenomena in hydrocarbon solvents and by the presence of a variety of ionic species in polar media. As a consequence, several simple addition reactions of alkyllithium initiators to vinyl-type compounds have been utilized as models for these initiation reactions without the added complication of concurrent propagation that occurs with vinyl monomers.

A. Ethylenation Reaction

Ethylene will add to secondary and tertiary alkyllithium compounds under mild conditions (1 atmosphere of ethylene pressure, -25°C) to form the corresponding primary organolithium compound as shown in Eq. 6.1[1–3].

(6.1)

This exothermic reaction should be thermodynamically favorable (∆H = -25.8 kcal/mol; 108 kJ/mol)[4] at low temperatures; however, no addition occurs in hydrocarbon solution in the absence of ethers or amines and no addition occurs for primary alkyllithiums under these conditions[2]. The effectiveness of Lewis bases as catalysts for the ethylenation reaction is in the order tetrahydrofuran >



1,4-dimethoxybutane (not 1,2-dimethoxyethane) > triethylamine > N-methylpyrrolidine > diethyl ether[3]. The addition of t-butyllithium, isopropyllithium and sec-butyllithium to ethylene occurs quantitatively and rapidly; addition is also observed for cyclohexyllithium[2]. The order of reactivity of the alkyllithiums is secondary > tertiary. No reaction was observed for cyclobutyllithium, cyclopropyllithium, phenyllithium, benzhydryllithium, or triphenylmethyllithium[2]. The kinetics of the ethylene addition reactions exhibits a first-order dependence on alkyllithium concentration, a first-order dependence on ethylene concentration, and a second-order dependence on diethyl ether concentration. The first-order kinetic dependence on alkyllithium concentration established that the rate is controlled by something that happens to the predominant RLi aggregate in solution, that was assumed to be the tetramer (see Chapter 1). Thus, in contrast to many other reactions of alkyllithium compounds[5], it is not the monomeric RLi that is reactive nor is it any other aggregate in equilibrium with the tetramer. It was concluded that the mechanism of the ethylenation reaction is a direct reaction between ethylene and an etherated alkyllithium tetramer, and that dissociation of the molecular aggregate is not the rate-determining step as represented in Scheme 6.1[3]. The structure of the dietherate complex is also shown. In this mechanistic

Scheme 6.1

scheme both the coordinating bases and the other lithium atoms have been omitted from the structures for simplicity. The role of the Lewis base is to coordinate with the lithium atoms, which will tend to increase the ionic character of the carbonlithium bond in the aggregate and activate the carbanion toward the insertion reaction with ethylene (see Chapter 1). It is interesting to note that addition of



tetrahydrofuran initially increases the rate of ethylenation of t-butyllithium, but the rate passes through a maximum and then steadily decreases as the concentration of tetrahydrofuran (THF) increases. This unusual kinetic effect can be rationalized by proposing that a vacant lithium site must be available for coordination with ethylene. In the presence of an excess of tetrahydrofuran, all of the lithium sites are effectively coordinated with THF molecules and the rate of reaction is reduced.

It is important to note that n-butyllithium in the presence of a chelating tertiary amine, such as N,N,N',N'-tetramethylethylenediamine, will polymerize ethylene at approximately 50 atm pressure and 100°C to yield polyethylene of reasonably high molecular weight [6–8].

B. Reaction with 1,1-Diphenylethylene

The kinetics of the addition reaction of alkyllithium compounds with 1,1-diphenylethylene has been extensively investigated (Eq. 6.2) [5]. The addition

(6.2)

reaction with 1,1-diphenylethylene is of interest as a model reaction because the reaction proceeds quantitatively to form the monoaddition product, presumably because of the steric requirements of the diphenylalkyl groups. Another advantage is that the diphenylalkyllithium adduct exhibits a strong ultraviolet (UV)-visible absorbance at λmax = 440 nm (ε = 15,700) [9] that is well separated from the absorptions due to the alkyllithium initiators. The kinetics exhibits a first-order dependence on 1,1-diphenylethylene concentration, but fractional kinetic-order dependencies for the alkyllithium compounds have generally been observed, as shown in Table 6.1. In many cases the fractional kinetic orders correspond approximately to the reciprocal of the degree of association of the organolithium compound. For example, in benzene a one-sixth-order dependence on n-butyllithium concentration has been reported [10] while a one-fourth order dependence has been observed for t-butyllithium [13]. Fractional kinetic-order dependencies for addition reactions of alkyllithium compounds in hydrocarbon solution have most simply been rationalized by proposing that dissociation of the organolithium aggregates to a kinetically active monomeric (unassociated) species is required prior to addition to 1,1-diphenylethylene, as shown in Scheme 6.2. It should be noted, however, that dissociation of hexamers and tetramers to unassociated species probably occurs stepwise via species with intermediate degrees of



Scheme 6.2

Table 6.1 Degrees of Aggregation of Alkyllithiums and Kinetic Reaction Orders for Addition of Alkyllithiums (in excess) with 1, 1-Diphenylethylene

RLi Na Solvent N Reference

n-C4H9Li 6 Benzene 0.18 10

C2H5Li 6 Benzene 0.1 11,12

t-C4H9Li 4 Benzene 0.28 13

ArCH2Lib 2c Toluene 0.5d 14

n-C4H9Li 0.4%Et2O/C6H6 0.50 12

C2H5Li 0.4%Et2O/C6H6 0.25 12

C2H5Li 4%Et2O/C6H6 0.41 12

C6H5CH2Li Et2O 1.2 15

CH3Li 4 Et2O 0.21 15

CH2=CHCH2Li 2 Et2O ˜1.3 15

C6H5Li Et2O 0.51 15

n-C4H9Li 4 Et2O 0.3 15

Vinyllithium THF 0.34 16,17

CH3Li 4 THF 0.27 16

C6H5CH2Li 1 THF 1.1 16

CH2=CHCH2Li THF ˜1 16

C6H5Li 2 THF 0.66 16,18

n-C4H9Li 2.4–2.8e THF ˜:0.4,0.5 12,16

aSee Chapter 1.

bAverage degree of association of p-t-amylbenzyllithium.



association (e.g., via tetramers and dimers); see Eqs. 6.5–6.8. As pointed out by Brown [19,20] and Wakefield [5], this explanation does not take into account the high stability of the aggregates; an alternative interpretation by Wakefield is that the rate-determining step involves coordination of a diphenylethylene molecule to one face of the polyhedral organolithium aggregate. However, the fact that the Hammett rho value for the additions of poly(styryl)lithium to a series of ring-substituted 1,1-diphenylethylenes in benzene is reported to be +1.8 [9] indicates that considerable anionic character is present on the diphenylmethyl carbon in the transition state, at least for this addition reaction (Eq. 6.3).

(6.3)

A further problem that complicates the kinetics of these addition reactions is the fact that cross-association of the diphenylalkyllithium species (dimeric in hydrocarbon solution) with the alkyllithium (tetrameric or hexameric association) would be expected to change the average association state of the initiator as the reaction proceeds. Thus, interpretable kinetic orders may only be available for the initial part of the kinetics; unfortunately, this factor has not always been recognized.

These two model reactions illustrate the extremes in the kinetic order dependencies for kinetics of addition reactions of alkyllithium compounds to carbon-carbon double bonds. As a consequence of the association of alkyllithium compounds in hydrocarbon solution (see Chapter 1), the kinetic order in [RLi]tot is observed to vary from unity to a fractional order, 1/n. It is assumed that kinetic orders between these two extremes could result from cross-association as the reaction proceeds or if there is a competition between different kinetically active species.

II. Styrene and Diene Initiation Kinetics for Alkyllithium Initiators

A. General Aspects

The kinetics of the alkyllithium initiation reactions for styrene and diene polymerization in hydrocarbon solution has been investigated extensively [21–23]; other alkali metal derivatives are not soluble in hydrocarbon media. Although initiation rates are faster in polar solvents such as ethers, decomposition reactions often occur in polar media (see Chapters 5 and 8) and, of course, diene microstructure is quite dependent on solvent (see Chapter 9).



The kinetics of the initiation reaction of n-butyllithium with styrene in benzene exhibits a first-order dependence on styrene concentration and a one-sixth-order dependence on n-butyllithium concentration, as shown in Eq. 6.4 [24].

Ri = kiKd[BuLi]1/6[M] (6.4)

Since n-butyllithium is aggregated predominantly into hexamers in hydrocarbon solution, the fractional kinetic order dependency of the initiation process on the total concentration of initiator was rationalized on the basis that the species that reacts with styrene monomer must be the unassociated form of the initiator and that this unassociated species is formed by the equilibrium dissociation of the hexamer, as shown in Scheme 6.3.

Scheme 6.3

The kinetic order for sec-butyllithium-initiated polymerization of styrene is close to 0.25 in benzene solution; this result is also consistent with reaction of the unassociated alkyllithium form since sec-butyllithium is associated predominantly into tetramers in benzene solution [25]. It should be noted that the actual dissociation process could involve the stepwise dissociation of tetramers to dimers, followed by dissociation of dimers to the unassociated alkyllithium initiating species.

As noted by Bywater and Worsfold [25], the frequent coincidence of the fractional order with the degree of association supports the postulate that the initiating species is a small amount of reactive monomeric alkyllithium in equilibrium with the much larger concentration of the unreactive aggregated species. However, the correctness of this interpretation, (i.e., direct dissociation to monomeric, unassociated species) has been questioned and defended by a number of groups [19,20,26,27]. The basic problem is that all evidence suggests that the experimentally observed energies of activation, (e.g., 18 kcal/mole for n-butyllithium initiation of styrene polymerization [24]) appear to be too low to include the enthalpy of complete dissociation of the aggregates [19], that is estimated to require approximately 108 kcal/mole [28]. An alternative possibility suggested by



theoretical calculations [28] is the incomplete or stepwise dissociation of the aggregate, for example, as shown in Eqs. 6.5–6.7 for hexamers; Eq. 6.8 plus Eq. 6.7 would apply for tetramers.

(6.5)

(6.6)

(6.7)

(6.8)

The use of aliphatic solvents causes profound changes in the observed kinetic behavior for the alkyllithium initiation reaction with styrene, butadiene, and isoprene: the inverse correspondence between reaction order dependence for alkyllithium and degree of organolithium aggregation is not observed [25]. This can be seen by an examination of the results collected in Table 6.2, that show that the use of an aliphatic solvent leads to kinetic orders that are unrelated, at least in a direct fashion, to the aggregation state of the initiating organolithium. Also, initial rates of initiation in aliphatic solvents were found to be several orders of magnitude less than those observed, under equivalent conditions, when the aliphatic solvent was replaced with benzene (see Figure 6.1) [23,25]. The effect of aromatic solvents is consistent with the known ability of these solvents to promote dissociation of organolithium aggregates [29]. Pronounced induction periods are also observed in aliphatic hydrocarbon solvent systems (see Figure 6.1). The rate of appearance of poly(styryl)lithium is a sigmoidal function with time, as illustrated in Figure 6.1 for the system sec-BuLi/styrene/cyclohexane [25]. Various workers have obtained similar data for the systems n-BuLi/butadiene/cyclohexane, n-BuLi/isoprene/cyclohexane, sec-BuLi/isoprene/n-hexane and n-BuLi/styrene/cyclohexane (see Table 6.2) [23]. It is interesting to note that in contrast to the kinetic behavior observed for other alkyllithium initiators, no induction periods are observed using menthyllithium in cyclohexane with either styrene or isoprene; furthermore, the rates of initiation are very similar in cyclohexane and in toluene [38]. It has been reported that menthyllithium is associated primarily into dimers in cyclohexane [39]. Initial rates determined during the slow induction period are reported in the literature by some workers, while other workers have measured the kinetic orders at the period of maximum rate, a region in which the system becomes very complicated as discussed below [23].

Thus, it would appear that in aliphatic vs. aromatic solvents the initiation process involves a different reaction mechanism involving the direct addition of monomer with aggregated organolithium species (Eq. 6.9) to form directly a

(6.9)



Table 6.2 Kinetic Orders for Alkyllithium Initiators in Hydrocarbon Solvents

Solvent

Monomer

Initiator

Degree ofAssociation

Kinetic Methoda

ReactionOrderb

Reference

Benzene Styrene n-C4H9Li 6 UV 0.16 24

Styrene n-C4H9Li 6 UV 0.33 30

Styrene sec-C4H9Li 4 UV 0.25 25

Butadienec n-C4H9Li 6 SEC 0.3–1.0 23

Butadiene sec-C4H9Li 4 IR 0.9 31

Isoprene sec-C4H9Li 4 UV 0.25 25

Cyclohexane Styrene n-C4H9Li 6 UV 0.5–1.0 32

Styrene n-C4H9Li 6 SEC 1.0 33

Styrene sec-C4H9Li 4 UV 1.4 25

Styrene sec-C4H9Li 4 SEC 1.0 33

Styrene t-C4H9Li 4 SEC 1.0 33

Styrene i-C4H9Li 4 SEC 1.0 33

Butadiene n-C4H9Li 6 UV 0.5–1.0 32

Butadiene sec-C4H9Li 4 SEC 1.0 33

Butadiene t-C4H9Li 4 SEC 1.0 33

Butadiene n-C4H9Li 6 SEC 1.0 33

Isoprene n-C4H9Li 6 UV 0.5–1.0 34

Isoprene n-C4H9Li 6 SEC 1.0 33



Isoprene sec-C4H9LI 4 UV 0.66 36,37

Isoprene sec-C4H9Li 4 SEC 1.0 33

Isoprene t-C4H9Li 4 UV 0.2–0.7 36,37

aMethod used to follow kinetics: UV, ultraviolet-visible spectroscopy; SEC, size exclusion chromatography; IR, infrared spectroscopy.

bKinetic order dependence of initiation rate on [RLi].



Figure 6.1 Typical curves for the appearance of the uv absorption

of poly(styryl)-lithium. A. Reaction of 1.09 × 10-3

M sec-butyllithium with 5.33 × 10-4 M styrene in benzene solution at 30°C.

B. Reaction of 1.34 × 10-3 M sec-butylithium with 8.67 × 10-2

M styrene in cyclohexane solution at 40°C. (From Ref. 25; reprinted by permission of Elsevier Science Ltd.)

prenyl)lithium in hexane, it was observed that the initiation step started with no indication of a slow initial rate (i.e., no induction period) and with enhanced rates relative to comparable normal initiation reactions [35]. It was estimated that intermolecular exchange was rapid and complete within a time scale of 1 min [35].

The effect of cross-association provides at least a partial explanation for the discrepancies reported in the literature for the kinetic orders on alkyllithium initiator concentration [22,23,26,35]; thus, only in the initial stages is it likely that a detailed interpretation of the mechanism is possible. Otherwise, the formation of time-dependent concentrations of a series of mixed aggregates (PLi)m

(BuLi)n will be formed, and each cross-associated species may react, with its own rate constants and kinetic order dependence. Thus, reaction orders determined at maximum rates of initiation would be expected to represent a composite of the reaction order for all of the mixed aggregated species present during the initiation reaction.

B. Solvent Effects

The rates of alkyllithium initiation reactions with monomers are faster in aromatic than aliphatic solvents. Thus, Roovers and Bywater [35] have reported that the rate of sec-butyllithium-initiated polymerization of isoprene is 2000 times faster



in benzene than hexane at a concentration of 10-3M. It is noteworthy that the rates of intermolecular exchange between alkyllithium compounds are reported to be as much as 103–104 times faster in toluene than in cyclopentane [44]. Benzene also decreases the average degree of aggregation of some alkyllithium compounds; thus, while trimethylsilylmethyllithium is associated predominantly into hexamers in cyclohexane, the average degree of association in benzene is four (< 0.06 m) [45]. As discussed in the previous section, the kinetic order dependence is generally first order in [alkyllithium] in aliphatic solvents, but fractional order dependencies are observed in aromatic solvents.

C. Lewis Base Effects

Lewis bases and alkali metal alkoxides have been used as additives to modify the initiation reaction with alkyllithium compounds. Worsfold and Bywater [46–48] have reported that in the presence of THF the initiation reaction of styrene with sec-butyllithium is first order in alkyllithium.

It has been discovered in a number of studies that Lewis bases such as ethers and amines, when present in amounts comparable to the initiator concentration, dramatically increase the relative rate of initiation of styrene and diene polymerizations relative to propagation [46–50]. For example, for styrene polymerization in benzene with n-butyllithium as initiator in the presence of 0.15M THF, the initiation step is completed virtually instantaneously on mixing of the reagents [46]. The propagation rate is also increased, but far less markedly [46]. Using small amounts of THF([THF]/[Li] = 2–30), narrow molecular weight distribution polystyrenes (Mw/Mn < 1.1) have been prepared even with n-butyllithium in benzene [49,50]. Similar increases in the relative rates of initiation have been observed for the n-butyllithium-initiated polymerization of isoprene in cyclohexane; using [THF]/[Li] = 10, the initiation is complete in 15 min versus hours in pure cyclohexane at 30°C [34].

D. Salt Effects

It is very important to consider the effect of lithium alkoxides on alkyllithium-initiated polymerizations because these salts are ubiquitously present to some extent as impurities formed by the reactions with oxygen [51,52] (Eq. 6.10) and hydroxylic impurities [53,54] (Eq. 6.11) [5,55]. In fact, it is common practice to

(6.10)

(6.11)

utilize excess butyllithium (i.e., more than the stoichiometric amount required to generate the required molecular weight; see Chapter 4) to scavenge impurities in the solvent and monomer feed [56].



The effects of lithium alkoxides on the rates of alkyllithium-initiation reactions depend on the solvent, the monomer, the alkoxide structure, the alkyllithium initiator, and the ratio of [RLi]/[LiOR'] [22]. For n-butyllithium initiation of styrene in cyclohexane (see Figure 6.2), the rate of initiation is increased at low relative concentrations of added lithium alkoxide, reaching a maximum rate at [BuLi]/[t-BuOLi] = 1/0.3–1/0.5 [57]. At a ratio of 1/1, the rate is essentially the same as the control without alkoxide; beyond this ratio, the rate decreased continuously with increasing relative concentration of lithium alkoxide. When the solvent is toluene or benzene, no initial increase in rate is observed; the initiation rate decreases with increasing relative concentrations of lithium alkoxide (see Figure 6.3) [57,58]. For example, a 10-fold excess of t-BuOLi decreased the intiation rate by a factor of 100 [58]. At high molar ratios of [lithium alkoxide] to [BuLi], the kinetic order dependence of the initiation rate on [BuLi] becomes one-half compared with the one-sixth order observed in the absence of alkoxide (see Table 6.2) [58].

In contrast to the rate-depressing effects observed for styrene, lithium alkoxides generally accelerate the rate of initiation by alkyllithiums (n-butyllithium and sec-butyllithium) for isoprene in hexane [35,37]. As a consequence of this effect, the induction time is reduced; in fact, at a ratio of [t-BuOLi]/[sec-BuLi] = 1.8, no induction period is observed (see Figure 6.4) [35]. It would be expected that these enhanced rates of initation would result in narrower molecular weight distributions (see Chapter 4) [22]. The exception is t-butyllithium for which results are

Figure 6.2 Rate of initiation Ri of styrene (0.86M) with n-BuLi

(2.7 × 10-3M) and variable t-BuOLi in cyclohexane at 30°C. (From Ref. 57; reprinted by permission of John Wiley & Sons, Inc.)



Figure 6.3 Rate of initiation Ri of styrene (0.44M) with n-BuLi

(2.7 × 10-3M) and variable t-BuOLi in toluene at 30°C. (From Ref. 57; reprinted by permission of John Wiley & Sons, Inc.)

Figure 6.4 Influence of added t-BuOLi on the rate of initiation of isoprene

in hexane at 30°C: [isoprene] = 5 × 10-2M; [sec-BuLi]o = (A) 1.1 × 10-3M, (B)

1.46 × 10-3M, (C) 1.46 × 10-3M; [t-BuOLi]/[sec-BuLi] = (A) 0.0, (B) 0.63, (C) 1.80.

(From Ref. 35; reprinted by permission of the American Chemical Society.)



more complex; although lithium alkoxides increase the initial rate, this is followed by rate decreases that generally result in longer times being required for complete initiation [37]. Lithium hydroxide formed by reaction with water was reported to decrease dramatically the rates of initiation for sec-butyllithium with isoprene in cyclohexane [36].

Aside from the empirical rate effects described above, it is difficult to develop a detailed understanding of the mechanism for the observed effects of added lithium alkoxides on alkyllithium initiation reactions. In this case, cross-association would involve three different organolithium species: the alkyllithium initiator, the polymeric organolithium compound formed by initiation, and the lithium alkoxide. Lithium alkoxides are associated in solution analogous to alkyllithium compounds (see Chapter 1). Thus, the degree of association of lithium t-butoxide is reported to be 5.8, 6.2, 4.1, and 4.0 in cyclohexane, benzene, tetrahydrofuran, and pyridine, respectively [59]. These results suggest that lithium alkoxides are more strongly associated than the corresponding alkyllithiums, since the high degree of association is maintained even in polar media such as pyridine.

The effect of lithium ethoxide on the association behavior of ethyllithium has been examined by freezing point depression measurements in cyclohexane [60]. Evidence for the following types of equilibria was obtained (Eqs. 6.12, 6.13).

(6.12)

(6.13)

Thus, it was concluded that lithium ethoxide coordinates to the intact hexameric aggregate, presumably by interacting with one of the empty faces of the octahedral structure of the hexamer. This is in contrast to the behavior of mixtures of organolithium compounds in which statistical mixtures of cross-associated species are obtained [61,62]. At a higher concentration of lithium alkoxides it was concluded that clusters of lithium alkoxide interact with the intact hexamer [60]. Thus, lithium alkoxides behave like Lewis bases in their interaction with hexameric ethyllithium. The interaction of lithium alkoxides with polymeric organolithiums will be discussed in Chapter 7.

III. Mixed Organometallic Initiators

Alkyllithium compounds react with organometallic compounds of different metals, most notably those of groups I, II and III, to form mixed organometallic compounds [63–65]. Wittig originally referred to these species as “ate” complexes and formulated their structures as ionic as represented schematically in Eq. 6.14, where Met represents a metal center [66,67]. X-ray [63,68,69] and NMR

(6.14)



[61,70] investigations indicate more complex, stoichiometry-dependent types of structures involving bridging alkyl groups between the two types of metal centers for aluminum, magnesium, and zinc systems as represented in Scheme 6.4.

Scheme 6.4 Structures of mixed organometallic complexes

The complexes of alkyllithiums and diethylzinc were reported to increase the rate of initiation for polymerization of butadiene and styrene in cyclohexane without changing the stoichiometric molecular weight based on [RLi] [71]. For example, t-butyllithium is reported to be an exceptionally slow initiator for styrene in cyclohexane, which results in polystyrenes with broad molecular weight distributions [32,72–74] (see however, ref. [75] for different results for t-butyllithium). However, styrene initiated with t-butyllithium complexed with diethylzinc in cyclohexane initiated almost immediately, whereas the control without diethyl-zinc took a long time to develop the characteristic yellow color of poly(styryl)-lithium (Figure 6.5). A corresponding decrease in molecular weight distribution from Mw/Mn = 2.07 (control) to Mw/Mn = 1.22 (1/1 Et2Zn complex) occurred, as shown in Figure 6.6. It was also observed that when diethylzine was added to an active polymer solution of α,ω-dilithiumpolybutadiene, the viscosity of the solution was observed to decrease with increasing amounts of diethylzinc. It is interesting to note that diethylzinc did not alter the propagation rate and no chain transfer was observed in cyclohexane; however, evidence for chain transfer was observed (i.e., decreasing molecular weight) in the presence of THF.

Dialkylmagnesium compounds are not active initiators for diene and styrene polymerization [76], but they participate in polymerization in polar media [77–79] or when complexed with an alkyllithium [80] or alkylsodium initiator [81]. This lack of initiator reactivity in hydrocarbarbon media is exploited advantageously by utilizing dialkylmagesium compounds to purify hydrocarbon monomers [74].



Figure 6.5 Effect of added diethylzinc on the rate of polymerization

of styrene. Styrene, 100 g; cyclohexane, 1 L; tert-BuLi, 1.8 mmole; 30°C. (From Ref. 71;

reprinted by permission of John Wiley & Sons, Inc.)

The addition of increasing amounts of dibutylmagnesium [(n-Bu)(sec-Bu)Mg] to a constant amount of sec-butyllithium in cyclohexane reduced the rate of styrene or butadiene polymerization and decreased the molecular weight without significant broadening of the molecular weight distribution or changing the polybutadiene microstructure [80]. Based on the observed molecular weights, it was estimated that 55–70% of the butyl groups in dibutylmagnesium participate in initiating polymerization; in the presence of 1 g THF per 100 g of monomer, 75–90% of the butyl groups were active. The narrow molecular weight distributions obtained are consistent with rapid intermolecular exchange of active alkyl/polymer groups between the mixed organometallic complexes, and also rates of initiation that are rapid or competitive with propagation (see Chapter 4).

The 1:1 complex of n-butylsodium and dibutylmagnesium [(n-Bu)(sec-Bu)Mg] initiates polymerization of styrene in benzene solution to produce poly-styrene with narrow molecular weight distribution and with molecular weights that correspond to only initiation by the complexed butylsodium moiety and not



Figure 6.6 SEC curves of polystyrene samples initiated by

tert-butyllithium with and without added diethylzinc. (From Ref. 71; reprinted by

permission of John Wiley & Sons, Inc.)

by the two butyl groups attributed to the dibutylmagnesium [81]. The initiator was less than ideal forthe polymerization of isoprene since polymers with broad molecular weight distributions (Mw/Mn = 1.26–1.79), high 3,4-microstructure (60%), and molecular weights lower than predicted based on butylsodium were obtained. It was tentatively concluded that both chain transfer and alkyl exchange with dibutylmagnesium groups were occurring.

IV. Initiator Reactivity

The rates of initiation for alkyllithium initiators are primarily dependent on the structure of the alkyllithium initiator and the solvent. The relative reactivities of alkyllithiums as polymerization initiators [22,23,32,37,38] are intimately linked to their degree of association as shown below with the average degree of



association in hydrocarbon solution, where known, indicated in parentheses (see Table 1.1):

Styrene Polymerization: menthyllithium (2) > sec-BuLi (4) > i-PrLi (4–6) >

i-BuLi > n-BuLi (6) > t-BuLi (4)

Diene Polymerization: menthyllithium (2) > sec-BuLi (4) > i-PrLi (4–6) >

t-BuLi (4) > i-BuLi > n-BuLi (6)

It is clear that, in general, the less associated alkyllithiums are more reactive as initiators than the more highly associated species. There is general agreement on these reactivity orders, except for the case of styrene initiation by t-butyllithium. Both Hsieh [32] and Morton and Fetters [73,74] have observed that t-butyllithium is a slow initiator, less reactive than n-butyllithium, for styrene in both cyclohexane and in benzene. Although an initial relatively rapid reaction was observed to occur between styrene and a fraction of the t-butyllithium, Morton and Fetters [74] provided evidence that some of the initiator was apparently cross-associated into a dormant state since a bimodal distribution resulted from addition of a second increment of styrene along with 1 vol % tetrahydrofuran. In contrast, Roovers and Bywater [75] have reported that freshly prepared t-butyllithium is almost as reactive as sec-butyllithium, for styrene polymerization in cyclohexane. They reported that the commercially available product contained an unidentified impurity (4–5%) that could not be removed by the normal short-path, vacuum sublimation employed by Morton and Fetters [73]. From a practical standpoint, the commercially available t-butyllithium is observed to be a very unreactive initiator for styrene polymerizaton in hydrocarbon media and its use is not recommended for this purpose.

The effect of solvent on initiator reactivity is also consistent with the importance of association phenomena. Aromatic solvents, that tend to decrease the average degree of association and promote dissociation processes of aggregates, are reported to lead to initiation rates 102–103 times faster than in aliphatic solvents [23,25]. In addition, different mechanisms of initiation are indicated by the different kinetic orders in alkyllithium initiator for aliphatic vs. aromatic solvents. Fractional kinetic orders suggesting initiation via a dissociated alkyllithium are observed in aromatic solvents; kinetic orders close to unity are observed in aliphatic solvents, suggesting direct reaction of the aggregated species with the monomer. These generalities are observed regardless of the monomer. However, it should be noted that exceptions have been observed. For example, Hsieh [32] reported first-order dependencies on [RLi] for initiation reactions in toluene. Roovers and Bywater [75] reported that the initiation rates for styrene with tert-butyllithium exhibited a first-order dependence on [t-BuLi] in benzene.

As a consequence of the low initiator reactivity of hexameric n-butyllithium, it is observed that depending on the relative concentration of monomer and initiator and the solvent, the monomer can be completely polymerized before all



of the initiator has reacted. This is illustrated in Figures 6.7–6.9 for butadiene, isoprene, and styrene, respectively, in cyclohexane at 50°C [32]. Since aromatic solvent accelerates the rate of initiation more than the rate of propagation [22], the amount of residual initiator is decreased in aromatic relative to aliphatic solvents as shown in Figure 6.10 compared to Figures 6.7 and 6.9 [32]. Using more reactive initiators likewise also eliminates the problem of slow initiation and incomplete consumption of monomer as shown in Figures 6.11 and 6.12, which also illustrate the relative reactivities of these initiators with diene and styrene monomers. The relative initiator reactivity of isomeric butyllithiums and the effect of temperature are illustrated in Table 6.3.

V. Summary

The rates of initiation of hydrocarbon monomers in hydrocarbon solvents using alkyllithium initiators depend on the monomer, the initiator, the solvent, and the temperature. The relative reactivity of initiators is generally inversely related to the degree of aggregation of the alkyllithium compound. For dienes with butyl-lithium initiators in hydrocarbon solution, the order is sec > tert > iso > normal.

Figure 6.7 Relationships between BuLi remaining and conversion

of polymerization of butadiene (1.7 mole/L) in cyclohexane at 50°C at various initial n-BuLi

concentrations: (A) 26 × 10-3 mole/L; (B) 8.7 × 10-3 mole/L; (C) 2.6 × 10-3 mole/L; (D)

0.9 × 10-3 mole/ L. (From Ref. 32; reprinted by permission of John Wiley & Sons, Inc.)




of polymerization of isoprene (1.3 mole/L) in cyclohexane at 50°C at various initial n-BuLi concentrations: (A) 26 × 10-3 mole/L; (B) 8.7 × 10-3 mole/L; (C) 2.6 × 10-3 mole/L;

(D) 0.9 × 10-3 mole/L; (From Ref. 32; reprinted by permission of John Wiley & Sons, Inc.)


of polymerization of styrene (1.0 mole/L) in cyclohexane at 50°C at various initial

n-BuLi concentrations: (A) 26 × 10-3 mole/L; (B) 8.7 × 10-3 mole/L; (C) 2.6 × 10-3 mole/L;

(D) 0.9 × 10-3 mole/L; (From Ref. 32; reprinted by permission of John Wiley & Sons, Inc.)



Figure 6.10 Effect of solvent on the rates of initiation by n-butyllithium at 50°C;

cyclohexane; n-hexane; toluene. (From Ref. 32; reprinted by permission of John Wiley & Sons, Inc.)

Figure 6.11 Effect of BuLi structure on the rate of initiation of butadiene and isoprene

in cyclohexane: n-BuLi; tert-BuLi; sec-BuLi. (From Ref. 32; reprinted by permission of John Wiley & Sons, Inc.)



Figure 6.12. Effect of BuLi structure on the rate of initiation of styrene

in cyclohexane at 50°C: n-BuLi;

tert-BuLi; sec-BuLi. (From Ref. 32; reprinted by permission of John Wiley & Sons, Inc.)

Table 6.3 Effect of Butyllithium Structure and Temperature on the Rates of Initiation in Cyclohexanea

Rate of Initiation (mole/L/min × 103)

Monomer

M (mole/L)

Temp. (°C)

n-BuLi

sec-BuLi

t-BuLi

i-BuLi

Butadiene 1.7 50 0.08 4.0 0.7 –

30 0.02 0.8 0.13 –

Isoprene 1.3 50 0.05 1.5 0.9 0.12

30 0.01 0.3 0.2 –

Styrene 1.0 50 0.09 5.2 0.04 0.19

30 0.02 1.2 – –

a[BuLi] = 2.6 × 10-3 mole/L.

Source: Ref. 32.



For styrene with butyllithium initiators in hydrocarbon solution, the order is sec > iso > normal > tert. The monomer reactivity order is generally styrene > butadiene > isoprene [32]; however, this relative order (styrene > dienes) depends on chain end concentration (see Chapter 7). The sensitivity to variation of the electronic nature of substituents in the aromatic ring for styrenes for n-butyllithium initiation in benzene has been determined using the Hammett linear free energy relationship [82]; the rho value (ρ) was determined to be +1.0. This relatively small value suggests that there is relatively little charge delocalization into the aromatic ring in the transition state for initiation. For any particular monomer-initiator combination, the initiation rate order is aromatic >> n-hexane > cyclohexane. Rates of initiation are generally accelerated by the addition of Lewis bases that facilitate dissociation of alkyllithium aggregates (see Chapter 1). Initiation rates for styrene are generally depressed by the addition of lithium alkoxides; however, lithium alkoxides accelerate the rate of initiation by alkyllithiums for isoprene in hexane, presumably because cross-association facilitates dissociation of the initiator aggregates.

References

1. P. D. Bartlett, S. Friedman, and M. Stiles, J. Am. Chem. Soc., 75, 1771 (1953).

2. P. D. Bartlett, S. J. Tauber, and W. P. Weber, J. Am. Chem. Soc., 91, 6362 (1969).

3. P. D. Bartlett, C. V. Goebel, and W. P. Weber, J. Am. Chem. Soc., 91, 7425 (1969).

4. W. K. Busfield, in Polymer Handbook, 3rd ed., J. Brandrup and E. H. Immergut, Eds., John Wiley, New York, 1989, p. II/297.

5. B. J. Wakefield, The Chemistry of Organolithium Compounds, Pergamon Press, Oxford, 1974.

6. W. B. Marshall, J. L. Brewbaker, and M. S. Delaney, J. Appl. Polym. Sci., 42, 533 (1991).

7. E. G. Eberhardt and W. R. Davis, J. Polym. Sci., Part A, 3, 3753 (1965).

8. A. W. Langer, Jr., Trans. N. Y. Acad. Sci., Ser., II, 27, 741 (1965).

9. R. Busson and M. Van Beylen, Macromolecules, 10, 1320 (1977).

10. A. G. Evans and D. B. George, J. Chem. Soc., 4653 (1961).

11. A. G. Evans, C. R. Gore, and N. H. Rees, J. Chem. Soc., 5110 (1965).

12. J. G. Carpenter, A. G. Evans, C. R. Gore, and N. H. Rees, J. Chem. Soc., (B), 908 (1969).

13. R. A. H. Casling, A. G. Evans, and N. H. Rees, J. Chem. Soc., B, 519 (1966).

14. R. Waack, M. A. Doran, and A. L. Gatzke, J. Organometal. Chem., 46, 1 (1972).

15. R. Waack and M. A. Doran, J. Organometal. Chem., 29, 329 (1971).

16. R. Waack and M. A. Doran, J. Am. Chem. Soc., 91, 2456 (1969).

17. R. Waack and P. E. Stevenson, J. Am. Chem. Soc., 87, 1183 (1965).

18. R. Waack, P. West, and M. A. Doran, Chem. Ind., 1035 (1966).

19. T. L. Brown, J. Organometal. Chem., 5, 191 (1966).

20. T. L. Brown, Adv. Organometal. Chem., 3, 365 (1965).



21. R. C. P. Cubbon and D. Margerison, in Progress in Reaction Kinetics, Vol. 3, G. Porter, Ed., Pergamon Press, New York, 1965, p. 405.

22. S. Bywater, in Comprehensive Chemical Kinetics, Vol. 15, Non-Radical Polymerization, C. H. Bamford and C. F. H. Tipper, Eds., Elsevier, New York, 1976, p. 1.

23. H. L. Hsieh and W. H. Glaze, Rubber Chem. Technol., 43, 22 (1970).

24. D. J. Worsfold and S. Bywater, Can. J. Chem., 38, 1891 (1960).

25. S. Bywater and D. J. Worsfold, J. Organometal. Chem., 10, 1 (1967).


27. M. Van Beylen, S. Bywater, G. Smets, M. Szwarc, and D. J. Worsfold, Adv. Polym. Sci., 86, 87 (1988).

28. G. Graham, S. Richtsmeier, and D. A. Dixon, J. Am. Chem. Soc., 102, 5759 (1980).

29. M. Y. Darensbourg, B. Y. Kimura, G. E. Hartwell, and T. L. Brown, J. Am. Chem. Soc., 92, 1236 (1970).

30. K. F. O'Driscoll, E. N. Ricchezza, and J. E. Clark, J. Polym. Sci., A3, 3241 (1965).

31. V. Zgonnik, E. Schadrina, K. Kalninsh, and B. Erussalimsky, Makromol. Chem., 174, 81 (1973).

32. H. Hsieh, J. Polym. Sci., A3, 163 (1965).

33. A. F. Johnson and D. J. Worsfold, J. Polym. Sci., Part A-3, 449 (1965).


35. J. E. L. Roovers and S. Bywater, Macromolecules, 1, 328 (1968).

36. A. Guyot and J. Vialle, J. Polym. Sci., Part B, 6, 403 (1968).

37. A. Guyot and J. Vialle, J. Macromol. Sci., Chem., A4, 79 (1970).

38. C. M. Selman and H. L. Hsieh, J. Polym. Sci., Part B, 9, 219 (1971).

39. W. H. Glaze and C. H. Freeman, J. Am. Chem. Soc., 91, 7198 (1969).


41. M. Morton, E. E. Bostick, and R. A. Livigni, Rubber Plast. Age, 42, 397 (1961).

42. F. Schue and S. Bywater, Macromolecules, 2, 458 (1969).

43. M. Morton, R. A. Pett, and L. J. Fetters, Macromolecules, 3, 333 (1970).

44. G. E. Hartwell and T. L. Brown, J. Am. Chem. Soc., 88, 4625 (1966).


46. S. Bywater and D. J. Worsfold, Can. J. Chem., 40, 1564 (1962).



58. J. E. L. Roovers and S. Bywater, Trans. Faraday Soc., 62, 1876 (1966).


60. T. L. Brown, J. A. Ladd, and G. N. Newman, J. Organometal. Chem., 3, 1 (1965).

61. T. L. Brown, Accounts. Chem. Res., 1, 23 (1968).


63. P. G. Williard, in Comprehensive Organic Synthesis, B. M. Trost, Ed., Pergamon, Oxford, UK, 1991, Vol. 1, Part 1.

64. W. E. Lindsell, in Comprehensive Organometallic Chemistry. The Synthesis, Reactions and Structures of Organometallic Compounds, G. Wilkinson, R. A. Stone, and E. W. Abel, Eds., Pergamon, New York, 1982, Vol. 1, p. 155.

65. J. Boersma, in Comprehensive Organometallic Chemistry. The Synthesis, Reactions and Structures of Organometallic Compounds, G. Wilkinson, R. A. Stone, and E. W. Abel, Eds., Pergamon, New York, 1982, Vol. 2, p. 823.

66. G. Wittig, F. J. Meyer, and G. Lange, Liebigs Ann. Chem., 571, 167 (1951).

67. G. Wittig, Proc. Robert A. Welch Found. Conf. Chem. Res., IX, 13 (1966).

68. B. Schubert and E. Weiss, Chem. Ber., 117, 366 (1984).

69. N. H. Buttrus, C. Eaborn, M. N. A. El-Kheli, P. B. Hitchcock, J. D. Smith, A. C. Sullivan, and K. Tavakkoli, J. Chem. Soc., Dalton Trans., 381 (1988).

70. J. P. Oliver, Adv. Organomet. Chem., 8, 167 (1970).

71. H. L. Hsieh, J. Polym. Sci., Polym. Chem. Ed., 14, 379 (1976).

72. H. L. Hsieh and O. F. McKinney, J. Poly. Sci., B4, 843 (1966).


74. M. Morton and L. J. Fetters, Rubber Chem. Tech., 48, 359 (1975).


76. D. B. Patterson and A. F. Halasa, Macromolecules, 24, 1583 (1991).

77. T. Narita, T. Yasumura, and T. Tsuruta, Polym. J., 4, 421 (1973).

78. T. Narita, A. Masaki, and T. Tsuruta, J. Macromol. Sci., Chem., A4, 277 (1970).

79. T. Narita and T. Tsuruta, J. Organometal. Chem., 30, 289 (1971).

80. H. L. Hsieh and I. W. Wang, Macromolecules, 19, 299 (1986).

81. M. Liu, C. Kamienski, M. Morton, and L. J. Fetters, J. Macromol. Sci. Chem., A23, 1387 (1986).

82. H. L. Hsieh, J. Polym. Sci., A3, 173 (1965).



7 Propagation Reactions: Kinetics and Mechanism for Styrenes and Dienes in Hydrocarbon Solvents with Lithium as Counterion

I. Introduction

The kinetics of propagation for styrene and diene monomers in hydrocarbon solvents with lithium as the counterion is complicated by the association of the chain end active centers [1–3], analogous to the kinetic effects of initiator association discussed in Chapter 6. However, the kinetics of propagation can be investigated independently of the initiation event and this simplifies the analysis relative to the complexities involved in delineating the kinetics of the initiation reactions using alkyllithium initiators, in which initiation and propagation occur simultaneously (see Chapter 6). Obviously, however, it is essential to ensure that the initiator has been completely consumed before propagation kinetics is investigated. The complete consumption of the initiator requires rather large amounts of monomer for less efficient alkyllithium initiators in hexane or cyclohexane.

In one of the first kinetic studies of n-butyllithium-initiated polymerization of styrene in benzene, Worsfold and Bywater [4] noted that certain minimum concentrations of monomer were required to effect complete consumption of the initiator. This is clearly illustrated in Figure 7.1, which shows that initiation is occurring continuously during the polymerization in benzene at 30°C for the concentrations indicated. Analysis of their kinetic data indicated that the initial monomer concentration must be 1.08 M to ensure complete consumption of the



YFP

Highlight

YFP

Underline

YFP

Underline

YFP

Underline

YFP

Underline

YFP

Underline

YFP

Highlight

YFP

Highlight

Figure 7.1 Typical reaction curves for initiation of styrene

polymerization with n-butyllithium in benzene at 30.3°C. Initially, [styrene] = 1.39 × 10-2

moles/L; [butyllithium] = 1.10 × 10-3 moles/L. A. Disappearance of styrene measured

at 291 µm. B. Appearance of polystyryl anion at 335 µm. (From Ref. 4; reprinted by permission of the National Research Council of Canada.)

initiator when [BuLi] = 0.01 M. The rate of initiation relative to propagation also depends on the solvent, the temperature and the structure of the initiator.

Hsieh [5] analyzed the consumption of initiator using gas chromatography to measure the amount of butane gas generated after hydrolysis of aliquots during polymerization using approximately molar concentrations of monomer at 50°C and the data are shown in Figures 6.7–6.12. These results clearly showed that for n-butyllithium in cyclohexane at concentrations higher than millimolar, residual initiator is present when all of the monomer (butadiene, isoprene, or styrene) has been consumed. As illustrated in Figure 6.11 and 6.12, rapid consumption of initiator occurs when either t-butyllithium or sec-butyllithium is used as initiators for dienes or when sec-butyllithium is used for styrene.

In early studies of the kinetics of alkyllithium-initiated polymerizations, “rates of polymerization” were often erroneously described as rates of propagation [2]. Thus, there was often disagreement between the results obtained by different investigators because the reported rates of polymerization included both initiation and propagation reactions. Artifacts of these misconceptions included reports



YFP

Highlight

YFP

Line

YFP

Polygonal Line

of rates of polymerization that were second order in monomer and independent of initiator concentration. Hsieh and Glaze[2] have reviewed this subject in detail.

II. Kinetics of Propagation

Unlike the situation encountered for the kinetics of initiation for styrene and dienes, the reaction order dependence of the propagation rate on active center concentration is independent of the identity of the hydrocarbon solvent, aromatic or aliphatic, although the relative propagation rates, under equivalent conditions, are faster in benzene than in an aliphatic solvent. The complication in propagation kinetics is associated with a controversy that exists regarding the relationship between the degree of association and the reaction order dependence on active center concentration. The problem is that there is disagreement with respect to the predominant degree of association of poly(dienyl)lithium chain ends in hydrocarbon solution[6–10] (see Chapter 1).

A. Styrene Monomer

The anionic propagation kinetics for styrene (S) polymerization with lithium as counterion are relatively unambiguous. The reaction order in monomer concentration is first order, as it is for polymerization of all styrene and diene monomers in heptane, cyclohexane, benzene, and toluene[1–3]. The reaction order dependence on total chain end concentration, [PSLi]o, is one-half as shown in Equation 7.1[4].

Rp = -d[S]/dt = kobs[PSLi]o 1/2[S] (7.1)

Representative kinetic order dependencies on chain end concentration for alkyllithium-initiated polymerization of styrene are listed in Table 7.1. The observed one-half kinetic order dependence on chain end concentration (see Table 7.1) can be explained in terms of the observation that poly(styryl)lithium is predominantly associated into dimers in hydrocarbon solution (see Chapter 1). Thus, if the unassociated poly(styryl)lithium is the reactive entity for monomer addition, a simple dissociative mechanism can be invoked (see Scheme 7.1). This mechanism

Scheme 7.1

leads to the kinetic equation shown in Equation 7.2. From Equations 7.1 and 7.2, it can be seen that the observed rate constant for propagation, kobs, is actually a

Rp = -d[S]/dt = kp[PSLi] [S]

= kp(Kd/2)1/2[PSLi]o 1/2[S]

= kobs[PSLi]o 1/2[S] (7.2)



Table 7.1 Kinetic Order Dependencies on Chain End Concentration for Alkyllithium-Initiated Polymerization of Styrene, Isoprene, and Butadiene in Hydrocarbon Solvents

Solvent

Initiator

[Initiator] (mol/L)

Kinetic OrderDependenceon [Initiator]

Reference

Styrene

Benzene EtLi (1–50) × 10-4 0.5 11

n-BuLi (8–70) × 10-3 0.5 12

1.6 × 10-5–3.9 × 10-2 0.5 4

Toluene sec-BuLi (3–40) × 10-4 0.5 13

Cyclohexane n-BuLi (5–100) × 10-5 0.5 14

sec-BuLi (7–70) × 10-4 0.5 13

Isoprene

n-Hexane n-BuLi (2.4–57) × 10-4 0.5 11,15,16

(2–140) × 10-5 0.5 17

(5–100) × 10-4 0.25 18

t-BuLi (1–10) × 10-3 0.5 19

n-Heptane n-BuLi 10-6–10-2 0.17–0.5 20–26

Cyclohexane EtLi (1–100) × 10-4 0.25 18

(0.6–4.1) × 10-3 0.5 27

n-BuLi (2.6–6.5) × 10-3 0.17 28

sec-BuLi (2.6–6.5) × 10-3 0.17 28

(1.5–6.0) × 10-2 0.5 13

(0.7–15) × 10-3 0.33 13

(0.06–16) × 10-3 0.25 29

t-BuLi (2.6–6.5) × 10-3 0.17 28

Benzene EtLi (5–350) × 10-4 0.25 18

n-BuLi (3–35) × 10-3 0.21–0.4 30

sec-BuLi (0.7–10) × 10-3 0.5 31

1,1-diphenylhexyllithium

(0.1–5) × 10-3 0.25 32

Rli 0.5 33



composite of the propagation rate constant, kp, and the equilibrium constant for dissociation of the dimeric aggregates, Kd, raised to the one-half power. Since the measurement of the dissociation constants of the aggregates is tenuous because of the low concentration of the unassociated species, it is generally difficult to obtain directly a value for the propagation rate constant, kp [7].

One of the interesting aspects of propagation kinetics for styrenes and dienes is that although simple organoalkali initiators of metals other than lithium are not soluble in hydrocarbon media, it is possible to prepare living polymers of these alkali metals, using, for example, alkali metal films, and to study their propagation kinetics [34,35]. A comparison of the observed propagation rate constants for styrene is shown in Table 7.2. Poly(styryl)sodium was presumably associated into dimers since kinetic orders of one-half were observed for the rate dependence on the active chain end concentration. Poly(styryl)potassium exhibits intermediate behavior; one-half order dependence at higher concentrations, but first-order dependence at low concentrations. Poly(styryl)rubidium and poly(styryl)cesium exhibit first-order dependencies and it was concluded that they are unassociated in cyclohexane. The strong counterion dependence (K+ > Na+ > Li+) for the complex observed rate constants (kobs) presumably reflects both the fact that the dissociation constant for the dimers increases with increasing cation size [1] and also the fact that the requisite energy associated with charge separation in the transition state would be less for the larger counterions (see Chapter 3).

B. Diene Monomers

Investigations of the kinetics of propagation for dienes have shown that although the rates exhibit first-order dependence on monomer concentration, fractional order dependencies are generally observed for the concentration of active centers

Table 7.2 Kinetic Parameters for Styrene Propagation in Hydrocarbon Solvents

Counterion

Solvent

Temperature(°C)

Kp(K/n) 1/n

(kobs)

kpa

Reference

Lithium Benzene 30 1.55 × 10-2 4,36

Lithium Cyclohexane 40 2.4 × 10-2 37

Sodium Benzene 30 17 × 10-2 34

Potassium Benzene 30 180 × 10-2 34

Potassium Cyclohexane 40 30 35

Rubidium Cyclohexane 40 22.5 35

Cesium Cyclohexane 40 19 35

aPropagation rate constant presumably for the unassociated species; first-order dependence on active chain end concentration observed.



as shown in Table 7.1. However, it is obvious from Table 7.1 that contradictory results have often been reported. Thus, although earlier kinetic studies reported one-half-order dependencies on initiator concentration, it has been noted by Morton [38] that when greater precautions were taken to eliminate impurities, isoprene exhibited one-fourth-order kinetics and butadiene exhibited one-fourth- or one-sixth-order dependencies. Recent kinetic studies for isoprene reported that the kinetic order dependence on active center concentration is 1/4 at [PLi] > 10-4 and 1/2 at [PLi] < 10-4M in benzene at 30°C [33]. These results were explained in terms of a concentration-dependent change in degree of assocation from predominantly tetrameric at higher concentrations to predominantly dimeric at lower concentrations as shown in Scheme 7.2. Observed changes in the ultra-

Scheme 7.2

violet (UV) spectra as a function of concentration were used to support the concentration-dependence of the degree of association.

Comparison of these kinetic orders with the degrees of association of the poly(dienyl)lithium chain ends is complicated by the lack of agreement regarding the predominant degree of association of these species in hydrocarbon solution (see Chapter 1). Predominant degrees of association of both two and four have been reported by different research groups using the same techniques (i.e., concentrated solution viscometry and light scattering) (see Chapter 1, Table 1.3). Furthermore, recent evaluation of the association states of poly(butadienyl)-lithium chain ends in benzene by small-angle neutron scattering, as well as both dynamic and static light scattering, indicates that higher order aggregates (n > 100) exist in equilibrium with dimeric species [8]. In view of this situation, it is worthwhile to review what is known about the association behavior of low-molecular-weight analogs of poly(dienyl)lithium chain ends.

Allyllithium is not soluble in hydrocarbon solvents; in diethyl ether, how



ever, it exhibits concentration-dependent degrees of association ranging from 2 to > 12 [39]. The adduct of t-butyllithium and butadiene (neopentylallyllithium) exhibits variable association numbers ranging from 2 to 4 that increase with increasing concentration in benzene solution [40]. Unfortunately, these results are complicated by the fact that the samples of neopentylallyllithium were contaminated with approximately 10% of t-butyllithium which is itself associated into tetramers [41].

Qualitative information regarding the degree of association of poly(dienyl)lithiums is available; for example, it has been reported that the viscosity of polymer solutions at 78°C in cyclohexane increases reversibly upon conversion of a poly(styryl)lithium chain end to a poly(dienyl)lithium chain end [42] (see Chapter 8 for a discussion of chain end stability at elevated temperatures). The relative viscosities were in the order poly(butadienyl)lithium > poly(isoprenyl)lithium > poly(styryl)lithium.

By light-scattering measurements in cyclohexane, Worsfold and Bywater [29] obtained an association number of 4 for poly(isoprenyl)lithium using polymer concentrations of 0.1–0.6% and with chain end concentrations of 10-5–10-4M. The degree of association of poly(dienyl)lithium chain ends obtained by end-capping the corresponding poly(styryl)lithium with either isoprene or butadiene was reinvestigated using both light scattering and concentrated solution viscosity measurements in cyclohexane [43]. For poly(butadienyl)lithium chain ends (Mn = 22 × 103 and 322 × 103 g/mol), an association number of 4 was obtained by light scattering for concentrations ranging from 4.2 × 10-6M to 7.6 × 10-4M. For poly(isoprenyl)lithium chain ends [(a) Mn = 11 × 103; (b) Mn = 24.7 × 103 g/mol; (c) Mw = 160 × 103 g/mol], the observed association numbers obtained by light scattering varied with concentration from 2.8 (4.9 × 10-6M) to 4.1 (8 × 10-4M). It was also observed that the concentrated solution viscosities increased upon converting poly(styryl)lithium chain ends to poly(isoprenyl)lithium chain ends by end-capping with isoprene. Sinn and co-workers [25] determined average degrees of association for poly(isoprenyl)lithium of 2.7 ([PILi] = 2 × 10-4M) and 4.4 ([PILi] = 1 × 10-3M)in heptane by the viscosity method. Makowski and Lynn [44] found that the average degree of association of poly(butadienyl)lithium was 2 in pentane at 25°C using viscosity measurements.

Morton and colleagues [11] reported association numbers of 2 for poly(styryl)lithium in benzene (0.23 × 10-3M), and for both poly(isoprenyl)lithium (7.5 × 10-3M), and poly(butadienyl)lithium (1.42 × 10-3M) in hexane using concentrated solution viscosity measurements. In addition, they determined that poly(isoprenyl)lithium and poly(butadienyl)lithium were unassociated in tetrahydrofuran (THF) solution (1.9 × 10-3M) using the same technique. Morton and colleagues [45] reinvestigated the association behavior of polymeric organolithiums. Poly(isoprenyl)lithium was found to be dimeric in n-hexane by concentrated solution viscosity (0.7 × 10-3 to 3.2 × 10-3M) and by light scattering measurements (10-3—



10-5M). End-capping of poly(styryl)lithium with either butadiene or isoprene (1–2%) did not result in a detectable increase in concentrated solution visocity or apparent degree of association. Reinvestigation of the degree of association of polymeric organolithiums using the concentrated solution viscosity method confirmed earlier results [11,45] and also included comparisons with in situ prepared dimers and star-branched polymers with three and four arms by linking reactions with the corresponding chlorosilanes [46]. Al-Jarrah and Young [47] also reported that a variety of poly(dienyl)lithiums exhibit predominant degrees of association of two in n-hexane as determined by concentrated solution viscosity measurements.

Sinn and co-workers [48] have examined the degree of association of poly(isoprenyl)lithium in heptane as a function of concentration using both viscosity and light-scattering measurements. A degree of association of two was determined to be consistent with the data; however, consideration of the expected differences in viscosities of star vs. linear molecules led to the conclusion that the degree of association increases with increasing concentration changing from 2 to 3 at approximately 10-3M. An extrapolated value of 4 was estimated for 10-2M.

Thus, different association numbers (2 and 4) have been reported for poly(butadienyl)lithium, both numbers were determined from the viscosity method as well as using light-scattering measurements. Similarly for poly(isoprenyl)lithium, association numbers of 2, 3 and 4 have been reported using these same methods. These discrepancies, therefore, cannot obviously or simply be attributed to experimental methods; nor does it seem likely that it is the result of the choice of solvents. Nevertheless, genuine differences exist, and a careful reexamination is warranted, especially in view of the recent scattering studies, which indicate that higher-order aggregates are in equilibrium with dimers [8].

It is also significant to note that temperature- and concentration-dependent UV spectra have been observed for poly(isoprenyl)lithium but not for poly(butadienyl)lithium [33,49]. These spectral variations have been interpreted in terms of tetramer—dimer and dimer-unassociated chain end equilibria, although this has not been independently confirmed. It is also important to note that a dependence of the apparent dissociation constant on molecular weight was noted (e.g., increased with increasing MW) [33].

In conclusion, poly(butadienyl)lithium and poly(isoprenyl)lithium chain ends are associated in solution to at least dimeric and perhaps higher degrees of association [8] that may depend on molecular weight [44,48], concentration, temperature, and solvent. In addition, evidence suggests that the degree of association and/or the association constant is larger for poly(butadienyl)lithium than poly(isoprenyl)lithium and poly(styryl)lithium. The relevance of these conclusions to the kinetics of propagation remains to be discussed.

If poly(isoprenyl)lithium molecules are indeed mostly associated into tetramers in hydrocarbon solution, the observed 1/4-kinetic order dependence on chain end concentration can be readily explained utilizing the association—dissociation



equilibria (see Scheme 7.2) and assuming that only the dissociated molecules propagate as originally proposed [29]. If, on the other hand, the association number is 2, we are forced to conclude that there is no general correlation between the kinetic order of these polymerizations and the state of association of the propagating chain ends [6,7]. A similar situation exists with respect to the mechanism of propagation for poly(butadienyl)lithium. Thus, at this time, the mechanistic significance of the 1/4-order kinetic order dependence on chain end concentration for isoprene and butadiene cannot be unambiguously described.

C. Relative Reactivities of Styrene and Dienes

It is generally stated that the relative rates of propagation for styrene and diene monomers are in the order styrene > isoprene > butadiene [2]. Thus, it is of interest to note that the copolymerization of styrene and diene monomers in hydrocarbon solution leads to the initial preferential incorporation of the diene, the less reactive monomer (see Chapter 10). However, styrene propagation exhibits a one-half order dependence on chain-end concentration, while both butadiene and isoprene propagations exhibit a one-fourth order dependence on chain-end concentration. Because of these differences in kinetic order dependence on chain-end concentration, at low chain-end concentrations it is possible for isoprene actually to propagate faster than styrene, as shown in Figure 7.2 [37,50]. Thus, the relative reactivities of dienes vs. styrenes depend on the chain-end concentrations.

Figure 7.2 The rate of chain propagation for poly(isoprenyl)lithium (——)

and poly(styryl)lithium (—.—.—.) at 40°C in cyclohexane as a function of chain-end concentration. (From Ref. 50.)



D. Effects of Lewis Bases

The addition of Lewis bases such as ethers and tertiary amines generally increases the rate of propagation in alkyllithium-initiated polymerizations, as reported by Welch [51], who noted the accelerating effects on overall rates of polymerization with added tetrahydrofuran, diethyl ether, and triethylamine. Addition of small amounts of ethers such as dioxane [52] and tetrahydrofuran [29,53–55] effect initial increases in propagation rate for isoprene or styrene polymerization in hydrocarbon solution; however, with increasing amounts of base, the rate passes through a maximum and then decreases as illustrated in Figure 7.3 [53].

Since it is known that Lewis bases decrease the average degree of association of organolithium aggregates (see Chapter 1, Tables 1.3 and 1.4), it can be assumed that the initial additions of ether to hydrocarbon solutions of polymeric organolithiums promote dissociation of these aggregates. Evidence for dissociation of the aggregates has been confirmed by concentration solution viscosity measurements [45,54]. Thus, it was determined that only trace amounts of THF ([THF]/[PSLi] 10) are required to dissociate poly(styryl)lithium completely [45], while relatively large amounts of THF ([THF]/[PILi] 2600) are required to eliminate

Figure 7.3 The effect of wide variations in the tetrahydrofuran concentrations

on the propagation rate at two median butyllithium concentrations: [BuLi] = 0.96-1.2

× 10-3 moles/L; [BuLi] = 1.2–1.8 × 10-4 moles/L. (From Ref.53; reprinted by permission of the National Research Council of Canada.)



association for poly(isoprenyl)lithium [54]. These results are consistent with the previous discussion in which at least “stronger” association, if not higher degrees of association, of poly(dienyl)lithium was indicated.

At higher concentrations of Lewis bases, it would be expected that the unassociated species would be complexed with base [55]. It may be that the monomer must compete with base for complexation with vacant sites on lithum at the organolithium chain end. However, it should also be noted that in polar media a wide variety of different ionic species may be involved such as contact and solvent-separated ion pairs, and free ions (see Chapter 2) [10,56].

E. Salt Effects

As discussed in Chapter 6, because of the ubiquitous presence of lithium alkoxides as impurities in organolithium-initiated polymerizations, it is important to consider their effects on propagation kinetics. A number of investigations have confirmed that lithium alkoxides decrease the rate of propagation for styrene [57,58], butadiene [59], and isoprene [28,59]. For example, the presence of t-BuOLi decreases the rate of propagation of PILi in hexane by a factor of 0.65 and 0.25 when the ratio of [LiOR]/[PILi] is 0.63 and 1.8, respectively [59]. However, the one-fourth-order dependence of the rate of propagation of PILi concentration is maintained in the presence of these amounts of t-BuOLi. For styrene propagation, the rate decreased by a factor of only 0.46 when the ratio of [LiOR]/[PSLi] was 10; the one-half order dependence of the rate of propagation on [PSLi] was also maintained in the presence of t-BuOLi [57]. Hsieh [58] reported that propagation rates for butadiene and styrene were reduced in both toluene and cyclohexane with increasing amounts of t-BuOLi as shown in Figures 7.4 and 7.5, respectively.

Lithium t-butoxide has been shown to be associated into hexamers in benzene solution [60]. Thus, it is concluded that this alkoxide is less hindered than the corresponding carbanion since t-butyllithium is tetrameric in hydrocarbon solution [41]. The effect of lithium ethoxide on the association behavior of ethyllithium has been examined by freezing point depression measurements in cyclohexane [61]. Evidence for the following types of equilibria was obtained (Eqs. 7.3, 7.4). Thus, it was concluded that

(7.3)

(7.4)

lithium ethoxide coordinates to the intact hexameric aggregate, presumably by interacting with one of the empty faces of the octahedral structure of the hexamer. This is in contrast to the behavior of mixtures of organolithium compounds in which statistical mixtures of cross-associated species are obtained [62]. At higher concentration of lithium alkoxides it was concluded that clusters of lithium



YFP

Highlight

YFP

Underline

YFP

Underline

YFP

Underline

YFP

Underline

YFP

Underline

YFP

Underline

YFP

Underline

YFP

Underline

YFP

Underline

YFP

Underline

YFP

Rectangle

YFP

Rectangle

YFP

Text Box

八面体

YFP

Line

YFP

Line

Figure 7.4 Rate of propagation (Rp) of butadiene (1.6M) preinitiated with

sec-BuLi (1.55 × 10-3M) and variable t-BuOLi at 30°C. (A) Toluene as solvent; (B)

cyclohexane as solvent. (From Ref. 58; reprinted by permission of John Wiley & Sons, Inc.)

Figure 7.5 Rate of propagation (Rp) of styrene (0.70M) preinitiated with

sec-BuLi (1.55 × 10-3M) and variable t-BuOLi at 30°C. (A) Toluene as solvent; (B)

cyclohexane as solvent. (From Ref. 58; reprinted by permission of John Wiley & Sons, Inc.)



alkoxide interact with the intact hexamer. Thus, lithium t-butoxides behave like Lewis bases in their interaction with hexameric ethyllithium.

The interaction with more resonance-stabilized organolithium compounds appears to be more complex. The interactions of the bidentate bases lithium 2-methoxyethoxide and lithium 2-dimethylaminoethoxide with 9-fluorenyllithium in cyclohexane have been investigated by UV-visible spectroscopy [63]. These alkoxides convert the contact ion pair (λmax = 349 nm) completely to the solvent-separated ion pair (λmax = 373 nm) at [base]/[Li] ratios of two. These spectral shifts are analogous to the effects of THF and lowering the temperature on ion-pair equilibria as discussed in Chapter 3. It has been reported that addition of lithium t-butoxide causes a shift in the UV spectrum of poly(styryl)lithium from λmax = 334 nm to λmax = 340 nm [57]; however, no effects on UV absorptions are observed for poly(isoprenyl)lithium upon addition of lithium t-butoxide [59]. It was also reported that the addition of lithium t-butoxide did not significantly affect the viscosity of poly(styryl)lithium [57]; thus, the dimeric state of association is unchanged, which is consistent with coordination of added lithium alkoxide to the dimeric aggregate.

It was suggested that several equilibria could be important in this system [57] (Scheme 7.3). The question arises as to whether added lithium alkoxide decreases

Scheme 7.3

the concentration of unaggregated, reactive species for propagation, that is, K' is smaller than K°, or if poly(styryl)lithium coordinated with lithium alkoxide is less reactive than uncoordinated poly(styryl)lithium: kp'' < kp°. No evidence is available to determine the relative importance of these factors to the rate effects of lithium alkoxides.

Hsieh and Wofford [64] have investigated the effect of other alkali metal alkoxides on the butyllithium-initiated polymerizations of styrene and butadiene. In general, the addition of sodium, potassium, rubidium, and cesium t-butoxides dramatically increased the rate of polymerization (8–100-fold) in amounts that increase with increasing ratio of [t-BuO- M+]/[PLi] up to a maximum value at a ratio of one as shown in Figures 7.6 and 7.7 for potassium t-butoxide in butadiene



Figure 7.6 Rate of polymerization of butadiene with potassium tert-butoxide-n-butyllithium at 30°C. (From Ref. 64;


and styrene polymerizations, respectively. The maximum rate increases were observed for potasium t-butoxide. It was proposed that the polymerization shifts toward a propagating species with the substituted alkali metal counterion. This was supported by the corresponding increase in 1,2-microstructure with increasing amounts of alkali metal alkoxide other than lithium (see Chapter 9).

III. Summary

The propagation kinetics for alkyllithium-initiated polymerization of styrene and diene monomers in hydrocarbon solution is complicated by the predominant association of the chain ends into dimeric, tetrameric, and perhaps even higher-order aggregates. Thus, fractional kinetic order dependencies on chain-end concentrations are observed, which generally have been interpreted in terms of a kinetic scheme in which the aggregates are relatively unreactive toward monomer addition whereas the small, equilibrium concentration of unassociated species is primarily responsible for chain propagation. Although this simple interpretation is



Figure 7.7 Rate of polymerization of styrene with potassium tert-butoxide

-n-butyllithium at 30°C. (From Ref. 65; reprinted by permission of John Wiley & Sons, Inc.)

unambiguously applied to styrene propagation kinetics, the situation for dienes is complicated by the lack of agreement with respect to the degree of aggregation of dienyllithium chain ends in hydrocarbon solution. Predominant association into both dimeric or tetrameric aggregates has been reported by different researchers using the same characterization methods. Lewis bases promote dissociation of organolithium aggregates and accelerate chain propagation, in general. The presence of lithium alkoxides as additives or as impurities from oxygen, hydroxylic, or other contaminants generally decrease the rates of propagation, although the effects are not large. In contrast, the addition of other alkali metal alkoxides generally accelerates the rate of propagation.

References

1. S. Bywater, in Comprehensive Chemical Kinetics, Vol. 15, Non-Radical Polymerization, C. H. Bamford and C. F. H. Tipper, Eds., Elsevier, New York, 1976, p. 1.


3. A. H. E. Müller, in Comprehensive Polymer Science, Vol. 3, Chain Polymerization I,



G. C. Eastmond, A. Ledwith, S. Russo, and P. Sigwalt, Eds., Pergamon Press, Oxford, 1989, p. 387.


5. H. Hsieh, J. Polym. Sci., A3, 163 (1965).


7. R. N. Young, L. J. Fetters, J. S. Huang, and R. Krishnamoorti, Polym. Int., 33, 217 (1994).

8. L. J. Fetters, N. P. Balsara, J. S. Huang, H. S. Jeon, K. Almdal, and M. Y. Lin, Macromolecules, 28, 4996 (1995).

9. M. Van Beylen, S. Bywater, G. Smets, M. Szwarc, and D. J. Worsfold, Adv. Polym. Sci., 86, 87 (1988).

10. M. Szwarc and M. Van Beylen, Ionic Polymerization and Living Polymers, Chapman and Hall, New York, 1993.

11. M. Morton, L. J. Fetters, and E. E. Bostick, J. Polym. Sci., Part C, 1, 311 (1963).

12. R. C. P. Cubbon and D. Margerison, Proc. Roy. Soc. London, Ser. A,. 268, 260 (1962); Polymer, 6, 102 (1965).

13. H. Hsieh, J. Polym. Sci., Part A, 3, 153, 173 (1965).

14. A. F. Johnson and D. J. Worsfold, J. Polym. Sci., Part A, 3, 449 (1965).

15. M. Morton, E. E. Bostick, and R. A. Livigni, Rubber Plast. Age, 42, 397 (1961).

16. M. Morton, E. E. Bostick, R. A. Livigni, and L. J. Fetters, J. Polym. Sci., Part A, 1, 1735 (1963).

17. L. J. Fetters, J. Res. Natl. Bur. Stand., Sect. A, 69, 159 (1965).

18. M. Morton, R. A. Pett, and J. F. Fellers, Prepr. IUPAC Macromol. Symp., Vol. 1, p. 69 (1966).

19. I. Kuntz, J. Polym. Sci., A2, 2827 (1964).

20. C. Lundborg and H. Sinn, Makromol. Chem., 41, 242 (1960).

21. H. Sinn and C. Lundborg, Makromol. Chem., 47, 86 (1961).

22. H. Sinn and O. T. Onsager, Makromol. Chem., 52, 246 (1962).

23. H. Sinn and W. Hofmann, Makromol. Chem., 56, 234 (1962).

24. H. Sinn and F. Bandermann, Makromol. Chem., 62, 134 (1963).

25. H. Sinn, C. Lundborg, and O. T. Onsager, Makromol. Chem., 70, 222 (1964).

26. W. Gebert, J. Hinz, and H. Sinn, Makromol. Chem., 144, 97 (1971).

27. B. Francois, V. Sinn, and J. Parrod, J. Polym. Sci., Part C, 4, 375 (1963).

28. A. Guyot and J. Vialle, J. Macromol. Sci., Chem., A4, 107 (1970).



38. M. Morton, Anionic Polymerization. Principles and Practice, Academic Press, New York, 1983, p. 114.

39. G. R. Brubaker and P. Beak, J. Organometal. Chem., 136, 147 (1977).

40. W. H. Glaze, J. E. Hanicak, M. L. Moore, and J. Chaudhuri, J. Organometal. Chem., 44, 39 (1972).

41. M. Weiner, C. Vogel, and R. West, Inorg. Chem., 1, 654 (1962).

42. H. L. Hsieh and A. G. Kitchen, in Initiation of Polymerization, ACS Symposium Series, No. 212, F. E. Bailey, Ed., American Chemical Society, Washington, D.C., 1983, p. 291.


44. H. S. Makowski and M. Lynn, J. Macromol. Chem., 1, 443 (1966).



47. M. M. F. Al-Jarrah and R. N. Young, Polymer, 21, 119 (1980).

48. A. Hernandez, J. Semel, H.-C. Broecker, H. G. Zachmann, and H. Sinn, Makromol. Chem., Rapid Commun., 1, 75 (1980).

49. J. E. L. Roovers and S. Bywater, Polymer, 14, 594 (1973).

50. Personal communication, L. J. Fetters.

51. F. J. Welch, J. Am. Chem. Soc., 82, 6000 (1960).

52. I. J. Alexander and S. Bywater, J. Polym. Sci., Part A 1, 6, 3407 (1968).


54. M. Morton and L. J. Fetters, J. Polym Sci., A, 2, 3311 (1964).

55. S. Bywater and D. J. Worsfold, J. Phys. Chem., 70, 162 (1966).


57. J. E. L. Roovers and S. Bywater, Trans. Faraday Soc., 62, 1876 (1966).

58. H. L. Hsieh, J. Polym. Sci., A-1, 8, 533 (1970).


60. V. Halaska, L Lochmann, and D. Lim, Coll. Czech. Chem. Commun., 33, 3245 (1968).



63. T. Narita and T. Tsuruta, J. Organometal. Chem., 30, 289 (1971).

64. H. L. Hsieh and C. F. Wofford, J. Polym. Sci., A1, 7, 449 (1969).



8 Termination and Chain Transfer Reactions

I. Introduction

It is important to delineate carefully the scope and limitations of the living nature of anionic polymerizations [1–3]. As noted by Szwarc [2], “The exclusion of natural death does not mean immortality either.” As discussed in Chapter 4, there has always been a tacit assumption that the categorization of a given system as living was based on results obtained on the laboratory time scale, that is, no chain termination or chain transfer reactions occurred within the normal time required to complete the polymerization and carry out any deliberate chemical reactions with the active polymer chain ends (see Chapter 4) [4]. Thus, in the absence of detailed kinetic investigations of the relative rates of propagation, termination and chain transfer, it is a matter of judgment with respect to this practical laboratory time scale. In this chapter, experimental information concerning the chemistry and kinetics of anionic termination and chain transfer reactions will be discussed to provide insight into the scope and limitations of the living nature of anionic polymerizations.



II. Chain Termination Reactions in Hydrocarbon Solvents: Thermal Decomposition

A. Alkyllithium Compounds

Simple alkyllithium compounds undergo spontaneous thermal decomposition at finite rates. The rates of decomposition depend on the structure of the alkyllithium, the temperature, and the concentration of lithium alkoxides. One of the first investigations of the stability of alkyllithium compounds in hydrocarbon solution was the work of Ziegler and Gellert [5], who investigated the thermal decomposition of n-butyllithium and ethyllithium. n-Butyllithium decomposes at temperatures above 100°C to yield 1-butene (90%), butane (8%), polymer materials, and lithium hydride [5]. The butane presumably is formed by the metalation of 1-butene by butyllithium since the yield is reduced to 1–2% when the reaction is carried out on a vacuum line to remove 1-butene as it is formed [6]. At ambient temperature, however, n-butyllithium is quite stable [7,8]; the rate of decomposition in hexane is reported to correspond to only a 0.06% decrease in active chain end concentration per month at 20°C [7]. With respect to the mechanism of decomposition, the rate of decomposition is first order in [BuLi] and a kinetic isotope effect (kH/kD) of 3–4 was reported [6]. This was interpreted as being consistent with a concerted, four-center type transition state (1) [6].

ing effect of lithium butoxide was also observed; the rate of decomposition was doubled by the presence of approximately 23% ([RLi]/[ROLi] = 76.5/23.5) of lithium alkoxide [6].

In general, branching decreases the thermal stability of alkyllithium compounds as shown in Table 8.1 [9–11]. The products observed from thermal decomposition of sec-butyllithium were 1-butene, cis-2-butene, and trans-2-butene, which accounted for 98% of the gaseous products [9,10]. In the absence of alkoxide, the formation of cis-2-butene is favored (cis-2 = 53%, trans-2 = 15%, and 1-butene = 32%), while in the presence of alkoxide 1-butene is favored (cis-2 = 22%, trans-2 = 27%, and 1-butene = 51%). The rate of pyrolysis of sec-butyllithium in decane solution in the presence of lithium alkoxides was reported to be one-half order in alkyllithium and first order in lithium alkoxide [11]. It was proposed that the sec-butyllithium tetramers are dissociated by the lithium alkoxide to form alkoxide-solvated dimers as shown in Equation 8.1. It should be noted

(8.1)



Table 8.1 Kinetic Results for the Pyrolysis of Neat Alkyllithium Compounds

RLi Temperature (°C) k × 105 (sec-1)

Ethyllithium 98 3.5

n-Butyllithium 98.9 3.3

Isopropyllithium 98.2 11

Sec-Butyllithium 87 5.8

that analogous studies of the interaction of lithium ethoxide with ethyllithium were interpreted in terms of a direct interaction of lithium alkoxide with the intact hexameric aggregate [12]. In contrast to n-butyllithium, sec-butyllithium decomposes at a finite rate at ambient temperatures; a 12% solution in cyclohexane at 20°C decomposes to the extent of 1.4% per month [7].

B. Polymeric Organolithium Compounds

It is generally observed that the color of polymeric organolithium compounds changes on standing [13].

Poly(styryl)lithium

At ambient temperature in hydrocarbon media, poly(styryl)lithium possesses good stability over the duration of the polymerizations and beyond (i.e., days). However, at elevated temperatures, it is observed that the initial UV absorption at 334 nm decreases and a new absorption is observed at 450 nm [13-15]. This transformation has been rationalized in terms of the mechanism shown in Scheme 8.1. The assignment of the absorption at 450 nm to a 1,3-diphenylallyllithium

Scheme 8.1



species is consistent with model studies of Burley and Young [16]. They reported that 3-methyl-1,3-diphenylallyllithium exhibits an absorption maximum in hexane at 481 nm; however, this anion was generated in diethyl ether and then the solvent was replaced by hexane. The presence of some residual ether may account for the discrepancy in the absorption maximum for this model compound compared to the proposed product (2) from the thermal decomposition of poly(styryl)lithium. In accordance with this mechanism, it was observed also that the concentration of active chain ends decreases upon heating [17]. The apparent first-order rate constants for decomposition of poly(styryl)lithium in cyclohexane were estimated to be 0.03 × 10-3 sec-1 at 65°C, 0.19 × 10-3 sec-1 at 93°C, and 0.36 × 10-3 sec-1 at 120°C [13,17]. These rates correspond to half-lives of 5.8 h, 1 h, and 0.5 h at 65°C, 93°C, and 120°C, respectively. Although decomposition is readily observed at a temperature of 150°C, the half-life for disappearance of the active chain ends is nearly 100 times greater than the half-life for propagation, which is of the order of a few seconds in cyclohexane [14].

Poly (α-methylstyryl) lithium

In contrast to poly(styryl)lithium, poly(α-methylstyryl)lithium [P(αMeS)Li] does not exhibit good stability for extended periods at ambient temperatures in hydrocarbon media [13]. Poly(α-methylstyryl)lithium exhibits an ultraviolet (UV) absorption maximum at 321.5 nm in neat monomer at 25°C [18]. The absorbance at this wavelength decreases significantly at 25°C; the observed half-lives are 5 h and a few minutes at 25 and 60°C, respectively. The increased rate of thermal decomposition of poly(α-methylstyryl)lithium to form a more substituted alkene (see Eq. 8.2) by elimination of lithium hydride compared to poly(styryl)lithium is

(8.2)

consistent with the effect of branching observed for simple alkyllithiums as shown in Table 8.1. Increased rates of elimination for more branched alkyllithiums are consistent with formation of a more stable, more substituted alkene (i.e., with significant double bond formation in the transition state) (see 1). A further complication in the mechanism of thermal decomposition of poly(α-methylstyryl)lithium is the report that oligomeric poly(α-methylstyryl)lithium decomposes to form lithium hydride and substituted indanes as shown in Scheme 8.2 [19].



Scheme 8.2

Poly(dienyl)lithiums

Like poly(styryl)lithium, the carbanionic active centers based on 1,3-butadiene and isoprene with lithium as the counterion generally possess good stability in hydrocarbon solvents at ambient temperatures. However, poly(dienyl)lithiums undergo complex decomposition reactions upon prolonged storage or heating at elevated temperatures. For example, it is reported that hydrocarbon solutions of poly(butadienyl)lithium chain ends exhibit continuously increasing flow times for concentrated solution viscosity measurements [20,21]. However, constant flow times are obtained when hydrocarbon solutions of polymers with butadienyllithium chain ends are formed by addition of a few units of butadiene to the end of either poly(isoprenyl)lithium or poly(styryl)lithium [21,22]. This was interpreted in terms of reaction of the active center with the polydiene chain units; for example, addition to the side-chain vinyl units or allylic metalation of the backbone [23].

Normally, hydrocarbon solutions of poly(butadienyl)lithium and poly(isoprenyl)lithium are colorless to light yellow; however, upon heating at elevated temperatures the colors change through dark yellow to dark amber [17]. It was reported that poly(butadienyl)lithium in ethylbenzene exhibited an absorption maximum at 300 nm that gradually decreases in intensity with the formation of absorption tails between 350 and 500 nm [24]. For example, approximately 20% of the active centers were destroyed in less than 3 h at 100°C in ethylbenzene. Pennisi and Fetters [23] presented results of thermal aging of poly(butadienyl)lithium at 75°C as shown in Figure 8.1; 10 wt% of the polymer linked to form higher-molecular-weight products after only 3 h, while 33 wt% linking was observed after 24 h. The apparent first-order rate constant for decomposition of poly(butadienyl)lithium in hexane was estimated to be 1.9 × 10-5 sec- at



Figure 8.1 SEC chromatograms of linear polybutadiene (upper left) and the

partially branched materials derived from that nonterminated sample by storage

at 75°C. The molecular weight (Mw) of the linear polybutadiene was 7.1 × 104 g/mol (Mz/Mw =

1.03 and Mw/Mn = 1.04). (From Ref. 23; reprinted by permission of the American Chemical Society.)

93°C and a chain end concentration of 2.2 milliequivalents of poly(butadienyl)-lithium per 100 grams of solution (25 wt% polymer) [17]; however, it was noted that the rate of decomposition decreased significantly with increasing chain end concentration. For example, the rate constant was estimated to be 1.3 × 10-4 sec-1 at a chain end concentration of 0.12 milliequivalents of poly(butadienyl)lithium per 100 grams of solution (10 wt% polymer) [17]. A similar increase in chain end stability with increasing polymer concentration was also noted in ethylbenzene [24]. The corresponding first-order rate constant for chain end decomposition of poly(isoprenyl)lithium at 93°C was estimated to be 6.7 × 10-5 sec-1 [17]. Although the differences are not large, the relative order of increasing stabilities of chain ends toward thermal degradation is poly(α-methylstyryl)lithium < poly(styryl)lithium < poly(isoprenyl)lithium < poly(butadienyl)lithium as estimated



by chain end titration data [17,18]. However, Pennisi and Fetters [23] have reported that poly(butadienyl)lithium is less stable to storage than either poly(styryl)lithium or poly(isoprenyl)lithium. This discrepancy could be related to different chain end or lithium alkoxide concentrations in the different studies.

With respect to the mechanism of thermal decomposition of poly(dienyl)-lithiums, size exclusion chromatographic (SEC) analysis of the thermal decomposition products of poly(dienyl)lithiums in heptane at 80°C has shown that the chain end decomposition is accompanied by formation of species that have double the molecular weight of the original living polymer [25]. After heating for 46 h at 80°C in heptane, a 12 wt% yield of coupled products was observed for poly(isoprenyl)lithium; after heating for 27 h at 80°C in heptane, a 19 wt% yield of coupled products was observed for poly(butadienyl)lithium [25]. Upon addition of further butadiene to the sample of poly(butadienyl)lithium that had been heated for 27 h, it was observed that approximately 65% of the initial carbon-lithium chain ends remained and that fractions of both uncoupled and coupled polymers were inactive toward further polymerization. Evidence for the formation of three armed star polymers during thermal decomposition of poly(isoprenyl)lithium was obtained by ultracentrifugation studies of the sample heated for 46 h. The reactions shown in Scheme 8.3 were proposed to explain the formation of these

Scheme 8.3

products. This scheme is based on the elimination of lithium hydride from poly(dienyl)lithiums to form a macrodiene, 3, which can react with another molecule of poly(dienyl)lithium to form the coupled product, 4. Repetition of this sequence will form the trimeric product, 6.



The effect of backbone composition on the thermal stability of poly(butadienyl)lithium as discussed by Pennisi and Fetters [23] suggests that metalation of the backbone can also occur as shown in Equation 8.3. Evidence for this type of

(8.3)

athermal (∆H 0) metalation reaction was previously provided by Antkowiak [26]. It would be expected that in-chain metalation (Eq. 8.3) coupled with elimination of lithium hydride would lead to in-chain diene units that would have even more reactive allylic hydrogens for further metalation-elimination-coupling sequences that would promote thermal decomposition, branching, and ultimately gel formation, as well as color formation in the polymers. Evidence for in-chain metalation and/or addition reactions has been obtained by analysis of the graft copolymers and the molecular weights of the polystyrene graft segments formed by addition of styrene monomer to poly(butadienyl)lithium solutions after thermolysis at 80°C [27]. The average number of active lithiated sites per polymer molecule was estimated to be between 11 and 15, as determined by this styrene grafting method for poly(butadienyl)lithium, which had been thermolyzed for 44 h at 80°C [27].

In conclusion, polymeric organolithium compounds exhibit good stability in hydrocarbon solution at ambient temperatures and for short periods of time at elevated temperatures. The principal mode of decomposition is loss of lithium hydride to form a double bond at the chain end; the resulting macromonomer can undergo further addition reactions with active chain ends to form coupled products. A clear elucidation of the effects of solvent, chain end concentration, and lithium alkoxides has not been presented, however. Another potentially important variable in laboratory synthesis is the possible role of light in catalyzing chain end decomposition reactions [28].

III. Chain Termination Reactions in Polar Solvents

A. Alkyllithium Compounds

It is well known that alkyllithium compounds react with ethers by cleavage reactions [29]. The general mode of reaction leads to the formation of an alkene and a lithium alkoxide as shown in Eq. 8.4 [29,30]. The rate of decomposition of

(8.4)

alkyllithium compounds in ethers depends on the structure of the alkyllithium, the structure of the ether and the temperature. The reactivity order for alkyllithiums



with ethers is 3° > 2° > 1°. For example, ethyllithium and n-butyllithium have half-lives in diethyl ether at reflux temperature (35°C) of 17 and 31 h, respectively [31,32]. In contrast, diethyl ether solutions of isopropyllithium, cyclohexyllithium, sec-butyllithium, and t-butyllithium decreased to one-half or less of the initial concentration within 30 min without heating [31]. With respect to the structure of the ethers, alkyllithium compounds exhibit decreasing stability in ethers in the order diethyl ether > tetrahydrofuran > ethylene glycol dimethyl ether (glyme). The stability of organolithium compounds is lower in tetra-hydrofuran (THF) compared to diethyl ether; thus, the molarity of a solution of n-butyllithium in THF decreased from 0.79M to 0.23M after 1 h at room temperature and completely decomposed after 2 h at 20°C [33]. The rates of decomposition of benzyllithium and α-methylbenzyllithium in THF at 25°C were reported to be -0.44%/hour and -2.14%/hour, respectively [34]. Tetrahydrofuran undergoes a cycloeliminative degradation reaction upon reaction with butyllithium at temperatures above 20°C, forming ethylene and the lithium enolate of acetaldehyde as shown in Scheme 8.4 [35–37]. The lithium enolate of acetaldehyde can be trapped

Scheme 8.4

with a trialkylsilyl chloride and the ethylene can be trapped by addition of t-butyllithium [36]. For dimethoxyethane, the half-lives for n-butyllithium, sec-butyllithium, and t-butyllithium at -20°C were 111, 2, and << 2 min, respectively; at -70°C, the half-life for t-butyllithium was 11 min [38]. These results suggest that reactive alkyllithium initiators possess only limited stabilities at room temperature or above in ether solvents.

B. Polymeric Organolithium Compounds

Polymeric organolithium compounds would be expected to be unstable in ether solvents analogous to alkyllithium compounds; a summary of relevant observations concerning the stability of anionic living polymers has been published [13]. The following conclusions were listed: living carbanionic polymers react with ether solvents such as THF in a pseudo-first-order decay process; the rate of reaction decreases in the order Li > Na > K [13]. Thus, it has been reported that for a 10-5M solution of poly(styryl)lithium in THF at 25°C, the rate of decay,



presumably due to reaction of the chain end with THF, was a few percent per minute, but poly(styryl)cesium was found to be exceptionally stable [39]. It was also noted that more concentrated solutions of living polystyryl carbanions were more stable [39]. The instability of poly(styryl)lithium in THF is regularly rediscovered [40]. The important conclusion is that both alkyllithium initiators and poly(styryl)lithium have limited stability in THF at room temperature; polymerizations in THF at ambient temperatures would be expected to lead to polymers with higher than stoichiometric molecular weights and with broadened molecular weight distributions [13]. The absorption maximum (λmax = 287 nm) in the UV spectrum of poly(isoprenyl)lithium decreases rapidly in THF [41]. In mixtures of hexane and THF (volume ratio of hexane/THF = 0.15), it has been reported that the absorption maximum for poly(butadienyl)lithium decreased in intensity by 60% over a period of 5 h and that the chain ends were more stable in the mixed solvent system compared to pure THF [42]. The relative stability of polymeric organolithium compounds in the presence of a mixed solvent of THF (74 vol % THF) and hexane at -2°C was demonstrated by the ability to prepare well-defined polyisoprene-block-polybutadiene diblock copolymers with Mn (calc)

Mn (osmometry) [42].

IV. Chain Transfer Reactions

A. General Aspects

Chain transfer in anionic polymerization (Scheme 8.5) is analogous to chain transfer reactions in other chain growth polymerization mechanisms. The basic requirements are that the chain transfer agent (A — X) has an atom or group, X,

Scheme 8.5

which can be transferred to the growing chain end (PLi) to terminate that chain's growth and form a new initiating species (A—Li). The new reactive intermediate, A—Li, should be capable of reinitiating growth of a polymer chain with a rate greater than or equal to the rate of propagation (ki > kp) so that the chain transfer process does not alter the rate of polymerization. In general, chain transfer agents in anionic polymerization are species which can function as a Brønsted acid, for example, A — H, by transferring a proton to the carbanionic chain end and forming the corresponding conjugate base (A — Li) that can reinitiate chain growth. The reactivity of a potential chain transfer agent can be deduced from its pKa value; in general, it should have a pKa value similar to that of the conjugate acid of the



propagating carbanionic chain end. If the acidity of the acid is too large, it will readily terminate the growth of an active chain; however, it will form a conjugate base that is too stable to reinitiate chain growth readily. The acidity of carbon acids and related compounds was discussed in Chapter 2.

B. Chain Transfer to Solvent

Ammonia

One of the first kinetic studies of anionic polymerization was the investigation of the potassium amide initiated polymerization of styrene in liquid ammonia [43]; this system is a model for an anionic polymerization proceeding with a chain transfer reaction. The polystyrene polymers prepared with potassium amide in liquid ammonia contained one nitrogen atom per molecule and exhibited molecular weights independent of the initiator concentration. The molecular weights were linearly dependent on the concentration of styrene and varied from 1.4 × 103 to 3.4 × 103 g/mol. The rate expression for consumption of styrene is shown in Eq. 8.5. A mechanism of polymerization consistent with the experimental

(8.5)

results and the rate expression is shown in Scheme 8.6. Thus, the free amide anion is the actual initiating species. This is consistent with the observation that no

Scheme 8.6



polymerization of styrene can be effected by sodium amide in diethyl ether that has too low of a dielectric constant to form a significant amount of free ions [44]. Efficient chain transfer by ammonia to the polystyryl carbanion (PS-) is expected based on the estimated pKa values of ammonia (41) and toluene (43) in DMSO [45] (see Chapter 2). Based on this mechanism, the rate expression can be rewritten as Equation 8.6. In addition, the number average degree of polymeriza-

(8.6)

(8.7)

tion can be expressed in terms of Equation 8.7. Thus, the degree of polymerization is predicted to be a linear function of styrene concentration [S], as observed. A value of kp/ktr of 4.28 × 103 was determined at -33.5°C. This corresponds to a calculated chain transfer constant (ktr/kp) of 2.34 × 10-4. It is interesting to note that in spite of this relatively small value for the chain transfer constant, the observed molecular weights are quite low (ca. 3 × 103 g/mol). This is a consequence of the fact that the chain transfer agent is the solvent whose concentration is approximately 48M (i.e., much larger than the concentration of monomer, 0.3M).

Toluene

Aromatic solvents are useful in promoting faster rates of initiation and propagation in alkyllithium polymerizations, as discussed in Chapters 6 and 7. However, if the aromatic solvent contains relatively acidic benzylic protons, chain transfer reactions can occur as indicated in Equations 8.8 and 8.9. The first study of anionic

(8.8)

(8.9)

chain transfer processes by Robertson and Marion [46] investigated the products of the sodium-initiated polymerizations of butadiene and isoprene in toluene after



termination with carbon dioxide. The oligomeric hydrocarbons and carboxylic acids corresponded to benzyl anion-derived oligomers were consistent with chain transfer reactions of toluene with the active centers to generate benzyl sodium as shown in Equations 8.10 and 8.11.

(8.10)

(8.11)

Gatzke [47] investigated the chain transfer reaction of poly(styryl)lithium with toluene at 60°C during the polymerization of styrene by using 14C-labeled toluene. The molecular weights and molecular weight distributions (ultracentrifuge determination) for polystyrenes of various calculated molecular weights prepared under these conditions are shown in Figure 8.2 and compared with calculated values based on the kinetics of the process. For the polymers with Mn(calc) < 100,000 g/mol, the effects of toluene as solvent are insignificant. However, it is clear that as the calculated molecular weights increase, there is significant tailing of the molecular weight distribution toward lower molecular weights, a corresponding broadening of the molecular weight distribution and observed molecular weights are smaller than the calculated molecular weights. The calculated chain transfer constant (CRH = ktr/kp) for poly(styryl)lithium transferring to toluene at 60°C was 5 × 10-6. Thus, even though this is a relatively small chain transfer constant, the effects of chain transfer to toluene as solvent can be dramatic especially at higher molecular weights as shown in Figure 8.2.

Under similar reaction conditions, it was reported that much lower observed molecular weights, relative to calculated molecular weights, are obtained with sodium as counterion compared to lithium as counterion [48]. Thus, observed molecular weights were one-third of the calculated values based on (grams of monomer)/(moles of initiator) when this ratio was 100,000 for the sodium α-methylstyrene tetramer-initiated polymerization of styrene in toluene at 20°C [48]. Using the radioactive tracer technique, the chain transfer constant to toluene for poly(styryl)sodium was calculated to be 1.3 × 10-4, i.e. approximately a factor of 100 larger than the corresponding value with lithium as counterion [47].

Gatzke [47] derived an equation to calculate the effect of the magnitude of the chain transfer constant, CRH, on the observed molecular weight as shown in Equation 8.12 in which x is the degree of monomer conversion, [M] is the monomer concentration, [I] is the concentration of initiator, and [RH] is the concentra-

(8.12)



P



P

Figure 8.2 Molecular weights and molecular weight distributions of polystyrenes at various monomer/initiator ratios for the

n-butyllithium-initiated polymerization of styrene in toluene at 60°C. (a) Ultracentrifuge determination. (b) Computer-calculated distributions based on kinetics. (From Ref. 47; reprinted by permission of John Wiley & Sons, I



tion of chain transfer agent. Using this equation, it is possible to predict the effect of the magnitude of the chain transfer constant on the discrepancy between the observed and calculated molecular weights as shown in Figure 8.3. With respect to living polymerizations and experimental criteria for determining if a given polymerization system is living (see Chapter 4), it is important to note that Figure 8.3 illustrates the fact that observable experimental effects of chain transfer reactions will not be detected for low values of the chain transfer constant unless experiments are designed to prepare high-molecular-weight polymers. Furthermore, it is important to recognize that the contribution of chain transfer processes depends on the counterion (e.g., Na+ > Li+), the temperature (less chain transfer is observed at lower temperatures [48]) and whether Lewis bases are added.

In general, Lewis bases promote chain transfer reactions [49]. In order to prepare oligomeric polybutadienes (Mn = 1000–8000 g/mole), for example, the alkyllithium-initiated polymerization of butadiene is effected in toluene at elevated temperatures in the presence of a strongly coordinating Lewis base such as N,N,N',N'-tetramethylethylenediamine [49]. Under these conditions as expected,

Figure 8.3 Predicted number-average degree of polymerization as a function of the monomer/initiator ratio for various values

of the chain transfer constant; [M] = 1.0M (monomer concentration), x = 0.9

(degree of conversion), [RH] = 10M (chain transfer agent). (From Ref. 47; reprinted by permission

of John Wiley & Sons, Inc.)



the molecular weight distributions of the polymers are broadened (Mw/Mn 1.6–2.2) and high 1,2-microstructure polybutadienes (40–50% 1,2) are obtained [49]. A chain transfer constant of 0.2 has been reported for the telomerization of butadiene initiated by metallic sodium in toluene (> 2.5M)-tetrahydrofuran mixtures at 40°C [50].

With respect to living polymerization and the ability to prepare polymers with well-defined structures in solvents such as toluene, the storage time and the temperature are both important variables. For example, it has been shown that poly(styryl)lithium undergoes significant amounts of chain transfer to toluene (ca. 13% of polymer) when stored for 1 day at 20°C [51].

It would be expected that the importance of chain transfer to solvent would be decreased by decreasing the Brønsted acidity of the solvent. Thus, ethylbenzene would be expected to be less acidic than toluene. Schue and co-workers [24] have examined the kinetics and mechanism of alkyllithium-initiated anionic polymerizations and chain end stability in ethylbenzene. Block copolymers were prepared in ethylbenzene by polymerizing styrene at 70°C for 2 h (Mn = 40,000 g/mole) and then blocking butadiene onto the poly(styryl)lithium at 90°C for 1 h. The resulting block copolymer (Mn = 55,000 g/mole) exhibited a monomodal molecular weight distribution and there was no observable broadening of the molecular weight distribution in the block copolymer compared to the original polystyrene block (Mw/Mn 1.2). However, it should be recognized that chain transfer effects may not be observable in this system because of the low molecular weights involved (see Figure 8.3).

Alkenes

It has been reported that alkenes or mixtures of alkenes (e.g., distillate fractions from petroleum refining) can be used as solvents for anionic polymerization [52–54]. In view of the expected athermal (∆H 0) nature of the metalation reaction of polymeric organolithium compounds with alkenes (see Eq. 8.3) and the high concentration of alkene when used as solvent, it is not surprising that chain transfer reactions can occur with these solvents. Significant decreases in observed molecular weights were observed when 1-alkenes were used as solvents compared to alkanes [52–54]. For example, the rate constant for chain transfer (ktr) to 1-hexene is 10-2 L/moleúmin at 20°C for poly(butadienyl)lithium [52,53]. The rate constant for chain transfer (ktr) to 1-pentene is 9 × 10-3 L/moleúmin at 20°C for poly(isoprenyl)lithium [52,53].

C. Chain Transfer Agents

Allenes and alkynes are regarded as impurities whose concentration cannot exceed certain minimum levels in monomer feed streams [55]. However, these same compounds are also added as modifiers in alkyllithium-initiated diene polymerizations to prevent thermal branching at higher temperatures via chain termination



and/or chain transfer reactions [52,55–58]. For example, it is reported that the alkyllithium-initiated polymerization of butadiene can be effected in the absence of solvent under reflux cooling to maintain the temperature below 38°C when 200–300 ppm of 1,2-butadiene is added to prevent gelation and thermal degradation; the reaction exothermed to 115°C when the condensing efficiency was reduced after 48% conversion [58].

The reactions of alkyllithium compounds with allenes and alkynes is complex. Both classes of compounds undergo metalation reactions with alkyllithium compounds [29]. Although the acidity of allenes has not been determined [59], the pKa of phenylacetylene in dimethylsulfoxide (DMSO) is reported to be 28.7 [45]. At low temperatures, allene can be metalated and undergoes reactions with electrophilic reagents to form the corresponding allene derivatives as illustrated in Eq. 8.13 [59,60]. At higher temperatures, isomerization to propynyllithium may take place as shown in Equation 8.14 [56,59,60].

(8.13)

(8.14)

The propynyl anion, 7, represents the global minimum energy structure for

this system [61]. However, the minimum energy structure of the allenic anion (H2CCCH-) corresponds to 8 in the gas phase. It should be recognized that 8 and

9 are not resonance structures. The minimum energy structure for allenyllithium (10) corresponds to a slightly bent structure ( C-C-C = 173°) with lithium



bridging. The interconversion from an allenyllithium to a propynyllithium may be preceeded by a tautomerism (Eq. 8.15) or the bridged structure, 10, may react

(8.15)

directly with a Brønsted acid to generate the alkyne intermediate (see Eq. 8.14). As expected, reactions of metalated alkynes and allenes with electrophiles can give both acetylenic and allenic derivatives [59,60].

The reactions of alkyllithium compounds with allenes and alkynes are further complicated by polylithiation reactions [29]. Allene reacts with two equivalents of n-butyllithium at -50°C in a 1/1 THF/hexane mixture to form 1,3-bis(trimethylsilyl)propyne in 63% yield after reaction with trimethylsilyl chloride [60]. Propyne reacts with four equivalents of n-butyllithium in hexane to form C3Li4 and liberates 4 equivalents of n-butane [62] (Eq. 8.16); reaction with trimethylsilyl chloride produces tetrakis(trimethylsilyl)allene as shown in Eq. 8.17 [62].

(8.16)

(8.17)

Thus, there is ample precedent for both allenes and acetylenes to behave as Brønsted acids and react with polymeric organolithium compounds to terminate chain growth and form the corresponding metalated derivatives as illustrated in Equation 8.18 for 1-butyne. However, the question is whether the resulting

(8.18)



metalated allenyllithium or alkynyllithium derivatives can reinitiate polymerization (i.e., function as chain transfer agents rather than simply as slowly reacting terminating agents). Evidence presented by Puskas [56] indicates that 1,2-butadiene reacts only slowly with poly(butadienyl)lithium at 50°C at [1,2-BD]/[Li] 5; there was no effect on either conversion or molecular weight. Other work suggests that chain transfer can occur at higher concentrations of 1,2-butadiene ([1,2-BD]/[Li] 100) [57]. For example, for butadiene polymerizations at temperatures > 100°C (peak temperature > 121°C), the polymer prepared in the presence of 75 ppm of 1,2-propadiene exhibited Mn = 100 × 103

g/mol, Mw = 223 × 103 g/mol (Mw/Mn = 2.23), while the pure monomer produced a polymer with Mn = 99 × 103 g/mol, Mw = 270 × 103 g/mol (Mw/Mn = 2.73). These results suggest that the effects of allenes depend on the polymerization temperature [63].

In contrast, relatively large effects are observed for alkynes. Alkynes dramatically reduce the observed molecular weight while not affecting conversion [52]. Thus, it appears that alkynes have a higher kinetic (and perhaps thermodynamic) acidity than allenes. It is also possible that a dimetalated acetylene or allene derivative, rather than the monometallic analog, can reinitiate chain growth [56,57].

D. Chain Transfer to Monomer

Monomers such as p-methylstyrene that contain acidic benzylic protons would be expected to participate in chain transfer reactions with growing polymeric organolithium compounds. However, poly(p-methylstyrene) homopolymers and copolymers with controlled, predictable molecular weight and narrow molecular weight distribution can be prepared when the temperature is maintained at room temperature or below [64–66]. Somewhat ambiguous results are available with respect to the incursion of chain transfer during the anionic equilibrium polymerization of p-isopropyl-α-methylstyrene since broad molecular weight distributions (Mw/Mn > 1.3) were obtained and there was evidence of low molecular weight tailing using n-butyllithium/N,N,N',N'-tetramethylethylenediamine as initiator at -25°C [67].

Chain transfer has been well documented for the anionic polymerization of 1,3-cyclohexadiene [68–75]. In general, the observed molecular weights are independent of conversion and do not depend on the monomer or initiator concentrations [69,70,72]. The results of kinetic studies in cyclohexane are summarized in Scheme 8.7 [75]. The chain transfer constant, Ctr (ktr/kp), was calculated to be 2.9 × 10-2 at 20°C and 9.5 × 10-3 at 5°C in cyclohexane [75]. These results correspond to polymer molecular weights of 2.8 × 103 and 8.6 × 103 g/mol at 20°C and 5°C, respectively. The ratio of rate constants for decomposition compared to reinitiation (kd/kri) was calculated to be 2 L1/2mol-1/2 at 20°C and 2.2 L1/2mol-1/2 at 5°C in cyclohexane [75]. The formation of benzene [70–72,75], cyclohexene



Scheme 8.7

[70–72], and 1,4-cyclohexadiene [75] has been observed during these polymerizations. It should be noted that 1,4-cyclohexadiene should be a very reactive chain transfer agent for alkyllithium-initiated polymerizations [71]; the ability of 1,4-cyclohexadiene to decrease strongly the observed molecular weights in the lithium naphthalenide-initiated polymerization of 1,3-cyclohexadiene has been demonstrated [72]. The chain transfer constant for 1,3-cyclohexadiene in THF at -20°C is reported to be 0.7 × 10-3M-1sec-1 [72]. Using difunctional alkali metal naphthalenides as initiators, molecular weights as high as 19 × 103 g/mol (Li+) and 38.7 × 103 g/mol (Na+) were obtained in dimethoxyethane in which the polymer precipitates [72].

V. Summary

Not all alkyllithium-initiated polymerizations of styrene and diene monomers are living. Both termination and chain transfer reactions can take place depending on the monomer, the solvent, the temperature, and the time. In general, hydrocarbon solutions of poly(dienyl)lithiums and poly(styryl)lithiums are stable for time



periods required for polymerization, block copolymerization, and functionalization reactions at reasonable temperatures. Decreased stability with respect to pyrolytic decomposition to form lithium hydride is observed at elevated temperatures; thus, prolonged heating at elevated temperatures (>50°C) should be avoided for synthesis of polymers with well-defined structures. Transmetalation reactions can occur at elevated temperatures also. In general, polar solvents and polar additives decrease chain end stability. Ethers are particularly susceptible to undergo chain termination reactions with polymeric organolithium compounds. Chain transfer reactions and/or termination reactions can occur with alkenes, dienes with allylic hydrogens, alkynes, allenes, and aromatic compounds with benzylic hydrogens.

References


2. M. Szwarc, Nature, 178, 1168 (1956).



5. K. Ziegler and H. G. Gellert, Ann. Chem., 567, 179 (1950).

6. R. A. Finnegan and H. W. Kutta, J. Org. Chem., 30, 4138 (1965).

7. R. J. Bauer, in Ullmann's Encyclopedia of Industrial Chemistry, VCH Verlagsgesellschaft, Weinheim, Germany, 1990, Vol A 15, p. 393.

8. R. Bach and J. R. Wasson, in Kirk-Othmer Encyclopedia of Chemical Technology, 3rd ed., Wiley-Interscience, New York, 1981, Vol. 14, p. 448.

9. W. H. Glaze, J. Lin, and E. G. Felton, J. Org. Chem., 30, 1258 (1965).

10. W. H. Glaze, J. Lin, and E. G. Felton, J. Org. Chem., 31, 2643 (1966).

11. W. H. Glaze and G. M. Adams, J. Am. Chem. Soc., 88, 4653 (1966).


13. M. D. Glasse, Prog. Polym. Sci., 9, 133 (1983).

14. S. Bywater and D. J. Worsfold, J. Polym. Sci., 58, 571 (1962).

15. S. S. Medvedev and A. R. Gantmakher, J. Polym. Sci., C4, 173 (1963).

16. J. W. Burley and R. N. Young, J. Chem. Soc. (B), 1018 (1971).

17. W. J. Kern, J. N. Anderson, H. E. Adams, T. C. Bouton, and T. W. Bethea, J. Appl. Polym. Sci., 16, 3123 (1972).

18. D. Ades, M. Fontanille, and J. Leonard, Can. J. Chem., 60, 564 (1982).

19. D. Margerison and V. A. Nyss, J. Chem. Soc. (C), 3065 (1968).

20. M. Morton and L. J. Fetters, J. Polym. Sci., Part A, 2, 3311 (1964).

21. R. Milner, R. N. Young, and A. Luxton, Polymer, 24, 5443 (1983).



27. J. N. Anderson, W. J. Kern, T. W. Bethea, and H. E. Adams, J. Appl. Polym, Sci., 16, 3133 (1972).

28. J. Comyn and M. D. Glasse, J. Polym. Sci., Polym. Chem. Ed., 21, 209, 227 (1983).

29. B. Wakefield, The Chemistry of Organolithium Compounds, Pergamon Press, 1974, p. 198.

30. P. D. Bartlett, S. Friedman, and M. Stiles, J. Am. Chem. Soc., 75, 1771 (1953).

31. H. Gilman, A. H. Haubein, and H. Hartzfeld, J. Org. Chem., 19, 1034 (1954).

32. D. Seyferth and H. M. Cohen, J. Organometal. Chem., 1, 15 (1963).

33. H. Gilman and B. J. Gaj, J. Org. Chem., 22, 1165 (1957).

34. H. Gilman and H. A. McNinch, J. Org. Chem., 27, 1889 (1962).

35. A. Rembaum, A.-P. Siao, and N. Indictor, J. Polym. Sci., 56, S17 (1962).

36. M. E. Jung and R. B. Blum, Tetrahedron Lett., No. 43, 3791 (1977).

37. R. B. Bates, L. M. Kroposki, and D. E. Potter, J. Org. Chem., 37, 560 (1972).

38. J. J. Fitt and H. W. Gschwend, J. Org. Chem., 49, 209 (1984).

39. D. N. Bhattacharyya, C. L. Lee, J. Smid, and M. Szwarc, J. Phys. Chem., 69, 612 (1965).

40. C. A. Ogle, F. H. Strickler, and B. Gordon III, Macromolecules, 26, 5803 (1993).

41. S. Bywater, A. F. Johnson, and D. J. Worsfold, Can. J. Chem., 42, 1255 (1964).

42. A. Gourdenne and P. Sigwalt, Eur. Polym. J., 3, 481 (1967).

43. W. C. E. Higginson and N. S. Wooding, J. Chem. Soc., 760 (1952).

44. J. J. Anderson and C. R. Hauser, J. Am. Chem. Soc., 71, 1595 (1949).

45. F. G. Bordwell, Accounts Chem. Res., 21, 456 (1988).

46. R. E. Robertson and L. Marion, Can. J. Res., Sect. B, 26, 657 (1948).

47. A. L. Gatzke, J. Polym. Sci., A-1, 7, 2281 (1969).

48. B. W. Brooks, Chem. Commun., 68 (1967).

49. A. R. Luxton, Rubber Chem. Technol., 54, 596 (1981).

50. S. Kume, A. Takahashi, G. Nishikawa, M. Hatano, and S. Kambara, Makromol. Chem., 84, 137 (1965).

51. L. S. Wang, J. C. Favier, and P. Sigwalt, Polym. Commun., 30, 248 (1989).

52. G. Heublein and D. Stadermann, Prog. Polym. Sci., 14, 195 (1989).

53. I. A. Lifshits and Y. B. Podolnyi, Vysokomolek. Soedin., B, 13, 149 (1971).

54. I. A. Lifshits and Y. B. Podolnyi, Vysokomolek. Soedin., A, 12, 2655 (1970).



64. J. Chen and L. J. Fetters, Polym. Bull., 4, 275 (1981).

65. R. P. Quirk, M. T. Sarkis, and D. J. Meier in Advances in Elastomers and Rubber Elasticity, J. Lal and J. E. Mark, Eds., Plenum Press, New York, 1986, p. 143.

66. J. W. Mays and N. Hadjichristidis, Polym. Bull., 22, 471 (1989).

67. J. Leonard and S. L. Malhotra, J. Macromol. Sci. Chem., A11, 2087 (1977).

68. G. Lefebvre and F. Dawans, J. Polym. Sci., A2, 3277 (1964).

69. P. E. Cassidy, C. S. Marvel, and S. Ray, J. Polym. Sci., A3, 1553 (1965).

70. H. Lussi and J. Barman, Helv. Chim. Acta, 50, 1233 (1967).

71. L. A. Mango and R. W. Lenz, Polym. Prepr., Am. Chem. Soc., Div. Polym. Chem., 12, 402 (1971).

72. Z. Sharaby, J. Jagur-Grodzinski, M. Martan, and D. Vofsi, J. Polym. Sci., Polym. Chem. Ed., 20, 901 (1982).

73. Z. Sharaby, M. Martan, and J. Jagur-Grodzinski, Macromolecules, 15, 1167 (1982).

74. X. F. Zhong and B. Francois, Makromol. Chem., 191, 2735 (1990).

75. B. Francois and X. F. Zhong, Makromol. Chem., 191, 2743 (1990).



9 Stereochemistry of Polymerization

I. Polydiene Stereochemistry

The importance of lithium as counterion for diene polymerization was first recognized in 1956. Scientists at the Firestone Tire and Rubber Company reported that polyisoprene produced by anionic polymerization using lithium metal initiation had a high (>90%) cis-1,4 microstructure analogous to natural rubber [1]. Since that time, the commercial importance of anionic polymerization has grown. One of the most important reasons for this growth is the ability to polymerize 1,3-dienes to yield homopolymers or copolymers (random or block) with various proportions of cis-1,4-, trans-1,4-, and vinyl microstructures and with controlled molecular weights; this leads to a wide range of useful polymer properties (see Chapters 15–22) [2–5]. The stereochemistry of anionic diene polymerization depends on the counterion, the solvent, the chain end concentration, the monomer concentration, the temperature, and the presence of Lewis base additives. These variables will be discussed in terms of the mechanism of diene polymerization and the control of microstructure.

A. Diene Microstructure

Conjugated 1,3-dienes [CH2=C(R)-CH=CH2] can polymerize to form four constitutional isomeric microstructures, ignoring chirality (chiral carbon atoms



YFP

Text Box

立体化学

YFP

Highlight

YFP

Pencil

Scheme 9.1 Polydiene microstructure

are indicated with an *), as illustrated in Scheme 9.1. When R=H, 1,2-enchainment is equivalent to 3,4-, for example, with butadiene.

The microstructure of polydienes has been investigated by infrared spectroscopy, Raman spectroscopy, 1H nuclear magnetic resonance (NMR) and 13C NMR spectroscopy. Although infrared (IR) spectroscopic analysis of polymers can be carried out relatively quickly, it is the least reliable method. Thus, IR absorption bands in the regions 970, 910, and 704 cm-1 result from contributions from all three microstructures for polybutadiene; analysis requires the solution of three simultaneous equations [6–8]. Inaccuracies are particularly evident at high vinyl contents [9]. IR absorption bands at 888 and 908 cm-1 are satisfactory for determination of 3,4- and 1,2- units, respectively, in polyisoprene [10]; however, the bands at 839 and 842 cm-1 for cis- and trans-1,4 units, respectively, are relatively weak and almost coincident [10].

Raman spectroscopy provides a reliable method for microstructure analysis; C=C bands corresponding to unobservable symmetrical stretches in the IR, for example, for trans-1,4-enchainments, are observed by Raman spectroscopy [8,11–13]. Thus, microstructure analysis of polybutadienes can be obtained from quantitative determination of υ(C=C) Raman bands at 1639, 1650, and 1664 cm-1 corresponding to 1,2-, cis-1,4- and trans-1,4-enchainments, respectively [8,11,13,14]. Similarly Raman bands for polyisoprene at 1662, 1641, and 1639 cm-1 for the 1,4-, 3,4- and 1,2-microstructures, respectively, can be used for quantitative microstructure analysis [12,13].

Both 1H and 13C NMR spectroscopy are reliable and reproducible characterization methods for determination of polydiene microstructure [8,9,15–18]. The 1H NMR of polybutadienes readily provides an estimate of the relative amounts of 1,4- vs. 1,2-enchainment by comparison of peak integrations at δ 5.4 and 4.9 ppm, respectively [8]; however, high fields are required to resolve 1,4-cis from 1,4-trans resonances [10] because routine 1H NMR analysis does not differentiate cis-1,4- from trans-1,4-microstructures. The 1H NMR spectra of polyisoprene exhibit significant solvent dependencies [17]. In CDCl3 the relative amounts of 1,4- vs. 1,2-enchainment can be determined by integration of the peaks at δ 5.1 and 4.75 ppm [8,18], respectively; separate resonances corresponding to methyl



YFP

Highlight

YFP

Pencil

YFP

Rectangle

YFP

Highlight

YFP

Highlight

YFP

Highlight

YFP

Line

YFP

Line

YFP

Polygonal Line

group protons in cis-1,4-, 3,4- and trans-1,4-units are observed at δ 1.68, 1.64, and 1.61 ppm [18], respectively.

13C NMR spectroscopy is the preferred method for microstructure analysis of polybutadienes [8,9,16,20]. Specific carbon resonances are observed for each type of microstructure, in addition to splittings due to sequence distributions. 13C NMR peaks corresponding to the CH2 = and =CH- carbons in 1,2-vinyl units are observed at δ 114 and 143 ppm [8], respectively. A complex resonance pattern between 127 and 133 ppm is assigned to the cis-1,4 and trans-1,4 olefinic carbons which are sensitive to triad sequence distributions [8]; complete microstructure analysis is possible based on the peak assignments in this region [16,21,22].

The 13C NMR spectra of polyisoprene are rich in microstructural information [23]. For example, the C1 (CH2) carbon is observed at δ32.08 and 39.70 ppm for

cis-1,4- and trans-1,4-microstructural units, respectively [24]. The C5 (CH3) resonances are observed at δ = 23.25, 15.76, and 18.74 ppm for the cis-1,4-, trans-1,4-, and 3,4 units, respectively [24]. The cis-1,4-units also exhibit resonances at δ = 134.85 and 124.65 ppm for the C2 and C3 carbons, respectively; analogous resonances for trans-1,4-units are assigned to peaks at δ134.38 and 123.87 ppm [25]. The C1 and C2 vinyl carbons in 3,4-units are observed at δ110.8 and 146.9 ppm, respectively [26–29]. The observed resonances provide additional information about the sequence distribution of the isomeric units [24,25].

Several classic investigations provided information regarding the effects of counterion on the stereochemistry of diene polymerization [30–34]; some of these data are listed in Tables 9.1 and 9.2. Although these data are somewhat unreliable in the sense that they are based only on infrared determinations of the polydiene microstructure and the sensitivity of microstructure to both monomer and chain-end concentration was not known, they display trends that are clear and useful. Furthermore, these results are not significantly different from recent results obtained by Morton and Rupert [35] and they are also comparable with data obtained at high monomer concentrations in cyclohexane solution [2], both of which are also listed for comparison.

One of the most important conclusions is that lithium is unique among the alkali metals in providing polydienes with high 1,4-microstructure. For isoprene,



YFP

Highlight

YFP

Text Box

32.08

YFP

Text Box

39.70

YFP

Text Box

23.25

YFP

Text Box

15.76

YFP

Text Box

18.74

YFP

Text Box

134.85

YFP

Text Box

124.65

YFP

Line

YFP

Line

YFP

Text Box

134.38

YFP

Text Box

123.87

YFP

Line

YFP

Line

YFP

Text Box

110.8

YFP

Text Box

146.9

YFP

Line

YFP

Line

YFP

Rectangle

YFP

Highlight

YFP

Highlight

YFP

Highlight

YFP

Rectangle

YFP

Polygonal Line

Table 9.1 Effect of Counterion on Polybutadiene Microstructure for Neat Polymerizations

Microstructure

Counterion

Temperature (°C)

1,4-cis

1,4-trans

1,2

Reference

Lithium 70 35 52 13 32

20 39 52 9 35

Sodium 50 10 25 65 32

Potassium 50 15 40 45 32

Rubidium 60 7 31 62 32

Cesium 60 6 35 59 32

this corresponds to > 90% cis-1,4-microstructure; for butadiene, the high 1,4-microstucture is a mixture of 39% cis and 52% trans content. The significance of high 1,4-microstructures for polydienes is that such polymers exhibit low glass transition temperatures, (e.g., — 64 to — 70°C for polyisoprene [36] and — 94°C for polybutadiene [11% 1,2]) [37]; these polydienes with high 1,4-microstructure exhibit good elastomeric properties at room temperature and above.

Table 9. 2 Effect of Counterion on Polyisoprene Microstructure for Neat Polymerizations

Microstructure

Temperature(°C)

1,4-cis

1,4-trans

1,2

3,4

Reference

Lithium 25 94 — — 6 31

94 — — 6 32

20 96 — 4 35

Sodium 25 — 45 7 48 31

— 43 6 51 32

15 44a 6 50 2

Potassium 25 — 52 8 40 31

25 — 52 8 40 32

15 59a 5 36 2

Rubidium 25 5 47 8 39 32

Cesium 25 4 51 8 37 32

15 69a 4 27 2

aTotal 1,4-content (cis + trans) in cyclohexane.



Several more recent studies have elucidated the variables that affect polydiene microstructure in hydrocarbon media with lithium as counterion and some of these results are summarized in Table 9.3.

Several general aspects of stereoregulation for diene polymerization can be gleaned from the data in Table 9.3. The highest cis-1,4-microstructures are obtained in the absence of solvent, that is, with neat monomer, at low concentrations of initiator (ca. 10-6 M): 96% cis-1,4- and 4% 3,4-polyisoprene; and 86% cis-1,4-,

Table 9.3 Microstructure of Polydienes in Hydrocarbon Media Using Organolithium Initiators

Initiator Microstructure

Concentration (M)

Solvent

Temperature(°C)

1,4-cis

1,4-

trans

3,4

Reference

Polyisoprene

6 × 10-3 Heptane -10 74 18 8 38

1 × 10-2 Heptane -10 78 17 5 38

1 × 10-4 Heptane -10 84 11 5 38

8 × 10-6 Heptane -10 97 — 3 38

5 × 10-6 Heptane 25 95 2 3 39

9 × 10-3 Benzene 20 69 25 6 35

4 × 10-5 Benzene 20 70 24 6 35

5 × 10-6 Benzene 25 72 20 8 39

1 × 10-2 Hexane 20 70 25 5 35

1 × 10-5 Hexane 20 86 11 3 35

3 × 10-3 None 20 77 18 5 35

8 × 10-6 None 20 96 — 4 35

5 × 10-6 None 25 98 — 2 39

Polybutadiene 1,4-cis

1,4-trans

1,2

5 × 10-1 Benzene 20 62a 38 40

5 × 10-2 Benzene 20 83a 17 40

5 × 10-3 Benzene 20 93a 7 40

5 × 10-1 Cyclohexane 20 53a 47 41

5 × 10-2 Cyclohexane 20 90a 10 41



9% trans-1,4- and 5% 1,2-polybutadiene. High cis-1,4-enchainment is also favored by the use of aliphatic vs. aromatic solvents at low concentrations of initiator; however, the total amount of 1,4-microstructure (cis + trans) is relatively insensitive to solvent and chain end concentration. There are very practical consequences associated with these effects. One problem arises in the preparation of a series of polydienes that have the same microstructure but with molecular weights that vary over a wide range. If all variables except initiator concentration are held constant, for example, the microstructure for polyisoprene prepared in heptane will vary from 78% cis-1,4 (10-3M initiator) to 97% cis-1,4 (8 × 10-6M initiator) for realistic molecular weight changes in the range of 100 × 103 g/mole to 1 × 106 g/mole. Thus, comparisons of the chemical or physical properties of these

Figure 9.1 Effect of temperature on the microstructure of polybutadiene

prepared by n-butyllithium-initiated polymerization in cyclohexane. (From Ref. 42;




polymers will reflect variations in microstructure as well as molecular weight. A practical method of solving this problem is to change solvent composition by adding increasing amounts of aromatic solvent with decreasing chain end concentration [2].

In general, temperature changes exert relatively minor influences on polydiene microstructure in hydrocarbon solution; however, relatively large effects of pressure have been reported. The effect of wide variations in temperature on polydiene microstructure is illustrated in Figure 9.1 [42]. These results have been confirmed in a number of studies [35,43]. It has been reported that the amount of cis-1,4-polyisoprene decreases from 78% to 44% when the pressure is increased from 1 bar to 14,000 bars in heptane with n-butyllithium as initiator [44]. Concurrently, the amount of 1,2-enchainment increases from 0 to 10%, the amount of 3,4-enchainment increases from 4 to 14% and the amount of trans-1,4-enchainment increases from 20 to 32%. In contrast, the major effect of pressure for the analogous polymerization of 2,3-dimethylbutadiene corresponds to a decrease in the amount of trans-1,4 enchainment and an increase in the amount of 1,2 micro-structure [45].

B. Mechanism of Diene Polymerization

A simple mechanism as shown in Scheme 9.2 was proposed initially to explain the cis-stereospecificity of diene polymerization with lithium as counterion [46]. This mechanism was described as follows: “The polyisoprenyllithium first complexes

Scheme 9.2



with isoprene in the cis-form. The complex subsequently rearranges to form a transition state in the form of a six-membered ring. The configuration of each monomer unit in the chain is thus fixed at its point of entry.” This mechanism is inadequate for a very fundamental reason: the configuration of a stereounit at the chain end is not fixed until the next monomer unit is added. Therefore, the fact that the addition of isoprene may occur via a six-membered ring transition state to form initially a cis-isoprenyllithium chain end configuration does not adequately explain the polydiene stereochemistry. This step fixes the stereochemistry of the previous chain end as shown, but does not fix that of the incoming monomer unit that becomes the chain end. Of course, this mechanism cannot explain the dependence of stereochemistry on the concentrations of chain end and monomer.

In order to explain the observed dependence of polybutadiene and polyisoprene microstructure on initiator concentration, Gerbert et al. [38] suggested that the initially formed cis-form of the poly(dienyl)lithium chain end can isomerize to the trans-form in competition with monomer addition. They also proposed that this isomerization occurs in the associated form of the poly(dienyl)lithium chain ends.

Worsfold and Bywater [47] have presented a comprehensive hypothesis based on model compound studies to explain the effects of the concentrations of active chain ends and monomer on polydiene microstructure. They examined the cis-trans isomerization kinetics for 2,5,5-trimethyl-2-hexenyllithium (1), 2-methyl-2-octenyllithium (2), and the monoaddition product of 1 with isoprene (3) to

determine if isomerization rates are indeed competitive with monomer addition as proposed by Gerbert et al. [38]. It is important to note that the regiospecificity of the addition of t-butyllithium to isoprene is predominantly 4,1-addition as expected for formation of the most stable allylic species, 4, vs. the less stable allylic anion, 5, resulting from 1,4-addition. The cis-trans isomerization rate of 1 (k = 2.8 × 10-4 sec-1 in heptane at 0°C; t1/2 = 40 minutes) was judged to be too slow to be competitive with propagation for which an average lifetime of an active center between monomer addition steps was estimated to be 1 sec. The half-life for cis-trans isomerization of the n-butyl analog, 2, was estimated to be less than 2 min, which indicates the sensitivity of the isomerization reaction to steric inter-



actions. Because of the fast rate of isomerization of 3 at 0°C (t1/2 ˜ 1 min), these isomerization kinetics were investigated at -20°C. Based on the observed isomerization rate constant (kisom = 1.04 ×10-3 sec-1), it was concluded that the isomerization rate at 30°C (kisom 1 sec-1) would be competitive with monomer addition. A further important conclusion was that the adduct 3 is formed with essentially all of the terminal active centers in the cis-form immediately after isoprene addition. However, at equilibrium in hydrocarbon solvents, these model compounds exist largely in the trans-form [2].

The kinetics of isoprene addition to 1 (cis- and trans-forms) relative to the rate of isomerization of the chain end has also been investigated. The ratio (R) of the rate of isoprene addition to the cis-1 units relative to the rate of isoprene addition to the trans-1 units was estimated to be 8.4 (kpcis/Kptrans).

Also, it was concluded, as expected, that cis-1 produces 3 with an internal cis unit and trans-1 produces 3 with an

internal trans unit.

Based on the results of these experiments and the known dependence of polydiene microstructure on diene monomer (D) and chain end concentrations as shown in Table 9.3, the mechanistic hypothesis shown in Scheme 9.3 was proposed by Worsfold and Bywater [47]. Thus, it was proposed that isomerization of the initially formed cis-form of the active chain end occurs competitively with



Scheme 9.3

monomer addition at each step of the reaction. Thus, when the concentration of monomer is high relative to the chain-end concentration, the first-order isomerization of the cis-form does not compete effectively with monomer addition; however, at low concentrations of monomer relative to chain ends, the isomerization does compete and significant amounts of the trans-form will be in equilibrium with the cis-form.

With this information and the value of R (kpcis/kptrans), the approximate amount of cis-enchainment can be

estimated using Equation 9.1 for an equilibrium

(9.1)

distribution of chain end configurations (70% trans, 30% cis). This simple calculation shows that the predominantly cis microstructure of polyisoprene is not because all poly(isoprenyl)lithium chain ends are in the cis-configuration, but it is primarily a consequence of the fact that the rate of monomer addition to the cis-chain end is eight times faster than the corresponding rate for the trans-isomer. However, in order to obtain higher cis-1,4-microstructure, a higher monomer concentration is required and this increases the second-order propagation step relative to the first-order isomerization reaction.

This simple scheme also provides an explanation for the differences in stereochemistry obtained for isoprene and butadiene under similar reaction conditions. For poly(butadienyl)lithium, the corresponding model compounds analogous to 1–3 isomerized too rapidly for analogous NMR kinetic analysis. However, it was estimated that at equilibrium, 25% of the chains ends are in the cis-form for poly(butadienyl)lithium based on studies with neopentylallyllithium. It was also estimated that R (kpcis/kptrans) = 2 [48]. Using these estimations, the microstructure



expected for polybutadiene can be calculated using Equation 9.1 for an equilibrium distribution of poly(butadienyl)lithium chain end configurations as shown below. Thus, the much lower cis-1,4-enchainment observed for polybutadiene

(9.1)

relative to polyisoprene is not a consequence of the active chain end stereochemistry, but can be ascribed to the much lower rate of monomer addition of the cis-form relative to the trans-form for poly (butadienyl)lithium compared to poly(isoprenyl)lithium; R is equal to 8 and 2 for poly(isoprenyl)lithium and poly(butadienyl)lithium, respectively. In addition, the rate of propagation for butadiene is slower than for isoprene, while the corresponding rate of isomerization of the active chain ends is faster for poly(butadienyl)lithium compared to poly(iso-prenyl)lithium.

A cautionary note is in order, however. This simple picture has totally neglected the role of chain end association, in contrast to the hypothesis of Gerbert and colleagues [38]. It is known that both poly(butadienyl)lithium and poly(isoprenyl)lithium are predominantly associated into at least dimers and perhaps tetramers or higher aggregates [49] in hydrocarbon solution (see Chapters 1 and 7). In addition, observed propagation rate constants are actually a composite of the equilibrium constants for dissociation of the aggregates, generally raised to some fractional power, times the rate constants for the elementary kinetic step of monomer addition (see Chapter 7). Therefore, the actual kinetic situation must be more complex than depicted in the Worsfold and Bywater hypothesis (Scheme 9.3). For example, it is reasonable to assume that isomerization occurs in the aggregated species rather than for the unassociated form as proposed by Gerbert and colleagues [38]. Support for isomerization occurring in the aggregates is provided by the kinetics of isomerization that exhibit first-order kinetic dependencies [47]. Since the isomerization reaction is first order in chain end concentration while the propagation step is approximately proportional to the one-fourth power, the rate of isomerization is increased relative to propagation by increasing the chain end concentration [3]. This is shown in Equation 9.2, in which the effect of an increase in chain end concentration of 104 on the relative rates of isomerization and propagation is calculated [3]. In addition, it is important to note that experimental evidence suggests that either poly(butadienyl)lithium chain ends have



(9.2)

higher degrees of association or lower aggregate dissociation constants compared to poly(isoprenyl)lithium (see Chapters 1 and 7). Thus, the higher preference for aggregation in poly(butadienyl)lithium vs. poly(isoprenyl)lithium would tend to promote isomerization that occurs in the aggregate, relative to propagation that presumably occurs in the unassociated state.

A mechanism involving isomerization in the aggregated species would eliminate the requisite charge separation that would be involved in isomerization of the s-cis (σ) form to the s-trans(σ) allylic form via a π-allyl species as shown in Scheme 9.4. In the aggregate a simple rearrangement (followed by bond rota-

Scheme 9.4

tion) is all that would be required for isomerization without the necessity of forming a π-allyl-type, charge-separated species as shown in Scheme 9.5.

One unexplained feature of alkyllithium-initiated anionic synthesis of polydienes is the fact that under most normal polymerization conditions there is 5–10% of vinyl side chain formation [3,4-polyisoprene or 1,2-polybutadiene]. However, it has been reported that 47% and 38% 1,2-microstructure can be obtained for polybutadiene at a chain end concentration of 0.5M in cyclohexane [41] and in benzene [35], respectively. The hypothesis of Worsfold and Bywater [47] for the dependence on diene stereochemistry on the relative concentrations of monomer to initiator does not adequately explain the formation and persistence of the vinyl side chain formation. It is possible that monomer can add to both the unassociated



Scheme 9.5

chain ends and directly with the aggregates. The effect of increasing chain end concentration on this competition would be analogous to the dependence illustrated in Equation 9.2 for isomerization vs. propagation.

C. Other Diene Monomers

The microstructure obtained for other diene monomers (see Table 9.4) provides further insight into the factors that affect diene microstructure. There seems to be a general trend that higher 1,4-microstructures are obtained with more sterically hindered diene monomers. Thus, using conditions that provide polyisoprene with 70% cis-1,4, 22% trans-1,4, and 7% 3,4 microstructure, 2-i-propyl-1,3-butadiene and 2-n-propyl-1,3-butadiene provide 86% and 91% cis-1,4 enchainment, respectively, as shown in Table 9.4 [54]. Furthermore, both 2-phenyl-1,3-butadiene (92% cis-1,4) and 2-(triethylsilyl)-1,3-butadiene (100% cis-1,4) also exhibit high cis-1,4-enchainment. These results could be a consequence of the preferred stereochemistry of the dienyllithium chain end (e.g., favoring a cis-type chain-end configuration); a faster rate of reaction of the cis-form vs. the trans-form; or it could be a consequence of the decreased state of aggregation of the more hindered chain end (assuming that chain end isomerization occurs primarily in the aggregated state). Thus, these variations can be explained in terms of the same rationale discussed with respect to the factors that affect the stereochemistry for isoprene and butadiene and the fact that isoprene gives higher cis-1,4-microstructure than butadiene.

D. Diene Microstructure in the Presence of Lewis Bases

One of the earliest discoveries in the anionic polymerization of dienes using alkali metals and organometallic compounds as initiators was the fact that polydiene



Table 9.4 Microstructure of Polydienes Prepared Using Organolithium Initiators

Microstructure

Solvent

Temperature (°C)

Cis-1,4 (%)

trans-1,4(%)

1,2 (%)

3,4 (%)

Reference

1,3-Pentadiene

Heptane 80 25.5 59.5 15 — 50

Hexane 22 49 40 11 — 51

1-Phenyl-1,3-butadiene

Hexane 20 28 49 — 23 52

Benzene 20 25 59 — 16 52

1-(4-Pyridyl)-1,3-butadiene

Hexane 25 — 90 — 10 53

2-Ethyl-1,3-butadiene

Heptane 40 78 14 — 8 54

2-n-Propyl-1,3-butadiene

Heptane 40 91 4 — 5 54

2-i-Propyl-1,3-butadiene

Heptane 40 86 10 — 4 54

2-n-Butyl-1,3-butadiene

Heptane 40 62 35 — 3 54

2-Phenyl-1,3-butadiene

Toluene 30 92 — 8 — 55

2-[(Trimethylsilyl)methyl]-1,3-butadiene

Hexane 25 47 23 — 30 56

2-(Triethylsilyl)-1,3-butadiene

Hexane 25 100 — — — 57

5-(N,N-Diisopropylamino)isoprene

Benzene 40 >95a — — 58

2,3-Dimethyl-1,3-butadiene

Benzene 30 87a 13 — 59

2-Methyl-1,3-pentadiene

Benzene 40 64 36 — — 60



aTotal, 14 content (mixture of cis-1,4 and trans-1,4).



Table 9.5 Effects of Polar Solvents on Polybutadiene Microstructurea

Temperature Microstructure (%)

Solvent Counterion (°C) 1,4-cis 1,4-trans 1,2 Reference

THF Lithium 0 6 6 88 61

THF Lithium -78 ~0 8 92 61

THF Sodium 0 6 14 80 61

THF Sodium -78 0 14 86 61

THF Potassium 0 or -78 5 28 67 61

Et2O Lithium 0 8 17 75 61

Et2O Sodium 0 7 23 70 61

Et2O Potassium 0 11 34 55 61

Dioxane Lithium 15 — 13 87 62

Dioxane Sodium 15 — 15 85 62

Dioxane Potassium 15 — 45 55 62

Dioxane Cesium 15 — 59 41 62

Dioxane Free ion 15 — 22 78 62

aFree ion formation was suppressed for the measurements in THF and Et2O by the addition of tetraphenylboride salts (triphenylcyanoboron for potassium).

microstructure is quite sensitive to the counterion (see Table 9.1) and solvent as shown in Tables 9.5 and 9.6. In polar media, the unique, high 1,4-stereospecificity of lithium observed in hydrocarbon media is lost and large amounts of 1,2-(poly-butadiene) and 3,4-(polyisoprene) enchainments are obtained. In fact, there is a tendency towards higher 1,4-content with increasing size of the counterion in polar media; for example, the highest 1,2-content in polybutadiene is observed with lithium and the highest 1,4-content is observed with cesium. Similar trends are observed for polyisoprene. There also seems to be a trend towards higher 1,4-content with decreasing solvent polarity (e.g., diethyl ether > tetrahydrofuran [THF]) [61].

There are several important structural differences for polydienyl anions in polar media vs. hydrocarbon solvents: chain ends are generally not associated into higher aggregates in polar media compared to hydrocarbon (see Chapter 1); the charge distribution of the allylic anion is a function of solvent and counterion; the kinetic and equilibrium distribution of chain end configurations can vary with solvent and counterion; and the distribution of contact ion pairs, solvent-separated ion pairs, and free ions can vary with solvent, counterion, and temperature.

It has been reported that the charge distribution in asymmetrical allylic carbanions varies with solvent and temperature. Using the hypothesis that the chemical shift per electron corresponds to 114 ppm/electron, the calculated charge distributions for neopentylallyl-alkali metal (I) and neopentylmethylallyl-alkali



Table 9.6 Effects of Polar Solvents on Polyisoprene Microstructure

Microstructure (%)

Temperature

Solvent Counterion (°C) 1,4-cis 1,4-trans 1,2 3,4 Reference

THF Lithium 30 (12) a 29 59 63

THF Sodium 0 (11)a 19 70 64

DMEb Li, Na, K, Cs

15 (24–26)a 28–33 44–48 65

Et2O Lithium 20 (35)a 13 52 66

Et2O Sodium 20 (17)a 22 61 66

Et2O Potassium 20 (38)a 19 43 66

Et2O Cesium 20 (52)a 16 32 66

Dioxane Lithium 15 3 11 18 68 62

Dioxane Potassium 15 4 32 14 50 62

Dioxane Free ion 15 <1 24 32 44 62

aTotal amount of 1,4-microstructure.

b1,2-dimethoxyethane.

Table 9.7 Calculated Charges on Allyl Carbon Atoms of (I) Neopentylallyl- and (II) Neopentylmethylallyl-Alkali Metal Compounds [67]

Calculated Charges on Allylic Carbon Atoms

Counterion Solvent α β γ Σ (total)

Neopentylallyl-alkali metal (I)

Li C6H60.79 -0.13 0.22 0.88

Et2O 0.69 -0.13 0.35 0.91

THF 0.69 -0.15 0.40 0.94

Na THF 0.65 -0.12 0.49 1.02

K THF 0.59 -0.11 0.53 1.01

Rb THF 0.55 -0.11 0.53 0.97

Cs THF 0.51 -0.12 0.52 0.91



Neopentylmethylallyl-alkali metal (II)

Li C6H60.80 -0.14 0.19 0.85

Et2O 0.72 -0.15 0.30 0.87

THF 0.73 -0.15 0.34 0.92

DMEa 0.72 -0.14 0.35 0.93

Na THF 0.69 -0.12 0.38 0.95

K THF 0.61 -0.09 0.44 0.96

Rb THF 0.58 -0.09 0.47 0.96

Cs THF 0.54 -0.10 0.45 0.89

a1,2-Dimethoxyethane



metal (II) compounds have been calculated and the results are shown in Table 9.7 [67]. In general, the chemical shifts for carbon atoms are shielded (i.e., shifted to higher field and lower values of chemical shift, δ, relative to TMS) in a carbanion relative to the corresponding hydrocarbon. Compounds I and II can be regarded as models for butadienyl and isoprenyl carbanionic species, respectively.

One important conclusion from the data in Table 9.7 is that while the negative charge is more localized on the alpha (α) carbon in hydrocarbon solution for the lithium derivatives, in polar media there is less charge on the alpha (α) carbon and more charge on the gamma (γ) carbon in these allyl organoalkali compounds. The presence of more negative charge on the gamma (γ) carbon provides a general explanation for the observation of predominantly side-chain vinyl microstructure in polar media for lithium as counterion (see Tables 9.5 and 9.6). Unfortunately, the data in Table 9.7 do not provide a simple explanation for the observed increase in 1,4-microstructure with increasing size of the cation since delocalization is greater with larger cations, but the vinyl content is lower.

It is noteworthy that the charge distributions shown in Table 9.7 indicate that these carbanions have unsymmetrical charge distributions as expected based on the destabilizing effect of an alkyl group at the γ position. Bywater [68] has suggested that a highly solvated lithium cation and a moderately solvated sodium cation in ether solvents situated closer to the alpha carbon in the allylic anion may block reaction with monomer at this position and lead to preferential attack at the gamma position.

Another possible variable that should be considered with respect to the interpretation of the effects of solvent and counterion on diene microstructure is the configurational distribution of the chain end. Since Worsfold and Bywater [47] were able to provide a rationalization for the effect of monomer and chain end concentration in terms of the equilibrium distribution of chain end configurations and their relative rates of reaction with monomer, it is prudent to consider these variables for polymerizations in polar media also.

NMR investigations have provided evidence for the presence of two types of chain end structures for polydienyl anions in polar media as shown in Scheme 9.6 [61]. The results are summarized in Table 9.8. In contrast to the results observed in hydrocarbon media, the most stable chain end configuration (except for Li/Et2O) is the cis form. It is obvious that this fact does not provide an explanation for the small amounts of 1,4-, and especially small amounts of cis-1,4-content in polydienes obtained in polar media.

Garton and Bywater [72] observed reversible changes in the ultraviolet (UV) absorption spectra for poly(butadienyl)lithium and poly(butadienyl)sodium as a



Scheme 9.6

function of temperature and monomer addition in THF. Poly(butadienyl)sodium generated by monomer addition at low temperature (-40°C) exhibited an absorption maximum initially at 350 nm. On standing, this spectrum gradually shifted to produce an absorption maximum at 310 nm. When the polymerization was carried out at higher temperatures, the shorter wavelength peak predominated. With lithium as counterion, similar spectral shifts were observed; at -40°C the initially formed chain end absorbs at 325 nm; however, on standing, the intensity of this peak diminishes, while the intensity of a shorter wavelength peak (285 nm) increases. By analogy with the NMR results, these spectral changes were interpreted in terms of two isomeric forms of the chain end. At low temperature, the cis chain ends are favored; however, during polymerization the trans form predominates as the temperature is lowered. It was concluded that the monomer addition in THF occurs preferentially to form a trans chain end structure initially. These results are all opposite to the results in hydrocarbon solution, wherein the trans form predominates at equilibrium, but the cis form is initially formed at the chain end by monomer addition.

Table 9.8 Equilibrium Percentage of trans Configuration in (I) Neopentylallyl- and (II) Neopentylmethylallyl-Alkali Metal Compounds [61, 67, 69–71]a

(I) Neopentylallyl-Alkali Metal (II) Neopentylmethylallyl-Alkali Metal

Counterion/Solvent % trans Chain End

Counterion/Solvent % trans Chain End

Li/benzene 77 Li/benzene 65

Li/Et2O 75 Li/Et2O 25

Li/THF 35 Li/THF 0

Na/THF 23 Na/THF 0

K/THF <10 K/THF 0

Cs/THF 0

a–20°C



In terms of the Worsfold and Bywater [47] mechanistic hypothesis for polydiene stereochemistry, it is also necessary to consider the relative rates of monomer addition versus isomerization of the chain ends [73,74]. The rates of isomerization (ktc) from trans to cis chain end structure were determined to be 16.7 × 10-4 sec-1 at -65°C for oligo(butadienyl)lithium, 5.1 × 10-4 sec-1 at -40.4°C for oligobutadienylsodium, and 5.0 × 10-4 sec-1 at 13.1°C for n-butylallylpotassium in THF [74]. The same relative rate order for isomerization (i.e., Li >> Na > K) was also observed for the corresponding neopentylallyl compounds [73]. These results were compared with ion-pair propagation rates for butadiene in THF [75] to rationalize the corresponding polymer microstructures [61]. For the potassium counterion, at 0°C, it was estimated that kp± = 7M-1 sec-1 corresponding to a half life with respect to monomer addition of less than a second, whereas the half life for isomerization would be more than an hour. Therefore, the first formed trans chain end would add monomer much faster than isomerization to the cis form; this provides a partial explanation for the polybutadiene microstructure since the 1,4-content (33%) corresponds to 84% trans and the microstructure does not change upon cooling to -78°C [61]. In contrast, with lithium as counterion, the isomerization half-life at 0°C was estimated to be 0.2 s while the half-life for monomer addition was calculated to be 83 s or 17 s at 0.2M or 1M monomer concentration, respectively. Since isomerization should be at least competitive with monomer addition, higher cis-content relative to other alkali metal counterions would be expected. Thus, in contrast to potassium, the cis-content at 0°C in THF is 50% (total 1,4-content is only 12.5%) [61]; at -78°C, the trans content is 90%. It was concluded that these results are only consistent with a greater reactivity of the initially formed trans form [74]. Kinetic studies indicate that for n-butylallylpotassium in THF the rate of butadiene addition is approximately three times faster for the trans chain end compared with the cis isomer [75,76].

Unfortunately, this rationalization fails to explain the formation of predominantly 1,2-microstructure observed for all alkali metal counterions in polar media except for cesium in dioxane (see Table 9.5). In order to explain the formation of such high amounts of 1,2-microstructures for polybutadiene in polar media, it was proposed that 1,2-units are formed from the trans chain end [75]. Furthermore, it has been postulated that the highly solvated cations (e.g., lithium and sodium) are situated closer to the alpha position and that this effectively blocks this position and favors reaction with monomer at the less hindered gamma carbon that would lead to 1,2-addition [68]. One other interesting variables to consider is the fact 13C NMR chemical shifts indicate that the charge distribution at the gamma position of allyllic organolithium compounds is higher in the trans form relative to the cis form in polar media [3,67]. For example, the chemical shift for the γ carbon in THF at -20°C is observed at δ79.6 ppm for the trans isomer and at δ81.9 ppm for the cis isomer of 5,5-dimethylhexenyl-2-lithium [67].



Another important variable with respect to chain end structure is the degree of solvation of the counterion and the concomitant effects on the ionic nature of the chain end. In terms of the Winstein-type spectrum of ionic chain end species (see Chapter 3), increasing solvent polarity would be expected to shift the equilibrium spectrum of aggregated species, unaggregated species, tight ion pairs, loose ion pairs, and free ions toward more ionic species.

(9.3)

In polar solvents such as amines and ethers, the chain ends exist primarily as unassociated species. However, this simplifying feature is countermanded by the fact that ion pairs and free ions can be present in polar media. Furthermore, because the more dissociated loose ion pairs and free ions are much more reactive than the tight ion pairs, the kinetics and stereochemistry can be dominated by small amounts of the later species. The ion pair dissociation constants for polymeric carbanions with alkali metal counterions depend on the counterion, the solvent and the temperature. For example, the dissociation constants for formation of free carbanions (5, Eq. 9.3) from the corresponding ion pairs (3,4, Eq. 9.3) range from 1.9 × 10-7M for lithium to 3.0 × 10-9M for cesium in THF at 25°C [77]. The extent of this dissociation is negligible in dioxane; for example, there is no kinetic effect of addition of the common ion sodium tetraphenylboride [77].

The stereochemistry of butadiene addition to the free polybutadienyl carbanion is reported to be 22% trans-1,4 and 78% 1,2 at + 15°C in THF; the 1,4-content decreases to 11% at -30°C [62]. These results are similar to the microstructure observed for the ion pair with sodium counterion under analogous conditions (see Table 9.5) [61]. It should be noted that the addition of a small amount of a common ion salt of high dissociation constant such as the tetraphenylboride salts will suppress the free anion contribution to rates and microstructure.

In hydrocarbon solvents it was noted that there was a trend toward higher 1,4-microstructures with more sterically hindered diene monomers as shown in Table 9.4. With respect to the variables that affect diene microstructure, it is significant to note that this tendency is also observed in polar media. For example, butyllithium-initiated polymerization of 2-isopropyl-1,3-butadiene in diethyl ether produces a polymer with 81% cis-1,4- and 19% trans-1,4 microstructure [54]. Butyllithium-initiated polymerization of 2-triethylsilyl-1,3-butadiene likewise provides a polymer with only 1,4-microstructure in THF at -25°C [57]; only 1,4-microstructure is also observed for anionic polymerization of 2-trimethoxysilyl-1,3-butadiene in THF at -78°C, regardless of the counterion [78]. High 1,4-microstructures are also obtained for the polymerization of 1-phenyl-1,3-butadiene (12% cis-1–4, 78% trans-1,4, 10% vinyl) [52], 1-pyridyl-1,3-butadiene



(90% trans-1,4, 9% 3,4) [53] and 2-phenyl-1,3-butadiene (90% cis-1,4 and 10% vinyl) [55] with alkyllithium initiators in THF at 0°C. Thus, thermodynamic and/or kinetic factors that favor 1,4-enchainment in hydrocarbon solution for these monomers are maintained in polar media.

E. Effects of Polar Modifiers (Additives) on Polydiene Microstructure

Small amounts of Lewis base additives in hydrocarbon media can exert dramatic effects on polydiene microstructure, even when present in amounts comparable to the lithium chain end concentration as shown by the data in Tables 9.9 and 9.10. In

Table 9.9 Effect of Temperature and Concentration of Lewis Base on Vinyl Content of Polybutadiene in Hexane

% 1,2 Microstructure

Base [Base]/[Li] 5°C 30°C 50°C 70°C Reference

Triethylamine 30 – 21 18 14 79

270 – 37 33 25

Diethyl ether 12 – 22 16 14 79

180 – 38 29 27

Tetrahydrofuran 5 – 44 25 20 79

85 – 73 49 46

Diglyme 0.1 – 51 24 14 79

0.8 – 78 64 40

TMEDAa 0.6 – 73 47 30 79

0.4 78 – – – 80

6.7 85 – – – 80

1.14 – 76 61 46 79

DIPIPb 0.5 91 50 44 21 81

1 99.99 99 68 31

BMEc 1 88 62 34 17 81

4 98 86 63 28

DIDIOXd 0.2 85 – – – 82

1 95–96 – – –

0.5 97 91 80 63 83

TMDCe 0.7 69 – – – 84

3 71 – – –



Table 9.10 Effects of Cation-Chelating Agents ([B]/[Li]= 1) on Diene Microstructure

Microstructure

Solvent

Temperature (°C)

Base

% 1,4

% 1,2

Reference

Butadiene

Bulk 0 [2.1.1]a 40 60 85

Cyclopentane 5 TMEDA 16 84 80

DIPIP 3 97 86

DIDIOX 4 96 83

Isoprene % 1,4 % 1,2 % 3,4

Bulk 20 [2.1.1] 30 20 50 85

Cyclohexane 22 PMDTb 52 6 42 87

TMEDA 31 13 55 87

21 12 67 88

DIPIP 20 12 68 88

aMacrobicyclic cryptand with two nitrogen atoms at the bridgeheads and having bridges between the nitrogens with two, one, and one oxygen atoms, respectively (see [89]).

bN.N.N',N',N''-Pentamethyldiethylenetriamine.

general, Lewis bases that interact most strongly with lithium produce the highest amount of 1,2-microstructure. For example, there is a correlation between the enthalpies of interaction of Lewis bases with polymeric organolithium compounds and the ability of these bases to promote 1,2-enchainment [3,90–92]. Thus, the highest vinyl contents for polybutadiene are obtained with the most strongly coordinating ligands such as the bidentate bases, N,N,N',N'-tetramethylethylenediamine (TMEDA) and bispiperidinoethane (DIPIP). To obtain significant amounts of vinyl microstructure with weak donor-type bases such as diethyl ether and triethylamine, they must be present in large amounts relative to lithium. In contrast, the strongly coordinating bases produce high vinyl polybutadiene microstructure at low base to lithium atom ratios (R = [base]/[Li] = 1–2).

The effects of polar additives can be explained in terms of the principles established for the effects of polar solvents on chain end structure and reactivity (see Chapter 1). Thus, it has been demonstrated that addition of Lewis bases such as TMEDA to hydrocarbon solutions of polymeric organolithium compounds promotes dissociation to form the unassociated species at low [TMEDA]/[Li] [93]. The dissociation of poly(butadienyl)lithium aggregates was reported to be complete at R

1.0 as deduced from concentrated solution viscosity measurements. A working hypothesis for the effects of Lewis bases on polydiene stereo-chemistry can be deduced by analogy with the mechanism proposed by Worsfold



Scheme 9.7

and Bywater [47] for hydrocarbon media (see Scheme 9.7). Thus, a competition between kinetic and thermodynamic factors is proposed, analogous to the mechanism proposed for stereochemical control in hydrocarbon media. However, it should be noted that it is proposed that the initially formed chain end configuration is trans, not cis as in hydrocarbon media; the equilibrium between cis and trans chain end configurations favors cis, not trans as in hydrocarbon media; and it is proposed that vinyl enchainment results from monomer addition to the trans chain end configuration. A variety of probes have been used to elucidate the factors affecting polydiene stereochemistry in polar media.

The effects of added bidentate bases on the 13NMR chemical shifts provide insight into the charge distribution changes which occur for poly(dienyl)lithium chain ends when complexed with bases. Table 9.11 illustrates the effects of DIPIP and TMEDA on chain end configuration and charge distributions. Addition of increasing amounts of DIPIP to a butadienyllithium chain end increases the amount of trans chain end configuration; TMEDA is much less effective in favoring the trans chain end configuration. In the presence of DIPIP, the charge distribution difference between the γ carbons for the cis and trans chain ends is quite large (ca. 8–11 ppm). Thus, a self-consistent proposal is that high vinyl polybutadiene microstructure is promoted by trans chain end configurations that have more negative charge on the γ carbons (more shielded, lower δ) than do the cis chain end configurations [75]. This proposal also rationalizes the fact that the 1,2-content in polybutadiene in the presence of TMEDA levels out at 85% at 5°C even when R = 6.7, in contrast to DIPIP, which generates 99% 1,2-enchainment at R = 2 [80]. Thus, even at high R values, a substantial fraction of the chain ends are in the cis configuration in the presence of TMEDA [80], whereas most of the chains (93%) appear to be in the trans configuration at R = 2 for DIPIP [94].

The behavior of the isoprenyllithium chain end is quite different from the butadienyllithium chain end. Addition of both DIPIP and TMEDA converts the predominantly trans chain end configuration to the cis chain end configuration. However, once again there is an unusually large chemical shift difference between the γ-carbon chemical shifts for the cis and trans chain end configurations.

It is also interesting to note that as the 1,2-content in the polybutadiene increases, the rate of polymerization also increases with increasing amounts of



Table 9.11 13C-NMR Characterization of Neopentylallyllithiuma and Neopentylmethylallyllithiumb,c [94]

% trans γ-Carbon Chemical

[Base]/ Chain End Shift (ppm) % 1,2

Base [Li] Configuration Trans Cis in Polymer

Reference

Butadiene

DIPIP

0 77 101.6 102.0

0.15 77 98 Broad 98 Broad 62 (R = 0.1)

0.61 76 82.4 89.7 90 (R = 0.5)

1.09 86 74.0 85.4

2.10 93 70.9 82.2 99 (R = 2)

TMEDA

2 61 72.2 72.2

Overlapped overlapped

DIDIOXd 0.25 65 99 Broad 99 Broad 85 (R = 0.2)

82

0.6 59 89.2 89.2 95–96 (R = 1)

82

3.0 59 88.3 83.4 82

THFe 35 79.6 81.9

Isoprene

DIPIP

0 66 104 102

0.13 60 100 Broad 100 Broad

0.58 19 84.2 90.8

0.95 15 n.m.f 90.3 62% vinyl at R = 1.2

1.80 20 72.0 90.0

3.65 32 69.9 89.7

TMEDA 2 0 n.m.f n.m.f



THFd 0 — 83.9

aAdduct of t-butyllithium and butadiene.

bAdduct of t-butyllithium and isoprene.

cCyclopentane, O°C.

d [Li] = 10-2 M.

eBulk THF.

fNot measured.



Table 9.12 Propagation Rates and Microstructure for Butadiene as a Function of Increasing Amounts of Bispiperidinoethane (DIPIP)a

[DIPIP]/[Li] Kp × 106 % Vinyl [DIPIP]/[Li] Kp × 106 % Vinyl

0 2.59 11 0.663 24.7 93

0.095 5.33 62 0.948 40.8 97

0.190 8.11 77 1.42 55.2 98

0.284 12.0 82 1.90 59.2 99

0.497 17.9 89

aActive chain end concentration of 1.2 × 10-3M, 5°C.

Source: Ref. 95.

DIPIP as shown by the data in Table 9.12. In fact, it is reported that there is a linear correlation between the rate of polymerization and the 1,2-content [94,95]. At a [DIPIP]/[Li] ratio of 2.0, the rate of propagation is first order in active chain end concentration that is consistent with the predominant species that adds monomer being the unassociated chain end [94].

Addition of strongly coordinating bases such as DIPIP and TMEDA also causes changes in the ultraviolet (UV) spectra of poly(butadienyl)lithium. The original absorption maximum at 276 nm decreases in intensity and is replaced by a new absorption band at 328 nm [94–97]. Analogous UV spectral shifts are observed for oligomeric poly(isoprenyl)lithium chain ends upon addition of TMEDA [98,99]; the initial absorption at 273 nm is replaced initially by a new species at 257 nm (R = 0.5; trans to cis chain end isomerization) and simultaneously a new absorption appears between 320 and 325 nm. Complete conversion to the longer-wavelength species required R values of 2 for DIPIP. It is proposed that the final unassociated species that adds monomer to form predominantly polybutadiene with high 1,2-content is a strongly solvated, ion-paired species whose exact nature has not been established [94]. The observation that very high 1,2-microstructure polybutadienes are produced even at relatively low R values (90% 1,2 at R = 0.5 for DIPIP; see Table 9.11) where a variety of associated and unassociated, base-complexed and uncomplexed species exist, suggests either that the unassociated, base-complexed species is much more reactive towards monomer addition to give 1,2-content units, or that other species also can generate these vinyl units [2]. A further curious aspect of the effect of Lewis bases is the fact that in the presence of strongly coordinating bases, 1,2-units are observed for polyisoprene (see Table 9.10). The formation of 1,2-units requires the formation of the less stable 1,4-chain end units (see structures 4 vs. 5) that is only observed at higher R values [49,99].

One interesting and surprising phenomenon observed in alkyllithium-initiated polymerization of butadiene in the presence of TMEDA is that cycliza-



tion to form in-chain vinylcyclopentane units (up to 60%) is observed when the butadiene monomer is introduced into the reactor at low rates (Eq. 9.4) [100,101].

(9.4)

Under such conditions, propagation does not effectively compete with cyclization; similar results are obtained with DIPIP [102]. With respect to the mechanistic requirements for this type of cyclization, it was reported that batch polymerization in THF/TMEDA (92/2, v/v) at 0°C showed no evidence of these cyclic units although the vinyl content was almost 90% [100,103]. This reaction forms a relatively unstable secondary alkyllithium from a resonance-stabilized allyllic lithium, which would appear to be energetically unfavorable (see Chapter 2). However, it should be noted that this process also converts a π-bond into a more stable sigma bond as in any vinyl polymerization. The generality of this cyclization process was demonstrated by showing that significant amounts of cyclization are observed using sodium as counterion in the presence of TMEDA and also with lithium only complexed with THF [101].

The ability to prepare polydienes with variable microstructures is an important aspect of alkyllithium-initiated anionic polymerization. For example, the glass transition temperature of polybutadiene is an almost linear function of the % 1,2 configuration in the chain as shown in Figure 9.2 [9,104]. Thus, while cis-1,4-polybutadiene has a glass transition temperature of -113°C, 1,2-polybutadiene has a glass transition temperature of -4°C [105]. This has practical consequences because the properties of polybutadienes with medium vinyl contents (e.g. 50%) have glass transition temperatures (ca. -60°C) and properties analogous to styrene-butadiene rubber (SBR) [104–107].

F. Temperature Effects on Diene Microstructure

Another interesting feature of the effect of Lewis base additives on diene microstructure is the fact that the vinyl microstructure generally decreases with increasing temperature as shown in Table 9.9 and illustrated in Figure 9.3 [42,79]. The effects of temperature on polydiene stereochemistry can be expressed by an Arrhenius-type equation and the resulting thermodynamic parameters are listed in Table 9.13. The relative insensitivity of diene microstructure to temperature in hydrocarbon media as illustrated in Figure 9.1 is confirmed by the small enthalpy and entropy differences for 1,2 vs. 1,4 and cis-1,4 vs. trans-1,4, especially for butadiene. The larger dependence of microstructure on temperature in polar media is shown by the larger differences in activation parameters. A simple explanation for the temperature dependence of vinyl microstructure, compared to the lack of



Figure 9.2. Variation of Tg with vinyl (1,2) content for polybutadiene. (From

Ref. 104; reprinted by permission of Plenum Press.)

dependence of microstructure in hydrocarbon media with lithium as counterion, is that high vinyl microstructure is associated with the addition of monomer to a base-coordinated chain end and this base coordination is reversed (less favorable) at higher temperatures.

The sensitivity of the microstructure to polymerization temperature depends on the Lewis base and the R value ([base]/[Li]) as shown in Table 9.9. Although the strongly chelating bidentate bases promote 1,2-polybutadiene microstructure at low temperatures (see Table 9.10), they generally exhibit a dramatic decrease in their ability to promote vinyl microstructure at elevated temperatures as shown in Table 9.9. An apparent exception is DIDIOX, which exhibits less dependence and is an effective promoter even at higher temperatures [83]. This temperature dependence presents a particular problem in high temperature processes (e.g., commercial batch or continuous) in which medium vinyl polybutadienes are desired [104–106].

G. Salt Effects

One of the most ubiquitous impurities in alkyllithium-initiated polymerizations is the corresponding lithium alkoxide or lithium hydroxide that can be formed by



Table 9.13 Enthalpy and Entropy Differences for Different Modes of Diene Addition in Anionic Polymerization as a Function of Solventa

∆v-∆1,4b ∆ cis-1,4-∆trans1,4

Monomer Solvent ∆H ∆S ∆H ∆S Reference

Butadiene Cyclohexane 1.16 0.38 0.90 0.18 42

Cyclohexane/THFc -3.67 11.8 -0.11 1.22 42

Dioxane -2.9 -5.5 62

Isoprene Benzene 0.5 4.0 -2.0 -4.0 108

Dioxane -5.7 -14.7 108

-3.3 -8.3 1.0 4.0 62

a∆H in kcal/mole; ∆S in cal/mole °K.

b∆v refers to the sum of 1,2 and 3,4 addition for polyisoprene.

cTHF concentration = 1 phm (per 100 monomer).

Figure 9.3. Effect of temperature on the microstructure of polybutadiene

prepared by n-butyllithium-initiated polymerization in cyclohexane in the presence

of 1.0 phm THF. (From Ref. 42; reprinted by permission of John Wiley & Sons, Inc.)



reaction of initiator with peroxides, molecular oxygen, or water, respectively [109]. The question then arises as to what the effect of this potential impurity is on the stereochemistry of anionic diene polymerization. At equivalent ratios of lithium sec-butoxide to active chain ends, increased vinyl contents have been observed [110]. Systematic studies of the effects of lithium t-butoxide on the stereochemistry and rates have been reported by Hsieh [111]. In cyclohexane, there is little effect of temperature or R ([LiOBu]/[PLi]) on the amount of vinyl incorporation into the polymer as shown by the data in Figure 9.4 [111]. However, in toluene at 30°C, relatively high vinyl enchainment (>30%) was reported as shown in Figure 9.4 [111]. Analogous to the effects of other weak polar bases, the vinyl content decreases at 70°C compared to 30°C as shown in Figure 9.4. Somewhat higher vinyl contents ( 60%) were reported for polymerizations in heptane at 25°C by Makowski and Lynn [112], although the polymers were quite low in molecular weight; the effect of lithium alkoxide was also shown to decrease dramatically with increasing molecular weight. The effect of added lithium alkox-

Figure 9.4. Effect of lithium t-butoxide on the vinyl unsaturation of

polybutadiene initiated with n-butyllithium: (A) at 30°C in toluene; (B) at 70°C in toluene; (C) at

70°C in cyclohexane; (D) at 30°C in cyclohexane. (From Ref. 111; reprinted by




ide leveled off at an R value of approximately 4. It was also reported that lithium phenoxide, lithium hydride, lithium hydroxide, and the products from the reaction of butyllithium and phenylacetylene and 1,6-heptadiyne had little effect on poly-butadiene microstructure, although large effects were observed for added hexynyllithium (55% 1,2 at R = 2) [112]. Worsfold and Bywater [113] reported that added lithium hydroxide and lithium butoxide exert relatively minor effects on polyisoprene microstructure. Thus, lithium alkoxides exert relatively small effects on the amount of 1,4-microstructure for polybutadienes and polyisoprenes at nominal molecular weights.

In contrast to the small effects of added lithium alkoxide, addition of other alkali metal alkoxides increase the amount of vinyl microstructure analogous to the microstructure obtained with the corresponding alkali metal counterions [114]. Thus, the maximum vinyl contents (%) were 67, 48, 55, and 53 for the sodium, potassium, rubidium, and cesium alkoxides, respectively, at R values of approximately one (see Figures 9.5–9.8) [114]. However, the effect of added alkali metal alkoxide decreased with increasing temperature for every alkali metal alkoxide analogous to the effect of increasing temperature in the presence of added Lewis bases (see Figure 9.3); for example as shown by comparison of the data in Figures 9.9 with Figure 9.5, at an [Na]/[Li] ratio of 0.5, >60% vinyl polybutadiene was obtained at 30°C, but <30% at 50°C.

Figure 9.5. Microstructure of polybutadiene obtained with sodium

t-butoxide-n-butyllithium at 30°C in cyclohexane. (From Ref. 114; reprinted by




Figure 9.6. Microstructure of polybutadiene obtained with potassium

t-butoxide-n-butyllithium at 30°C in cyclohexane. (From Ref. 114;


Figure 9.7. Microstructure of polybutadiene obtained with rubidium





Figure 9.8. Microstructure of polybutadiene obtained with cesium



H. trans-1,4-Polybutadiene

The preparation of high trans-1,4-polybutadiene is of current interest because it has been reported that these polymers can exhibit strain-induced crystallization analogous to natural rubber [115]. The trans-1,4 content of polybutadiene prepared in hydrocarbon solvents with alkyllithium initiators is in the range of 55–60%, which is not sufficiently stereoregular to undergo strain-induced crystallization (ca. 80% trans-1,4). One class of initiators that forms high trans-1,4-polybutadiene contain barium salts [104,116]. For example, a polybutadiene with 79% trans-1,4 and 7% vinyl (Tg = 91°C) was prepared from an initiator from n-butyllithium with 0.5 equivalents of a barium (t-butoxide-hydroxide) salt with 9 mole % hydroxide ion in toluene at 30°C [116]. The trans-1,4-content decreased at higher temperatures. A more recent modification utilizes a catalyst formed from a barium(t-alkoxide-hydroxide) or a barium(t-alkoxide)2 salt with a complex of a dialkylmagnesium and a trialkylaluminum [104,115,116]. Polybutadienes with 90% 1,4-content were prepared when the ratio of [Ba]/[Mg] was approximately 0.20 and the [Mg]/[Al] ratio was 6 for polymerizations in cyclohexane at 50°C [116]. The trans-content decreased with increasing temperature, but was as high



Figure 9.9. Microstructure of polybutadiene obtained with sodium

t-butoxide-n-butyllithium at 50°C in cyclohexane. (From Ref. 114; reprinted by permission of

John Wiley & Sons, Inc.)

as 79% even at 80°C. In contrast to the Ba/Li systems that undergo chain transfer in toluene, the Ba/Mg/Al systems were reported to exhibit the chacteristics of a living polymerization.

Another initiator system for preparation of high trans-1,4-polybutadiene is based on the complex formed from dibutylmagnesium and potassium t-amyloxide in hydrocarbon solvent [117]. Trans-1,4-polybutadiene was prepared in variable yields (50–100%) with [KOAm]/[R2Mg] 1, [KOAm]/[Bu3MgK] = 1.5–2.3, and [KOAm]/[R3MgNa] 2.4–10. With respect to the mechanism of stereoregulation, propagation via an organopotassium species was deduced based on the observation that trans-1,4-polybutadiene could also be generated from an initiator formed from potassium metal and potassium t-amyloxide. It should be noted that, in general, two polymer fractions were obtained: one a soluble, high 1,2-polymer and the other fraction was the insoluble trans-1,4-polybutadiene. The insolubility of the trans polymer prepared using these initiators is in contrast to the solubility of the trans-polybutadienes based on the barium systems. High trans-1,4-polybutadienes were also obtained with initiators based on a mixture of potassium



t-amyloxide and n-butyllithium ([KOAm]/[RLi] > 4) [118]. Two polymer fractions, soluble (high vinyl) and insoluble (high trans-1,4), were observed with this initiator as with the corresponding magnesium-based systems.

I. Polystyrene Stereochemistry

The homogeneous alkyllithium-initiated polymerization of styrene in hydrocarbon media was reported to produce polystyrene with an almost random (i.e., atactic) microstructure [2]. Thus, it was reported that the racemic diad fraction (Pr) was 0.53 for the butyllithium/toluene system [119]. The effect of counterion, solvent, and temperature on polystyrene stereochemistry is shown in Table 9.14. The principal conclusion from these results is that the stereoregularity of polystyrenes prepared by anionic polymerization is predominantly syndiotactic (Pr = 0.66–0.71) and that the stereoregularity is surprisingly independent of the nature of the cation, the solvent, and the temperature, in contrast to the sensitivity of diene stereochemistry to these variables. In view of these results, it is interesting to note that isotactic polystyrene was reportedly formed from butyllithium-initiated polymerizations in toluene, benzene or hexane at -30°C [122–124]. Subsequent careful studies by Worsfold and Bywater [125] failed to reproduce these results using rigorously dried monomer and solvent. However, when small amounts of water were deliberately added to the butyllithium, it was possible to prepare polystyrene with as much as 85% insoluble in refluxing methyl ethyl ketone and identified as isotactic polystyrene by X-ray crystallography. It is also noteworthy that these polymerizations were quite slow under these conditions.

Table 9.14. Stereoregularity of Polystyrenes Prepared with Anionic Initiators

Stereochemistry

Counterion

Solvent

Temperature (°C)

mm

mr

rr

Pr

Li THF -78 0.10 0.32 0.58 0.74

20 0.12 0.37 0.51 0.69

Toluene -20 0.13 0.42 0.45 0.66

20 0.07 0.41 0.52 0.73

K THF -78 0.09 0.34 0.57 0.74

Cs THF -78 0.14 0.35 0.51 0.69

Na Toluene 10 0.15 0.40 0.45 0.65

K 0.22 0.37 0.41 0.59

Rb 0.21 0.44 0.35 0.57

Cs 0.24 0.41 0.35 0.56

Source: Refs. 120, 121.



Recent studies have shown that isotactic polystyrene (10–22 % crystalline) can be prepared when lithium t-butoxide is added to n-BuLi initiator and the polymerization in hexane (styrene/hexane = 1) is effected at -30°C [126,127]. This polymerization was reported to become heterogeneous and to be quite slow (after 2–5 days, 50% monomer conversion; 13–20% conversion to isotactic polymer).

II. Conclusions

The stereochemistry of polydienes prepared by anionic polymerization depends on the counterion, the solvent, the temperature, chain end concentration, monomer concentration, and the presence of polar additives including salts. Lithium-based initiators in hydrocarbon media are unique in producing polydienes with high 1,4-microstructure. The highest cis-1,4-polydienes are obtained at high monomer concentrations and low chain end concentrations. The use of other alkali metal counterions, Lewis base additives, and polar solvents decreases the 1,4-content of polydienes and can result in polymers with high vinyl microstructures.

References

1. F. W. Stavely, F. C. Forster, L. L. Binder, and L. E. Forman, Ind. Eng. Chem., 48, 778 (1956).

2. S. Bywater, in Comprehensive Polymer Science, Vol. 3, Chain Polymerization I, G. C. Eastmond, A. Ledwith, S. Russo, and P. Sigwalt, Eds., Pergamon Press, Elmsford, New York, 1989, Chapter 28, p. 433.



5. H. L. Hsieh, Chemtech, 626 (1981).

6. R. S. Silas, J. Yates, and V. Thornton, Anal. Chem., 31, 529 (1959).

7. J. Haslam, H. A. Willis, and D. C. M. Squirrel, Identification and Analysis of Plastics, 2nd ed., Butterworth, London, 1972.

8. J. A. Frankland, H. G. M. Edwards, A. F. Johnson, I. R. Lewis, and S. Poshyachinda, Spectrochim. Acta (A), 47, 1511 (1991).

9. A. Yoshioka, K. Komuro, A. Ueda, H. Watanabe, S. Akita, T. Masuda, and A. Nakajima, Pure Appl. Chem., 58, 1697 (1986).

10. R. R. Hampton, Rubber Chem. Technol., 41, 47 (1968).

11. S. W. Cornell and J. L. Koenig, Rubber Chem. Technol., 43, 322 (1970).

12. S. W. Cornell and J. L. Koenig, Rubber Chem. Technol., 43, 313 (1970).

13. J. L. Koenig, Chemical Microstructure of Polymer Chains, Wiley-Interscience, New York, 1980.

14. S. Poshyachinda, H. G. M. Edwards, and A. F. Johnson, Polymer, 32, 338 (1991).

15. H. J. Harwood, Rubber Chem. Technol., 55, 769 (1982).



16. G. van der Velden, C. Didden, T. Veermans, and J. Beulen, Macromolecules, 20, 1252 (1987).

17. H. Y. Chen, Anal. Chem., 34, 1793 (1962).

18. H. Sato and Y. Tanaka, J. Polym. Sci., Polym. Chem., Ed., 17, 3551 (1979).

19. H. Y. Chen, Anal. Chem., 34, 1134 (1962).

20. K.-F. Elgert, G. Quack, and B. Stutzel, Polymer, 16, 154 (1975).

21. H. Sato, K. Takebayashi, and Y. Tanaka, Macromolecules, 20, 2418 (1987).

22. G. P. M. van der Velden and L. J. Fetters, Macromolecules, 23, 2470 (1990).

23. M. W. Duch and D. M. Grant, Macromolecules, 3, 165 (1970).

24. H. Sato, A. Ono, and Y. Tanaka, Polymer, 18, 580 (1977).

25. Y. Tanaka and H. Sato, Polymer, 17, 113 (1976).

26. A. S. Khatchaturov, E. R. Dolinskaya, L. K. Prozenko, E. L. Abramenko, and V. A. Kormer, Polymer, 18, 871 (1977).

27. D. H. Beebe, Polymer, 19, 231 (1978).

28. K. Hayashi, M. Tachibana, and S. Okamura, J. Polym. Sci., Chem. Ed., 18, 3381 (1980).

29. Q. T. Pham, R. Petiaud, H. Waton, and M.-F. Llauro-Darricades, Proton and Carbon NMR Spectra of Polymers, CRC Press, Boca Raton, FL, 1991, p. 158.

30. H. Morita and A. V. Tobolsky, J. Am. Chem. Soc., 79, 5853 (1957).

31. A. V. Tobolsky and C. E. Rogers, J. Polym. Sci., 40, 73 (1959).

32. F. C. Foster and J. L. Binder, Advances in Chemistry Series No. 19, Am. Chem. Soc., 1957, p. 26.

33. R. S. Stearns and L. E. Forman, J. Polym. Sci., 41, 381 (1959).

34. L. E. Forman, in Polymer Chemistry of Synthetic Elastomers, J. P. Kennedy and E. Tornqvist, Eds., Part II, Wiley-Interscience, New York, p. 522.

35. M. Morton and J. R. Rupert, in Initiation of Polymerization, F. E. Bailey, Jr., Ed., ACS Symposium Series 212, American Chemical Society, Washington, D.C., 1983, p. 284.

36. G. C. Meyer and J. M. Widmaier, Macromolecules, 14, 823 (1981).

37. A. E. Oberster, T. C. Bouton, and J. K. Valaitis, Angew. Makromol. Chem., 29/30, 291 (1973).

38. W. Gerbert, J. Hinz, and H. Sinn, Makromol. Chem., 144, 97 (1971).

39. M. M. F. Al-Jarrah, R. L. Apikian, and E. Ahmed, Polym. Bull., 12, 433 (1984).

40. R. Milner and R. N. Young, Polymer, 23, 1636 (1982).

41. S. Bywater, D. J. Worsfold, and G. Hollingsworth, Macromolecules, 5, 389 (1972).



50. K.-F. Elgert and W. Ritter, Makromol. Chem., 777, 2021 (1976).

51. J. Inomata, Makromol. Chem., 135, 113 (1970).

52. T. Suzuki, Y. Tsuji, and Y. Takegami, Macromolecules, 11, 639 (1978).

53. S. Kohjiya, S. Fujiwara, K. Fujii, and S. Yamashita, Makromol. Chem., 183, 163 (1982).

54. R. Ohno, Y. Tanaka, and M. Kawakami, Polym. J., 4, 56 (1973).

55. T. Suzuki, Y. Tsuji, Y. Takegami, and H. J. Harwood, Macromolecules, 12, 234 (1979).

56. Y.-X. Ding and W. P. Weber, Macromolecules, 21, 2672 (1988).

57. Y.-X. Ding and W. P. Weber, Macromolecules, 21, 530 (1988).

58. C. Petzhold, R. Stadler, and H. Frauenrath, Makromol. Chem., Rapid Commun., 14, 33 (1993).

59. H. Yuki, Y. Okamoto, and H. Takano, Polym. J., 2, 663 (1971).

60. X. Zhongde, J. Mays, C. Xuexin, N. Hadjichristidis, F. C. Shilling, H. E. Bair, D. S. Pearson, and L. J. Fetters, Macromolecules, 18, 2560 (1985).

61. S. Bywater, Y. Firat, and P. E. Black, J. Polym. Sci., Polym. Chem. Ed., 22, 669 (1984).

62. R. Salle and Q. T. Pham, J. Polym. Sci., Polym. Chem. Ed., 15, 1799 (1977).


64. Unpublished work of S. Bywater, quoted in ref. 2.

65. A. Essel and T. Pham, J. Polym. Sci., A-1, 10, 2793 (1972).

66. C. J. Dyball, D. J. Worsfold, and S. Bywater, Macromolecules, 12, 819 (1979).

67. S. Bywater and D. J. Worsfold, J. Organometal. Chem., 159, 229 (1978).

68. S. Bywater, in Anionic Polymerization. Kinetics, Mechanisms, and Synthesis, J. E. McGrath, Ed., ACS Symposium Series 166, American Chemical Society, Washington, D.C., 1981, p. 71.

69. S. Brownstein, S., Bywater, and D. J. Worsfold, Macromolecules, 6, 715 (1973).

70. P. Lachance and D. J. Worsfold, J. Polym. Sci., Polym. Chem. Ed., 11, 2295 (1973).

71. W. H. Glaze, J. E. Hanicak, J. Chaudhuri, M. L. Moore, and D. P. Duncan, J. Organometal. Chem., 51, 13 (1973).

72. A. Garton and S. Bywater, Macromolecules, 8, 694 (1975).

73. R. T. McDonald and S. Bywater, Organometallics, 5, 1529 (1986).

74. R. T. McDonald, S. Bywater, and P. Black, Macromolecules, 20, 1196 (1987).

75. S. Bywater, in Recent Advances in Anionic Polymerization, T. E. Hogen-Esch and J. Smid, Eds., Elsevier, New York, 1987, p. 187.

76. R. T. McDonald and S. Bywater, Br. Polym. J., 19, 387 (1987).



83. J. E. Hall (assigned to Firestone Tire & Rubber Company), U.S. patent 4,520,123 (1985).

84. S. Bywater, P. Lachance, and P. E. Black, Eur. Polymer J., 25, 801 (1989).

85. S. Raynal, J. Macromol. Sci., Chem., A19, 1049 (1983).

86. S. Bywater, D. H. Mackerron, D. J. Worsfold, and F. Schue, J. Polym. Sci., Polym. Chem. Ed., 23, 1997 (1985).

87. S. Dumas, J. Sledz, F. Schue, S. Raynal, S. Bywater, and D. J. Worsfold, Polymer, 24, 1291 (1983).

88. P. Bres, M. Viguier, J. Sledz, F. Schue, P. E. Black, D. J. Worsfold, and S. Bywater, Macromolecules, 19, 1325 (1986).

89. J.-M. Lehn, Account. Chem. Res., 11, 49 (1978).

90. R. P. Quirk, in Anionic Polymerization: Kinetics, Mechanisms, and Synthesis, J. E. McGrath, Ed., ACS Symposium Seriess 166, Am. Chem. Soc., Washington, D.C., 1981, p. 117.

91. R. P. Quirk and D. McFay, J. Polym. Sci., Polym Chem. Ed., 19, 1445 (1981).

92. R. P. Quirk and D. McFay, J. Polym. Sci., Polym. Chem. Ed., 24, 827 (1986).

93. R. Milner, R. N. Young, and A. R. Luxton, Polymer, 24, 543 (1983).

94. D. J. Worsfold, S. Bywater, F. Schue, J. Sledz, and V. Marti-Collet, Makromol. Chem., Rapid Commun, 3, 239 (1982).

95. S. Bywater and D. J. Worsfold, in Advances in Elastomers and Rubber Elasticity, J. Lal and J. E. Mark, Eds., Plenum, New York, 1986, p. 37.

96. S. Poshyachinda, H. G. M. Edwards, and A. F. Johnson, Polymer, 32, 930 (1991).

97. S. W. Medina, L. J. Fetters, and R. N. Young, Eur. Polym. J., 22, 267 (1986).

98. V. Collet-Marti, S. Dumas, J. Sledz, and F. Schue, Macromolecules, 15, 251 (1982).

99. P. Bres, M. Viguier, J. Sledz, F. Schue, P. E. Black, D. J. Worsfold, and S. Bywater, Macromolecules, 19, 1325 (1986).

100. G. Quack and L. J. Fetters, Macromolecules, 11, 369 (1978).

101. A. R. Luxton, M. Burrage, G. Quack, and L. J. Fetters, Polymer, 22, 382 (1981).

102. R. Milner and R. N. Young, unpublished observations reported in ref. 43.

103. L. J. Fetters and E. Firer, Polymer, 18, 306 (1977).

104. S. L. Aggarwal, T. G. Hargis, R. A. Livigni, H. J. Fabris, and L. F. Marker, in Advances in Elastomers and Rubber Elasticity, J. Lal and J. E. Mark, Eds., Plenum, New York, 1986, p. 17.

105. I. G. Hargis, R. A. Livigni, and S. L. Aggarwal, in Developments in Rubber Technology-4, A. Whelan and K. S. Lee, Eds., Elsevier, Essex, UK, 1987, p. 1.

106. E. W. Duck, in Molecular Behavior and Development of Polymeric Materials, A. Ledwith and



112. H. S. Makowski and M. Lynn, J. Macromol. Sci. Chem., A2, 683 (1968).



115. R. A. Livigni, I. G. Hargis, H. J. Fabris, and J. A. Wilson, J. Appl. Polym. Sci., Appl. Polym. Symp., 44, 11 (1989).

116. I. G. Hargis, R. A. Livigni, and S. L. Aggarwal, in Synthesis and Properties of Diene Elastomers with Controlled Structure, J. Mark and J. Lal, Eds., ACS Symposium Series 193, American Chemical Society, Washington, D.C., 1982, p. 73.

117. D. B. Patterson and A. F. Halasa, Macromolecules, 24, 1583 (1991).

118. D. B. Patterson and A. F. Halasa, Macromolcules, 24, 4489 (1991).

119. T. Kawamura, T. Uryu, and K. Matsuzaki, Makromol. Chem., Rapid Commun., 3, 661 (1982).

120. S. Suparno, J. Lacoste, S. Raynal, J. Sledz, and F. Schue, Polym. J., 13, 313 (1981).

121. S. Raynal, Polym. J., 15, 585 (1983).

122. R. J. Kern, Nature, 187, 410 (1960).

123. D. Braun, W. Betz, and W. Kern, Makromol. Chem., 42, 89 (1960).

124. D. Braun, in The Stereochemistry of Macromolecules, Band II, A. D. Ketley, Ed., Marcel Dekker, New York, 1967, p. 1.

125. D. J. Worsfold and S. Bywater, Makromol. Chem., 65, 245 (1963).

126. L. Cazzaniga and R. E. Cohen, Macromolecules, 22, 4125 (1989).

127. L. Cazzaniga and R. E. Cohen, Macromolecules, 24, 5817 (1991).



10 Copolymerization

I. Introduction

The ability to copolymerize monomers by chain reaction polymerization provides a powerful method for the variation and control of polymer properties. If two monomers can be polymerized in a random fashion, the resulting copolymer will possess properties intermediate between the properties of the two corresponding homopolymers. Unfortunately, few polymerizations proceed in a random fashion to incorporate both monomers uniformly into polymer chains throughout the course of the copolymerization. More often, one monomer is preferentially incorporated into the polymer chain initially; when that monomer is depleted in the monomer feed, the other monomer is incorporated into the polymer chains. This situation for a typical chain reaction polymerization proceeding with concurrent initiation, propagation, and termination results in copolymers with compositional heterogeneity between the copolymer chains formed initially (and terminated) compared to the copolymer chains formed at the later stages of the copolymerization.

The situation is quite different for living chain copolymerizations that proceed in the absence of chain termination and chain transfer. These differences can be most easily deduced by consideration of the Mayo-Lewis copolymerization equation, which is general for all chain reaction polymerizations regardless of whether the mechanism involves anionic, cationic, radical, or organometallic



reactive chain species; furthermore, it is independent of the mode of termination of chain growth [1].

The copolymerization of two monomers can be uniquely defined by four elementary kinetic steps (Eqs. 10.1–10.4), if it is assumed that the reactivity of the

(10.1)

(10.2)

(10.3)

(10.4)

chain end depends only on the last unit added to the chain end (i.e., there are no penultimate effects). The instantaneous copolymer composition equation deduced from these four equations is shown in Equation 10.5, where Mi refers to the

(10.5)

amount of monomer i incorporated into the copolymer at a time t, [mi] refers to the concentration of monomer i in the feed, and [Pi-] refers to the concentration of active anionic chain ends with monomer i at the chain end. If the steady-state assumption is made for chain ends P1- and P2-, as shown in Equation 10.6, and the monomer reactivity ratios defined in Equations 10.7 and 10.8 are substituted for the corresponding ratios of rate constants, the familiar instantaneous copolymerization equation (Eq. 10.9) is obtained that does not contain the concentration of active propagating chain ends.

(10.6)

(10.7)

(10.8)

(10.9)

At equal molar concentrations of monomer, Equation 10.9 is converted to Equation 10.10, from which initial copolymer compositions can easily be calculated recognizing that M1 + M2 = 1.



(10.10)

It is useful to review several important aspects of monomer reactivity ratios and copolymerization behavior, which are summarized as follows [2]:

Monomer reactivity ratios

Monomer reactivity ratios

r1 = 0 No homopolymerization of monomer 1; only crossover of P1- with monomer 2.

r1 = 1 Equal probability of P1- adding monomer 1 or monomer 2.

r1 >> 1 strong tendency for P1- to preferentially add monomer 1.

Copolymerization behavior

r1r2 = 0 Tendency to obtain alternating copolymer

r1 r2 = 1 Tendency to obtain random copolymer

r1 r2 >> 1

Tendency to obtain block copolymer

Scheme 10.1

Furthermore, it is important to note that the copolymer composition will tend to drift (i.e., dM1/dM2 will change with conversion), except in those special cases in which r1 = r2 = 1, a so-called ideal copolymerization [2]. When r1r2=1, then k11/k12=k21/k22 and monomer incorporation depends only on the relative reactivity of the monomers and is independent of the nature of the chain end (i.e., a random copolymer structure will be obtained).

II. Copolymerization in Hydrocarbon Solvents

With this general background information, it is possible to understand the anionic copolymerization behavior of a variety of monomer systems [3–6]. The monomer reactivity ratios for anionic copolymerization of several pairs of monomers are listed in Table 10.1. In general, monomer reactivity ratios are consistent with expectations based on the expected effects of substituents on stability of carbanions. Thus, styrene (M1) prefers to add styrene monomer (r1 > 1) relative to substituted styrene monomers (M2), which have electron-donating substituents that would be expected to decrease monomer reactivity and carbanion stability (r2 < 1). These electron-donating substituents include methyl, methoxy, and t-butyl groups. In contrast, electron-withdrawing groups



copolymerization of DPE with styrene compared to DPE copolymerizations with dienes and the copolymerization of dienes with styrene. Both isoprenyllithium and the butadienyllithium chain ends are quite unreactive with respect to crossover to 1,1-diphenylethylene (r1 >> 1) compared to the analogous reaction of the styryllithium chain end with 1,1-diphenylethylene (r1 < 1). The styrene-diene copolymerizations will be considered in detail.

In view of the similar stabilities of the corresponding carbanionic chain ends (see Chapter 2), the large disparity between monomer reactivity ratios observed for copolymerization of styrene and dienes is surprising. In hydrocarbon solution the alkyllithium-initiated copolymerizations of styrene with butadiene or isoprene are quite interesting and unusual and provide an excellent example of the importance of crossover reactions relative to homopolymerization reactivity in determining copolymer composition. The homopolymerization rate constants for styrene, isoprene and butadiene are 1.6 × 10-2(L/mole)1/2s+1, 1.0 × 10-3 (L/mole)1/4s-1 and 2.3 × 10-4 (L/mole)1/4s-1, respectively [3]. However, it is observed that butadiene or isoprene is incorporated into the copolymer more rapidly than styrene. Because of this and the fact that these are living polymerizations, copolymerization of mixtures of a diene and styrene in hydrocarbon solvent yields an unusual type of structure with compositional heterogeneity incorporated intra-molecularly along the polymer chain. This type of copolymer composition is described as either a tapered block copolymer or a graded block copolymer. The monomer sequence distribution can be described by the structures below (1): first

there is a diene-rich block; a middle block follows that is initially richer in butadiene with a gradual change in composition until eventually it becomes richer in styrene; a final block of styrene completes the structure. It is also interesting and important to note that addition of a small amount of Lewis base can give essentially a random copolymer (see Tables 10.2 and 10.3).

From a phenomenological perspective for a typical copolymerization of styrene and butadiene (25/75, mol/mol), the solution is initially almost colorless, corresponding to the dienyllithium chain ends, and the rate of polymerization is slower than the hompolymerization rate of styrene [6]. After approximately 70–80% conversion, the solution changes to orange-yellow, which is characteristic of styryllithium chain ends. At the same time, the overall rate of polymerization increases (inflection point); this has been described as an “inversion phenomenon”. This behavior is illustrated in Figures 10.1 and 10.2. In Figure 10.1, the rate acceleration after the inflection point is clearly indicated as well as the expected dependence on conversion. In Figure 10.2, the effects of solvent on the rates of



Figure 10.1. Copolymerization of butadiene and styrene in cyclohexane

at 50°C. (From Ref. 6; reprinted by permission of the Rubber Division of the

American Chemical Society.)

Figure 10.2. Copolymerization of butadiene and styrene in different solvents





polymerization and the time required to reach the inflection point are shown. It is noteworthy that the percentage conversion at which the inflection point is observed does not appear to depend on solvent even though the time to reach this percentage conversion is quite solvent-dependent. Analysis of the copolymer composition indicates that the total percentage styrene in the copolymer is less than 5% up to approximately 75% conversion (see Figure 10.3) [6,11]. For copolymerization at 40°C in toluene, at 75.6% conversion, the styrene content is 6.9 mole % [11]. After 80% conversion, corresponding to the inflection point in the conversion-time curve, the amount of styrene increases very rapidly. For example, at 89.3% conversion the percentage styrene has increased to 13.0 mole % [11]. Furthermore, when these samples are analyzed by oxidative degradation by ozonolysis, polystyrene segments (corresponding to polystyrene blocks in the copolymer) are recovered only after the inflection point is reached as shown in Figures 10.3 and 10.4 [6,11]. It is clear that the major portion of the styrene charged was not polymerized until the less reactive (with respect to homopolymerization reactivity) diene monomer was exhausted, but then it homopolymerized with concomitant increase in rate to form the polystyrene block. However, at low chain end concentrations (e.g., < 10-4M) it can be estimated that the rate of isoprene homopolymerization will be faster than the rate of styrene homopolymerization because of the difference in reaction order dependence on chain end concentrations for these two systems [3].

Figure 10.3. Styrene incorporation as a function of conversion for copolymerization of butadiene and styrene (75/25)





Figure 10.4. Ozonolysis-GPC curves of copolymer

samples at various conversions: (A) conversion 23.6% (styrene, 1.7 mol %); (B)

conversion 65.6% (styrene, 4.9 mol %); (C) conversion 75.6% (styrene, 6.9 mol %); (D)

conversion 89.3% (styrene, 13.0 mol %); (E) converison

100% (styrene, 20.6 mol %). (From Ref. 11; reprinted by permission of

the American Chemical Society.)

These results are easily accounted for in terms of the monomer reactivity ratios which are listed in Table 10.1. Although there is rather large scatter in the experimental results, it is apparent that in hydrocarbon solution rdiene > 1 and rstyrene < 1. The values for rdiene are in the range of 7–15, while the values for rstyrene are < 0.2. Using Equation 10.9, it is apparent that the copolymer initially formed from an equimolar feed ratio will correspond to > 90% diene content. Since the product (r1r2) 1.0, this corresponds to a random copolymerization [2]; thus, the placement of the styrene units into this initially formed polybutadiene-rich segment is expected to be random. For a polymerization proceeding with concurrent initiation, propagation, and termination, the polymer chains formed initially would be rich in diene monomer and the chains formed at the end of the copolymerization would be rich in styrene. Thus, there would be considerable inter-



Pa

Table 10.1 Anionic Copolymerization Parameters in Hydrocarbon Solution with Alkyllithium Initiators

M1

M2

Solvent

Temperature(°C)

r1

r2

Refere

Butadiene Styrene None 25 11.2 0.04 7

Benzene 30 10 0.035 8

25 10.8 0.04 7

Toluene 20 12.9 0.004 9

20 11.3 0.04 10

40 13 0.16 11

Cyclohexane 25 15.5 0.04 7

Cyclohexane 50 15.1 0.025 12

Heptane 30 7 0.1 13

Hexane 0 13.3 0.03 7

25 12.5 0.03 7

50 11.8 0.04 7

Isoprene Hexane 50 3.38 0.47 14

20 2.72 0.42 15

30 1.72 0.35 15

40 2.18 0.35 15

Benzene 40 3.6 0.5 16

1,1-Diphenylethylene Benzene 40 54 ˜0 17

(table continued on next page)



Pa

(table continued from previous page)

M1

M2

Solvent

Temperature(°C)

r1

r2

Refer

Isoprene Styrene Benzene 30 7.7 0.13 18

Toluene 27 9.5 0.25 19

Cyclohexane 40 16.6 0.046 20

1,1-Diphenylethylene Benzene 40 37 ˜0 21

Styrene α-Methylstyrene α-Methylstyrene 100 60 0.01 22

p-Methylstyrene Benzene 20 0.72 1.09 23

30 2.5 0.4 24

Toluene 0 2.5 0.26 25

p-t-Butylstyrene Benzene 20 1.34 0.86 23

p-Methoxystyrene Toluene 0 11 0.05 24

p-Divinylbenzene Benzenec 30 0.094 10.0 26

1,1-Diphenylethylene Benzene 30 0.7 ˜0 27

Vinyl trimethylsilane Neat monomers 23 18 0.06 28

o-MeO-styrene 1,1-Diphenylethylene Benzene 40 20 ˜0 29

p-divinylbenzene 1,1-Diphenylethylene Toluene -20 16 ˜0 30

m-Divinylbenzene 1,1-Diphenylethylene Toluene -20 2.5 ˜0 30

2,3-Dimethylbutadiene

1,1-Diphenylethylene Benzene 40 0.23 ˜0 31

a[PLi] = 0.001M.

b[PLi] = 0.36M.

c2% tetrahydrofuran.



molecular compositional heterogeneity. However, in an alkyllithium-initiated copolymerization of a diene and styrene, there is no termination or chain transfer occurring. This living polymerization will incorporate all of the compositional heterogeneity within each polymer chain (i.e., there will be intramolecular compositional heterogeneity). As a result, the structure of the copolymer will correspond to the tapered (or graded) block copolymer as shown previously (1). It is note-worthy that the extensive studies of Morton and Huang[4,7] clearly showed that the styrene—butadiene copolymerization parameters are relatively insensitive to temperature and hydrocarbon solvent, as expected for ratios of rate constants.

Another monomer pair that copolymerizes analogously to the styrene—diene pairs to form tapered block copolymers is the styrene—vinyltrimethylsilane system[28]. There is some tendency to form a tapered block-type structure also in butadiene—isoprene copolymerizations[14,15]. Thus, butadiene is preferentially incorporated initially even though it is the less reactive monomer; furthermore, the rate of copolymerization is retarded relative to the corresponding homopolymerization of isoprene.

A different situation is encountered in the copolymerizations of either styrene or diene monomers with 1,1-diphenylethylene. Because of steric effects, 1,1-diphenylethylene does not undergo homopolymerization (r2 = 0) and thus its monomer reactivity ratio is equal to zero. As a consequence, copolymerizations with styrene and dienes can be formally described as alternating, since the product (r1r2) = 0[2]. Because of the low ceiling temperature of α-methylstyrene (Tceil = 61°C)[32], it is possible to prepare alternating copolymers (r1r2 = 0) at temperatures > 61°C by using α-methylstyrene as solvent and adding comonomer in a continuous process[22,33].

A. Kinetics in Hydrocarbon Solution

One question that arises with regard to the copolymerization behavior of styrene and dienes is why there is an inversion phenomenon in hydrocarbon solution, that is, preferential incorporation of the less reactive (with respect to homopolymerization) monomer, butadiene or isoprene, which is reversed in polar solvents (see Tables 10.2 and 10.3). One straightforward explanation is available from kinetic studies of these copolymerizations. One interesting and important aspect of living anionic copolymerizations is that stable carbanionic chain ends can be generated and the rates of their crossover reactions with other monomers measured independently of the copolymerization reaction. Two of the four rate constants involved in copolymerization correspond at least superficially to the two homopolymerization reactions of butadiene and styrene (e.g., kBB and kSS) respectively. The other two rate constants can be measured independently as shown in Equations 10.11 and 10.12.

(10.11)

(10.12)



Table 10.2 Anionic Copolymerization Parameters in Polar Solvents with Alkyllithium Initiators

M1

M2

Solvent

Temperature (°C)

r1

r2

Butadiene Styrene THF -78 0.04 11.0

0 0.2 5.3

25 0.3 4.0

-35 0.2 8

Diethyl ether 25 1.7 0.4

30 1.78 0.11

Triethylamine 25 3.5 0.5

Anisole 25 3.4 0.3

Diphenyl ether 25 2.8 0.1

1,1-Diphenylethylene THF 0 0.13 ˜0

Isoprene Styrene 27 0.1 9

Triethylamine 27 1.0 0.8

1,1-Diphenylethylene THF 0 0.12 ˜0

Styrene p-Methylstyrene 0 1.3 0.9

p-Methoxystyrene 0 2.9 0.23

1,1-Diphenylethylene 30 0.13 ˜0

p-Methylstyrene p-Methoxystyrene 0 1.9 . 72

p-Divinylbenzene 1,1-Diphenylethylene -20 2.5 ˜0

m-Divinylbenezene 1,1-Diphenylethylene -20 1.2 ˜0



Table 10.3 Effect of Solvents on the Microstructure of Polybutadiene and on the Initially Formed Copolymers (Styrene-Isoprene) Prepared with Butyllithium Initiators at 25°C

Solvent

% 1,2-polybutadiene [7]

wt % Styrene in Copolymer [37]

Bulk 9 –

Benzene 9 15

Hexane 8 –

Cyclohexane 6 –

THF 70 80

Diethyl ether 52 68

Triethylamine 48 59

Anisole 50 –

Diphenyl ether 26 –

aWeight % of styrene incorporated into copolymer from an equimolar feed of styrene and isoprene (60/40 wt/wt).

Before describing the results of these kinetic investigations, it is important to consider the limitations and ambiguities associated with these types of measurements. First, the nature of the chain ends in anionic copolymerization should be considered. In general, both poly(styryl)lithium and poly(butadienyl)lithium chain ends are associated in solution (see Chapter 1); furthermore, mixtures of these chain ends would be expected to cross-associate. One simplifying feature of these reactions may be the fact that evidence is available suggesting that the unassociated chains are the reactive species in propagation reactions (see Chapter 7). If this is the situation, the problem of the effects of cross-association would not necessarily appear, except possibly as changes in the fractional kinetic dependencies for the chain-end concentration and for the equilibrium constants for dissociation of the aggregates. However, the possibility definitely exists that aggregated species could react directly with monomer [5].

One method of answering unambiguously the question of the effect of cross-association is to carry out the kinetic investigations of the crossover reactions in the presence of the other chain end as shown in Equations 10.13 and 10.14. The effects of cross-association have been investigated in the kinetic studies of Ohlinger and Bandermann [9]. The results of a number of independent kinetic studies of these crossover reactions can be summarized as follows [4,9]:



(10.13)

(10.14)

The numbers in parentheses under the rate constants are relative values of the rate constants. This kinetic order contains the expected order of homopolymerization rates with kSS > kBB. The surprising result is that the fastest reaction rate is associated with the crossover reaction of the poly(styryl)lithium chain ends with butadiene monomer (kSB); conversely, the slowest reaction rate is associated with the crossover reaction of the poly(butadienyl)lithium chain ends with styrene monomer (kBS. It is also significant to note that the sensitivity of these rate constants to the presence of the other active chain ends was examined and found to produce no significant differences.

The kinetic parameters have been determined also for the copolymerization of styrene and isoprene in cyclohexane at 40°C [20]. It was reported that the kinetic order dependencies on chain end concentration were 0.213 and 0.289 for the reaction of poly(isoprenyl)lithium with isoprene and styrene, respectively. Poly(styryl)lithium reacted with both isoprene and styrene with a one-half order dependencies on chain end concentration. It was concluded that although the chain ends are associated (see Chapter 1), it is the unassociated polymeric organolithium species which reacts with monomer (see also Chapter 7). Preliminary light-scattering measurements indicated that the degree of association of poly(iso-prenyl)lithium was between 3 and 4, and usually nearer 3. The results of these kinetic studies can be summarized as follows:

The form of these apparent rate constants includes the proposed dissociation constant (Ki) for the chain ends, that is, kij, (apparent) = kij Ki1/n in which kij

is the actual propagation rate constant and n is the proposed degree of association of the chain ends. It should be noted again, however, that because of the different dependencies on chain end concentration, kII>kSS at very low chain end concentrations [3]. Thus, the proposed mechanism of styrene-isoprene copolymerization is shown in Scheme 10.2, where the degree of association of poly(isoprenyl)-lithium is indicated to be four, in accord with the copolymerization kinetic reaction orders, but it should be noted that there is disagreement in the literature with respect to the degree of association (see Chapters 1 and 7) [5].

An early hypothesis to explain the unusual inversion of copolymer reactivity relative to homopolymerization rates in hydrocarbon media was that the copolymerization behavior is governed by preferential complexation of active chain



Scheme 10.2

ends with diene monomer. This rationalization was rendered unnecessary by the observations of no dependence of observed rates (e.g., Eqs. 10.13, 10.14) on excess diene monomer [3,20].

In conclusion, the copolymerization behavior of styrene and dienes is a classic example of the fact that the copolymerization behavior of two monomers cannot be deduced from the corresponding relative reactivities in homopolymerization. These carbanionic copolymerizations are dominated by the relative rates of the crossover reactions: thus, the chain end with the least reactive (homopolymerization) monomer, butadiene, reacts only sluggishly with the more reactive (homopolymerization) styrene monomer, while the chain ends with the most reactive monomer react very rapidly with the less reactive monomer.

It was mentioned that there is a wide range of values of monomer reactivity ratios in the literature. In general, most of these values are obtained using the copolymerization equation (Eq. 10.9) by analysis of the instantaneous copolymer composition at low degrees of conversion. Because there can be quite different rates of initiation compared to homopolymerization and crossover reactions, it is possible that these initial composition data may be in error. It is known, for example, that the less reactive monomer in these copolymerizations, styrene, reacts much more rapidly with the butyllithium initiator than do the dienes. A simplified model system dealing with this problem has been analyzed by Fueno and Furukawa [34] and their conclusion was that it is important to obtain data at higher degrees of polymerization (DP > 100) to avoid this problem. Nevertheless, analyses based on copolymerization composition data at low degrees of conversion give monomer reactivity ratios that are in reasonable agreement with those values determined by actual kinetic studies [3].

III. Copolymerization in Polar Solvents

Alkyllithium-initiated copolymerizations can also be carried out in polar solvents and representative corresponding monomer reactivity ratios are shown in Table



10.2. What is interesting about this data is that the “anomalous” copolymerization reactivity for styrenes and dienes observed in hydrocarbon solvents disappears in polar solvents such as tetrahydrofuran (i.e., styrene that homopolymerizes faster than dienes is preferentially incorporated into the copolymer initially). Thus, an “inverted” (relative to copolymerization in hydrocarbon) type of tapered block copolymer structure (2) would be expected for both isoprene and butadiene. There

is, however, another more important consideration; in polar solvents, the high 1,4-stereospecificity of lithium as counterion is lost and polydiene enchainments with high vinyl contents are formed (see Chapter 9). The vinyl content of polybutadienes formed in various polar media using alkyllithium initiators are listed in Table 10.3. It is apparent that although polar solvents such as THF can alter the copolymerization behavior for styrenes and dienes as shown by the effects of these solvents on styrene incorporation for styrene-isoprene copolymerizations, they have the disadvantage of concurrently increasing the amount of vinyl microstructure for polybutadiene. In fact, a positive correlation has been noted between the percentage of styrene in the copolymer and the amount of 1,2 (or 3,4 for isoprene) in the polydiene produced under the same conditions of solvent and initiator [38]. Because of this undesirable correlation, there was a need to find suitable additives that would promote random copolymerization of styrene and dienes without significantly increasing the vinyl microstructure of polydienes.

IV. Copolymerization With Polar Additives

A. Lewis Bases

If dieneis preferentially incorporated into the initially formed copolymer in hydrocarbon solution and styrene is preferentially incorporated into the initially formed copolymer in polar solvents such as THF, it might reasonably be expected that addition of appropriate, small amounts of Lewis bases would result in formation of a random copolymer with relatively low levels of vinyl microstructure. It has been reported that addition of polar modifiers to alkyllithium-initiated styrene-butadiene copolymerizations causes an increase in the rate of styrene incorporation resulting in a more random (i.e., less blocky) styrene-butadiene copolymer [39,40]. For example, for a 65/35 weight ratio of BD/S, whereas in the absence of modifier approximately 5% styrene was initially incorporated into the copolymer, this increased to 26% and 29% in the presence of one equivalent of diglyme and TMEDA, respectively, at 32°C. These values were 21% and 26% at 66°C. Monomer reactivity ratios for styrene (r1 = 0.91) and



butadiene (r2 = 0.86) have been reported for copolymerizations in the presence of one equivalent of TMEDA ([TMEDA]/[Li] = 1) [41]. At these levels of polar modifier, no polystyrene blocks are detected by oxidative degradation. It should be noted, however, that this more random styrene-butadiene copolymer will incorporate the butadiene predominantly as 1,2-type units. In the presence of one equivalent of base at 32°C (or 66°C), there was 76% (55%) and 65% (50%) 1,2-enchainment of butadiene for diglyme and TMEDA, respectively [39]. Even though the amount of 1,2-enchainment decreases with increasing temperature and the styrene incorporation is relatively insensitive to temperature, the amount of 1,2-enchainment is still quite high with these bases even at elevated temperatures.

A more delicate balance of copolymerization behavior and polydiene microstructure is apparently possible using tetrahydrofuran (THF) as polar modifier [13]. For example, using an equimolar mixture of styrene and butadiene in heptane at 30°C with a [THF]/[Li] ratio of only 0.42, the initially formed copolymer had 34 mole % styrene and only 15 % 1,2-addition; at higher relative concentrations of THF, no significant increases in styrene incorporation were observed, but the amount of 1,2-addition increased proportionately.

B. Alkali Metal Alkoxides

Addition of alkali metal alkoxides other than lithium is another effective method of altering the copolymerization behavior without simultaneously increasing the amount of 1,2-polydiene microstructure for styrene-butadiene copolymers [42]. Addition of alkali metal t-butoxides to n-butyllithium-initiated copolymerizations of butadiene and styrene increased the rate of copolymerization and eliminated the break in the time vs. conversion curve characteristic of tapered block copolymerizations as shown in Figure 10.5. In contrast, lithium t-butoxide retarded the rate of copolymerization while retaining the break. As shown in Figure 10.6, the addition of potassium t-butoxide in a molar ratio of 0.067 relative to the butyllithium initiator provided a copolymer from a butadiene/styrene (BD/S) mixture (75/25 by wt.) that contained approximately 24 wt.% styrene at 1% conversion and the amount of styrene did not vary significantly with conversion [42]. Equally important is the observation that potassium t-butoxide at this ratio produces a polybutadiene microstructure that contains only approximately 15% 1,2-microstructure at 50°C as shown in Figure 10.7 (see Chapter 9) [43]. It has been proposed that there is rapid exchange of the alkali metal counterions at the chain end when alkali metal alkoxides are added to butyllithium-initiated copolymerizations [42]. This is supported by the observation that the potassium t-butoxide-modified copolymerizations behave analogously to a phenyl potassium-initiated copolymerization in terms of the percentage of bound styrene, as shown in Figure 10.8. Since alkoxides can function as Lewis bases and coordinate to organolithium aggregates [44], their interaction with the propagating chain end could be to modify its ionic character through this coordination.



Figure 10.5. Time—conversion curves for copolymerization of butadiene

(75 parts) and styrene (25 parts) with equimolar mixtures of alkali metal tert-butoxides and n-butyllithium: (A) t-BuOK/BuLi at

30°C; (B) t-BuORb/BuLi at 30°C; (C) t-BuONa/BuLi at 30°C; (D) t-BuOCs/BuLi at 30°C; (E)

BuLi control at 50°C; (F) t-BuOLi/BuLi at 50°C. (From Ref. 41; reprinted by permission

of John Wiley & Sons, Inc.)

Figure 10.6. Sytrene (25 wt %) incorporation with potassium

tert-butoxide-butyllithium at 50°C at various t-BuOK/n-BuLi ratios: (A) 0.38; (B) 0.24;

(C) 0.067; (D) 0.027; (E) 0.022. (From Ref. 41; reprinted by permission of John Wiley & Sons, Inc.)



Figure 10.7. Microstructure of polybutadiene obtained with potassium

tert-butoxide/n-butyllithium at 50°C. (From Ref. 42; reprinted by permission of John Wiley & Sons, Inc.)

Figure 10.8. Styrene incorporation in butadiene/styrene (75/25) copolymerization in cyclohexane. (From Ref. 6;

reprinted by permission of the Rubber Division of the American Chemical Society.)



In contrast to the ineffectiveness of lithium alkoxides as promoters for styrene-diene copolymerizations, it has been reported that the lithium salt of ethylene glycol monomethyl ether promoted 34% styrene incorporation at 40% conversion (styrene/butadiene = 30/70 mole ratio) at the level of one equivalent relative to the chain end concentration at 40°C in toluene [45]. The analogous 2-dimethylamino-substituted lithium ethoxide was less effective in promoting styrene incorporation (14% styrene incorporation at 53% conversion) [46]. However, no information on polybutadiene microstructure was provided for these modifiers.

Bu and Ying [47] have combined the two modifiers, THF and potassium t-amyloxide, in varying ratios in cyclohexane at 50°C to control the copolymerization and composition of styrene—butadiene copolymers. By adjusting the ratio of K/Li and the ratio of THF/Li, copolymers with compositions identical to the feed ratios have been obtained.

V. Copolymer Structure Analyses

The structures of these “graded block” and “randomized” copolymers have been investigated by a number of techniques. Using a combination of high resolution 1H nuclear magnetic resonance (NMR), infrared (IR), dynamic mechanical analysis (DMA), and transmission electron microscopy (TEM), it has been concluded that the addition of 0.05 moles t-BuOK/mole n-butyllithium is sufficient to produce random copolymers with uniform composition distribution [48]. In general, the 1,2-vinyl content of the copolymers increased with increasing molar ratios of t-BuOK/n-butyllithium, varying from 11 to 18% at a ratio of 0.05 to 33% at a ratio of 0.1.

The structure and properties of tapered block copolymers of styrene and isoprene have been examined by a combination of 1H NMR, pyrolysis—gas chromatography, and electron microscopy [49]. Because of weaker repulsive interaction between the two block segments, a lower critical temperature and unique mechanical properties were observed for the tapered block copolymers than with the ideal block copolymers.

Kraus and co-workers [50] have concluded that dynamic storage and loss moduli or loss tangents determined by DMA are far more discriminating than T measurements in elucidating the structures of graded and random block copolymers.

Tanaka and co-workers [11,51] have investigated the sequence distribution of styrene-butadiene “graded” block copolymers by SEC measurements of ozonolysis products. The value of this method is that it provides direct information about the distribution of the long and short sequences of the styrene and 1,2-butadiene units. The general structure for the ozonolysis products (3) is shown below. At the initial stages of the polymerization, the styrene units exist predominantly as isolated sequences. Diad and triad sequences of styrene increased with increasing conversion up to 75%. At the final stage of polymerization, long block



sequences of styrene (number average sequence lengths of 43–65) were formed plus diad and triad sequences (see Fig. 10.4). The fact that the number-average length of the observed block styrene sequences was about 1.6 times that expected from the ratio of initial butyllithium and residual styrene was attributed to termination reactions due to small amounts of impurities such as water or oxygen and the transmetalation to toluene.

VI. Conclusions

The alkyllithium-initiated copolymerization of styrenes and dienes is complicated by the disparity of monomer reactivity ratios. In hydrocarbon solution, dienes are preferentially incorporated into the initially formed polymer resulting ultimately in a tapered or graded block copolymer structure. Analogously, in polar media such as THF, styrene is preferentially incorporated and an inverted type of tapered block structure is obtained. With the addition of small amounts of polar compounds, random copolymerization can be achieved. However, polar compounds promote increases in vinyl microstructure, which raises the Tg of the copolymer. Alkali metal alkoxides such as potassium t-amyloxide are hydrocarbon-soluble additives that promote random copolymerization without significant increases in vinyl microstructure provided they are added at low molar ratios ([KOR]/[Li] << 1).

References

1. F. R. Mayo and C. Walling, Chem. Rev., 46, 191 (1950).

2. G. Odian, Principles of Polymerization, 3rd ed., Wiley-Interscience, New York, 1991, p. 452.

3. S. Bywater, in Comprehensive Chemical Kinetics, Vol. 15, C. H. Bamford and C. F. H. Tipper, Eds., Elsevier, Amsterdam, 1976, Chapter 1.

4. M. Morton, Anionic Polymerization, Principles and Practice, Academic Press, Inc., New York, 1983.



7. L.-K. Huang, Ph.D. Dissertation, University of Akron, Akron, Ohio, 1979; quoted in ref. 4, p. 142.

8. A. A. Korotkov, Angew. Chem., 70, 85 (1958).

9. R. Ohlinger and F. Bandermann, Makromol. Chem., 181, 1935 (1980).

10. V. P. Shatalov and I. Y. Kirchevskaya, Vysokomol. Soedin, Ser. B:14, 602 (1972).



11. Y. Tanaka, H. Sato, Y. Nakafutami, and Y. Kashiwazaki, Macromolecules, 16, 1925 (1983).

12. V. D. Mochel, Rubber Chem. Technol., 40, 1200 (1967).

13. I. Kuntz, J. Polym. Sci., 54, 569 (1961).

14. G. V. Kakova and A. A. Korotkov, Dokl. Akad. Nauk SSSR, 119, 982 (1958); Rubber Chem. Technol., 33, 623 (1960).

15. D. J. T. Hill, J. H. O'Donnell, P. W. O'Sullivan, J. E. McGrath, I. C. Wang, and T. C. Ward, Polym. Bull., 9, 292 (1983).

16. J. Furukawa, T. Saegusa, and K. Irako, Kagaku Zasshi, 65, 2029 (1962); Chem. Abstr., 58, 11553 (1963).

17. H. Yuki and Y Okamoto, Bull. Chem. Soc. Jpn., 43, 148 (1970).

18. S. L. Agarwall, R. A. Livigni, L. F. Marker, and T. J. Dudek, in Block and Graft Copolymers, J. J. Burke and V. Weiss, Eds., Syracuse University Press, Syracuse, New York, 1973, p. 157.

19. Yu. L. Spirin, D. K. Polyakov, A. R. Gantmakher, and S. S. Medvedev, Polym. Sci. USSR, 3, 233 (1963).

20. D. J. Worsfold, J. Polym. Sci., A-1, 5, 2783 (1967).

21. H. Yuki and Y. Okamoto, Bull. Chem. Soc. Jpn., 42, 1644 (1969).

22. D. B. Priddy, T. D. Traugott, and R. H. Seiss, Polym. Prep., Am. Chem. Soc., Div. Polym. Chem., 30(2), 195 (1989).

23. J. Chen and L. J. Fetters, Polym. Bull., 4, 275 (1981).

24. K. F. O'Driscoll and R. Patsiga, J. Polym. Sci., Part A, 3, 1037 (1965).

25. A. V. Tobolsky and R. J. Boudreau, J. Polym. Sci., 51, s53 (1961).


27. H. Yuki, J. Hotta, Y. Okamato, and S. Murahaski, Bull. Chem. Soc. Jap., 40, 2659 (1967).

28. G. K. Rickle, J. Polym. Sci., A, Polym. Chem., 31, 113 (1993).

29. H. Yuki and Y. Okamoto, Polym. J., 1, 13 (1970).

30. K. Hatada, Y. Okamoto, T. Kitayaja, and S. Sasaki, Polym. Bull., 9, 220 (1983).

31. H. Yuki, Y. Okamoto, and K. Sadamoto, Bull. Chem. Soc. Jpn., 42, 1754 (1969).

32. H. W. McCormick, J. Polym. Sci., 25, 488 (1957).

33. L. H. Tung and G. Y. Lo, in Advances in Elastomers and Rubber Elasticity, J. Lal and J. E. Mark, Eds., Plenum, New York, 1986, p. 129.

34. T. Fueno and J. Furukawa, J. Polym. Sci., Part A, 2, 3681 (1964).

35. Yu. L. Spirin, A. A. Arest-Yakubovich, D. K. Polyakov, A. A. Gantmakher, and S. S.






45. T. Narita, A. Masaki, and T. Tsuruta, J. Macromol. Sci.-Chem., A4, 277 (1970).

46. T. Narita, M. Kazato, and T. Tsuruta, J. Macromol. Sci.-Chem., A4, 885 (1970).

47. L. Bu and S. Ying, J. Appl. Polym. Sci., 44, 1499 (1992).

48. K. Sardelis, H. J. Michels, and G. Allen, Polymer, 25, 1011 (1984).

49. Y. Tsukahara, N. Nakamura, T. Hashimoto, H. Kawai, T. Nagaya, Y. Sugimura, and S. Tsuge, Polym. J., 12, 455 (1980).

50. G. Kraus, C. W. Childers, and T. T. Gruver, J. Appl. Polym. Sci., 11, 1581 (1967).

51. Y. Tanaka, H. Sato, and Y. Nakafutami, Polymer, 22, 1721 (1981).



IV ANIONIC SYNTHESIS OF POLYMERS WITH WELL-DEFINED STRUCTURES



11 Functionalized Polymers and Macromonomers

I. Introduction

There has been growing interest and research on new synthetic methods for the preparation of well-defined polymers with in-chain and chain-end functional groups [1–14]. These functional groups in polymers can participate in reversible ionic association; chain extension, branching, or crosslinking reactions with poly-functional reagents; coupling and linking with reactive groups on other oligomer or polymer chains; and initiation of polymerization of other monomers. In order to exploit this unique potential of functionalized polymers, it is important to consider the scope and limitations of current functionalization methodology using anionic polymerization.

The methodology of living anionic polymerization, especially alkyllithium initiated polymerizations of styrene and diene monomers, is particularly suitable for the synthesis of functionalized polymers with well-defined structures for systems that proceed in the absence of chain termination and chain transfer reactions [15–23]. Since these living polymerizations generate stable, anionic polymer chain ends when all of the monomer has been consumed, post-polymerization reactions with a variety of electrophilic species can be used to generate a diverse array of functional groups [10,24–34]. The literature abounds with tabular and text descriptions of functionalization reactions for living polymeric carbanions with a variety of electrophilic species [1,15,24–27]. Unfortunately, many



YFP

Highlight

YFP

Highlight

YFP

Highlight

YFP

Highlight

YFP

Underline

of these functionalization reactions have not been well characterized [10,16]. Thus, it is not obvious to the uninitiated which specific procedures would be suitable for introduction of many functional groups. This chapter will provide insight into the state of the art with regard to anionic synthesis of functionalized polymers.

II. Specific Functionalization Reactions

A. End Functionalization with Electrophilic Reagents

The traditional approach to the anionic synthesis of chain-end functionalized polymers utilizes the post-polymerization reactions of the living anionic polymers with specific electrophiles for each different functional group as depicted in Equation 11.1. Thus, it is necessary to develop, analyze and optimize new proce-

(11.1)

dures for each different functional group. Optimization procedures often utilize variables such as chain-end structure, solvent, temperature, concentration, stoichiometry, mode of addition of reagents, and addition of polar additives since these factors can have dramatic effects on yield and product distributions. This section will provide a critical overview of the use of specific functionalization reactions to prepare polymers labeled with carboxyl, hydroxyl, amino, and sulfonate end groups primarily via alkyllithium-initiated polymerization methods.

B. Functionalized Initiators

An alternative procedure for the preparation of end-functionalized polymers is to use a functionalized initiator as shown in Equation 11.2 [2]. Because most

(11.2)

functional groups of interest are not stable in the presence of either simple or polymeric organolithium reagents, it is generally necessary to use suitable protecting groups in the initiator. A suitable protecting group is one that is not only stable to the anionic chain ends but is also readily removed upon completion of the polymerization (see Chapter 5). There are several distinct advantages in the use of functionalized initiators. For a living polymerization, each initiator molecule generates one macromolecule; therefore, each functional initiator molecule will generate an end-functionalized macromolecule regardless of molecular weight.



YFP

Highlight

YFP

Highlight

YFP

Underline

YFP

Underline

Further advantages relative to functionalization by termination are that it is not necessary to be concerned with efficient and rapid mixing of reagents with viscous polymers or with the stability of the chain end that is of concern at the elevated polymerization conditions often employed (see Chapter 8). As shown in Equation 11.2, the use of a functionalized initiator also provides a methodology for the preparation of telechelic polymers by termination with a suitable electrophilic reagent. An alternative procedure is to couple two living polymer chains with a suitable difunctional coupling agent (see Chapter 13) as illustrated in Equation 11.3. Both of these procedures for syntheses of telechelic polymers avoid physical

(11.3)

gelation problems normally associated with other direct syntheses of telechelic polymers using dilithium initiators [2]. The principal limitation on the use of functionalized initiators is that they are not readily available and often exhibit only limited solubility in hydrocarbon solution [2].

C. Carbonation

Carbon Dioxide

The carbonation of polymeric carbanions using carbon dioxide is one of the most useful and widely used functionalization reactions. However, there are special problems associated with the simple carbonation of polymeric organolithium compounds [35–39]. For example, when carbonations with high-purity, gaseous carbon dioxide are carried out in benzene solution at room temperature using standard high vacuum techniques, the carboxylated polymer is obtained in only 27–66% yields for poly(styryl)lithium, poly(isoprenyl)lithium, and poly(styrene-b-isoprenyl)lithium. The functionalized polymer is contaminated with dimeric ketone (23–27%) and trimeric alcohol (7–50%) as shown in Equation 11.4, where P represents a polymer chain. It was proposed that the formation of these side-

(11.4)

products is favored relative to the desired carboxylated polymer by aggregation of the chain ends in hydrocarbon solution [37]. For example, it is known that poly(styryl)lithium is primarily associated into dimers in benzene solution while the degree of association of poly(dienyl)lithiums in hydrocarbon solvents is at least dimeric (see Chapters 1 and 7) [16]. It has been reported that the addition of sufficient quantities of Lewis bases such as tetrahydrofuran (THF) [40–42] and



YFP

Underline

YFP

Highlight

YFP

Underline

YFP

Rectangle

N,N,N',N'-tetramethylethylendiamine (TMEDA) [43] can reduce or even eliminate the association of polymeric organolithium chain ends. In accord with these considerations, it was found that addition of large amounts of either THF (25 vol%) [37–39] or TMEDA ([TMEDA]/[PLi] = 1–46) [37–39] was effective in favoring the carbonation reaction to the extent that the carboxylated polymer was obtained in yields > 99% for poly(styryl)lithium, poly(isoprenyl)lithium, and poly(butadienyl)lithium. In general, a higher molar ratio of [TMEDA]/[PLi] was required to eliminate dimer and trimer formation for poly(dienyl)lithiums compared to poly(styryl)lithium. This is consistent with the general observations that the dienyllithium chain ends are either more highly associated than styryllithium chain ends, or their association constant is larger (see Chapters 1 and 7) [44–46]. The actual mechanism of this carboxylation reaction is complex, since the product distribution is dependent on chain end concentration, solvent, temperature, the rate of stirring, and the pressure of carbon dioxide [38,39]. For example, the yield of carboxylated polystyrene is reduced from > 90% to 60% when the solution is stirred during the addition of carbon dioxide even in the presence of 25 vol% THF [38]. The yield of carboxylated polymer is increased by lowering the chain end concentration and by increasing the pressure of carbon dioxide [38].

In general, it is observed that the amount of dimer and trimer contaminants is higher for poly(dienyl)lithiums versus poly(styryl)lithium. Thus, under conditions in which the yields of carboxylated polymer, dimer, and trimer are 47, 27, and 26%, respectively, for poly(styryl)lithium, the corresponding yields are 27, 23, and 50% for the analogous poly(styrene-b-butadienyl)lithium [39]. These results are once again consistent with evidence suggesting that poly(dienyl)lithiums are either more highly associated or more strongly associated compared to poly(styryl)lithium [44–46].

The effect of chain end structure (stability and steric requirements) has also been investigated [39]. The steric and electronic nature of the anionic chain end can be modified by reaction with 1,1-diphenylethylene as shown in Equation 11.5 [47–54]. When the direct carbonation is effected in benzene at room temperature with the diphenylalkyllithium species formed by addition of poly(styryl)lithium (Mn = 2.0 × 103 g/mol) to 1,1-diphenylethylene (Eq. 11.5), the carboxylated

(11.5)

polymer can be isolated in 98% yield compared to only a 47% yield for the analogous poly(styryl)lithium without end-capping under the same conditions



YFP

Underline

YFP

Underline

YFP

Underline

[39]. These 1,1-diphenylalkyllithium species are reported to be associated into dimers in hydrocarbon solution [55], although it would be anticipated that the strength of the dimeric association (e.g., Kassoc) would be decreased by the increased steric requirements of the chain end. It is tentatively concluded that the competing reaction to form dimeric (and trimeric) side-products is quite sensitive to the steric requirements of the chain end.

The important conclusion is that the carbonation reaction of polymeric organolithium compounds in hydrocarbon solution with gaseous carbon dioxide can be carried out in essentially quantitative yield by adding sufficient quantities of Lewis bases such as tetrahydrofuran or TMEDA prior to the functionalization reaction. It is particularly important to note that this procedure ensures that functionalized polydienes with high 1,4-enchainment can be prepared since the Lewis base is not present during the diene polymerization [16,19,56,57].

A rather specialized solid-state carbonation procedure can be used to carbonate poly(styryl)lithium and other living polymers with backbones that have glass transition temperatures significantly above room temperature. Thus, freeze-drying of benzene solutions of poly(styryl)lithium generates a porous solid that can be carbonated in the solid state to give minimal amounts of dimeric ketone products (1–2%) [38,39]. In addition, essentially quantitative yields of carboxylated polystyrene were obtained from freeze-dried solutions of poly(styryl)lithium complexed with 1–2 molar equivalents of N,N,N',N'-tetramethylethylenediamine (TMEDA) [38]. No dimer was detected by size exclusion chromatography (SEC) or thin-layer chromatography (TLC) analyses. In contrast, a freeze-dried sample of poly(styrene-b-butadienyl)lithium (PBD block Mn = 450 g/mole) complexed with 3 molar equivalents of TMEDA formed the corresponding carboxylated polymer in 93% yield [39].

From a practical, synthetic point of view, it should be noted that it has been reported that > 90% yields of carboxylated polymers can be obtained simply by pouring a hydrocarbon/THF (99.5/0.5, vol/vol) solution of the poly(styryl)lithium onto solid carbon dioxide [36]. For an analogous carboxylation in the absence of THF, a 78% yield of carboxylated polymer was reported [36]. Conversion to the corresponding Grignard reagent prior to gaseous CO2 termination has also been reported to produce > 90% yields of the carboxylated polymer [36]. It is also noteworthy that essentially quantitative carboxylation has been reported when potassium is the counterion in THF [58] or when gaseous CO2 is added to a THF solution of poly(styryl)lithium at -78°C [59].

The carbonation of α,ω-dilithiumpolymers is complicated by the occurrence of physical gelation phenomena that produce severe mixing problems [60]. In general, heteroatom derivatives of lithium such as lithium carboxylate salts are highly associated in solution [61]; therefore, the polymeric α,ω-dicarboxylate salts will form an insoluble, three-dimensional network during the functionalization reaction. A variety of procedures have been described to minimize these



YFP

Highlight

YFP

Underline

YFP

Underline

effects, including the use of solvents with low solubility parameters (<7.2) [62], reaction in a T-tube mixer [63] and the use of a two-substance jet with a high flow rate and high CO2/PLi ratio [60].

The carbonation reaction is somewhat ideal since it is possible to analyze the reaction products using a variety of probes including osmometry, SEC, end group titration, 13C nuclear magnetic resonance (NMR), FTIR, and TLC. For example, 13C NMR analysis showed the presence of unexpected ring-carboxylation products from the reaction of poly(styryl)lithium in benzene with gaseous carbon dioxide [64]. In addition, the pure functionalized polymer can be separated from unfunctionalized polymer and dimeric ketone products by SiO2 chromatography using toluene as eluent. For example, column chromatography has been used to separate about 1% unfunctionalized polybutadiene with Mn = 98 × 103 g/mol from the corresponding carboxyl-functionalized polymer using this technique [65]. Furthermore, it was possible to detect < 1% of the unfunctionalized polybutadiene by SiO2 TLC using toluene as eluent [65].

Termination with 4-Bromo-1,1,1-Trimethoxybutane

The reaction of poly(styryl)lithium and other alkali metal polystyrene derivatives with 4-bromo-1,1,1-trimethoxybutane at -78°C in THF is reported to yield the corresponding ortho ester derivative that can be hydrolyzed to the carboxyl-func

Scheme 11.1

tionalized polymer as shown in Scheme 11.1 [66]. It has not been determined that analogous high yield functionalization reactions can be effected in hydrocarbon solution at room temperature or above.

Protected Oxazoline Initiator

The use of an organolithium initiator with an oxazoline protecting group has been examined for the preparation of α-carboxyl-functionalized polymers [67]. The initiator was prepared by lithiation of 2,4,4-trimethyl-2-oxazoline as shown in Equation 11.6. This initiator solution was used without further analysis or charac-



YFP

Text Box

呃唑啉

(11.6)

terization to initiate the polymerization of styrene, 2-vinylpyridine, and methyl methacrylate at -78°C in tetrahydrofuran. Low initiator efficiencies (4% for styrene) and broad molecular weight distributions (Mw/Mn = 2 for polystyrene) were observed for methyl methacrylate and styrene polymerizations. The polymerization of 2-vinylpyridine with this initiator provided the stoichiometric molecular weight, but a very broad molecular weight distribution (Mw/Mw = 3). These results were interpreted as an indication of the “weakness” of the initiator. This may result from association effects or because the initiator is too stable to be an effective initiator for these monomers.

D. Hydroxylation

Ethylene Oxide

The preparation of hydroxyl-terminated polymers from polymeric organolithium compounds by reaction with ethylene oxide is one of the few simple, efficient functionalization reactions. It also serves to emphasize the uniqueness of organolithium-initiated polymerizations. The direct reaction of poly(styryl)lithium with excess ethylene oxide in benzene solution produces the corresponding hydroxyethylated polymer in quantitative yield without formation of detectable amounts of oligomeric ethylene oxide blocks (Eq. 11.7) [68]. For example, 13C-NMR

(11.7)

analysis of a hydroxyethylated polystyrene (Mn = 1.3 × 103 g/mol, MwMn = 1.08) showed no evidence for the formation of any ether linkages expected for oligomerization of ethylene oxide. This result is surprising in view of the steric strain [69] and intrinsic reactivity of ethylene oxide toward nucleophiles [70,71]. Although organolithium compounds react the fastest with ethylene oxide itself compared to other organoalkali compounds [72], the lithium alkoxides are the least reactive among the alkali metal alkoxides for anionic polymerization of ethylene and propylene oxide [73–76]. Apparently the high degree of aggregation of lithium alkoxides and the strength of this association even in polar solvents render them unreactive [77]. However, it was recently reported that poly(ethylene oxide) is the major product from the reaction of ethylene oxide with the metalation products from the interaction of ferrocene with n-butyllithium in hexane in the



YFP

Highlight

YFP

Highlight

presence of an equivalent amount of N,N,N',N'-tetramethylethylenediamine (TMEDA); poly(ethylene oxide) was also reportedly formed from the direct reaction of ethylene oxide with n-BuLi/TMEDA in hexane [78]. Lithium alkoxides will also polymerize ethylene oxide in a dipolar aprotic solvent such as dimethylsulfoxide at elevated temperature [79–81].

Telechelic dihydroxypolymers can be prepared from the corresponding α,ω-dilithium polymers by treatment with ethylene oxide. For example, functionalities of 1.9–2.0 have been reported for the ethylene oxide termination reaction for α,ω-dilithiumpolyisoprene [82]. It should be noted, however, that termination of α,ω-dilithium polymers with ethylene oxide in hydrocarbon solvents resulted in physical gel formation that required periods of 1–4 days to achieve complete termination.

Protected Initiators

Hydroxyl-terminated polymers have also been prepared using organolithium initiators with protected hydroxyl functionality [2,83]. Thus, using initiators such as 2-(6-lithio-n-hexoxy)tetrahydropyran (1) and ethyl 6-lithiohexyl acetaldehyde acetal (2), it was possible to prepare narrow molecular weight distribution poly-

butadiene polymers (Mw/Mn = 1.05–1.08) with hydroxyl functionalities of 0.87–1.02 per chain after mild acid hydrolysis of the acetal groups. However, since these initiators were prepared in diethyl ether, the polybutadienes had relatively high 1,2-microstructures (36–54%). Difunctional α,ω-dihydroxylpolybutadienes (functionality = 1.77–2.04) were formed either by terminating with ethylene oxide or coupling with dichlorodimethylsilane followed by mild acid hydrolysis. Either of these approaches avoided the gel formation that occurs when α,ω-dilithiumpolymers are reacted with ethylene oxide [82,84]. These initiators are reported to be insoluble in hexane, but are soluble in benzene or diethyl ether.

Carbonyl Compounds

The review literature generally describes the reactions of polymeric organolithium compounds with carbonyl compounds in terms of the simple addition reaction shown in Eq. 11.8 [24,27]. However, our investigations suggest that these reac-

(11.8)



tions do not proceed simply to provide quantitative yields of the desired hydroxyl-terminated polymers. For example, the addition reaction of polymeric organolithium compounds with p-dimethylaminobenzaldehyde was described as a simple reaction, relatively free of side reactions because of the absence of enolizable α-protons [85]. Careful reexamination of this reaction showed that the yellow polymeric reaction products included carbonyl chain end functionality and significant dinner formation (10–33%) for polymers with Mn values in the range of 0.5 × 103-6.4 × 103 g/mol [86]. A Cannizzaro reaction of the initially formed carbonyl addition product (3) to form the corresponding acetophenone-type end functionalized polymer (5), was suggested based on the isolation and identification of the expected reduction product, p-dimethylaminobenzyl alcohol, (6) (see Scheme 11.2).

Scheme 11.2



The formation of dimer product (7) results from the addition of a second mole of polymeric organolithium compound to the acetophenone-functionalized polymer (5).

Another presumably simple reaction is the addition of polymeric organolithium compounds to benzophenone and its derivatives as shown in Equation 11.9. Preliminary investigation of the reaction of poly(styryl)lithium with benzo-

(11.9)

phenone (0–0.5 molar excess) in benzene indicated that the desired hydroxyl-functionalized polymer was formed in 82–94% yield, a dimeric product was formed in 4–10% yield, accompanied by 2–8% unfunctionalized polymer [87]. The formation of significant amounts of dimer and unfunctionalized polymer is consistent with the intervention of an electron-transfer pathway [88,89] for this functionalization reaction as shown in Scheme 11.3 where SH represents a solvent

Scheme 11.3

molecule. The highest yield of functionalized polymer was obtained using a 0.2 molar excess of benzophenone at a temperature of 50°C. These results were especially interesting in view of the observations that functionalization of poly(butadienyl)lithium with 4,4'-bis(diethylamino)benzophenone effects significant improvement in the properties of tires utilizing the functionalized polymer in their formulation compared to the corresponding unfunctionalized polybutadiene



[90]. The reaction of poly(styryl)lithium (Mn = 2.63 × 103 g/mol, Mw/Mn = 1.03) in benzene with a 10% molar excess of 4,4-bis(4,4'-diethylamino)benzophenone in benzene was carried out at 25°C [87]. The SEC chromatogram of the reaction products showed no distinct peak corresponding to a dimeric product in contrast to the benzophenone functionalization reactions. After silica gel chromatography in the presence of triethylamine, a 94% yield of the functionalized polymer was isolated in addition to 6% of the unfunctionalized polymer and 1% of the dimer. The functionalized polymer tended to form yellow-colored side products on storage or in the presence of trace amounts of acid.

E. Amination

Protected Imines

The primary amination of polymeric organolithium compounds is a challenge because of the acidity of nitrogen-hydrogen bonds [91,92]. Thus, chain-end amination reactions require indirect methods such as the use of protecting groups [93]. Nakahama and co-workers [94,95] have reported that high yields (96–100%) of primary amine-functionalized polymers could be obtained by reaction of poly-(styryl)lithium with 1.5–2 molar equivalents of the protected imine, N-(benzylidene)-trimethylsilylamine, in benzene at room temperature (Eq. 11.10). The analogous

(11.10)

functionalization reaction of poly(isoprenyl)lithium in cyclohexane proceeded in 90% yield.

Attempts to reproduce these results in our laboratories with poly(styryl)lithium were not successful [96]. The aminated polystyrenes were contaminated with dimeric products (15–19% yields). In fact, the aminated polystyrene was obtained in only 69% yield for poly(styryl)lithium with Mn = 3 × 103 and an acetophenone-type functionalized polymer (12% yield) was formed in addition to the dimeric product (19% yield). It was proposed that the acetophenone-type functionality and the dimeric side reaction product results from a Cannizzaro-type reaction (Scheme 11.4) as previously described for p-dimethylaminobenzaldehyde under the carbonyl functionalization reactions (see Scheme 11.2).

Methoxyamine/Methyllithium

Primary amine-functionalized polystyrenes have been prepared by the reaction of poly(styryl)lithium with the product of the reaction of methoxyamine and methyl



Scheme 11.4

lithium at low temperatures as shown in Eq. 11.11 [97–99]. Using a twofold excess of the methoxyamine/methyllithium reagent at -78°C in a mixture of

(11.11)

THF/benzene/hexane, poly(styryl)lithium (Mn = 2 × 103 g/mol) was aminated in 92% yield after methanol work-up [97]. The telechelic diamine, α,ω-diamino-polystyrene, was prepared in 80% yield by amination of α,ω-dilithiumpolystyrene (Mn = 10× 103 g/mol) using similar procedures [98]. It was possible to isolate pure amine-functionalized polymers using silica gel column chromatography.

Functionalized and Protected Initiators

Eisenbach and colleagues [100] first reported the use of the dimethylaminopropyl-lithium initiator for the preparation of tertiary amine-functionalized polymers. Although the initiator produced polystyrenes with high amine functionalities, their procedure was limited because the initiator was prepared in a polar ethereal



solvent. Stewart and co-workers [101] recently described a procedure for the preparation of a benzene solution of 3-dimethylaminopropyllithium in > 90% yields by lithiation of the corresponding chloride in hexane at 20°C followed by replacement of the solvent by benzene. Use of this initiator solution for butadiene polymerization in hexane produced the corresponding α-dimethylaminopoly butadienes with high 1,4-polybutadiene microstructure (76–86%) but unspecified molecular weight distributions. These workers noted that the amount of vinyl content in the resulting polybutadiene increased with the amount of diamine formed by Wurtz coupling during the synthesis of the initiator.

Primary amine-terminated polymers have also been prepared using organolithium initiators with protected amine functionality. The p-lithio-N,N-bis(trimethylsilyl)aniline initiator was prepared from the corresponding aryl bromide by reaction with lithium metal in diethyl ether [93]. A major limitation of this initiator is the fact that it is insoluble in hydrocarbon solvents and therefore diene polymerizations must be carried out in mixtures of hexane and diethyl ether. Using this initiator, relatively narrow molecular weight distribution (MwIMn = 1.06-1.25) polybutadienes were prepared with functionalities of 0.69-1.0 as determined by titration after acid hydrolysis of the amine protecting group. However, the use of diethyl ether for the initiator solution produced polybutadienes and polyisoprenes with 39–43% and 45–50% vinyl microstructures, respectively. Termination of corresponding isoprene polymerizations with dichlorodimethylsilane followed by hydrolysis of the protecting group generated α,ω-diaminopolyisoprenes with functionalities of 1.7–1.9 and relatively broad molecular weight distributions (MwIMn = 1.49–2.22).

It has recently been reported that a useful protected primary amine initiator can be generated by the reaction of sec-butyllithium with p-bis(trimethylsilyl)-aminostyrene in benzene or cyclohexane solution at 25°C by careful control of the stoichiometry of the reaction as shown in Eq. 11.12[102]. Although oligomeriza

(11.12)

tion was observed when THF was added or when n-butyllithium was used as the initiator, it was reported that no oligomerization was observed for sec-butyllithium in hydrocarbon as deduced by a combination of gas chromatography and 1H NMR analysis of the acetic acid-quenched reaction. This initiator was used to prepare primary amine functionalized poly(dimethylsiloxane) using the cyclic trimer, D3,



as monomer in the presence of the promoter, hexamethylphosphoramide, followed by acid hydrolysis with dilute aqueous hydrochloric acid. Reported amine functionalities were 0.94–0.97 for the α-aminopolydimethylsiloxanes and 1.9 for the corresponding telechelic polymers obtained by coupling with dichlorodi-methylsilane. It remains to be demonstrated that a useful protected primary amine functionalized initiator for styrene and diene polymerization can be generated using these procedures.

ω-Halo-α-Aminoalkanes

Richards et al. [103] have reported that poly(butadienyl)lithium can be functionalized efficiently by reaction with N-(3-chloropropyl)dialkylamines as shown in Eq. 11.13. The reported yields were as high as 95%. Teyssie and co-workers

(11.13)

[104] have applied this procedure for the preparation of α,ω-bis(dimethylamino)-polyisoprene. Sodium naphthalene was used as the difunctional initiator to prepare α,ω-disodiumpolyisoprene at -78°C. Termination with excess 1-chloro-3-(dimethylamino)propane produced α,ω,-bis(dimethylamino)polyisoprene with MwIMn = 1.2, presumably high vinyl microstructure, and a functionality of 1.85 as determined by perchloric acid titration. All of these results are surprising in view of the fact that normally the coupling reactions of alkyl halides with organolithium compounds are complicated by side reactions such as lithium-halogen exchange and dehydrohalogenation (elimination) reactions [29].

Nakahama and co-workers [105] have recently extended this approach to prepare primary-amine functionalized polymers by terminating with α-halo-ω-aminoalkanes containing a protected amino functionality as shown in Eq. 11.14. It

(11.14)

was reported that the yields of > 94% were readily obtained for the functionalization reactions of poly(styryl)lithium, poly(isoprenyl)lithium, and α,ω-dipotassium-polyisoprene with the α-bromo- and α-chloro-ω-silyl-protected aminopropanes and the α-bromo-ω-silyl-protected aminoethane to produce the corresponding mono- and telechelic amine-functionalized polymers. Precipitation into methanol was sufficient to remove the silyl-protecting groups. Evidence for Wurtz-coupling reactions leading to the formation of dimer was observed for the α-iodo derivative. The typical procedure involved reaction with excess reagent at -78°C in the presence of THF, followed by standing for 1 h at 25°C. The products were



SecLover

Highlight

characterized by SEC analysis of the benzoyl derivatives, TLC, TLC with flameionization detection, VPO, and end-group titration with HClO4. In addition, chain-extension reactions of the telechelic α,ω-diaminopolystyrene (Mn = 5.35 × 103 g/mol) with 2,4-tolyldiisocyanate produced a segmented polyurea with Mn = 230 × 103 g/mol and no base polymer was detected by SEC. This result is consistent with high functionality for the diamine.

High amine functionality (> 94%) was also observed for analogous reactions of poly(styryl)lithium and poly(isoprenyl)lithium in benzene solution; THF (17 vol%) was added prior to the functionalization reaction.

F. Sulfonation

Direct Reaction

The simplest procedure for the sulfonation of polymeric organolithium compounds is the direct reaction with sultones. Thus, the functionalization of poly(styryl)lithium (Mn = 2.7–4.7 × 103 g/mol) with generally a twofold molar excess of 1,3-propane sultone (Eq. 11.15) was investigated as a function of solvent and

(11.15)

temperature [106]. At temperatures from 3–80°C in benzene solution, only 24–30% of the terminally sulfonated polymer was formed; the remainder of the product was the unfunctionalized polymer. At -78°C in THF using a molar excess of sultone, a yield of 53% of the terminally-sulfonated polystyrene was obtained compared to the yields of 67–72% reported by Omeis and colleagues [107] using a fivefold excess of sultone. It is assumed that the relatively low yields of sulfonated polymer are a result of the competing metalation reaction of poly(styryl)lithium with the acidic α-hydrogens of the sultone as shown in Eq. 11.16 [108]. Analogous reactions of butyllithium with 1,3-propane sultone produce the metalated sultone in 65–85% yields [108,109].

(11.16)

These results with poly(styryl)lithium are quite different from the results of Eisenbach and colleagues [100], who reported sulfonation yields of > 90% for the reaction of poly(α-methylstyryl)lithium with 1,3-propane sultone in THF at -78°C. To examine this effect of chain end structure, poly(styryl)lithium (Mn = 4.7 × 103 g/mol; Mw/Mn = 1.03) was reacted with α-methylstyrene at -78°C in THF to form poly(styrene-block-α-methylstyryl)lithium [Mn(SEC, apparent)



= 6.0 × 103 g/mol; Mw/Mn = 1.16]. When this block copolymeric organolithium compound was reacted with a molar excess of 1,3-propane sultone, the corresponding sulfonated diblock copolymer was obtained in 94% yield [106]. This is in excellent agreement with the reported results of Eisenbach and colleagues [100]. However, when the sulfonation reaction of an analogous diblock copolymeric organolithium compound was effected in toluene under analogous conditions with almost a threefold excess of sultone, the corresponding sulfonated product was obtained in only 28% yield. It appears that the competition between ring-opening and metalation is quite sensitive to solvent and chain-end structure. These results indicate that although the direct sulfonation of poly(α-methylstyryl)lithium with 1,3-propane sultone is an efficient reaction in THF at -78°C [100], this is not a generally useful procedure for poly(styryl)lithium since only 53–72% yields can be obtained [106,107]. This dramatic effect of chain end structure suggested that other methods of attenuating the reactivity of the polymeric organolithium chain end might also be effective. A sulfonation procedure that could be effected at room temperature was also desirable.

1,1-Diphenylethylene End-Capping

End-capping of poly(styryl)lithium with 1,1-diphenylethylene (see Eq. 11.5) prior to reaction with a sultone (see Scheme 11.5) was investigated as a method for

Scheme 11.5

increasing the efficiency of the sulfonation reaction by both increasing the steric congestion around the carbanion and by decreasing the basicity by conversion to a more stable, delocalized anion. The sulfonation reactions of the polymeric diphenylalkyllithium compounds with 1,3-propane sultone and 1,4-butane sultone were investigated [106]. Using a 1/6 (vol/vol) ratio of THF/benzene and at least a 0.5-fold excess of 1,3-propane sultone at 25–30°C, a 93% yield of the sulfonated



polymer could be obtained via the 1,1-diphenylethylene end-capping route for poly(styryl)lithium. Lower sulfonation yields (54–76%) were obtained in benzene solution and the yield decreased when a large excess of sultone (ninefold) was used. Sulfonation yields with 1,4-butane sultone were generally lower than the corresponding sulfonations with 1,3-propane sultone under analogous conditions; however, the effects of solvent and excess sultone were similar to the effects observed with 1,3-propane sultone. These procedures have been applied to the synthesis of α,ω-disulfonatopolystyrenes; the sulfonation yields were reported to be > 95% [110].

G. Oxidation

The reaction of poly(styryl)lithium with molecular oxygen produces a complex mixture of products [111–114]. The products and the range of yields obtained by allowing molecular oxygen to diffuse into an unstirred solution at room temperature under various conditions are shown in Eq. 11.17 [114]. Within the yield ranges indicated in Eq. 11.17, the product distribution was relatively insensitive

(11.17)

to the presence of polar additives such as THF or TMEDA in benzene solution [114]. It is noteworthy that the expected formation of the dimeric peroxide by a free radical chain oxidation mechanism [115] has been generally overlooked [25]. In contrast to the solution oxidation results, the oxidation of poly(styryl)lithium complexed with one molar equivalent of TMEDA in the solid state obtained by freeze-drying the corresponding benzene solution produced ω-hydroperoxypoly-(styrene) in 95% yield [111,114]. This functionalization reaction generates a potentially useful polymeric free radical initiator that could be used to generate new block copolymers.

In addition to the synthetic aspects of the oxidation of polymeric organolithium compounds, the presence of oxygen as an impurity during polymerization, functionalization or termination can give rise to significant amounts of dimer formation [113]. Furthermore, the presence of lithium alkoxides can affect the kinetics of initiation (see Chapter 6) and propagation (see Chapter 7), as well as the stereochemistry of polymerization (see Chapter 9).

H. Aldehyde Functionalization

ω-Formyl-functionalized polystyrenes can be synthesized in quantitative yield by reacting poly(styryl)lithium in benzene with 4-morpholinecarboxaldehyde followed by methanol termination and precipitation into methanol as shown in Eq.



(11.18)

11.18 [116]. The absence of dimer formation as expected if aldehyde is formed in the presence of poly(styryl)lithium was rationalized by proposing that the direct reaction product is a tetrahedral α-amino alkoxide intermediate (see Scheme 11.6), which is stable under the reaction conditions until work-up with methanol. Evidence for this intermediate was obtained by trapping experiments with diphenylphosphinic chloride.

Scheme 11.6

I. Summary of Specific Functionalization Reactions

This discussion of chain-end functionalizations using specific reactions for each functional group illustrates some of the complexities of effecting the quantitative preparation of functionalized polymers with a variety of electrophilic species. The ability to prepare chain-end functionalized or labelled polymers is not limited by the range of potential and reported functionalization and labelling reactions for living carbanionic chain ends. It is limited by the fact that many functionalization reactions reported in the literature have not been adequately characterized or optimized for general utility [10,11]. Another limitation is the necessity of developing, optimizing, and characterizing new reactions for each different functional group as illustrated by the previous discussion.

III. General Functionalization Reactions

General functionalization reactions are reactions of organolithium compounds that proceed efficiently to introduce a variety of different functional groups. The most useful reactions can be utilized at elevated temperatures in hydrocarbon solution so that the unique characteristics of organolithium-initiated polymerizations can be preserved. Two examples of general functionalization reactions are the coupling reactions with substituted silyl halides and the addition reactions with substituted 1, 1-diphenylethylenes.



A. Functionalization Reactions with Silyl Halides

Perfluoroalkyl Group Functionalization

As discussed in Chapter 13, the reaction of polymeric organolithium compounds with silyl halides is a very efficient reaction that is not complicated by competing side reactions. Therefore, these reactions provide the opportunity to prepare a variety of end-functionalized polymers by reactions with silyl halides containing either functional groups or protected functional groups. For example, the reaction of poly(styryl)lithium with (tridecafluoro-1,1,2,2-tetrahydrooctyl)-1-dimethyl-chlorosilane has been used to prepare perfluoroalkyl-terminated polystyrenes as shown in Eq. 11.19 [117].

(11.19)

Trialkoxysilane Functionalization

The reaction of poly(styryl)lithium with p-(chloromethylphenyl)trimethoxysilane is reported to provide an efficient method for the synthesis of trimethoxysilane-terminated polystyrenes as shown in Eq. 11.20 [118]. A combination of elemental

(11.20)

analysis, 29Si NMR, TLC, and the products of hydrolysis [119] was consistent with quantitative end-group functionalization under these conditions. However, the use of -78°C in THF limits the utility of this functionalization reaction and results in molecular weight distributions that are not narrow (MwIMn > 1.1).

B. 1,1-Diphenylethylene Functionalization Reaction

Background

The reaction of polymeric organolithium compounds with substituted 1,1-diphenylethylene derivatives is an excellent system for development of a general



(11.21)

functionalization reaction (Eq. 11.21) because these addition reactions are simple and quantitative; only monoaddition (i.e., no oligomerization) has been reported; the rate and efficiency of the crossover reaction can be monitored by ultraviolet-visible spectroscopy; copolymerization of substituted 1,1-diphenylethylene derivatives with other monomers will result in polymers with multiple functional groups along the polymer chain; these addition reactions take place readily in hydrocarbon solution at room temperature and above; and a variety of substituted 1,1-diphenylethylenes with functional groups on the aromatic ring can be prepared readily [10,11,18,19,29,50–54,120].

Unlike most electrophilic functionalization reactions, this reaction is not in itself a termination reaction. The product of the addition reaction of a simple or polymeric organolithium compound to a substituted 1,1-diphenylethylene is a carbanionic species (1,1-diphenylalkyllithium; see 8 in Eq. 11.21) that can initiate

Scheme 11.7



anionic polymerization of an additional monomer such as isoprene [121] or methyl methacrylate [47–49] to extend the chain or form a new block. Thus, this procedure can be described as a living functionalization reaction. This method can be used to prepare polymers with functional groups at the initiating end (α) of the polymer chain (see 11 in Scheme 11.7) or within the polymer chain (see 12 in Scheme 11.8), in addition to the terminating end (ω) (Eq. 11.21). Thus, it offers

Scheme 11.8

the potential of providing a versatile, general anionic functionalization procedure with which one can rationally design and place functional groups at essentially any position in a polymer molecule. Furthermore, this methodology can be applied to the synthesis of polymers with two or more functional groups at the initiating end, at the interface between two blocks, or at the terminating end, as shown in Equation 11.22 for terminal functionalization, where X and Y can either be the same or different functional groups [122]. The α,α- or ωω-difunctional products of these functionalization reactions could behave as macromolecular monomers for condensation-type copolymerization reactions [4,123].

The following sections describe the applications of 1,1-diphenylethylene for the preparation of phenol, amino, and carboxyl functionalized polymers as well as for the preparation of polymers with aromatic functional groups that provide fluorescent labels. In general, the methodologies described in Eqs. 11.21 and 11.22, as well as in Schemes 11.7 and 11.8, have been realized. However, the ability to prepare either polymers with labels at the initiating end or at the interface



(11.22)

between two blocks is limited by the reactivity of the diphenylalkyllithium species as an initiator. Since diphenylmethane, toluene and propene have estimated pKa values of 32, 42, and 43 (see Chapter 2) [32,92,124,125], respectively, it is surprising that diphenylalkyllithium species can initiate the polymerization of styrene and diene monomers [121]. However, we have observed that diphenyl-alkyllithiums, either unsubstituted or with electron-donating substituents on the aromatic ring, function as effective initiators for the polymerization of styrene, diene, and methacrylate monomers (i.e., polymers with predictable molecular weights and relatively narrow molecular weight distributions [Mw/Mn < 1.1 [126]) are obtained) [87,127,128]. However, when electron-withdrawing functional groups such as the oxazoline and tertiary N,N,-dialkylamide-protecting groups are substituted on the aromatic ring, broad molecular weight distribution polymers are obtained (M/Mn > 1.1) [129]. These results have been ascribed to the effects of the conjugated electron-withdrawing groups in stabilizing the intermediate 1,1-diphenylalkyllithium species, which reduces their reactivity as initiators for styrene and diene monomers. Nakahama and co-workers [130,131] have observed similar limitations on the ability to utilize living polymers formed from styrene monomers with these electron-withdrawing groups for block co-polymer formation with styrene and α-methylstyrene. Aside from these limitations, anionic functionalizations utilizing substituted 1,1-diphenylethylenes as outlined herein provide a versatile and general functionalization methodology for the quantitative introduction of functional groups at the chain ends and within the polymer chain by design.

Phenol Functionality

The first application of this methodology was directed to the preparation of a phenol-terminated polymer [127]. The tert-butyldimethylsilyl-protecting group was chosen for the aromatic hydroxyl group based on the results reported by Nakahama and co-workers [132–134]. Poly(styryl)lithiums with Mn = 1.5 × 103 to 13.1 × 103 g/mol were functionalized with 1-(4-tert-butyldimethylsiloxyphenyl)-1-phenylethylene (14) to form the corresponding phenol-end-functionalized poly



styrenes in > 99% yield after hydrolysis with 1% HCI in tetrahydrofuran. The efficiency of these functionalization reactions was evaluated by end-group titration, elemental analyses, 1H NMR, and 13C NMR analyses, as well as thin-layer chromatography [127]. All of the available evidence suggests that this is an essentially quantitative functionalization reaction. It is noteworthy that Heitz and Hocker [122] have carried out similar functionalizations using 1,1-(4,4'-dimethoxyphenyl)ethylene.

Amine Functionality

In an analogous fashion, 1-(4-dimethylaminophenyl)-1-phenylethylene (15) can be used to prepare amine-terminated polymers [87,128]. This substituted di-

phenylethylene was readily prepared from the corresponding benzophenone derivative via the Wittig reaction. The reaction of poly(styryl)lithiums (Mn = 3.1 × 103 - 10.2 × 103 g/mol) in benzene solution with 15 could be monitored by ultraviolet-visible spectroscopy at 406 nm. Analyses of these functionalization reactions showed that the amine-terminated polymers were obtained in > 99% yields [128]. The functionalization of poly(butadienyl)lithium in benzene was likewise effected in the presence of THF ([THF]/[PBDLi] = 21) to form the corresponding ω-dimethylamino-functionalized polybutadiene with a functionality of 0.94 and high 1,4-microstructure (90.5%) [87]. It is noteworthy that in the absence of THF no addition of poly(butadienyl)lithium to 15 was detected by ultraviolet (UV) spectroscopy; this is in accord with the lack of reactivity of poly(dienyl)lithium compounds with respect to addition to 1,1-diphenylethylenes (see also Chapter 10) [135].

Telechelic α,ω-bis(dimethylamino)-terminated polystyrene was prepared by



first reacting sec-butyllithium with 15 to form an amine-functionalized initiator, (16); α-dimethylaminopoly(styryl)lithium (17 was prepared by initiating styrene polymerization with 16 as shown in Scheme 11.9. An aliquot of 17 was removed

Scheme 11.9

from the reactor and terminated with methanol to form the corresponding α-dimethylamino-functionalized polystyrene (18). This polymer had Mn = 2.05 × 104 g/mol (SEC), Mw/Mn = 1.04 (SEC), and an amine functionality of 1.2 as determined by amine end group titration. The remainder of the living polymer (17) was then functionalized by addition of 15 to form the corresponding α,ω-bis-dimethylaminopolystyrene (18) after methanol termination as shown in Scheme 11.9. The α,ω-bis(dimethylamino)polystyrene exhibited Mn(SEC) = 2.05 × 104 g/mol, Mn(VPO) = 2.19 × 104 g/mol, and Mw/Mn = 1.04. The amine group functionality of this telechelic polymer was 2.1 as determined by end-group titration [128]. The functionalized initiator (16) has also been used to prepare α-4-dimethylaminophenyl-functionalized polybutadiene (Mn = 2.65 × 105 g/mol; Mw/Mn = 1.11) in benzene with 90.5% 1,4-microstructure [87]. The functionality of this



high-molecular-weight polybutadiene was determined by UV-visible spectroscopy using absorptions at λmax = 262 and 303 nm. The average Mn value determined by UV spectroscopy was 2.84 × 105 g/mol, which corresponds to a functionality of 0.93. The ability to determine the functionality of high-molecular-weight polydienes is an added advantage of this 1,1-diphenylethylene functionalization methodology.

The ability to place functional groups within polymer chains was demonstrated by first functionalizing poly(styryl)lithium with 15 (Scheme 11.10) [128].

Scheme 11.10

A portion of this functionalized, living polymer (19) was removed from the reactor and terminated with methanol to give a monofunctional ω-dimethylamino-functionalized polystyrene. The functionality of this polymer as calculated from the end-group titration was 0.96. This amine-functionalized polystyrene exhibited only one spot by TLC analysis and it was estimated that less than 1% unfunctionalized polystyrene was formed. The living, functionalized polymer (19) was chain extended by addition of butadiene monomer to form a diblock polymer, poly(styrene-block-butadiene) (21,) with the dimethylamino functional group at the interface between the blocks after termination as shown in Scheme 11.10. The block copolymer (21) exhibited Mn (membrane osmometry) = 2.13 × 104 g/mol, Mn (titration) = 2.19 × 104 g/mol with Mw/Mn (SEC) = 1.01. The functionality calculated from the titration results was 0.97 and the polymer exhibited only one



spot by TLC analysis and no functionalized polystyrene base polymer was detected. These results show that functionalized diphenylalkyllithiums with electron-Donating substituents are effective initiators for anionic polymerization of styrene and diene monomers (i.e., polymers with narrow molecular weight distributions are obtained).

ω-Primary amine-functionalized polystyrenes (24) were synthesized by reacting poly(styryl)lithium with 1-[4-[N,N-bis(trimethylsilyl)amino]phenyl]-1-phenylethylene (22) in benzene at room temperature followed by acid-catalyzed hydrolysis and neutralization, as shown in Scheme 11.11 [136]. The isolated

Scheme 11.11

amine-functionalized polymer had an amine functionality of 98.5%. Characterization by 1H and 13C NMR and FTIR spectroscopy was consistent with the presence of the primary aromatic amine functionality; less than 0.5 wt% of the unfunctionalized polymer was isolated by column chromatography.

Carboxyl Functionality

The use of substituted 1,1-diphenylethylenes to prepare end-functionalized polymers has also been utilized to prepare carboxyl-functionalized polymers. The carboxyl functionality has been protected using the oxazoline group [137]. The oxazoline-substituted 1,1-diphenylethylene was not stable to the anionic chain end at room temperature, however [129]. The functionalization reaction was effected in toluene/THF mixtures (4/1, vol/vol) at -78°C to produce the carboxyl-functionalized polystyrene (Mn = 2.4 × 103 - 14.6 × 103 g/mol) in quantitative yield after acid hydrolysis as shown in Eq. 11.23 [129].



(11.23)

The carboxyl group has also been protected using the diisopropylamide derivative [138]. The corresponding diisopropylamide-functionalized 1,1-diphenylethylene was synthesized and characterized by elemental analysis and a variety of spectroscopic methods. The diisopropylamide-functionalized 1,1-diphenylethylene was not stable to the anionic chain end at room temperature. The functionalization reaction was effected in toluene/THF mixtures, (4/1, vol/vol) at -78°C to produce the amide-functionalized polystyrene in 92–100% yields as shown in Eq. 11.24 (Mn = 2.3 × 103 - 12.6 × 103 g/mol). Although it was

(11.24)

somewhat difficult to hydrolyze the amide, heating under reflux in toluene with toluenesulfonic acid was found to be effective in generating the carboxyl-functionalized polymers [129].

Condensation Macromonomers: Diphenol Functionality

Another advantage of the 1,1-diphenylethylene functionalization methodology is that it can be used to prepare condensation macromonomers (25), which are polymers with two polymerizable functional groups at one chain end [4,123]. This type of macromonomer can participate in step-growth (condensation) polymerization



with other difunctional monomers to form model comb-type, branched condensation copolymers [4,123]. ω,ω-Diphenolpolystyrenes (28) can be synthesized in quantitative yields by reacting poly(styryl)lithium with 1,1-bis(4-t-butyldimethyl-siloxyphenyl)ethylene (26) followed by methanol termination and hydrolysis with dilute acid as shown in Scheme 11.12 [139]. No unfunctional polystyrene was

Scheme 11.12

detected by TLC analysis of the functionalized polymers (27, 28), which were also characterized by UV-visible, 1H and 13C NMR spectroscopy.

Copolymerization

The anionic copolymerization of substituted 1,1-diphenylethylene derivatives with copolymerizable monomers (M) will result in polymers with multiple functional groups along the polymer chain, as illustrated in Eq. 11.25. The number of functional groups per polymer molecule can be controlled by the monomer feed ratio and the molecular weight.



(11.25)

The anionic copolymerization of styrene and 1-(4-dimethylaminophenyl)-1-phenylethylene in benzene has been investigated [140]. Yuki and co-workers [135,141–143] have developed the formalism for analyzing the kinetics of copolymerization of 1,1-diphenylethylene (M2) with styrene and diene monomers (M1). It is assumed that the 1,1-diphenylethylene derivative, M2, does not add to itself due to steric effects (i.e., k22 = 0). Thus, the monomer reactivity ratio for M2 is zero (i.e., r2 = k22/k21 = 0). It is also assumed that the styrene monomer is completely consumed at the end of the polymerization [(M1) = 0] and that M2 is in excess, that is, there is still unreacted 1-(4-dimethylaminophenyl)-1-phenylethylene after the copolymerization. The resulting copolymerization equation is shown in Eq. 11.26, where r1 is not unity and (M1)0, (M2)0 and (M1), (M2) are the initial and final monomer concentrations, respectively.

(11.26)

The copolymerization of a 1.5 molar excess of styrene with 1-(4-dimethylaminophenyl)-1-phenylethylene produced a copolymer (Mn = 1.6 × 104 g/mol) with 24 amine groups per chain [140]. In analogous fashion, the copolymerization of a 3.6 molar excess of styrene with 1-(4-dimethylaminophenyl)-1-phenylethylene produced a copolymer (Mn = 3.8 × 104 g/mol) with 37 amine groups per chain. These copolymers did not move on TLC plates in toluene (i.e., Rf = 0) due to the large number of amine groups in the polymer molecules. The average value for r1 was determined to be 5.6, which means that styrene is 5.6 times as reactive as the amine-substituted 1,1-diphenylethylene towards the poly(styryl)lithium anion. The corresponding value of r1 is 0.71 for the copolymerization of styrene and 1,1-diphenylethylene in benzene at 30°C [143]. These results are reasonable, since the addition reaction of 1,1-diphenylethylene derivatives to poly(styryl)lithium in benzene at room temperature has a Hammett rho value of +1.8 [51]. The crossover reaction rate of poly(styryl)lithium with 1-(4-dimethyl-aminophenyl)-1-phenylethylene (k12) is expected to be much less than that of



poly(styryl)lithium to 1,1-diphenylethylene, due to the strong electron-donating effect of the dimethylamino group (σ = -0.83) [144]. Thus, the monomer reactivity ratio, r1 (r1 = k11/k12), is expected to be larger for the copolymerization of styrene with 1-(4-dimethylaminophenyl)-1-phenylethylene than in the copolymerization of styrene and 1,1-diphenylethylene as observed.

It was anticipated that the copolymerization of substituted 1,1-diphenylethylenes with dienes such as butadiene and isoprene would be complicated by the very unfavorable monomer reactivity ratio for the addition of poly(dienyl)lithium compounds to 1,1-diphenylethylene [142,143]. Yuki and co-workers [142,143] calculated values of r1 = 54 and r1 = 29 in hydrocarbon solutions for the copolymerization of 1,1-diphenylethylene (M2) with butadiene (M1 and isoprene (M1), respectively. Although the corresponding values in THF are r1(butadiene) = 0.13 and r1(isoprene) = 0.12, this would not be an acceptable solution since THF is known to form polymers with high 1,2-microstructures (see Chapter 9) [15,19,56,57]. Anionic copolymerizations of butadiene (M1) with excess 1-(4-dimethylaminophenyl)-1-phenylethylene (M2) were conducted in benzene at room temperature for 24–48 h using sec-butyllithium as initiator. Anisole, triethylamine, and t-butyl methyl ether were added in ratios of [B]/[RLi] = 60, 20, 30, respectively, to promote copolymerization. Narrow molecular weight distribution copolymers with Mn = 14 × 103-32 × 103 g/mol (Mw/Mn = 1.02–1.03) and 8, 12, and 30 amine groups per chain for anisole, triethylamine, and t-butyl methyl ether, respectively, were obtained. The butadiene monomer reactivity ratios (r1) and microstructures (%1,2) were 42 (14), 33 (29), and 14 (58) for anisole, triethylamine, and t-butyl methyl ether, respectively. It was anticipated that the addition of alkali metal (Mt) salts, for example, alkoxides, at low levels ([MtOR]/[Li] << 1) would promote copolymerization without significant increases in vinyl microstructure (see Chapter 10) [146,147]. Copolymerizations of 1-(4-dimethylaminophenyl)-1-phenylethylene (M2) with butadiene in the presence of potassium t-amyloxide ([KOR]/[Li] = 0.025) readily incorporated amine groups into the polybutadiene copolymer (r1 = 1.35) and the % 1,2-microstructure was 28% [148].

Fluorescent Group Labeling

Anionic functionalization methodology based on addition reactions to 1,1-diphenylethylenes and their analogs can also be utilized for the preparation of polymers labeled with fluorescent groups. Thus, poly(styryl)lithium can be quantitatively labeled with a fluorescent naphthyl end group via the reaction with 1-(2-naphthyl)-1-phenylethylene [149]. The adduct of 1-(2-naphthyl)-1-phenylethylene with butyllithium has been used to initiate the anionic polymerization of methyl methacrylate at low temperature in THF; the degree of labeling was reported to be 93% [150]. This method has also been utilized for the preparation of pyrene end-labeled polymers [151] as shown in Equation 11.27. Spectroscopic analyses [λmax(dioxane) 330, 346 nm; 1H and 13C NMR] indicate that this reaction



(11.27)

proceeds quantitatively to produce the polymers with fluorescent labels [151]. Since this fluorescent labeling methodology is a living functionalization reaction, the resulting living fluorescent-labeled polymer can be used to initiate the polymerization of a second monomer to produce block copolymer with the label at the block interface. For example, this procedure has been used to prepare polystyrene-block-poly(ethylene oxide) copolymers with both pyrene and naphthalene fluorescent groups at the interface between the two blocks [80]. This methodology has been extended to the preparation of anthracene and phenanthrene-labeled polymers [152,153].

Conclusions Regarding 1,1-Diphenylethylene Functionalizations

These results show that the addition reactions of simple and polymeric organolithium compounds with substituted 1,1-diarylethylene derivatives provide a general method for the synthesis of functionalized and labeled polymers. With this method in conjunction with appropriate protecting groups and reaction conditions, a wide variety of well-characterized, quantitatively functionalized, and labeled polymers and copolymers can now be prepared with diverse molecular structures as illustrated in Figure 11.1.

IV. Macromonomers

A. Introduction

Macromonomers are linear macromolecules carrying some polymerizable functional group at their chain end; the polymerizable functional groups can be at one chain end or at both chain ends [4,12,122,154–161]. Macromonomers are macro-molecular monomers, often referred to as “Macromers” [162]. The important feature of macromonomers is that they can undergo copolymerization with other monomers by a variety of mechanisms to form comb-type, graft copolymers [163,164], as shown in Scheme 11.13. This aspect of macromonomers differentiates them from telechelic (α,ω-difunctional) polymers [4,159]; telechelic polymers undergo step-growth type chain extension reactions with other monomers to form linear macromolecules, not branched structures. The polymerizable func



Scheme 11.13

tional group at the chain end of a macromonomer is often a vinyl group (28), but it can also be a heterocyclic ring such as an oxirane (epoxide) functionality (29). These functional groups participate in chain reaction polymerizations with other vinyl or heterocyclic monomers, respectively. A condensation-type macromonomer has two functional groups at one chain end (25) that can participate in step-growth (condensation) polymerization with other difunctional monomers as discussed previously in this chapter; for example, the functional group, X, could be hydroxyl, amino, carboxyl, or isocyanate [4,122,160]. A general method for

Figure 11.1 Molecular architecture for functionalized polymers prepared

by anionic polymerization.



anionic synthesis of condensation macromonomers has been described recently (see Eq. 11.22 and Scheme 11.12) [139].

Macromonomers provide a unique method of preparing graft copolymers with control of the branch structure (see Chapter 14). Graft copolymers formed from macromonomers (see Scheme 11.13) are often classified as comb-type, graft copolymers to differentiate them from the normal type, which have a distribution of branch lengths [164]. If a well-defined macromonomer can be prepared with a low degree of compositional heterogeneity, then copolymerization of this macromonomer with other monomers will form a graft copolymer in which the structure of the graft branches is also well defined. However, the incorporation of the macromonomer will be governed by the statistics of the copolymerization process (i.e., there will be compositional heterogeneity associated with the number and distribution of the graft branches in the graft copolymers) (see Chapter 14).

The molecular weights of macromonomers are generally in the range of 5 × 102- × 104 g/mole [5]. The choice of molecular weight is often a compromise between copolymerizability and the effect of branch length on physical properties. It is obvious that the concentration (number/gram) of functional groups decreases with increasing molecular weight of the macromonomer; thus, the copolymerization efficiency will decrease with increasing molecular weight. The graft copolymers formed from macromonomers are generally heterophase materials whose morphology and physical properties are dependent on their composition [164]. The optimization of morphology and physical properties requires the investigation of the effects of branch content and branch molecular weight.

B. Synthesis of Macromonomers

The concept of macromonomers as conceived and developed by Milkovich [154–156,162] is most clearly represented by the synthesis of a methacryloyl-terminated polystyrene as shown in Scheme 11.14. This synthesis illustrates some of the complexities and attributes of anionic macromonomer synthesis. First, poly(styryl)lithium (PSLi) can be synthesized readily with molecular weights ranging from less than 1 × 103 g/mole to the upper ranges of useful macromonomers (20–30 × 103 g/mole) [165]. However, poly(styryl)lithium is too reactive for direct



Scheme 11.14

reaction with methacryloyl chloride; vinyl addition competes with acylation [4]. The reactivity of the chain end was attenuated by first end-capping with ethylene oxide to form the corresponding lithium alkoxide (see Eq. 11.7 and previous discussion in this chapter). The polymeric alkoxide was then reacted with methacryloyl chloride to form the methacrylate-functionalized macromonomer. High functionality (> 90%) can be obtained for these reactions even using large scale equipment [155].

It is important to determine quantitatively the functionality of macromonomers. A common procedure is to subject the macromonomer to copolymerization with other monomers (e.g., using free-radical polymerization) and then determine the amount of “macromonomer” that does not undergo copolymerization by SEC; this method assumes that the amount of polymer remaining corresponds to the maximum amount of unfunctionalized macromonomer [154–156]. The vinyl functionality can be determined by titration with bromine [166] or mercuric acetate [167,168] and this can be used to estimate Mn; comparison of this value with either Mn(SEC) or Mn (osmometry) provides an estimate of functionality. The 1H NMR spectrum of the methacrylate-functionalized polystyrene exhibits resonances at δ = 5.4 and 5.8 (5.9) ppm for the vinyl protons (=CH2), at δ = 3.75 ppm for the oxymethylene group (-OCH2), and at δ = 1.8 ppm for the vinyl methyl group (=C-CH3) [166,168]. In general, high functionalities have been reported for the methacrylate-terminated polymers using these analytical methods.

This same methodology can be used to prepare diblock macromonomers as shown in Scheme 11.15. The living diblock copolymer polystyrene-block-poly(isoprenyl)lithium was first prepared by sequential monomer addition, followed by end-capping with ethylene oxide and then termination with methacryloyl chloride [169].

Macromonomer synthesis via the direct reaction of polymeric organolithium compounds with unsaturated alkyl halides (A, Scheme 11.16) is complicated by side reactions such as lithium-halogen exchange (B), coupling reactions to form dimeric species (C,D) and elimination reactions when a hydrogen is located in a vicinal position relative to the halogen [29]. For example, direct addition of a



Scheme 11.15

Scheme 11.16



benzene solution of poly(styryl)lithium (Mn = 6 × 103 g/mol) to a THF solution of p-vinylbenzyl chloride (8 molar excess) at 0°C yields only a 50% yield of the desired macromonomer; a 50% yield of dimer (SEC analysis) is also obtained [170]. However, when THF (17 vol %) was added to poly(styryl)lithium before the addition to p-vinylbenzyl chloride (13 molar excess), the macromonomer was obtained in quantitative yield. The macromonomer functionality (f = 0.9–1.1) was determined by UV analysis utilizing the high molar extinction coefficient of the styryl unit (ε = 1.64 × 1041/mol cm) compared to polystyrene (ε = 1.33 × 1021/mol cm) at λ = 250 nm. Other recent studies have confirmed the efficiency of this procedure [171]; functionalities of 95–99% were determined using 1H NMR by integrating the resonance for the methyl groups from the sec-butyl initiator fragment relative to the methylene group of the vinyl benzyl unit at the chain end (Mn = 2 × 103 to 17.5 × 103 g/mol). Reactions of poly(styryl)lithium with p-vinylbenzyl chloride at -78°C in a toluene/THF mixture provided macromonomers with functionalities of 90% by 1H NMR analysis and 71–81% by UV analysis [172]. Macromonomer functionalities (UV analysis) of only 72% have been reported for termination reactions polystyrene-block-poly(isoprenyl)lithium with p-vinylbenzyl chloride in a toluene/THF (100:1, vol/vol) mixture at -78°C. The analogous functionalizations of poly(α-methylstyrene)-block-poly(isoprenyl)lithium were reported to be quantitative [173].

It has been reported that macromonomer synthesis can also be effected by termination with p-vinylbenzyl tosylate [174]. The direct reaction of poly(isoprenyl)lithium with p-vinylbenzyl tosylate in either benzene or a benzene/THF mixture produced the corresponding macromonomer with reported functionalities of 96–100% and no dimer formation.

An extension of the concept of deactivation of the chain end to reduce side reactions is the use of 1,1-diphenylethylene end-capping as shown in Scheme 11.17 [175]. Both polymeric and simple organometallic compounds add quantitatively and relatively rapidly with 1,1-diphenylethylene to produce the corresponding 1,1-diphenylalkyl carbanion, as discussed previously in this chapter. The chain end reactivity is attenuated to minimize addition reactions to the double bonds in the terminating agents.

Since the desired substitution reactions may not occur to the exclusion of addition to the double bonds or lithium-halogen exchange reactions that lead to coupled polymers, it is important to carry out adequate characterization to establish the end-group functionality of the macromonomers. This vinylbenzyl-functionalized polystyrene was titrated using mercuric acetate [167] for end-group analysis of double bonds. This estimate of number average molecular weight was in good agreement with Mn (SEC). It is noteworthy that the use of 1-phenylethyl potassium as initiator in THF at -70°C resulted in polymers exhibiting somewhat broad molecular weight distributions (Mw/Mn = 1.1–1.15) [176].

Macromonomer syntheses by termination reactions with other alkyl halides have also been reported. For example, poly(styryl)lithium was terminated with



Scheme 11.17

allyl chloride [177]. Although direct characterization of the resulting macromonomer was not reported, copolymerization with a mixture of ethylene and propylene using vanadium Ziegler-Natta catalysts (e.g., VOCl3/Et3Al2Cl3) incorporated the macromonomer into the resulting polymer with a reported efficiency of 80%. It has been reported that termination reactions of living carbanions with both vinyidimethylchlorosilane and vinyl(chloromethyl)dimethylsilane are quantitative in THF over a wide temperature range, although rather minimal characterization of functionality was provided [178].

An alternative procedure for the synthesis of polystyrene macromonomers is to use an initiator with either a protected functional group or a polymerizable functional group stable to the anionic polymerization conditions. However, there are few examples of this type of macromonomer synthesis, primarily because most unsaturated groups are not stable to polystyryl carbanions.

Waack and Doran [179,180] have described the use of allyllithium and vinyllithium in THF as initiators for the anionic polymerization of styrene to produce the corresponding vinyl-substituted styrenes. Unfortunately, both of these initiators are relatively inefficient. Only 11% of allyllithium reacted with styrene, and only 5% of vinyllithium reacted; the result was that polystyrenes with molecular weights much higher than predicted by stoichiometry were obtained.

It would be expected that initiators with styryl units would not be useful for styrene polymerization because the styryl group would undergo competitive addition reactions with the growing carbanionic chain end. However, it has been



reported that 4-vinylbenzyllithium can be used to initiate the polymerization of styrene without addition to the initiator double bond [181]. 4-Methylstyrene was metalated with lithium diisopropylamide, as shown in Scheme 11.18. Unfortu-

Scheme 11.18

nately, the equilibrium metalation reaction does not proceed to form the desired 4-vinylbenzyllithium in very high yield; consequently, the residual 4-methylstyrene undergoes competing metalation during the polymerization reaction. However, it was reported that the resulting polystyrene macromonomer (Xn < 15) was monofunctional based on both UV-visible analysis (λ = 296 nm) and 1H NMR analysis of double bond functionality. No evidence for 4-methylphenyl groups was observed by 1H NMR. It should be noted that although 4-methylstyrene should be less reactive than styrene, it would not be expected to be unreactive. Competitive addition to the styrene unit at the chain end and branching would be expected if a higher molecular weight macromonomer synthesis was attempted. For example, the monomer reactivity ratio for styrene in copolymerizations with 4-methylstyrene is 0.72 in benzene [182]. An analogous functionalized initiator was formed from 4-trimethylsilylmethylstyrene and lithium diisopropylamide, which polymerized 4-trimethylsilylmethylstyrene to form a complex type of macromonomer because of competing polymerization and metalation [183].

Termination reactions of living poly(butadienyl)lithium with dimethylfulvene have been reported to incorporate cyclopentadiene chain-end functionality as shown in Scheme 11.19 [184]. The resulting cyclopentadienyl-functionalized



Scheme 11.19

polybutadiene was reacted with various dienophiles, but no other characterization was reported.

Polystyrene macromonomers with terminal 1,1-diphenylethylene functionality have been prepared by the reaction of one equivalent of poly(styryl)lithium with two equivalents of 1,4-bis(1-phenylethenyl)benzene in the presence of small amounts of THF ([THF]/[Li] = 20), as shown in Scheme 11.20 [185,186]. The macromonomer was characterized by UV spectroscopy utilizing the strong absorption of the 1,1-diphenylethylene group at 260 nm (ε 1.18 × 104) compared

Scheme 11.20



with the weak absorbance at this wavelength for polystyrene (ε = 182). In addition, for a macromonomer with Mn = 5.4 × 103 g/mole, 1H NMR spectroscopy could also be used to evaluate the chain end functionality. The macromonomer exhibited characteristic resonances for the terminal diphenylalkyl methine proton and vinyl protons at δ = 3.5 ppm and 5.4 ppm, respectively. Both of these methods indicate that the functionality is approximately 98% for addition reactions effected at 5–8°C in benzene; less than 1.4 % of the dimer adduct is obtained under these conditions.

The synthesis of macromonomers using the anionic polymerization of polar monomers will be discussed in Chapter 23.

V. Conclusions

The chemistry of living alkyllithium-initiated polymerization provides an excellent methodology for the synthesis of end-functionalized and in-chain functionalized polymers with high functionality and low degrees of compositional heterogeneity. The incorporation of one initiator residue per macromolecule provides a reliable method for the preparation of end-functionalized polymers by using functionalized initiators; however, this method is limited by the availability of functionalized initiators. The synthesis of functionalized polymers by post-polymerization reactions of living anionic polymers with specific electrophiles for different functional groups is general and useful. However, many of the functionalization reactions reported in the literature have not been adequately characterized or optimized for general utility. General functionalization methods based on addition or substitution reactions independent of the nature of the functional group provide the most versatile methods. Reactions of living polymers with substituted silyl chlorides or substituted 1,1-diphenylethylenes have been shown to provide useful, general methods for the synthesis of a wide variety of functionalized polymers. For all functionalization reactions, regardless of precedent, it is necessary to characterize adequately the functionalized polymer.

References

1. Telechelic Polymers: Synthesis and Applications, E. J. Goethals, Ed., CRC Press, Boca Raton, Florida, 1989.

2. D. N. Schulz, J. C. Sanda, and B. G. Willoughby, in Anionic Polymerization. Kinetics, Mechanisms, and Synthesis, J. E. McGrath, Ed., ACS Symposium Series No. 166, American Chemical Society, Washington, D.C., 1981, p. 427.

3. Reactive Oligomers, F. W. Harris and H. J. Spinelli, Eds., ACS Symposium Series No. 282, American Chemical Society, Washington, D.C., 1985.

4. P. F. Rempp and E. Franta, Adv. Polym. Sci., 58, 1 (1984).



5. J.-C. Brosse, D. Derouet, F. Epaillard, J.-C. Soutif, G. Legeay, and K. Dusek, Adv. Polym. Sci., 81, 167 (1987).

6. A. Akelah, J. Material Sci., 21, 2977 (1986).

7. G. Paulus, R. Jerome, and P. Teyssie, Brit. Polym. J., 19, 361 (1987).

8. Functional Polymers, D. E. Bergbreiter and C. R. Martin, Eds., Plenum Press, New York, 1989.

9. R. P. Quirk, J. Yin, S.-H. Guo, X.-W. Hu, G. Summers, J. Kim, L. Zhu, and L. E. Schock, Makromol. Chem. Macromol. Symp., 32, 47 (1990).

10. R. P. Quirk, in Comprehensive Polymer Science, First Supplement, S. L. Aggarwal and S. Russo, Eds., Pergamon Press, Elmsford, New York, 1992, p. 83.

11. R. P. Quirk, J. Yin, S.-H. Guo, X.-W. Hu, G. Summers, J. Kim, L.-F. Zhu, J.-J. Ma, T. Takizawa, and T. Lynch, Rubber Chem. Tech., 64, 648 (1991).

12. Y. Gnanou, Ind. J. Technol., 31, 317 (1993).

13. E. J. Goethals, in Ring-Opening Polymerization. Mechanisms, Catalysis, Structure, Utility, D. J. Brunelle, Ed., Hanser, New York, 1993, p. 295.

14. R. P. Quirk and J. Kim, in Ring-Opening Polymerization. Mechanisms, Catalysis, Structure, Utility, D. J. Brunelle, Ed., Hanser, New York, 1993, p. 263.

15. M. Morton, Anionic Polymerization: Principles and Practice, Academic Press, New York, 1983, Chapt. 11.




19. S. Bywater, in Encyclopedia of Polymer Science and Engineering, 2nd ed., J. I. Kroschwitz, Ed., Wiley-Interscience, New York, 1985, Vol. 2, p. 1.

20. M. van Beylen, S. Bywater, G. Smets, M. Szwarc, and D. J. Worsfold, Adv. Polym. Sci., 86, 87 (1988).

21. Recent Advances in Anionic Polymerization, T.E. Hogen-Esch and J. Smid, Eds., Elsevier, New York, 1987.

22. Anionic Polymerization. Kinetics, Mechanisms, and Synthesis, J. E. McGrath, Ed., ACS Symposium Series No. 166, American Chemical Society, Washington, D.C., 1981.

23. M. Szwarc and M. Van Beylen, Ionic Polymerization and Living Polymers, Chapman and Hall, New York, 1993.

24. D. H. Richards, G. C. Eastmond, and M. J. Stewart, in Telechelic Polymers: Synthesis and Applications, E. J. Goethals, Ed., CRC Press, Boca Raton, Florida, 1989, p. 33.

25. M. Fontanille, in Comprehensive Polymer Science, Vol. 3, Chain Polymerization I, G. C. Eastmond, A. Ledwith, S. Russo, and P. Sigwalt, Eds., Pergamon Press, Elmsford, New York, Chapter 27, 1989, p. 425.



31. L. Brandsma and H. D. Verkruijsse, Preparative Polar Organometallic Chemistry, Vol. 1, Springer-Verlag, Berlin, 1987.

32. R. B. Bates and C. A. Ogle, Carbanion Chemistry, Springer-Verlag, Berlin, 1983.

33. J. L. Wardell, in The Chemistry of the Metal-Carbon Bond, Vol. 4, F.R. Hartley, Ed., Wiley, New York, 1987, p. 1.

34. B. J. Wakefield, in Comprehensive Organometallic Chemistry, G. Wilkinson, Ed., Pergamon, Oxford, 1982, Chapter 44.

35. D. P. Wyman, V. R. Allen, and R. Altares, J. Polym. Sci., Part A, 2, 4545 (1964).

36. P. Mansson, J. Polym. Sci., Polym. Chem., Ed., 18, 1945 (1980).

37. R. P. Quirk and W.-C. Chen, Makromol. Chem., 183, 2071 (1982).

38. R. P. Quirk, J. Yin, and L. J. Fetters, Macromolecules, 22, 85 (1989).

39. R. P. Quirk and J. Yin, J. Polym. Sci., Part A: Polym. Chem., 30, 2349 (1992).



42. M. Morton, L. J. Fetters, J. Polym. Sci., Part A, 2, 3311 (1964).

43. R. Milner, R. N. Young, and A. R. Luxton, Polymer, 24, 543 (1983).

44. H. L. Hsieh and A. G. Kitchen, in Initiation of Polymerization, F. E. Bailey, Jr., Ed., ACS Symposium Series No. 212, American Chemical Society, Washington, D.C., 1983, p. 291.


46. A. Hernandez, J. Semel, H.-C. Broecker, H. G. Zachmann, and H. Sinn, Makromol. Chem., Rapid Commun., 1, 75 (1980).

47. D. Freyss, P. Rempp, and H. Benoit, J. Polym. Sci., Polym. Lett. Ed., B2, 217 (1964).

48. D. M. Wiles and S. Bywater, J. Polym. Sci., Polym. Lett. Ed., B2, 1175 (1964).


50. Z. Laita and M. Szwarc, Macromolecules, 2, 412 (1969).

51. R. Busson and M. van Beylen, Macromolecules, 10, 1320 (1977).

52. K. Ziegler and H. G. Gellert, Justus Liebigs Ann. Chem., 567, 179 (1950).

53. G. Kobrich and I. Stober, Chem. Ber., 103, 2744 (1970).

54. G. Wittig and U. Schollkopf, Chem. Ber., 87, 1318 (1954).

55. L. J. Fetters and R. N. Young, in Anionic Polymerization: Kinetics, Mechanisms, and Synthesis, J. E. McGrath, Ed., ACS Symp. Series 166, Am. Chem. Soc., Washington, D.C., 1981, p. 95.

56. S. Bywater, in Comprehensive Polymer Science, Vol. 3, Chain Polymerization I, G. C.



64. R. P. Quirk, J. Yin, L. J. Fetters, and R. V. Kastrup, Macromolecules, 25, 2262 (1992).

65. Unpublished work of Jian Yin, University of Akron.

66. A. Hirao, H. Nagahama, T. Ishizone, and S. Nakahama, Macromolecules, 26, 2145 (1993).

67. G. Broze, R. Jerome, and P. Teyssie, Makromol Chem., 179, 1383 (1978).

68. R. P. Quirk and J.-J. Ma, J. Polym. Sci. Part A: Polym. Chem., 26, 2031 (1988).

69. H. Sawada, J. Macromol. Sci.-Revs. Macromol. Chem., C5(1), 151 (1970).

70. F. E. Bailey and J. V. Koleske, Poly(ethylene Oxide), Academic Press, New York, 1976.

71. S. Patai, Ed., The Chemistry of the Ether Linkage, Interscience, New York, 1967.

72. C. J. Chang, R. F. Kiesel, and T. E. Hogen-Esch, J. Am. Chem. Soc., 95, 8446 (1973).

73. L. E. St. Pierre and C. C. Price, J. Am. Chem. Soc., 78, 3432 (1956).

74. N. N. Lebedev and Y. I. Baranov, Polym. Sci. USSR, 8, 211 (1966).

75. N. P. Doroshenko and Y. L. Spirin, Polym. Sci. USSR, 12, 2812 (1970).

76. I. Cabasso and A. Zilkha, J. Macromol. Sci.-Chem., A8(8), 1313 (1974).


78. K. Gonsalves and M. D. Rausch, J. Polym. Sci., Polym. Chem. Ed., 24, 1419 (1986).

79. R. P. Quirk and N.S. Seung, in Ring-Opening Polymerization: Kinetics, Mechanisms, and Synthesis, J. E. McGrath, Ed., ACS Symposium Series No. 286, American Chemical Society, Washington, D.C., Chapter 3, 1985, p. 37.

80. R. P. Quirk, J. Kim, K. Rodrigues, and W. L. Mattice, Makromol. Chem., Macromol. Symp., 42/43, 463 (1991).

81. S. Kobayashi, M. Kaku, T. Mizutani, and T. Saegusa, Polym. Bull., 9, 169 (1983).

82. M. Morton, L. J. Fetters, J. Inomata, D. C. Rubio, and R. N. Young, Rubber Chem. Tech., 49, 303 (1976).

83. D. N. Schulz, A. F. Halasa, and A. E. Oberster, J. Polym. Sci., Polym. Chem. Ed., 1974, 12, 153.

84. S. F. Reed, J. Polym. Sci., A-1, 10, 649 (1972).

85. W. J. Trepka, Macromolecules, 77, 497 (1984).

86. R. P. Quirk and M. Alsamarraie, Ind. Eng. Chem., Prod. Res. Dev, 25, 381 (1986).

87. R. P. Quirk, T. Takizawa, G. Lizarraga, and L.-F. Zhu, J. Appl. Polym. Sci., Polym. Symp., 50, 23 (1992).

88. L. E. Eberson, Electron Transfer Reactions in Organic Chemistry, Springer-Verlag, Heidelberg, 1987.



95. I. Hattori, A. Hirao, K. Yamaguchi, S. Nakahama, and N. Yamazaki, Makromol. Chem., 184, 1355 (1983).

96. R. P. Quirk and G. J. Summers, Brit. Polym. J., 22, 249 (1990).

97. R. P. Quirk and P. L. Cheng, Macromolecules, 19, 1291 (1986).

98. R. P. Quirk, W.-C. Chen, and P.-L. Cheng, in Reactive Oligomers, F. W. Harris and H. J. Spinelli, Eds., ACS Symposium Series 282, American Chemical Society, Washington, D.C., 1985, Chapter 12, p. 139.

99. P. Beak and B. J. Kokko, J. Org. Chem., 47, 2822 (1982).

100. C. D. Eisenbach, H. Schnecko, and W. Kern, Makromol. Chem., 776, 1587 (1975).

101. M. Stewart, N. Shepherd, and D. M. Service, Brit. Polym. J., 22, 319 (1990).

102. W. H. Dickstein and C. P. Lillya, Macromolecules, 22, 3882 (1989).

103. D. H. Richards, D. M. Service, and M. J. Stewart, Brit. Polym. J., 16, 117 (1984).

104. P. Charlier, R. Jerome, and P. Teyssie, Macromolecules, 23, 1831 (1990).

105. K. Ueda, A. Hirao, and S. Nakahama, Macromolecules, 23, 939 (1990).

106. R. P. Quirk and J. Kim, Macromolecules, 24, 4515 (1991).

107. J. Omeis, E. Muhleisen, and M. Moller, Polym. Prepr., Am. Chem. Soc., Div. Polym. Chem., 27(1), 213 (1986).

108. T. Durst and J. D. Manoir, Can. J. Chem., 47, 1230 (1969).

109. W. E. Truce and D. J. Vencur, Can. J. Chem., 47, 860 (1969).

110. P. Vanhoorne, G. Van den Bossche, F. Fontaine, R. Sobry, R. Jerome, and M. Stamm, Macromolecules, 27, 838 (1994).

111. J.-M. Catala, J. F. Boscato, E. Franta, and J. Brossas, in Anionic Polymerization. Kinetics, Mechanisms, and Synthesis, J. E. McGrath, Ed., ACS Symposium Series No. 166, American Chemical Society, Washington, D.C., 1981, p. 483.

112. J.-M. Catala, G., Reiss, and J. Brossas, Makromol. Chem., 178, 1249 (1977).

113. L. J. Fetters and E. R. Firer, Polymer, 18, 306 (1977).

114. R. P. Quirk and W.-C. Chen, J. Polym. Sci., Polym. Chem. Ed., 22, 2993 (1984).

115. E. J. Panek and G. M. Whitesides, J. Am. Chem. Soc., 94, 8768 (1972).

116. R. P. Quirk and J. Kuang, Polym. Internat., 33, 181 (1994).

117. M. O. Hunt, Jr., A. M. Belu, R. W. Linton, and J. M. DeSimone, Macromolecules, 26, 4854 (1993).

118. T. E. Long, L. W. Kelts, S. R. Turner, J. A. Wesson, and T. H. Mourey, Macromolecules, 24, 1431 (1991).



125. F. G. Bordwell, D. Algrim, and N. R. Vanier, J. Org. Chem., 42, 1817 (1977).

126. L. J. Fetters, in Encyclopedia of Polymer Science and Engineering, 2nd. ed., J. I. Kroschwitz, Ed., Wiley-Interscience, New York, 1987, Vol. 10, p. 19.

127. R. P. Quirk and L. Zhu, Makromol. Chem., 190, 487 (1989).

128. R. P. Quirk and L. Zhu, Brit. Polym. J., 23, 47 (1990).

129. G. Summers, Ph.D. thesis, University of Akron, 1991.



132. A. Hirao, K. Yamaguchi, K. Takenaka, K. Suzuki, S. Nakahama, and N. Yamazaki, Makromol. Chem., Rapid Commun., 3, 941 (1982).

133. A. Hirao, K. Takenaka, S. Packirisamy, K. Yamaguchi, and S. Nakahama, Makromol. Chem., 186, 1157 (1985).

134. A. Hirao, K. Kitamura, K. Takenaka, and S. Nakahama, Macromolecules, 26, 4995 (1993).


136. R. P. Quirk and T. Lynch, Macromolecules, 26, 1206 (1993).

137. A. Hirao, Y. Ishino, and S. Nakahana, Macromolecules, 21, 561 (1988).

138. A. Hirao and S. Nakahama, Polymer, 27, 309 (1986).

139. R. P. Quirk and Y. Wang, Polym. Int., 31, 51 (1993).

140. R. P. Quirk and L. Zhu, Polym. Int., 27, 1 (1992).

141. H. Yuki, J. Hotta, Y. Okamoto, and S. Murahashi, Bull. Chem. Soc. Jpn., 40, 2659 (1967).



144. J. A. Dean, Lang's Handbook of Chemistry, 13th ed., McGraw-Hill Book Company, New York, 1985, 3–136.

145. R. P. Quirk and J. Kuang, Makromol. Chem., Macromol. Symp., 85, 267 (1994).

146. H. L. Hsieh and C. F. Wofford, J. Polym. Sci., A-1, 7, 449 (1969).


148. J. Kuang, Ph.D. thesis, University of Akron, 1994.

149. R. P. Quirk, S. Perry, F. Mendicuti, and W. L. Mattice, Macromolecules, 21, 2294 (1988).

150. L. Chen, M. A. Winnik, E. T. B. Al-Takrity, A. D. Jenkins, and D. R. M. Walton, Makromol. Chem., 188, 2621 (1987).



158. P. Rempp, P. Lutz, P. Masson, P. Chaumont, and E. Franta, Makromol. Chem., Suppl., 13, 47 (1985).

159. P. Rempp, E. Franta, P. Masson, and P. Lutz, Progr. Colloid Polym. Sci., 72, 112 (1986).

160. Y. Kawakami, in Encyclopedia of Polymer Science and Engineering, J. I. Kroschwitz, Ed., Wiley-Interscience, New York, 1987, Vol. 9, p. 195.

161. V. Percec, C. Pugh, O. Nuyken, and S.D. Pask, in Comprehensive Polymer Science, Vol. 6, Polymer Reactions, G. C. Eastmond, A. Ledwith, S. Russo, and P. Sigwalt, Eds., Pergamon Press, Elmsford, New York, 1989, p. 281.

162. Macromer® T is a trademark of CPC International Inc.; R. Milkovich and M. T. Chiang, U.S. Patent 3,786,116 (1974).

163. Basic definitions of terms relating to polymers, Pure Appl. Chem., 40, 482 (1974).

164. P. Dreyfuss and R. P. Quirk, in Encyclopedia of Polymer Science and Engineering, J. I. Kroschwitz, Ed., Wiley, New York, Vol. 7, 1986, p. 551.

165. M. Morton and L. J. Fetters, Rubber Chem. Technol., 48, 359 (1975).

166. K. Ito, Y. Masuda, T. Shintani, T. Kitano, and Y. Yamashita, Polym. J., 15, 443 (1983).

167. J. B. Johnson, and J. P. Fletcher, Anal. Chem., 31, 1563 (1959).

168. P. Masson, E. Franta, and P. Rempp, Makromol. Chem., Rapid Commun., 3, 499 (1982).

169. G. O. Schulz and R. Milkovich, Ind. Eng. Chem. Prod. Res. Dev., 25, 148 (1986).

170. R. Asami, M. Takaki, and H. Hanahata, Macromolecules, 16, 628 (1983).

171. M. Arnold, W. Frank, and G. Reinhold, Makromol. Chem., 192, 285 (1991).

172. Y. Tsukahara, K. Tsutsumi, Y. Yamashsita, and S. Shimada, Macromolecules, 23, 5201 (1990).

173. K. Ishizu, K. Shimomura, and T. Fukutomi, J. Polym. Sci., Polym. Chem., 29, 923 (1991).

174. R. Asami and M. Takaki, Makromol. Chem., Suppl., 12, 163 (1985).

175. Y. Gnanou and P. Lutz, Makromol. Chem., 190, 577 (1989).

176. P. Rempp and E. Franta, in Recent Advances in Anionic Polymerization, T. E. Hogen-Esch and J. Smid, Eds., Elsevier, New York, 1987, p. 353.

177. J.-J. Ma, D. Pang, and B. Huang, J. Polym. Sci., Polym. Chem., 24, 2853 (1986).

178. P. Chaumont, J. Herz, and P. Rempp, Eur. Polym. J., 15, 537 (1979).

179. R. Waack and M. A. Doran, Polymer, 2, 365 (1961).

180. R. Waack and M. A. Doran, J. Org. Chem., 32, 3395 (1967).

181. Y. Nagasaki and T. Tsuruta, Makromol. Chem., 187, 1583 (1986).



12 Block Copolymers

I. Introduction

Block copolymers are macromolecules composed of linear arrangements of blocks; a block is composed of units that are chemically different, with respect to either chemical constitution or stereochemistry, from adjacent portions of the macromolecule [1–3]. Block copolymers can be considered as a combination of two of more macromolecules joined end-to-end. Block copolymers often exhibit unique and useful properties in solution and in the solid state as a consequence of the general thermodynamic incompatibility of the blocks which results in microphase separation into domains [4–9]. The properties of thermoplastic elastomers, which are generally block copolymers, are a direct consequence of the composition-dependent morphology of these polymers.

In the following sequence arrangements (1)—AAAAAAAAA-BBBBBBBBB— (2)—AAAAAAAAA-BBBBBBBBB-AAAAAAAA— (3)—AABABAAABB-AAAAAAAAA-BBBBBBBBB — the sequences — AAAAAAAAA —, — BBBBBBBBB — and — AABABAAABB— are blocks as described in the IUPAC recommendations for Source-Based Nomenclature for Copolymers [1]. The block sequence arrangements of (1), (2), and (3) are represented by Ak-block-Bm, Ak-block-Bm-block-Aj and (A-stat-B)-block-Ak-block-Bm, respectively, where k, m, and j correspond to the degrees of polymerization in the respective blocks. The corresponding polymers are named polyA-block-polyB, polyA-block-



Administrator

Highlight

Administrator

Pencil

Table 12.1 Block Copolymer Notations

Type of Block

Polymer Structure Notation Copolymer

-AAAAAAAA-BBBBBBBBB- A-B Diblock

-AAAAAAAA-BBBBBBBBB-AAAAAAAA- A-B-A Triblock

-AAAAAAAA-BBBBBBBBB-CCCCCCCCC- A-B-C Triblock

-AAAA-BBBB-(AAAA-BBBB)n-2-AAAA-BBBB-

(A-B)nMultiblock

(AAAAAAAA-BBBBBBBBB)n-X (A-B)n-X Stara

aThe subject of star-branched block copolymers is discussed in Chapter 13.

polyB-block-polyA, and poly(A-stat-B)-block-polyA-block-polyB, respectively. The order of citation of the block names corresponds to the order of succession of the blocks in the chain as written from left to right. For example, a styrene-butadiene-methyl methacrylate triblock copolymer would be named polystyrene-block-polybutadiene-block-poly(methyl methacrylate). If no ambiguity arises, a long dash may be used to designate block connections as follows: polyA—polyB. For complex cases, the use of -block- rather than the long dash is suggested. Another system in common usage is to substitute the abbreviation -b- in place of -block-.

A common shorthand notation system for representing block copolymers is shown in Table 12.1, where a letter represents a block of the corresponding monomer units. With this system, polystyrene-block-polybutadiene-block-polystyrene would be represented as S-B-S (often simply shown as SBS).

It is important to note that the present IUPAC nomenclature recommendations for block copolymers do not specifically designate the length of a block (i.e., the minimum number of repeating units required). However, it is stated in these recommendations that although short sequence lengths are not strictly embraced within the definitions of “block”, the same device may be usefully employed by using the general prefix “oligo” or the appropriate specific prefix (e.g., tri). Thus, this question of minimum block length is deferred to the distinction between a polymer and an oligomer, which is encompassed in earlier recommendations [2].

II. Synthesis of Block Copolymers

One of the most important synthetic applications of living polymerizations is the synthesis of block copolymers by sequential monomer addition [10,11]. In fact, this aspect of living polymerization was described and illustrated in the seminal paper of Szwarc et al. [12], which introduced the concept of living polymerization. Although block copolymers can be prepared by end-linking reactions of polymers



Administrator

Highlight

Administrator

Highlight

with reactive functional end groups, the methodology of living polymerization, especially living anionic polymerization, provides the best technique for preparation of well-defined block copolymers [8,13,14]. The ability to prepare block copolymers is a direct consequence of the stability of the carbanionic chain ends on the laboratory time scale, as discussed in Chapter 4. The effects of chain end structure, solvent, and temperature on the stability of carbanionic chain ends were discussed in Chapter 8. Since a living polymerization and the ability to prepare block copolymers requires the absence of chain termination and chain transfer reactions (see Chap. 4), monomer purity and the absence of side reactions with the monomer are requirements (see Chap. 5).

A further consideration for successful design and synthesis of block copolymers is the order of monomer addition. In general, a carbanionic chain end formed from one monomer will crossover to form the chain end of another monomer and initiate polymerization of this monomer, provided that the resulting carbanion is either of comparable stability or more stable than the original carbanion. The pKa values of the conjugate acids of carbanions provide a valuable guide to the relative stabilities of carbanions, as discussed in Chapter 2. Thus, crossover will generally occur to monomers that have conjugate acid pKa values the same or smaller than the pKa of the conjugate acids corresponding to the initiating carbanionic chain ends. For example, to prepare a block copolymer of methyl methacrylate and styrene it is necessary first to polymerize styrene (estimated pKa of toluene = 43) [15] and then add methyl methacrylate (estimated pKa of ethyl acetate = 30–31) [15] to form the second block; the ester enolate anion formed from methyl methacrylate cannot initiate polymerization of styrene. With these limitations in mind, living anionic polymerization provides a powerful synthetic method for preparing block copolymers with well-defined structures, including copolymer composition, block molecular weights, block molecular weight distributions, block sequence, and with low degrees of compositional heterogeneity [16].

There are three general methods for anionic synthesis of triblock copolymers: (1) three-step sequential monomer addition; (2) two-step sequential addition followed by coupling reactions; and (3) difunctional initiation and two-step sequential monomer addition [7–9,13,14,16–20]. Each of these methods has certain advantages and limitations that will be discussed in the following sections. The scope of this chapter will be limited to linear block copolymers. The subject of star-branched block copolymers will be considered in Chapter 13.

A. Block Copolymer Synthesis by Three-Step Sequential Monomer Addition

The preparation of block copolymers by sequential addition of monomers using living anionic polymerization and a monofunctional initiator is the most direct method for preparing well-defined block copolymers. Detailed laboratory proce-



dures for anionic synthesis of block copolymers are available [21–25]. Several important aspects of these syntheses can be illustrated by considering the preparation of an important class of block copolymers, the polystyrene-block-polydiene-block-polystyrene triblock copolymers, as shown in Scheme 12.1. The goal

Scheme 12.1

of each step in this sequence is to prepare a block segment with predictable, known molecular weight and narrow molecular weight distribution without incursion of chain termination or transfer. The first step is simple alkyllithium-initiated polymerization of styrene [26]. As indicated in Scheme 12.1, a hydrocarbon solvent is required to obtain polydiene block segments with high 1,4-microstructures and low glass transition temperatures (see Chap. 9).

For living anionic polymerization at complete monomer conversion, the number average molecular weight is uniquely defined by the simple relationship shown in Equation 12.1. Thus, to prepare 100 g of a polystyrene block with Mn = 15,000 g/mole, the required amount of butyllithium initiator would correspond



Administrator

Highlight

Administrator

Underline

(12.1)

to 6.7 × 10-3 moles. It is apparent that impurity levels for solvent, monomer, and equipment must be reduced to lower than millimolar amounts to obtain the desired molecular weights. It is also important to note that the level of impurities that can be tolerated decreases with increasing molecular weight of the desired polymer because of the decreasing amounts of initiator required (10-

6–10-7M concentrations for Mn > 106 g/mole) [27]. To obtain a block segment with narrow molecular weight distribution (Mw/Mn < 1.1[28]) in a living polymerization, it is necessary to use a reactive initiator that effects a rate of initiation competitive with or faster than propagation (see Chap. 4) [29]. Hsieh and McKinney [30] have shown that this condition is fulfilled for sec-butyllithium, but not for n-butyllithium, with styrene and diene monomers. In a three-stage block copolymer system, at the completion of each step a sample of polymer can be removed and characterized independently with respect to molecular weight, molecular weight distribution, and composition.

The second step in the three-step synthesis of a triblock copolymer by sequential monomer addition requires that the carbanionic chain end of the first block initiates polymerization of the second (diene) monomer. The monomer added at this step must be very pure to prevent significant termination of the active poly(styryl)lithium chain ends. To the extent that this occurs, the final product will be contaminated with polystyrene homopolymer and the molecular weight of the second block will be increased because of a decrease in chain end concentration, in accord with Equation 12.1. The carbanion formed from addition to the second monomer must be either more stable or of comparable stability relative to the propagating carbanionic chain end corresponding to the first block segment. Thus, styrene and diene monomers can crossover to each other and also to a variety of polar vinyl and heterocyclic monomers, as described in Chapter 5. To obtain a narrow molecular weight distribution for the second block, the rate of crossover to the second monomer (i.e., the initiation reaction for the second block) must be competitive with or faster than propagation. This can be a problem for polar monomers that exhibit rapid rates of propagation. The crossover reaction of poly(styryl)lithium to butadiene to form a poly(butadienyl)lithium chain end is a very fast reaction compared to the rate of propagation for butadiene monomer, as discussed in Chapter 10 with respect to copolymerization. The order of reaction rate constants is kSB > kSS > kBB > kBS [31]. An analogous kinetic situation exists for the order of reaction rate constants for styrene and isoprene [32]. Thus, it is relatively easy to obtain polystyrene-block-poly(butadienyl)lithium diblock segments with controlled, predictable block molecular weights and narrow molecular weight distributions. An example of diblock copolymer synthesis is shown in Figure 12.1 [29]. The first block segment of poly(styryl)lithium (Mn = 1.9 × 103



Administrator

Underline

Administrator

Underline

Administrator

Highlight

Administrator

Highlight

Administrator

Highlight

Administrator

Highlight

Administrator

Highlight

Figure 12.1 SEC curves for polystyrene-block-polybutadiene

synthesis: ———, polystyrene base polymer; ——, polystyrene-block-polybutadiene. (From Ref.

29; reprinted by permission of the Society of Chemical Industry, London, UK.)

g/mole; Mw/Mn = 1.04) was crossed over with butadiene to form a diblock copolymer with Mn = 11.3 × 103 g/mole and Mw/Mn = 1.02. The absence of a peak corresponding to the first polystyrene segment in the final diblock copolymer is an indication of the absence of chain termination or chain transfer reactions; this requires the use of high purity monomers. The overall narrow molecular weight distribution is consistent with the rapid rate of crossover of poly(styryl)lithium to butadiene (initiation) relative to butadiene propagation.

In the third step, styrene monomer is added, usually in an amount that corresponds to formation of an end block with the same molecular weight as the first polystyrene block. The purity of styrene monomer is critical at this step also, since if impurities are present the final triblock copolymer will be contaminated with the diblock copolymer, polystyrene-block-polybutadiene. In addition, impurities will lead to an increase in the molecular weight of the third block because of a decrease in chain end concentration in accord with Equation 12.1. Unfortunately, the rate of the crossover reaction of poly(butadienyl)lithium to styrene monomer to form a poly(styryl)lithium chain end is slow compared to the rate of styrene propagation, as discussed in Chapter 10 and in the previous section. Because of the slow rate of styrene initiation relative to propagation, a broad molecular weight distribution would be expected for the final polystyrene block segment. To obtain polystyrene end blocks with narrow molecular weight distributions, a Lewis base such as an ether or amine is often added before styrene



Administrator

Highlight

monomer addition in this third stage of the triblock copolymer synthesis [13,33–37]. An example of a three-step synthesis of a polystyrene-block-polydiene-block-polystyrene triblock copolymer with a myrcene (7-methyl-3-methylene-1,6-octadiene) center block is shown in Figure 12.2 [38]. It is difficult to evaluate the purity of the final product with respect to diblock copolymer contamination in the triblock copolymer because size exclusion chromatography (SEC) generally cannot separate these two components.

B. Block Copolymer Synthesis by Two-Step Sequential Monomer Addition and Coupling

Another general method for synthesis of A-B-A triblock copolymers involves a two-step sequential monomer addition sequence followed by a coupling reaction with a difunctional electrophilic reagent, as illustrated in Scheme 12.2. For the

Scheme 12.2

two-step synthesis of a polystyrene-block-polybutadiene-block-polystyrene triblock copolymer analogous to the polymer prepared by the three-step sequential monomer addition process, the first polystyrene block synthesis and the crossover reaction with butadiene monomer would be analogous to the three-step process.



Administrator

Highlight

Administrator

Highlight

Administrator

Highlight

Figure 12.2 SEC curves of poly(styrene-block-myrcene-block-styrene)

(55,000–80,000–55,000) and each block segment. (From Ref. 38; reprinted by

permission of Plenum Press, New York.)



However, the amount of butadiene monomer added in the second step would correspond to only half of the amount required for the three-step process. The coupling reaction can be effected efficiently with a variety of difunctional linking agents such as α,α'-dichloro-p-xylene, which is illustrated in Scheme 12.2. A list of useful difunctional coupling agents is shown in Table 12.2.

This two-step method with coupling offers many advantages over the three-step sequential monomer addition method. From a practical point of view, the polymerization time is reduced to one-half that required for the three-step synthesis of a triblock copolymer with the same molecular weight and composition[42]. One problem avoided in the two-step process is that the final crossover step from poly(butadienyl)lithium to styrene is eliminated. As discussed previously, this rate of crossover (initiation) from poly(butadienyl)lithium to styrene is slow compared to the rate of styrene propagation, which results in a broadening of the molecular weight distribution of the final polystyrene end block [37]. Although ethers or amines can be added before styrene addition (but after diene addition to prevent formation of high vinyl polydiene microstructure), these polar additives decrease the stability of the chain end (see Chap. 8) and can be a problem in commercial processes that utilize solvent recycling [37]. The elimination of the third monomer addition step also decreases the possibility of termination by impurities in a third monomer addition.

Table 12.2 Difunctional Coupling Agents for Triblock Copolymer Synthesis

Coupling Agent Coupling Efficiency (%)

α,α'Dibromo-p-xylene 94

α,α'Dichloro-p-xylene 94

bis(Chloromethyl)ether 95

Methylene iodide 94

Iodine 93

1,4-Dibromo-2-butene 91

1,4-Diiodo-2-butene 90

Ethyl acetate 90

Terephthalaldehyde 83

Anthraquinone 81

Phosgene 74 [3:6]

Dichlorodimethylsilane 70–90 [4:0]

Methylene bromide —[19] a

1,2-Dibromoethane 80[41]

a No yields reported.

Source: Ref. 39.



Administrator

Highlight

Administrator

Highlight

Administrator

Pencil

Another advantage of the two-step process is that it is more versatile with respect to the chemical composition of the center block. With the two-step method, the center block can be a more reactive monomer that would not be capable of reinitiating polymerization of styrene because of the increased stability of the chain end. For example, a polystyrene-block-poly(methyl methacryloyl)lithium diblock could be coupled with αα'-dibromo-p-xylene [43] to form a polystyrene-block-poly(methyl methacrylate)-block-polystyrene triblock; this tri-block copolymer could not be synthesized directly by three-step sequential monomer addition sequence. Another example is the synthesis of poly(α-methylstyrene)-block-poly(propylene sulfide)-block-poly(α-methylstyrene) by the stepwise poly-merization of (a) α-methylstyrene followed by (b) propylene sulfide and then coupling the active lithium thiolate chain ends with phosgene [44]. Only 5% of the uncoupled diblock copolymer was present in the final product. The synthesis of poly (α-methylstyrene)-block-polydimethylsiloxane-block-poly(α -methylstyrene) was effected by sequential polymerization of α-methylstyrene followed by crossover to the cyclic trimer of dimethylsiloxane (D3) after first capping the chain ends with a few units of butadiene to promote the crossover reaction [45]. The resulting diblock copolymer was coupled with dimethyldichlorosilane to form the desired triblock copolymer with only 5% of the diblock copolymer remaining after the coupling reaction [42,45].

The limitations of the two-step monomer addition plus coupling procedure are primarily associated with the coupling reaction. The efficacy of the coupling step requires both an efficient coupling reaction and high precision in controlling the stoichiometry of this reaction [42]. For example, it is useful to add the coupling reagent in increments, particularly if the anionic chain end exhibits an ultraviolet (UV)-visible absorption that can be monitored. However, in practice, it is difficult to control the stoichiometry of the coupling reaction and many two-step syntheses yield triblock copolymers with significant amounts of uncoupled diblock contaminants [42,43,46]. It is reported that commercial Kraton triblock copolymers possess bimodal molecular weight distributions with 15–20 wt% of apparently diblock polymers with one-half of the molecular weight of the triblock copolymer [42,46]. In general, the presence of diblock material affects the triblock copolymer morphology [46], which has a detrimental effect on the physical properties of triblock copolymers [36,42]. However, the presence of diblock material reduces the viscosity of the polymer that facilitates processing.

C. Block Copolymers by Difunctional Initiation and Two-Step Sequential Monomer Addition

One of the most versatile methods for the synthesis of A-B-A triblock copolymers is the use of a difunctional initiator with a two-step sequential monomer addition sequence as illustrated in Scheme 12.3. A practical two-step synthesis of a



Administrator

Pencil

Administrator

Text Box

光气, 碳酰氯

Administrator

Underline

Administrator

Underline

Administrator

Underline

Administrator

Highlight

Administrator

Underline

polystyrene-block-polybutadiene-block-polystyrene triblock copolymer requires the use of a hydrocarbon-soluble, dilithium initiator such as the initiator shown in Scheme 12.3, which is formed by the addition of 2 moles of sec-butyllithium to

Scheme 12.3

1,3-bis(1-phenylethenyl)benzene (DDPE) [47–49] (see Chap. 5 for a discussion of dilithium initiators). A hydrocarbon-soluble dilithium initiator ensures that the polydiene center block will have a high 1,4-microstructure and a correspondingly low glass transition temperature. The number average molecular weights of the center diene block and the polystyrene end blocks of the copolymer are uniquely determined by the stoichiometry of the reaction, in accord with Equation 12.2.



Administrator

Underline

Administrator

Underline

Administrator

Highlight

Administrator

Underline

Administrator

Underline

(12.2)

One of the advantages of using a difunctional initiator and a two-step sequential monomer addition is the fact that, analogous to the two-step coupling method, the elimination of a third monomer addition step decreases the possibility of termination by impurities in a third monomer addition. Also, the dilithium initiator is the most versatile method with respect to the chemical composition of the end blocks. The α,ω-dilithiumpolydiene center block can be used to initiate the polymerization of polar monomers to form both end blocks simultaneously. For example, this initiator has been utilized to prepare poly(t-butyl methacrylate)-block-polyisoprene-block-poly(t-butyl methacrylate) triblock copolymers [47]. This type of triblock copolymer with polar end blocks cannot be prepared either by the three-step sequential monomer addition process or the two-step monomer addition with coupling method.

An example of the use of this dilithium initiator to prepare a polystyrene-block-polybutadiene-block-polystyrene triblock copolymer is shown in Figure 12.3 [48]. Noteworthy about this synthesis is that although the dilithium initiator is soluble in hydrocarbon media for extended periods of time, it produces polystyrenes and polybutadienes that exhibit bimodal molecular weight distributions [48]. To obtain polymers with relatively narrow molecular weight distributions it was necessary to add two equivalents of lithium sec-butoxide per equivalent of dilithium initiator. The addition of lithium alkoxide increased the amount of 1,2 microstructure in the polybutadiene block to 12–14%. Because of the relatively slow cross-over rate of poly(butadienyl)lithium to styrene monomer, the addition of tetrahydrofuran (THF) is recommended to ensure that the polystyrene end blocks have narrow molecular weight distributions [50].

It should be noted that the classic difunctional initiators are aromatic radical anions such as sodium naphthalene [51–53]. These initiators are of limited utility for the preparation of elastomeric block copolymers because they are prepared and utilized in polar solvents such as tetrahydrofuran that result in polydienes with high vinyl microstructures and relatively high Tg values (see Chap. 9).

The main limitation with respect to the synthesis of triblock copolymers using dilithium initiators is the fact that most dilithium initiators are not soluble in hydrocarbon media and require the addition of polar additives such as ethers and amines for solubilization. However, a useful dilithium initiator is formed by the dimerization of 1,1-diphenylethylene with lithium in cyclohexane in the presence of anisole (15 vol %) [54]. The polystyrene-block-polyisoprene-block-polystyrene and poly(α-methylstyrene)-block-polyisoprene-block-poly(α-methylstyrene) triblock copolymers formed using this initiator contained polyisoprene center blocks with > 90% 1,4-microstructure (65–70 % cis-1,4; 20–25% trans-1,4). Further-



Administrator

Underline

Administrator

Underline

Administrator

Underline

Administrator

Highlight

Administrator

Highlight

Administrator

Underline

Administrator

Underline

Administrator

Highlight

Administrator

Highlight

Administrator

Underline

Figure 12.3 SEC curves for polybutadiene [Mn(calc) =

58.9 × 103 g/mol] prepared with the dilithium initiator, DDPELi2

(Scheme 12.3), and lithium sec-butoxide ([BuOLi]/[DDPELi2] = 1.08) (A) and a

poly(styrene—block—butadiene—block—styrene) triblock copolymer [Mn (calc) =

98.8 × 103 g/mol] prepared by subsequently adding styrene to this α,ω-dilithiumpolybutadiene (B).

(From Ref. 48; reprinted by permission of the Society of Chemical Industry, London, UK.)

more, the polymers had predictable molecular weights and narrow molecular weight distributions (Mw/Mn = 1.07). Another problem with respect to dilithium initiators is due to the association of the polymeric organolithium chain ends (see Chap. 1), which can result is higher viscosities of the dilithium polymer solutions compared to the monolithium analogs.

D. Block Copolymers by Two-Step Synthesis with Copolymerization

The alkyllithium-initiated copolymerization of a mixture of styrene and diene monomers in hydrocarbon solution produces a tapered, or graded, block copolymer structure (1) as described in Chapter 10; for example, the monomer reactivity



Administrator

Underline

Administrator

Highlight

Administrator

Underline

Administrator

Underline

ratios are 15.5 and 0.04 for butadiene and styrene, respectively, in cyclohexane at 25°C [55]. As a consequence of these copolymerization parameters, the less reactive butadiene monomer is preferentially polymerized first, then as butadiene is depleted a segment with mixed composition is formed, followed by a homopolymer block of polystyrene at the end. A type of triblock copolymer can be prepared by a two-step synthesis utilizing this copolymerization behavior. In the first step, a block segment of poly(styryl)lithium is prepared in hydrocarbon solution by the usual methodology (see Scheme 12.1). In the second step, a mixture of styrene and diene is added to form the final two block segments, as shown in Scheme 12.4.

Scheme 12.4

An alternative procedure utilizing a dilithium initiator involves only one comonomer addition step. Thus, the reaction of a dilithium initiator with a mixture of diene and styrene monomers in hydrocarbon solution forms the center poly-butadiene block first, followed by two tapered block regions and finally the two polystyrene end blocks. The dilithium initiator shown in Scheme 12.3 has been used for such a triblock copolymer synthesis utilizing a mixture of butadiene and styrene dissolved in α-methylstyrene as solvent [49]. The terminal end blocks were composed of a copolymer of styrene and α-methylstyrene. The direct polymerization of isoprene in α-methylstyrene as solvent at 50°C produced polyisoprene that contained less than 2% incorporation of α-methylstyrene. It was observed that the properties of such tapered block copolymers were comparable to those of the triblock copolymer formed by sequential monomer addition. Commercial dilithium initiators based primarily on 1,3-diisopropenylbenzene have also been utilized to prepare analogous tapered triblock copolymers in α-methylstyrene [56].



Administrator

Underline

Administrator

Underline

The advantages of this method include the two-step monomer addition process that minimizes termination by impurities that can occur in a three-stage polymerization process. The disadvantages are the broad compositional interfaces that result in the second copolymerization step; broad compositional interfaces (e.g., [BD/S]) tend to promote interfacial mixing and this can have a deleterious effect on ultimate properties of the triblock copolymer [57,58].

E. Styrene and Diene Monomers

From a practical point of view, the most important types of hydrocarbon monomers that undergo living alkyllithium-initiated polymerization to form block copolymers with well-defined structures are styrenes and dienes. These monomers can be anionically polymerized over a wide range of useful temperatures from room temperature to elevated temperatures (see Chap. 8). Furthermore, alkyllithium-initiated polymerizations of dienes in hydrocarbon solution produce elastomeric block segments with low glass transition temperatures, which are useful as elastomers and adhesives. The block copolymers of a variety of styrenes and dienes have consequently been intensively and extensively investigated to probe the limits of the synthetic methods and to delineate the interrelationships among structure, morphology, and properties. A representative list of the types of block copolymers that can be prepared from these monomers using the different anionic polymerization methods is shown in Table 12.3. More extensive tabulations can be found in the review literature [6–9,13,17,20,59–61]. Summaries have been presented by Nakahama and co-workers [62,63] that describe the extensive investigations of the homopolymerization and block copolymerizations of styrene monomers substituted with protected, polar functional groups.

F. Polar Monomers

The range of polar monomers that undergoes controlled anionic polymerization is limited because of the occurrence of side reactions of the initiators or the growing anionic chain ends with the polar functionality in the monomer or polymer. Another side reaction is the enolization of acidic protons adjacent to unsaturated, polar functional groups such as carbonyl, ester, nitrile, nitro, and sulfonyl groups. These enolization reactions have until recently prevented the controlled anionic polymerization of monomers such as acrylates, acrylonitrile, and related compounds, compared to methacrylates and methacrylonitriles. A detailed discussion of the anionic polymerization of polar monomers is presented in Chapter 23.

The controlled, living anionic polymerization of alkyl methacrylates can be performed at low temperatures (-78°C) [[85]. The proper choice of initiator is critical, however, since the use of an initiator that is too reactive causes side reactions with the monomer and the growing polymer chains (see Chap. 5). For example, when n-butyllithium is used as initiator for methyl methacrylate poly



Administrator

Underline

Administrator

Underline

Administrator

Text Box

烯醇化(作用)

Administrator

Highlight

Administrator

Pencil

Table 12.3 Block Copolymers of Styrenes and Dienes

A B C Initiator Solvent

Styrene Butadiene sec-BuLi C6H6

a

C6H6/C6H12b

n-BuLic C6H6

Styrene EtLi C6H14

DiLid C6H6d

Butadiene styrenee

n-BuLi and sec-BuLi

C6H6

Styrene-co-α- methylstyrene

Butadiene Styrene-co-α-methylstyrene

sec-BuLi or DDPELi2f

C6H12g

Styrene Butadiene Styrene DDPELi2fC6H12

Styrene Isoprene sec-BuLi C6H6/C6H12b

C6H6

n-BuLi

sec-BuLi THF

Styrene C6H12

EtLi C6H6c

n-BuLi C6H6

Styrene-b-isopreneh sec-BuLi

Styrene Myrcene Styrene sec-BuLi C6H6





AB C Initiator Solvent

Butadienei Butadienej n-BuLic C6hH6j

Isoprene Butadienej sec-BuLi C6H12j

p-t-Butylstyrene Butadiene p-t-Butylstyrene sec-BuLi C6H6

p-t-Butylstyrene Isoprene p-t-Butylstyrene

p-t- Butadiene Styrene

p-Methylstyrene Butadiene p-Methylstyrene

αMethylstyrene Isoprene αMethylstyrene Li2TPBm C6H12n

Butadiene 2-Isopropenylnaphthalene

n-BuLi Toluene

2-Vinylnaphthalene N,N-Dimethylaminostyrene

C6H6

aTriethylamine added prior to butadiene monomer.

b10/90, vol/vol.

cAnisole added to promote initiation ([anisole]/[Li] = 3-10).

dCommercial dilithium initiator from Lithium Corporation; solvent contained 8% dimethyl ether.

eStyrene addition followed by addition of both butadiene and styrene to form a tapered block copolymer.

fAdduct of 2 moles of sec-butyllithium with 1,3-bis(1-phenylethenyl)benzene.

gα-Methylstyrene as cosolvent for some polymerizations.

h(Polystyrene-block-polyisoprene) n multiblock copolymer.

i89%, 1,4

j98% 1,2; 2 equivalents of 1,2-dipiperidinoethane added prior to second butadiene addition.

kAfter hydrogenation

lSamples described as ''near-monodisperse.''

m1,4-Dilithio-1,1,4,4-tetraphenylbutane as initiator in the presence of 15 vol % anisole.

nDiethyl ether (250/90, vol/vol, Et2O/cyclohexane) was added prior to α-methylstyrene polymerization.

Also see ref. 13.




AB C Initiator Solvent

Butadiene Hexamethyl- cyclotrisiloxane

BuLi RHi

2-Isopropenyl- naphthalene

Hexamethyl- cyclotrisiloxane

BuLi THF

α-Methylstyrene Hexamethyl- cyclotrisiloxane

α-Methylstyrene

sec-BuLi THFj

2-Vinylpyridine Hexamethy- cyclotrisiloxanek

Sodium naphthalene THF

Methyl methacrylate Glycidyl methacrylate BuLi/DPEc,d THF

Methyl methacrylate

Butadiene ω-Caprolactone 1) n-BuLil C6H12m

2) Et2AlCk

Butadiene ε-Caprolactone

Styrene ε-Caprolactone Sec-BuLin Toluene

Styrene 2,3-Dimethyl- trimethylene- carbonate

Pivalolactone Isoprene Pivalolactone 1) DIPLi 1) C6H12

2) CO22) THF

3) TBAOHo

Styrene 2-Vinylpyridine Butadienep C6H6; THFq

Isoprene Styrene 2-Vinylpyridine Cumyl potassium THF

Propylene sulfide Isoprene Propylene sulfide

Li2 TPBs C6H14 or C6H6

α-Methylstyrene Propylene sulfidet α-Methylstyrene

THF

Footnotes on p. 326.

(Table continued on next page)



Table 12.4 Block Copolymers of Heterocyclic and Polar Vinyl Monomers

A B C Initiator Solvent

Styrene Ethylene oxide sec-BuLi C6H6/DMSOa

Cumyl potassium THF

Ethylene oxide Styrene Ethylene oxide Potassium naphthalene THF

Styrene t-Butyl acrylate LiαMeSb THFc

BuLi/DPEd THFc

t-Butyl acrylate Styrene t-Butyl acrylate Lithium naphthalene

Styrene Stearyl methacrylate LiαMeSb

t-Butyl methacrylate Cumyl potassium THF

t-Butyl methacrylate Isoprene t-Butyl methacrylate

DDPELi2e 1) C6H12

2) THF

t-Butyl methacrylate S-b-1-b-Sf t-Butyl methacrylate

Styrene Butadiene Methyl methacrylate

BuLig THF

Methyl methacrylate t-Butyl acrylate LiαMeSb,c THF

t-Butyl acrylate Methyl methacrylate t-Butyl acrylate TPBLi2h THFc

2-Vinylpyridine t-Butyl acrylate BuLi/DPEd or TPBLi2h

THF




(Table continued from previous page)

Footnotes for Table 12.4

a DMSO (30–60 vol %) was added for ethylene oxide polymerization step.

b Oligomeric poly(α-methylstyryl)lithium.

c Lithium chloride added; [LiCl]/[initiator] = 5 (ref. 5, 17; variable refs. 6, 13).

d Addition product of butyllithium and 1,1-diphenylethylene.

e Adduct of 2 moles of sec-butyllithium with 1,3-bis(1-phenylethenyl)benzene.

f The center block was a polystyrene-block-polyisoprene-block-polystyrene triblock segment.

g “Lithium alkoxides” were generated prior to polymerization by decomposition of butyllithium in THF (see Chap. 8).

h 1,1,4,4-Tetraphenyl-1,4-dilithiobutane was obtained from lithium naphthalide and 1,1-diphenylethylene.

i Hydrocarbon solvent not specified; siloxane polymerization effected in the presence of THF as promoter.

j A small amount of butadiene was added prior to the siloxane monomer addition; triblock copolymer formed by couplindichlorodimethylsilane.

k Polymers named as poly(4-vinylpyridine)- b-poly(dimethylsiloxane); however, sodium naphthalene initiator should ge(4-vinylpyridine)- b- poly(dimethylsiloxane).

l Poly(butadienyl)lithium was end-capped with ethylene oxide prior to lactone polymerization.

m THF as a promoter for lactone polymerization.

n Poly(styryl)lithium was end-capped with ethylene oxide, formaldehyde, or ω-caprolactone before the second stage of p

o DIPLi is a difunctional initiator prepared from 2 moles of sec-butyllithium and m-diisopropenylbenzene in the presencby addition of 5 equiv. isoprene. The chain ends were carboxylated and then converted to the tetrabutylammonium salts (TBAOH).

p Final block prepared by coupling of poly(butadienyl)lithium with the excess p-xylene dichloride-functionalized dibloc1,1-diphenylethylene.

q Benzene was solvent for styrene and butadiene polymerizations; THF was the solvent for 2-vinylpyridine polymerizati

r After fractionation.

s 1,4-Dilithio-1,1,4,4-tetraphenylbutane as initiator.

t Triblock copolymer obtained by coupling diblock with phosgene; approximately 5% uncoupled diblock remains in trib



merization, 52% of the initiator is destroyed by reaction with the ester functionality to produce unreactive lithium ethoxide [86]. The use of less reactive, hindered initiators such as 1,1-diphenylhexyllithium and low temperatures permits the synthesis of poly(alkyl methacrylates) with control of the major structural variables [87]. With the use of these techniques, block copolymers of styrenes with alkyl methacrylates have been prepared by reacting the anionic styryl chain end with 1,1-diphenylethylene prior to addition of the methacrylate monomer [88–90].

Teyssie and co-workers [91,92] have recently reported that the ability to prepare poly(alkyl methacrylates) and poly(t-butyl acrylates) with controlled molecular weight and relatively narrow molecular weight distribution can be enhanced by the addition of lithium chloride at -78°C in THF. This methodology has been applied to the synthesis of poly(methyl methacrylate)-b-poly(t-butyl acrylate) block copolymers with good control of block size and molecular weight distribution (Mw/Mn < 1.15) [93].

Anionic polymerization of heterocyclic monomers such as oxiranes (epoxides), thiiranes (episulfides), cyclic siloxanes, and lactones can be used to prepare block copolymers with controlled structures [94]. A representative list of the types of block copolymers that can also be prepared by anionic polymerization using polar vinyl and heterocyclic monomers is shown in Table 12.4. It is important to note that many of the polar monomers listed in Table 12.4 undergo side reactions during anionic polymerization that prevent precise control of molecular weight and structure for the corresponding block copolymers (see Chap. 5). The use of these monomers for block copolymer synthesis is often limited to the preparation of the last-formed block in the sequence [35]. The monomers in this category include acrylonitriles, methacrylonitriles, cyanoacrylates, 2-methyloxirane (propylene oxide), vinyl ketones, acrolein, vinyl silanes, vinyl sulfones, halogenated monomers, ketenes, nitroalkenes, vinylpyridines, and isocyanates.

References

1. IUPAC Commission on Macromolecular Nomenclature, Pure Appl. Chem., 57, 1427 (1985).

2. Basic definitions of terms relating to polymers (1974), Pure Appl. Chem., 40, 479 (1974).

3. M. V. Metanomski, in International Union of Pure and Applied Chemistry, Blackwell, Scientific Publications, Oxford, 1991.

4. F. S. Bates and G. H. Fredrickson, Annu. Rev. Phys. Chem., 41, 525 (1990).

5. M. J. Folkes, Ed., Processing, Structure and Properties of Block Copolymers, Elsevier, London, 1985.

6. D. J. Meier, Ed., Block Copolymers: Science and Technology, Harwood, New York, 1983.

7. A. Noshay and J. E. McGrath, Block Copolymers. Overview and Critical Survey, Academic Press, New York, 1977.



Administrator

Underline

Administrator

Underline

Administrator

Text Box

环硫化物

8. R. P. Quirk, D. J. Kinning, and L. J. Fetters, in Comprehensive Polymer Science, Vol. 7, Specialty Polymers and Polymer Processing, S. L. Agarwal, Ed., Pergamon, Oxford, 1989, p. 1.

9. G. Riess, G. Hurtrez, and P. Bahadur, in Encyclopedia of Polymer Science and Engineering, Vol. 2, Wiley, New York, 1985, p. 324.

10. O. W. Webster, Science, 251, 887 (1991).

11. L. J. Fetters and E. L. Thomas, in Material Science & Technology, Vol. 12, VCH Verlagsgesellschaft, Weinheim, Germany, 1993, p. 1.


13. M. Morton, Anionic Polymerization: Principles and Practice, Academic Press, New York, Chapt. 9, 1983, p. 179.

14. R. P. Quirk and J. Kim, Rubber Chem. Technol., 64, 450 (1991).



17. L. J. Fetters, J. Polym. Sci., Part C, 26, 1 (1969).

18. M. Morton, in Encyclopedia of Polymer Science and Technology, N. Bikales, Ed., Wiley, New York, Vol. 15, 1971, p. 508.

19. P. Dreyfuss, L. J. Fetters, and D. R. Hansen, Rubber Chem. Technol., 53, 728 (1980).

20. R. Jerome, R. Fayt, and T. Ouhadi, Prog. Polym. Sci., 10, 87 (1984).



23. J. C. Falk, M. A. Benedetto, J. Van Fleet, and L. Ciaglia, Macromol. Syn., 8, 61 (1982).

24. J. J. O'Malley and R. H. Marchessault, Macromolecular Synthesis, Coll. Vol. 1, J. A. Moore, Ed., Wiley, New York, 1977, p. 419.

25. D. Braun, H. Cherdron, and W. Kern, Practical Macromolecular Organic Chemistry, K. J. Ivin (translator), Harwood, New York, 1984.

26. L. J. Fetters and M. Morton, in Macromolecular Synthesis, Coll. Vol. 1, J. A. Moore, Ed., Wiley, New York, 1977, p. 463.

27. E. L. Slagowski, L. J. Fetters, and D. McIntyre, Macromolecules, 7, 394 (1974).

28. L. J. Fetters, in Encyclopedia of Polymer Science and Engineering, 2nd ed., Vol. 10, J. I. Kroschwitz, Ed., Wiley, New York, 1987, p. 19.


30. H. L. Hsieh and O. F. McKinney, J. Polym. Sci., Part B, Polym. Lett., 4, 843 (1966).

31. R. Ohlinger and F. Bandermann, Makromol. Chem., 181, 1935 (1980).



Administrator

Highlight

Administrator

Highlight

39. H. L. Hsieh, Rubber Chem. Technol., 49, 1305 (1976).

40. D. N. Schulz and A. F. Halasa, J. Polym. Sci., Polym. Chem. Ed., 15, 2401 (1977).

41. C. D. Han, D. M. Back, J. K. Kim, T. Hashimoto, and S. Okamoto, Macromolecules, 24, 408 (1991).

42. L. J. Fetters, in Block Copolymers: Science and Technology, D. J. Meier, Ed., Harwood, New York, 1983, p. 17.

43. R. P. Quirk and L.-F. Zhu, Br. Polym. J., 23, 47 (1990).

44. M. Morton, R. P. Kammereck, and L. J. Fetters, Macromolecules, 4, 11 (1971).

45. M. Morton, Y. Kesten, and L. J. Fetters, Appl. Polymer Symp., 26, 113 (1975).

46. L. J. Fetters, B. H. Meyer, and D. McIntyre, J. Appl. Polym. Sci., 16, 2079 (1972).

47. T. E. Long, A. D. Broske, D. J. Bradley, and J. E. McGrath, J. Polym. Sci., Polym. Chem., 27, 4001 (1989).


49. L. H. Tung and G. Y. Lo, in Advances in Elastomers and Rubber Elasticity, J. Lal and J. E. Mark, Eds., Plenum Press, New York, 1986, p. 129.

50. B. Lee, Ph.D. Dissertation, University of Akron, Akron, Ohio, 1991.


52. M. Szwarc, Carbanions, Living Polymers and Electron Transfer Processes, Inter-science, New York, 1968.

53. M. Szwarc and M. Van Beylen, Ionic Polymerization and Living Polymers, Chapman & Hall, New York, 1993.


55. L.-K. Huang, Ph.D. Dissertation, University of Akron, Akron, Ohio, 1979; quoted in ref. 13, p. 142.

56. R. E. Cunningham, J. Appl. Polym. Sci., 22, 2907 (1978).

57. T. Hasimoto, Y. Tsukahara, K. Tachi, and H. Kawai, Macromolecules, 16, 648 (1983).

58. R. E. Cunningham and M. R. Trieber, J. Appl. Polym. Sci., 12, 23 (1968).

59. W. H. Janes, in Block Copolymers, D. Allport and W. H. Janes, Eds., Halsted Press, New York, 1973.

60. I. Goodman, Ed., Developments in Block Copolymers, Applied Science, London, 1982, Vol. 1.

61. M. Morton and L. J. Fetters, Macromol. Revs., 2, 71 (1967).




Administrator

Underline

Administrator

Highlight

Administrator

Highlight

Administrator

Highlight


73. T. Hashimoto, N. Nakamura, M. Shibayama, A. Izumi, and H. Kawai, J. Macromol. Sci. Phys., B17, 389 (1980).

74. H. H. Winter, D. B. Scott, W. Gronski, S. Okamoto, and T. Hashimoto, Macromolecules, 26, 7236 (1993).

75. F. A. Morrison, H. H. Winter, W. Gronski, and J. D. Barnes, Macromolecules, 23, 4200 (1990).

76. J. E. McGrath, Ph.D. Dissertation, University of Akron, Akron, Ohio, 1967.

77. Y. Matsushita, Y. Mogi, H. Mukai, J. Watanabe, and I. Noda, Polymer, 35, 246 (1994).

78. F. S. Bates, H. E. Bair, and M. A. Hartney, Macromolecules, 17, 1987 (1984).

79. F. S. Bates, J. H. Rosedale, H. E. Bair, and T. P. Russell, Macromolecules, 22, 2557 (1989).

80. L. J. Fetters, E. M. Firer, and M. Dafauti, Macromolecules, 10, 1200 (1977).

81. R. P. Quirk, M. T. Sarkis, and D. J. Meier, in Advances in Elastomers and Rubber Elasticity, J. Lal and J. E. Mark, Eds., Plenum Press, New York, 1986, p. 143.

82. M. Morton, J. Polym. Sci., Polym. Symp., 60, 1 (1977).

83. H. L. Hsieh and I. W. Wang, in Recent Advances in Anionic Polymerization, T. E. Hogen-Esch, and J. Smid, Eds., Elsevier, New York, 1987, p. 411.

84. J. Lin and M. A. Fox, Macromolecules, 27, 902 (1994).

85. B. C. Anderson, D. G. Andrews, P. Arthur, Jr., H. W. Jacobson, L. R. Melby, A. J. Playtis, and W. H. Sharkey, Macromolecules, 14, 1599 (1981).

86. K. Hatada, T. Kitayama, K. Fumikawa, K. Ohta, and H. Yuki, in Anionic Polymerization. Kinetics, Mechanisms and Synthesis, J. E. McGrath, Ed., ACS Symp. Ser., No. 166, 1981, p. 327.

87. G. D. Andrews and L. R. Melby, in Monomers and Polymers, B. M. Culbertson and C. U. Pittman, Jr., Eds., Plenum, New York, 1984, p. 357.

88. C. Auschra and R. Stadler, Polym. Bull., 30, 257 (1993).


90. J. Chan. S. Fox, D. Kiserow, C. Ramireddy, P. Munk, and S. E. Webber, Macromolecules, 26, 7016 (1993).

91. J. P. Hautekeer, S. K. Varshney, R. Fayt, C. Jacobs, R. Jerome, and Ph. Teyssie, Macromolecules, 23, 3893 (1990).


93. S. K. Varshney, C. Jacobs, J.-P. Hautekeer, P. Bayard, R. Jerome, R. Fayt, and P. Teyssie, Macromolecules, 24, 4997 (1991).

94. D. J. Brunelle, Ed., Ring-Opening Polymerization, Hanser, New York, 1993.



99. X. F. Zhong, S. K. Varshney, and A. Eisenberg, Macromolecules, 25, 7160 (1992).

100. L. Leemans, R. Fayt, Ph. Teyssie, and H. Uytterhoeven, Polymer, 31, 106 (1990).

101. A. Qin, M. Tian, C. Ramireddy, S. E. Webber, and P. Munk, Macromolecules, 27, 120 (1994).

102. W. Li and B. Huang, J. Polym. Sci., Polym. Phys., 30, 727 (1992).

103. T. Rhein and R. C. Schulz, Makromol. Chem., 186, 2301 (1985).

104. N. Nugay, Z. Kucukyavuz, and S. Kucukyavuz, Polym. Int., 32, 93 (1993).

105. G. Hild, J. P. Lamps, and P. Rempp, Polymer, 34, 2875 (1993).

106. H. L. Hsieh and I. W. Wang, in Ring-Opening Polymerization, J. E. McGrath, Ed., ACS Symposium Series No. 286, 1985, p. 161.

107. H. Keul and H. Höcker, Makromol. Chem., 187, 2833 (1986).

108. R. P. Foss, H. W. Jacobson, H. N. Cripps, and W. H. Sharkey, Macromolecules, 9, 373 (1976).

109. H. Watanabe, T. Shimura, T. Kotaka, and M. Tirrell, Macromolecules, 26, 6338 (1993).

110. Y. Mogi, H. Kotsuji, Y. Kaneko, K. Mori, Y. Matsushita, and I. Noda, Macromolecules, 25, 5408 (1992).

111. M. Morton and S. L. Mikesell, J. Macromol. Sci. Chem., A7, 1391 (1973).

112. M. Morton, R. F. Kammereck, and L. J. Fetters, Br. Polym. J., 3, 120 (1971).



13 Star Polymers

I. Introduction

A branched polymer is comprised of molecules with more than one backbone chain (i.e., it is a nonlinear polymers) [1]. It is characterized by the presence of branch points (junction points): atoms or a small group from which more than two long chains emanate. The branch chain is attached to the main chain with one of its chain ends at a branch point. A branched polymer is also characterized by the presence of more than two end groups. A graft polymer can be considered as a type of branched polymer. In contrast to a branched polymer, a graft polymer is a polymer comprising molecules with one or more species of block connected to the main chain as side chains, these side chains having constitutional or configurational features that differ from those of the main chain (see Chap. 14) [2]. In a branched polymer the side chain (branch chain) has the same constitutional and configuration features as the main backbone chain. A star-branched polymer consists of several linear chains linked together at one end of each chain by a single branch or junction point [1–4].

The formation of branched polymers is a common feature in many chain growth polymerizations and also in step-growth polymerizations in which the average functionality of the monomers is greater than two. A classic example is the high-temperature, high-pressure free-radical chain polymerization of ethylene, which forms low-density polyethylene with a variety of both short- and



long-chain branched structures [5]. These branches contribute to the dramatic difference in properties of low-density polyethylene compared to linear, high-density polyethylene [6]. Thus, branching affects the crystallinity, physical properties, and both the solution and melt viscosities of polymers [1,3–8]. However, it is difficult to understand and predict the relationships between branching and properties based on the behavior of this type of polymer because the branching reaction generally occurs in a random fashion. As a consequence, the number and types of branches per macromolecule are difficult to define except on an average basis.

Fundamental understanding regarding the effect of chain branching on polymer properties requires the availability of branched polymers with well-defined structures and low degrees of compositional heterogeneity [9,10]. This chapter will consider the methods developed to synthesize regular star-branched polymers and heteroarm star-branched polymers. Regular star-branched polymers have a single branch point and all arms exhibit low degrees of compositional heterogeneity with respect to molecular weight and molecular weight distribution. Heteroarm star-branched polymers [11,12], also described as mikto-arm star polymers [13], also have a single branch point, but the arms differ in either molecular weight or composition. When the arms differ in composition, heteroarm star-branched polymers can be considered as a special type of graft copolymer, as defined above.

II. Synthesis

The methodology of living anionic polymerization is ideally suited for the preparation of regular star-branched polymers and copolymers with well-defined structures. The absence of termination and chain transfer reactions allows the preparation of chain segments and polymers with predictable molecular weights and narrow molecular weight distributions [14–18]. After anionic polymerization of a given monomer is completed, the resulting living polymer with a reactive carbanionic chain-end can be reacted with a variety of linking reagents to generate the corresponding star-branched polymer with uniform arm length; these polymers are often referred to as “radial” polymers [19]. These techniques have been used to prepare model polymers to investigate the quantitative effects of chain branching on polymer properties and to test theoretical predictions [1,3,4,9]. After analogous sequential polymerization of two or more monomers, linking reactions with the resulting living diblock, or multiblock copolymers will generate star-branched block copolymers with uniform block arm length [20]. It is noteworthy that, in principle, each arm is of uniform block copolymer composition with precise block molecular weights and narrow molecular weight distributions (i.e., with low degrees of compositional heterogeneity), as discussed in Chapter 12. Table 13.1 contains a list of examples of the various types of linking agents that



YFP

Highlight

YFP

Highlight

Table 13.1 Linking Agents for Anionic Synthesis of Star-Branched Polymers

Linking Agent Theoretical Functionality

Dimethyl phthalate 4

Phosphorous trichloride 3

Methyltrichlorosilane 3

Silicon tetrachloride 4

Hexachlorodisilane 6

p- and m-Divinylbenzene variable (3–56)

Tetra(methyldichlorosilylethane)silane 8

Si(CH2CH2SiCl3)412

(Cl3SiCH2CH2)3SiCH2CH2Si(CH2CH2SiCl3)318

CH3Cl2SiCH2CH2SiCl2CH3 4

Cl3SiCH2CH2SiCl3 6

1,2,4-tris(Chloromethyl)benzene 3

Hexachlorocyclotriphosphazene 6

1,2,4,5-tetra(Chloromethyl)benzene 4

Hexa[p-(chloromethyl)phenyl]benzene 6

Tin tetrachloride 4

1,1,4,4-Tetraphenyl-1,4-bis-(diallyloxytriazine)butane 4

1-(Methyldichlorosilyl)-2-(trichlorosilyl)ethane 5

1,3-bis(1-Phenylethenyl)benzene 2,4a

Tetrakis[(phenyl-1-vinyl)-4-phenyl]plumbane 4,8a

aThe first number represents the linking functionality for preformed living carbanionic polymer chains. The second number represents the total number of arms formed by the living linking reaction followed by addition of monomer to form new growing arms.

Source: Refs. 19, 20

have been used to prepare star-branched polymers, as well as their theoretical functionality.

In the following sections, the general methods for synthesis of regular star-branched polymers and heteroarm star-branched polymers will be described. Each method has its own advantages and limitations, and these will be presented.

A. Divinylbenzene Linking Reactions, Copolymerization, and Multifunctional Initiators

Divinylbenzene Linking Reactions



YFP

Stamp

tions: (a) crossover to DVB; (b) block polymerization of DVB; and (c) linking reactions of carbanionic chain ends with pendant vinyl groups in the DVB block [poly(4-vinylstyrene)] [1,3,4,21]. These reactions are illustrated in Scheme 13.1.

Scheme 13.1

The uniformity of the lengths of the DVB blocks depends on the relative rate of the crossover reaction (a) compared to the block polymerization of DVB (b) and the linking reactions (c) (see Chaps. 4, 5). This block copolymerization-linking process has been described as formation of a DVB microgel nodule that serves as the branch point for the star-shaped polymer [4]. In principle, j molecules of divinylbenzene could link together (j + 1) polymer chains [4]. Although the number of arms in the star depends on the ratio of DVB to polymeric organolithium compound, the degree of linking obtained for this reaction is a complex



function of reaction variables [3,4,21–23]. It should be noted that these linking reactions are effected with technical grades of DVB that have been variously reported to consist of (a) 33 % DVB (11% p-divinylbenzene, 22% m-divinyl-benzene) and 66% o-, m- and p-ethylvinylbenzene (EVB) [4,24]; (b) 78% DVB (meta/para = 2.6), 22% ethylvinylbenzene isomers [22]; (c) 56% DVB, 44% EVB isomers [23]; (d) 18 mole % p-divinylbenzene, 39 mole % m-divinylbenzene, 10 mole % p-ethylvinylbenzene, and 33 mole % m-ethylvinylbenzene [25]. These differences suggest that it is prudent to analyze the composition of DVB samples since the degree of linking depends on the molar amount of DVB added relative to active chain end concentration. The purity of DVB is critical also, because impurities can terminate the active chain ends and this will also alter the effective molar ratio of DVB to active chain end concentration.

These systems incorporate the complexities associated with (a) organolithium chain-end association effects on kinetics for dienes vs. styrenes (see Chap. 7); and (b) the relative rates of crossover to DVB vs. DVB homopolymerization for poly(styryl)lithium chain ends vs. poly(dienyl)lithium chain ends. Therefore, it is instructive to examine separately the DVB linking reactions for poly(styryl)lithium and poly(dienyl)lithiums.

Poly(styryl)lithium The linking reactions of poly(styryl)lithium with DVB can be considered as a block copolymerization of DVB followed by linking reactions of carbanionic chain ends with pendant vinyl groups in the DVB block [poly(4-vinylstyrene)]. Worsfold [26] has measured the rate constant for crossover of poly(styryl)lithium with para-divinylbenzene (step a in Scheme 13.1); this rate constant is 10.6 times faster than homopolymerization of styrene and is the same order of magnitude as the rate constant for homopolymerization of para-divinyl-benzene (step b in Scheme 13.1). Furthermore, both of these rate constants are approximately 10 times faster than the rate constant for reaction of styryllithium chain ends with the pendant double bonds in poly(4-vinylstyrene) (step c in Scheme 13.1). Thus, it would be expected that the DVB block formed by cross-over from poly(styryl)lithium would be relatively uniform and that the linking reaction would generally occur after the formation of the DVB block. Representative data for the formation of polystyrene stars by DVB linking reactions with poly(styryl)lithium are shown in Table 13.2. It can be seen that, in general, the number of arms, as indicated by the arm functionality of the star, F, increases with increasing ratios of [DVB]/[PSLi], R. It is also important to note that the efficiency of the DVB linking reaction for poly(styryl)lithium is reported to be quite high except at very low ratios of [DVB]/[PSLi]; for example, when [DVB]/[PSLi] = 7 (F = 16.8; Sample 11 in Table 13.2), it was reported that only 4 wt % unlinked polymer remained after 24 hours at 45°C and that this fraction corresponded to dimer (see Fig. 13.1) [25]. However, when R = 1.8 (Sample 1, Table 13.2), the linking efficiency is low (75%); although the arm functionality, F, was 5, 29% dimer formation was reported (see Fig. 13.2) [27]. It is noteworthy that narrow



Table 13.2 Linking Reactions of Poly(styryl)lithiums with Divinylbenzenes

Sample

Mn (arm)

(g/mol × 10-3) [DVB]/[PSLi]

Fa

Reference

1 12 1.8 5 27

2 6.6 2.7 10 27

3 3.5 3.5 15 28

4 21 4.0 3 27

5 52 4.1 4 27

6 62.3 5.5 12.9 29

7 6.5 6 12 27

8 10.8 6 15 28

9 11 6 11 28

10 12 6 11 28

11 19 7 16.8 25

12 50.2 7 14.7 29

13 12.7 8 17 28

14 11.7 8 19.1 29

15 34.3 8 23.1 29

16 82.1 8 15.1 29

17 113 8 14.2 29

18 179 8 14.9 29

19 11.4 9 11 27

20 13.7 9 12 28

21 56.2 10 22.1 29

22 32 10 4 27

23 27 10.3 5 27

24 26 12 7 27

25 52.5 13 29.7 29

26 60.1 20 32.1 29

27 56.8 30 38.7 29

aDegree of branching. Average number of arms per star-branched macro-molecule

molecular weight distributions (Mw/Mn < 1.1; size exclusion chromatography [SEC]) have been



reported [29]. Thus, the linking reactions of poly(styryl)lithiums with DVB are relatively well-behaved with respect to the ability to obtain branched polymers with high linking efficiency, a wide variation in branching functionality (up to 40 arms per star), and with narrow molecular weight distributions.



Figure 13.1 Size exclusion chromatographs of unfractionated

divinylbenzene-linked polystyrene stars; PS/DVB-5 corresponds to sample 11,

Table 13.2. (From Ref. 25; reprinted by permission of the American Chemical Society, Washington, D.C.)

Figure 13.2 Size exclusion chromatogram of unfractionated divinylbenzene-linked polystyrene star polymer;

sample 1, Table 13.2 (From Ref. 27; reprinted by permission of Hüthig & Wepf Verlag,

Zug, Switzerland.)



Poly(dienyl)lithium The linking reactions of poly(dienyl)lithium compounds with DVB are more complicated and less efficient than the corresponding linking reactions with poly(styryl)lithium. From copolymerization studies (see Chap. 10), it is known that the rate of crossover of poly(dienyl)lithium chain ends to styrene monomer is slow compared to diene (D) or styrene (S) homopolymerization, with the order of reaction rate constants kSD > kSS > kDD > kDS. Because the rate of homopolymerization of DVB is 10 times faster than styrene homo-polymerization, it would be expected that kDVD-DVB >> kDVD-DVB, that is, the rate of homopolymerization of DVB will be much faster than the rate of crossover of dienyllithium chain ends to DVB. Thus, for a given chain end concentration and a given ratio of [DVB]/[PLi], the average block length of DVB should be higher for dienyllithium chain ends than styryllithium chain ends. A longer block length for a few chains that crossover to DVB would be expected ultimately to lead to a higher degree of branching and lower linking efficiencies for dienes vs. styrenes; the arm functionality, F, as a function of R for dienes is shown in Table 13.3.

The kinetics of reactions of poly(dienyl)lithiums with isomers of divinylbenzene has been investigated by Young and Fetters [21]. It was reported that the rate of reaction with either DVB isomer by poly(butadienyl)lithium is much slower (by approximately a factor of 10) than the corresponding reactions of poly(isoprenyl)lithium. As expected based on the relative stabilities of the corresponding benzyl carbanions (1 > 4), the rate of addition to p-divinylbenzene was

approximately two times the corresponding rate for addition to m-divinylbenzene. However, more efficient linking was observed with the meta isomer [21].

The consequences of the relatively slow rate of crossover of dienylithiums to DVB are especially evident at low ratios of [DVB]/[PLi] = R. When R is approximately 3, it has been reported that the linking efficiency is only 70–80% for both poly(isoprenyl)lithium and poly(butadienyl)lithium (see Samples 2,6,24–26 in Table 13.3); however, the arm functionalities for these systems are quite high (F = 8–23) [21–23]. Furthermore, the lower molecular weight fractions are composed primarily of dimer as illustrated in Figure 13.3 (Sample 25, Table 13.3) [23]. These observations are consistent with a relatively slow rate of crossover to DVB compared to DVB homopolymerization, which results in a few chains with longer poly(4-vinylstyrene) (4VS) blocks. Linking of active chains with these



Table 13.3 Linking Reactions of Poly(dienyl)lithiums with Divinylbenzenes (DVB)

Sample

Mn (arm)

(g/mol × 10-

3)

[DVB]/[PDLi]

Fa

Reference

Poly(isoprenyl)lithium

1 70 3 5.1 30

2 27.6 3 7.7b 23

3 51 3.2 6 30

4 39.8 3.2 9.8 23

5 66 3.3 6.8 30

6 140 3.6 10.8c 22

7 78 4.4 6.2 30

8 54 4.5 6.8 30

9 970 5 12 22

10 41.2 5.3 12.7 23

11 666 5.5 9.6 22

12 127 6 9 30

13 70 6.1 9 30

14 51.0 6.1 13.1 23

15 55.7 6.7 9.8 23

16 71 8 15 30

17 960 8.1 16 30

18 140 8.5 17.9 30

19 140 11.4 22.5 22

20 37.7 11.9 12.1 23

21 72 12.6 6.9 30

22 970 20 32.7 22

23 140 41.4 55.8 22

Poly(butadienyl)lithium

24 34.8 2.8 13.4d 23

25 32.0 3 18e 23

26 35.2 3.2 23.4f 23



27 30 4 6 30

28 21 4 5.7 30

29 64.0 4.9 15 23

30 14 5 9.3 30

31 35.0 6.8 23.8 23

32 20 7 15 30

33 30 11 29 30

34 35.0 12.1 16g 23

aDegree of branching. Average number of arms per star-branched macro-molecule.

b23% linear.

c78% linking efficiency.

d26% dimer.

e36% dimer.

f22% dimer.

g1% dimer.



Figure 13.3 Size exclusion chromatogram of unfractionated

divinylbenzene-linked polybutadiene star polymer; sample 25, Table 13.3.

(From Ref. 23; reprinted by permission of the American Chemical Society, Washington, D.C.)

long 4VS blocks leads to high arm functionality but low linking efficiency regardless of reaction time and temperature. However, with higher values of R (R > 4–5), elevated temperatures (45–60°C), and relatively long reaction times (24 h), linking efficiencies > 90% have been reported as illustrated in Figure 13.4 (Sample 34, Table 13.3) [21–23]. As shown in Table 13.3, the arm functionalities increase with increasing values of R, and arm functionalities as high as 56 have been obtained for poly(isoprenyl)lithium when R = 41 [22]. It should be noted that even after the linking reactions are complete, it was reported that the residual vinyl group concentrations were 10–15% for m-divinylbenzene and 20–25% for the para isomer [21].

Because of the random nature of the linking reactions for DVB star-branched polymers, it would be expected that relatively broad molecular weight distributions would be obtained. In contrast, narrow molecular weight distributions (MwIMn < 1.1) have been reported for polystyrene stars obtained with R values varying from 5.5 to 30 and the corresponding arm functionalities (F) varying from 13 to 39 [29]. Even for DVB linking of poly(dienyl)lithiums, narrow molecular weight distributions (MwIMn < 1.1) are obtained for R values in the range of 5–6.5 with arm functionalities of 9–13 [22]. However, when higher values of R are used, the molecular weight distributions obtained for polyisoprene stars increase to 1.2–1.4 [22,23].

In conclusion, the end-linking reactions of polymeric organolithium com-



Figure 13.4 Size exclusion chromatogram of

unfractionated divinylbenzene-linked polybutadiene star polymer; sample 34, Table 13.3. (From Ref. 23;

reprinted by permission of the American Chemical Society, Washington, D.C.)

pounds with divinylbenzenes are useful for preparation of star-branched polymers. Efficient linking is observed beyond a minimum value of R ([DVB]/[PLi] > 4) and the arm functionalities can be varied over a broad range (F = 6–56). However, this method lacks the precision associated with other syntheses of well-defined branched polymers using living anionic polymerization.

Copolymerizations with Divinylbenzenes

The copolymerization of divinylbenzenes with styrene or diene monomers can form in-chain vinylstyrene units that can react with polymeric organolithium chain ends to form long-chain branched structures, as illustrated in Scheme 13.2. Of course, if the amount of difunctional DVB monomer is too high, a three-dimensional gel will be formed instead of a branched polymer [24,26,31,32].

It has been determined that the copolymerization of styrene with the commercial mixture of DVB will form macroscopic gels when the ratio [DVB]/[initiator] is larger than 15.1 [33]. The copolymerization of styrene (M1) with



Scheme 13.2

p-divinylbenzene (M2) has been investigated in benzene in the presence of 2% tetrahydrofuran (THF)[26]. The calculated values of the monomer reactivity ratios were r1 = 0.09 and r2 = 10. The product of r1r2 = 0.94 indicates that the copolymerization exhibits close to ideal copolymerization behavior [34]; thus, the relative rates of incorporation of the two monomers into the copolymer would be expected to be independent of the identity of the chain end unit. The higher reactivity of DVB will be balanced by the very low molar ratio of DVB relative to styrene ([DVB]/[styrene] 0.01–0.03) for polymerizations yielding branched and not gelled products [24]. For example, using the instantaneous copolymerization equation (see Eq. 10.9) with [M1] = 100, [M]2{] = 2, it can be calculated that the initially formed polymer will contain 82 mole % styrene and 18 mole % DVB. Therefore, even with a relatively small amount of DVB, DVB will be preferentially incorporated into the initially formed chain segments and the amount of DVB in the monomer feed will rapidly decrease. It should be noted that the data of Worsfold [26] indicate that the rate constant for reaction of poly(styryl)lithium with the in-chain vinylstyrene units resulting from DVB copolymerization is comparable to the rate constant for styrene homopolymerization.

The copolymerization of dienes with DVB would be expected to exhibit behavior analogous to the copolymerizations of dienes with styrene (i.e., kDVB-D >> kD-DVB), as discussed in Chapter 10. Therefore, the copolymerization would be expected to exhibit strong compositional heterogeneity along the chain with preferential incorporation of diene initially, followed by “blocking” of DVB when the concentration of diene decreases significantly. Thus, a tapered-type of block copolymer structure would be expected; therefore, branching would tend to occur late in the copolymerization after significant incorporation of DVB has occurred.

Multifunctional Initiators from DVB

The ability to effect the concurrent polymerization and branching reactions of polymeric organolithium compounds with DVB and form a soluble, nongelled product was extended by Burchard and co-workers [35] to form a multifunctional initiator. DVB was first polymerized using butyllithium in benzene to form soluble microgels of high molecular weight. These microgels with their attendant anionic groups were used as multifunctional initiators to polymerize monomers such as



styrene. For example, when the ratio of [DVB]/[BuLi] was 2, the microgel exhibited Mn = 1.9 × 103 g/mol (Mw/Mn = 16.8) and the resulting polystyrene star had a calculated star arm functionality of 6 [35]. When the ratio of [DVB]/[BuLi] was 3, the microgel exhibited Mn = 4.5 × 103 g/mol (Mw/Mn = 10.7) and the resulting polystyrene star had a calculated star arm functionality of 10.

This method has been extended by Rempp and co-workers [36,37] as a general “core-first” method to prepare star-branched polymers. A schematic representation of this star synthesis is shown in Scheme 13.3. The “plurifunc

Scheme 13.3

tional” metalorganic initiator (5) was prepared by potassium naphthalene-initiated polymerization of DVB in THF at -40°C with [DVB]/[K+] ratios of 0.5–3. Microgel formation was reported outside of this stoichiometric range or when m-DVB was used instead of either p-DVB or the commercial mixture. Within the prescribed stoichiometric ratios, star polymers (6) with arm functionalities varying from 8 to 42 were reported. In addition, block copolymers were formed by first polymerizing styrene and then adding ethylene oxide [36]. As expected for this type of process, the polydispersities were described as being quite broad and attributed primarily to a random distribution of core sizes and functionalities. Using lithium naphthalene as initiator in the presence of lithium chloride, the core-first method was used to prepare star-branched polystyrene-b-



poly(t-butyl acrylate) polymers. After the formation of the poly(styryl)lithium arms was initiated at -40°C, 1,1-diphenylethylene was added to attenuate the reactivity of the chains for efficient crossover to t-butyl acrylate at -55°C (see Chap. 23). The dramatic effect of added lithium chloride in promoting the controlled polymerization of acrylates and methacrylates was first reported by Teyssie and co-workers (see Chap. 23) [38,39]. Arm functionalities for the star-block copolymers were reported to range from 22 to 1300 and the size distributions (SEC) were broad and multimodal.

Heteroarm Star-Branched Polymers by Living Linking Reactions with DVB

Eschwey and Burchard [40] recognized that the products of the reaction of DVB with polymeric organolithium compounds are living polymers that could function as polyfunctional, macromolecular initiators as indicated in Scheme 13.4 for

Scheme 13.4

linking of five polymeric organolithium chains (Pa). This type of process has been described as a living linking reaction [41]. Thus, after DVB linking of short poly(styryl)lithium chains to form the living, linked star polymer, 7, additional styrene or isoprene monomer (Mb) was added to double the number of arms, in principle, and form a heteroarm, star-branched polymer, 8. Preliminary viscosity and light-scattering data were consistent with heteroarm, star polymer formation.



It has been reported that initiation is slow and incomplete with both styrene and diene monomers in hydrocarbon solvents using this type of polyfunctional anionic initiator [4], presumably because of steric effects, but association phenomena could also inhibit efficient initiation [42].

This “core-first” heteroarm star synthesis method based on DVB linking of poly(styryl)lithium has been applied to the synthesis of star-polystyrene-star-poly(n-butyl methacrylate) [27] and star-polystyrene-star-poly(t-butyl methacrylate) [28,37]. Although the final molecular weights were consistent with the expected star structures, no demonstration of the uniformity of the branches was provided. Furthermore, the authors assume that the polymers possess equal number of branches of either type. Independent work with trifunctional anionic macromolecular initiators suggests that it is difficult to obtain uniform growth of poly(alkyl methacrylate) arms under analogous conditions [43].

Thus, the DVB linking methodology for the synthesis of both regular star-branched polymers and heteroarm, star-branched polymers presents an interesting technology. However, it lacks the precision and control necessary for the synthesis of polymers with well-defined structures and low degrees of compositional heterogeneity.

B. Coupling Reactions with Silyl Halides

Regular Star-Branched Polymers

The most general methods for the preparation of regular star polymers have been developed based on linking reactions of polymeric organolithium compounds with multifunctional electrophilic species such as silicon tetrachloride as shown in Equation 13.1. Analogous linking reactions of polymeric organolithium com-

(13.1)

pounds with multifunctional halogenated hydrocarbons are complicated by side reactions such as lithium-halogen exchange and elimination reactions [44]. Complexities can also result from the fact that the mechanism for these reactions may involve radical intermediates under certain conditions [45,46]. Lithium-halogen exchange and subsequent coupling of two species derived from the multifunctional linking reagent have been invoked to explain the formation of star-branched polymers with functionalities higher than the functionality of the organohalogen linking agents (e.g., as high as two or three times the functionality of the halogenated hydrocarbon) [1,3,4,47]. In contrast to these complexities for organolithium reactions with polyfunctional halogenated hydrocarbon-type linking agents, the reactions with chlorosilane compounds are very efficient and uncomplicated by analogous side reactions such as lithium-halogen exchange [48,49]. However, the extents and efficiencies of these linking reactions are dependent upon the steric requirements of the carbanionic chain end [4].



SecLover

Highlight

In general, for a given multifunctional silicon halide, the efficiency of the linking reaction decreases in the order poly(butadienyl)lithium > poly(isoprenyl)lithium > poly(styryl)lithium. For example, the reaction of poly(styryl)lithium (Mn = 60.6 × 103 g/mol; Mw/Mn = 1.06) with a less than stoichiometric amount (exact amount unspecified) of silicon tetrachloride in benzene at 50°C for 48 h produced a polymer product with Mw = 1.93 × 105 g/mol [50]. After fractionation of this product, a four-armed star polymer (Mw = 2.57 × 105 g/mol; Mw/Mn = 1.09) and a three-armed star polymer (Mw

= 1.70 × 105 g/mol; Mw/Mn 1.0) were isolated in weight fraction amounts corresponding to 0.252 and 0.349, respectively. From the molecular weight of the first-formed polymer mixture, it can be estimated that the product contained 26% of the four-armed star and 74% of the three-armed star products, if it is assumed that these were the only components in the mixture. Thus, even though a mixture of products was produced and the linking efficiency to form the four-armed star polymer was low, a narrow molecular weight, four-armed, star-branched product could be obtained after fractionation. Analogous linking reactions in a mixture of THF and benzene produced product mixtures with reported average linking functionalities of 2.6–3.74 [51].

In contrast to the results of inefficient linking for poly(styryl)lithiums, the linking reactions of poly(butadienyl)lithiums with methyltrichlorosilane and silicon tetrachloride in cyclohexane at 50°C for 3 h were reported to proceed in high efficiency to form the corresponding three-arm and four-arm stars, respectively, as established by Mw (light scattering), intrinsic viscosity-molecular weight correlations and fractionation data [52]. The main impurity was the corresponding linear polybutadiene, presumably formed by termination from impurities.

The linking efficiency of poly(isoprenyl)lithium with silicon tetrachloride is not high, analogous to poly(styryl)lithium. The stoichiometric reaction of poly(isoprenyl)lithium with silicon tetrachloride is reported to form predominantly the three-armed star product [53]. However, high linking efficiency to form the three-armed star is obtained with methyltrichlorosilane [53].

Two general approaches have been used to increase the efficiency of linking reactions of polymeric organolithium compounds with multifunctional silyl halides. The first procedure is to add a few units of butadiene to either the poly(styryl)lithium or poly(isoprenyl)lithium chain ends effectively to convert them to the corresponding less sterically hindered poly(butadienyl)lithium chain ends. For example, after crossover to butadienyllithium chain ends, the efficiency of forming a four-armed star polyisoprene with silicon tetrachloride was essentially quantitative in cyclohexane [53].

The second method is to utilize a polychlorosilane compound in which the silyl halide units are more separated to reduce the steric repulsions in the linked product. Thus, a four-armed, star-branched polystyrene can be prepared by using 1,2-bis(methyldichlorosilyl)ethane [54]. The direct reaction of poly(styryl)lithium with 1,2-bis(methyldichlorosilyl)ethane in benzene forms the dimer adduct within



the mixing time of the reagents; however, the addition of the third arm occurs more slowly over a period of 8–24 h. The addition of the fourth arm requires heating at 48°C for period of 4 days to 2 weeks depending on molecular weight. A four-armed star polymer with Mn = 5 × 105 g/mole can be obtained in 85% yield with 15% of the trifunctional polymer using these procedures. When poly(styryl)lithium is treated with 3 equivalents of isoprene, the three-armed star is formed in less than 3 h at 30°C and the four-armed star is completed in 1–2 days with no contamination by three-armed star (Mn up to 1.4 × 106 g/mole). In actual practice, an excess ([PLi]/[Si-Cl] = 1.2–1.8) [54] of the polymeric organolithium arm is employed to provide reasonable reaction times; therefore, most of these star syntheses require at least one fractionation step to obtain the desired narrow molecular weight distribution star-branched polymer (Mw/Mn < 1.1).

The formation of narrow molecular weight distribution, well-defined six-armed star polystyrenes has been effected by linking reactions of excess (20%) isoprene-tipped poly(styryl)lithium with 1,2-bis(trichlorosilyl)ethane in benzene at 30°C [55]. As expected from these results, four-armed and six-armed star-branched polyisoprenes that are homogeneous in degree of branching and molecular weight can also be formed by reaction of poly(isoprenyl)lithium with 1,2-bis-(methyldichlorosilyl)ethane and 1,2-bis(trichlorosilyl)ethane, respectively [56]. Using a small excess of polymeric organolithium, no evidence for intermediate degrees of linking were observed.

The efficiency of the linking reactions of polychlorosilanes with poly-(dienyl)lithium compounds has been documented by synthesis of narrow molecular weight distribution 8-, 12- and 18-armed polyisoprenes [57,58], 18-armed polybutadiene [59], and 12- and 18-armed polystyrene [60], by linking reactions with octachlorosilane, dodecachlorosilane, and decaoctachlorosilane, respectively. It should be noted that linking reactions of poly(isoprenyl)lithium were sometimes effected by end-capping the chain ends with several units of butadiene to increase the rate of linking, although the degree of linking was not affected [58]. The SEC curve for a representative unfractionated 18-arm polyisoprene star product is shown in Figure 13.5. For the synthesis of an 18-armed star polybutadiene, triethylamine was added to increase the linking rate [59]. Analogous star-block copolymers can be prepared using these same linking agents and procedures [61].

The linking reactions of poly(butadienyl)lithium (Mn = 5.3–89.6 × 103 g/mol) with carbosilane dendrimers [62] with 32, 64, and 128 Si-Cl bonds is reported to proceed smoothly at room temperature for periods up to 3 weeks [63,64]. The arm functionality as determined by light-scattering measurements varied from 31.5 to 35.7 for 32 Si-Cl [63], 56–62 for the 64 Si-Cl [64], and 114–127 for the 128 Si-Cl [64].

A method with the potential of generating stars with multiple hundreds of arms is based on the hydrosilation reaction of polybutadiene with high 1,2-



microstructure as shown in Scheme 13.5 [65]. Using the hydrosilation product from poly(1,2-butadiene) with Mn = 9.04 × 103 g/mole, star-branched polybuta-

Scheme 13.5

dienes with 263–278 arms were prepared by reaction with poly(butadienyl)lithiums (Mn = 5–17 × 103 g/mol) in benzene for 1 week followed by 1–2 weeks in the presence of triethylamine.

Heteroarm Star-Branched Polymers

Pennisi and Fetters [66,67] have prepared hetero, three-armed star polystyrenes and polybutadienes using the reactions shown in Equations 13.2 and 13.3. This

(13.2)

(13.3)

procedure is based on the decreased reactivity of poly(styryl)lithium compared to poly(dienyl)lithiums with respect to linking reactions with polyhalosilanes, which has been ascribed primarily to chain-end steric effects. The first step utilized a 10-fold excess of methyltrichlorosilane ([Si-Cl]/[Li] = 30) to prevent coupling and linking reactions; this excess was removed by freeze-drying and heat treatment (50°C) of the methyldichlorosilane end-functionalized polystyrene. In the second step, a 20% excess of polymeric organolithium compound was reacted with the dichlorosilane-functionalized polystyrene; these reactions required days for com-



Figure 13.5 Size exclusion chromatogram of unfractionated decaoctachlorosilane-linked polyisoprene star

polymer. (From Ref. 58; reprinted by permission of the American Chemical Society,

Washington, D.C.)

pletion even in the presence of added Lewis base promoter (triethylamine). Fractionation was required to remove the excess nonlinked polymer chains. It is important to note that each arm component was individually characterized and that the final hetereoarm star polymers exhibited quite narrow molecular weight distributions (Mw/Mn < 1.1) as illustrated in Figure 13.6. The arm molecular weights varied from 0.2 to 14 × 104 g/mol. A similar approach was used to synthesize an A2B star, Ps(PI)2, which is a model graft copolymer, poly(isoprene-g-polystryene)[68]. The second step of linking a slight excess of poly(isoprenyl)lithium (Mn = 18.1 × 103 g/mol) with the dichlorosilane-functionalized polystyrene (Mn = 61.7 × 103 g/mol) was conducted at 40°C for 72 h. After fractionation, the model graft copolymer exhibited a narrow molecular weight distribution (Mw/Mn = 1.04).

This methodology has been extended recently to the preparation of A2B2, ABCD, and ABC heteroarm star polymers[13,69,70]. The ABC star terpolymer was prepared by (a) reaction of poly(isoprenyl)lithium with an excess of methyltrichlorosilane ([Si-Cl]/[Li] = 60) and removal of the excess methyltrichlorosilane; (b) stoichiometric addition (titration) of poly(styryl)lithium; and (c) addition of a small excess of poly(butadienyl)lithium[13]. The resulting star polymer was fractionated to remove the last unlinked polymer fraction; the final product exhibited a narrow molecular weight distribution (Mw/Mn = 1.03).

The synthesis of A2B2 heteroarm stars was effected by first reacting two



Figure 13.6 SEC chromatographs of hetero, three-armed polystyrene star

polymer (arm Mn = 2000; 2000; 5000 g/mol) (From Ref. 66; reprinted by permission

of the American Chemical Society, Washington, D.C.)

equivalents of poly(styryl)lithium with SiCl4 for 12 h at 40°C[70]. The absence of three-armed star was attributed to the “rapidly escalating steric difficulty” of multiple polystyrene linking beyond 2; the relatively rapid addition of the first two polystyrene arms was noted in previous studies[54,55]. The reaction with a large excess of poly(isoprenyl)lithium was effected in the presence of triethylamine for one week at 40°C. Two fractionations produced pure star polymer.

The extension of this methodology based on differential reactivity to the synthesis of ABCD four-armed, star-branched polymers[69] is problematic. The fourth linking reaction with poly(butadienyl)lithium was reported to require 2 months for completion. The product [polystyrene, poly(4-methylstyrene), polyisoprene and polybutadiene arms] was multimodal as shown by the SEC curve in Figure 13.7. After fractionation, a narrow molecular weight distribution product was obtained (Mw/Mn = 1.06) and characterized by SEC and light scattering. However, all of the precursor arms had narrow molecular weight distributions and approximately the same molecular weight; therefore, the fact that the overall composition was as predicted does not convincingly demonstrate that the molecules are compositionally homogeneous. Unfortunately, polymers obtained at intermediate stages were not separately characterized. It remains to be convincingly demonstrated that this is a reliable method for synthesis of a wide variety of well-defined, compositionally homogeneous, ABCD-type, heteroarm, star-branched polymers.



Figure 13.7 SEC chromatograms of the ABCD hetero, four-armed star

branched polymer [polystyrene, poly(4-methylstyrene), polyisoprene and polybutadiene

arms]. (From Ref. 69; reprinted by permission of the American Chemical Society, Washington, D.C.)

A series of H-shaped star-branched polymers have been prepared by linking reactions of an α,ω-disodiumpolystyrene with the product of the reaction of two moles of poly(styryl)lithium with methyltrichlorosilane [(PS)2Si(CH3)C1] [71]. Ultracentrifugation analysis of the product prior to fractionation exhibited four peaks corresponding to all possible intermediate star products and precursors.

C. 1,1-Diphenylethylene-Based Living Linking Reactions

A living linking agent is a species that can react with polymers that have carbanionic chain-ends to generate a linked polymer product that retains the active center stoichiometry to initiate polymerization of additional monomer [41]. The general reaction scheme is shown below in Scheme 13.6, where C is a living linking agent with a linking functionality of n, P1Li is a well-defined, living carbanionic polymer chain and M2 is a second monomer which can be anionically



Scheme 13.6

polymerized by the active carbanionic centers in the intermediate living linked product, 10. It is obvious from this reaction scheme that the use of living linking agents provides a method for preparing heteroarm star-branched polymers, 11. Several criteria must be satisfied for a species to be useful as a living linking agent to prepare well-defined, compositionally homogeneous, heteroarm, star-branched polymers [41]:

1. the living linking agent must react quantitatively with living carbanionic chain ends without oligomerization;

2. the coupled product must retain the active centers stoichiometrically; and

3. the living coupled product must be capable of reinitiating polymer chain growth rapidly (relative to propagation) and stoichiometrically.

Living linking agents generally have two or more reactive vinyl substituents that can undergo addition reactions with polymeric carbanions, but that do not homopolymerize anionically due to steric hindrance. Divinylbenzenes and m-di-isopropenylbenzene could be regarded as potential living linking agents; however, the homopolymerization and oligomerization, respectively, of these divinyl compounds limit their effectiveness for the synthesis of well-defined star-branched polymers, as discussed for DVB in the first part of this chapter.

Living linking methods have been developed for the synthesis of hetero four-armed star-branched homopolymers and copolymers [11,12,41] and for symmetrical three-armed star polymers [72] utilizing 1,1-diphenylethylene-derived living linking agents. As an extension of these investigations, a hetero three-armed star branched polymer was prepared using a 1,1-diphenylethylene-functionalized macromonomer [73].

Four-Armed Star Polymers

The methodology of using 1,3-bis (1-phenylethenyl)benzene (MDDPE) to prepare hetero four-arm star polymers via anionic polymerization has been extensively investigated [11,12,41,70]. This process involves two reactions: a coupling (addition) reaction and a crossover reaction to a second monomer. This reaction sequence is outlined in Scheme 13.7. In the first step, two living carbanionic



SecLover

Highlight

SecLover

Pencil

Scheme 13.7

polymer chains (P1Li) are coupled with the living coupling agent, MDDPE, to generate the corresponding dianion, [(P1)2MDDPE(Li)2] (12). In the second step, this polymeric dianion (12) reinitiates the polymerization of a second monomer (M2) to generate two new polymer chains (P2Li), thus forming a hetero four-arm star branched homopolymer or block copolymer (14) after termination. Functionalized heteroarm, star-branched polymers (15) can be prepared by quenching the dianion (13) with an appropriate electrophilic species (see Chap. 11) [74].

The coupling reactions can be monitored using ultraviolet (UV)-visible spectroscopy by observing the increase in absorbance of the diphenylalkyllithium species (12) at 438 nm [41]. Size exclusion chromatography (SEC) was used to follow the course of the coupling reaction as shown in Figure 13.8 [12] and also evaluate the efficiency of the coupling reaction. The coupling reaction of PSLi of various molecular weights with MDDPE in benzene is a very efficient reaction when the stoichiometry of the reaction is carefully controlled by determining the exact chain-end concentration. Coupling reactions of MDDPE with poly(styryl)-lithium produced the coupled product in excellent yield (> 96%) as shown in



Figure 13.8 SEC chromatograms of the coupling reaction of

MDDPE with poly(styryl)lithium (Mn = 1000 g/mol) in cyclohexane; base polymer (A), 55%

coupling (B) and 100% coupling (C). (From Ref. 12; reprinted by permission of Marcel Dekker,

New York.)

Table 13.4; the efficiency decreased slightly with increasing molecular weight. The efficient addition of two equivalents of low-molecular-weight poly(styryl)-lithiums with MDDPE and related compounds has recently been examined as a method to prepare hydrocarbon-soluble dilithium initiators [75].

The high reactivity of the second diphenylethylene unit in MDDPE with respect to formation of the coupled product (12) is further exemplified by the observation that, when using cyclohexane as solvent at room temperature, the

Table 13.4 Molecular Weight Characterization Data and Coupling Efficiencies in the Reaction of Stoichiometric Amounts of PSLi with MDDPE in Benzene as Determined by SEC

PS (PS)2

Mn

(g/mol)

Mw/Mn

Mn (Coupled) (g/mol)

Mw/Mn

Coupling Efficiency

(%)

1800 1.06 3700 1.05 99

3000 1.04 6300 1.04 99

4600 1.04 9650 1.04 97

9800 1.02 23,500 1.02 96

14,600 1.03 30,250 1.03 96

Source: Ref. 12.



addition reaction of poly(styryl)lithium (Mn = 3600 g/mol) with excess MDDPE ([MDDPE]/[PSLi] = 4) proceeded to give the dimeric adduct in 87% yield [76]. It should be noted that these results and previously published data [11,12,41] stand in sharp contrast to the results of Ikker and Möller [77], who reported an inability to obtain high coupling efficiencies for the reactions of poly(styryl)lithium with MDDPE in benzene using the cryptand [2.1.1].

One limitation of this methodology as a general procedure for synthesis of a wide variety of A2B2 star-branched polymers is the relative lack of reactivity of poly(dienyl)lithium compounds in addition reactions with 1,1-diphenylethylene units [78]. It is necessary to add a small amount of Lewis base such as THF ([THF]/[PLi] = 40) to accelerate the addition reaction; the coupling efficiency of poly(butadienyl)lithium (Mn = 1.9 × 103 g/mol) with MDDPE in cyclohexane was > 99% in the presence of added THF. The required addition of a Lewis base to promote this dienyllithium coupling reaction unfortunately limits the possibility of further diene polymerization for the second type of arms formed by the crossover reaction; in the presence of THF, the polydiene exhibits relatively high 1,2-microstructure and, consequently, a higher Tg (see Chap. 9) [79]. A further complication in the observed coupling reaction of poly(butadienyl)lithium with MDDPE is that unpolymerized butadiene (or isoprene [70]) monomer can cause the formation of small amounts of high-molecular-weight components in the coupled product because the butadiene will copolymerize with 1,1-diphenylethylene units to generate branch functionalities higher than two.

A second limitation with respect to the coupling reaction is the requirement that the living carbanionic polymers utilized must be sufficiently reactive to undergo facile addition reactions to 1,1-diphenylethylene units. In practice, this means that the first arms for the linking reaction are limited primarily to styrene- and diene-type monomers.

Crossover Reactions Of critical importance for the synthesis of hetero-arm star-branched polymers are reaction conditions that accelerate the rate of the crossover reaction of the diphenylalkyllithium sites (see 12) with monomer (initiation) (k2) relative to the rate of propagation (kp), as described by the relative rate constant ratio, k2/kp = R. The values of R for crossover to styrene monomer in benzene were determined to be 0.07 and 0.10 at 25°C and 5°C, respectively [80]. The analogous value of R at 25°C in cyclohexane was 0.12 [80]. Using the molecular weight distribution as an approximate experimental criterion for R, it was found that the narrowest molecular weight distributions for the polymers obtained by crossover experiments of polymeric diphenylmethyllithiums with styrene were obtained at 5°C (vs. 25 or 45°C) in cyclohexane (vs. benzene). Diphenylalkyllithium can initiate styrene monomer polymerization with a reasonably rapid rate, even though it is slower than its homopolymerization rate (R < 1). Stoichiometry is very important; when excess MDDPE was present in the crossover reaction, the star polymer product exhibited multimodal distribution by SEC



SecLover

Highlight

analysis (with branching functionality greater than four arms). Even under stoichiometric coupling reaction conditions, the star-branched polymer obtained from the crossover reaction with styrene to form growing polystyrene arms with Mn = 3.0 × 103 g/mol in each arm in cyclohexane solution showed bimodal molecular weight distribution as illustrated in Figure 13.9; the degree of bimodality reflects the relative number of molecules growing by propagation with two reactive sites versus one reactive site. Whenever bimodality was observed, the UV-visible spectrum showed the existence of residual absorption for unreacted diphenylalkyllithium sites at 438 nm.

It was considered that the observed bimodality resulted from unusual chain-end association behavior for these dilithium species in hydrocarbon solution. Analogous to other polymeric organolithium compounds (see Chapter 1) [16], the dilithium adduct would be expected to be aggregated at least into dimers in hydrocarbon solution. A cyclic dimeric ring form has been suggested by Szwarc and co-workers [81] for these difunctional diphenylalkyllithium aggregates. The unique association behavior of these species may complicate the crossover reaction compared to monofunctional diphenylalkyllithium species, which, although also dimeric in hydrocarbon solution [16,82], exhibit no analogous inefficient crossover reactions to styrene monomer.

Addition of tetrahydrofuran (THF) to a coupled product (12) prior to addition of styrene produces narrow-molecular-weight distribution, star-branched polymers even for polystyrene growing arm molecular weight/s as low as 3.0 × 103 g/mol as shown in Figure 13.10. However, when the growing arm molecular

Figure 13.9 SEC chromatograms of four-arm,

star-branched polystyrene, [PS(Mn = 3000 g/mol)]2-

MDDPE-[PS(Mn = 3000 g/mol)]2, prepared in cyclohexane: (A) at 5°C; (B) at 25°C; (C) at 45°C

(From Ref. 12; reprinted by permission of Marcel Dekker, New York.)



Figure 13.10 SEC chromatograms of

base polystyrene, Mn = 12,000 g/mol (A);

coupled product, (PS)2-MDDPE (B); and

(PS)2-MDDPE-(PS)2 star polymer with

arm-out Mn (PS) = 3000 g/mol (C). (From Ref. 12; reprinted

by permission of Marcel Dekker, New York.)

weight was only 2 × 103 g/mol, the presence of residual diphenylalkyllithium species was detected by UV-visible absorption at 438 nm [41]. Thus, the minimum arm lengths required for complete crossover to styrene monomer in the presence of THF are 3 × 103 g/mol. As discussed previously in connection with the coupling reaction of poly(butadienyl)lithium with MDDPE, the required addition of THF was not a satisfactory solution to prevent bimodal molecular weight distributions for the crossover reaction to form elastomeric polydiene blocks, because it is known that diene microstructure is dramatically changed from high 1,4- to high 1,2-microstructure for alkyllithium-initiated diene polymerization in the presence of small amounts of THF (see Chapter 9) [79]. It was found that the addition of sufficient amounts of lithium sec-butoxide ([sec-BuOLi]/[PLi] = 1) to the living coupled product (12) prior to addition of butadiene monomer produced monomodal, heteroarm star-branched polymers, even with relatively low molecular weights for the polybutadiene arms that grow out from the coupled product; however, the molecular weight distribution was somewhat broad. This effect of lithium alkoxides was also observed for the dilithium initiator formed by the addition of two moles of sec-butyllithium to MDDPE [42]; this effect has recently been confirmed for isoprene polymerization [83]. Polymer products with minimum polybutadiene (PBD) arm molecular weights (Mn = 10.0 × 103 g/mol) exhibited a reasonably narrow molecular weight distribution without significant



effect on polybutadiene microstructure as shown by the characterization data in Table 13.5.

There are fewer restrictions on the monomers that can be used in the cross-over reaction than for the linking reaction. In essence, any monomer that undergoes living anionic polymerization can be added to the dilithium adduct (12) to grow arms and form heteroarm star-branched polymers with well-defined structures and low degrees of compositional heterogeneity. Since the growing arms are living, block copolymer arms can also be prepared and these polymers exhibit unique and interesting physical and rheological properties [84].

Three-Armed Star Polymers

As discussed previously, 1,3-bis(1-phenylethenyl)benzene (MDDPE) reacts rapidly with two equivalents of an organolithium compound in hydrocarbon solution to form the corresponding dilithium adduct. In fact, it has been reported that the rate constants for the first and second addition step of MDDPE with sec-BuLi are both 1.98 × 10-2 L/mol s in toluene [85]; for the para analog, 1,4-bis(1-phenylethenyl)benzene (PDDPE), the rate constant for the first addition is 2.06 × 10-2 L/mol s and that for the second addition is 1.51 × 10-3 L/mol s [85]. However, McGrath and co-workers [86] reported that the monoadduct of MDDPE and sec-butyllithium could be prepared in THF at -78°C. Similar observations confirming the dramatic effect of solvents on the course of these addition reactions were made by J.-J. Ma [87] for the corresponding reaction with poly(styryl)lithium and recently confirmed by Ikker and Möller [77]. This interesting effect of THF has been utilized to prepare macromonomers with a terminal 1,1-diphenylethylene functionality by addition of one equivalent of polymeric organolithium with either

Table 13.5 Molecular Weight Characteristics and Microstructures of Star-Branched (PS)2-MDDPE-(PB)2 Styrene-Butadiene Block Copolymers Prepared in Cyclohexane in the Presence of Lithium sec-Butoxide

Microstructure of PB Block

Number Mn× 10-3 (g/mol)

Calculateda

1H-NMRb Mw/Mn1,4- 1,2-

S-1 22–2.52 8.7 1.14 85 15

S-2 22–102 24.5 1.05 90 10

S-3 102–11.52 42.5 92 8

aSuperscripts indicate the number of arms.

bCalculated using the polystyrene molecular weight (SEC) and the relative integration areas for the aromatic protons relative to the vinyl protons from the diene units.

Source: Ref. 12.



MDDPE [1,3-bis(1-phenylethenyl)benzene] or PDDPE [1,4-bis(1-phenylethenyl)-benzene] as illustrated in Scheme 13.8 [12,73,76]. The resulting nonhomopolymerizable macromonomers, 16, provide the starting materials for the rational synthesis of hetero three-armed, star-branched polymers as illustrated in Scheme

Scheme 13.9

13.9 for synthesis of a three-armed, heteroarm, star-branched polymer with three polystyrene arms with different molecular weights [12,73].

Although the addition reactions of poly(styryl)lithium with both MDDPE and PDDPE were investigated for preparing these macromonomers, PDDPE exhibited less tendency to form the corresponding diadduct in both hydrocarbon solution and in the presence of THF. It is presumed that dimer formation is less favorable in the case of the para-substituted PDDPE than the meta-substituted MDDPE because the negative charge in the monoadduct can be delocalized into



Figure 13.11 SEC chromatograms of (A) polystyrene macromonomer (Mn

= 5.4 × 103 g/mol); (B) polystyrene base polymer (Mn = 15.3 ×

103 g/mol) for second arm; (C) coupled product of PSLi with polystyrene macromonomer

(Mn = 20.6 × 103 g/mol); (D) hetero three-armed star polymer PS(1)-PDDPE-

branch-PS(2)PS(3) [Mn (calc.) = 50

× 103 g/mol] and (E) fractionated star polymer. (From Ref. 73; reprinted by permission of Springer-Verlag,

Heidelberg, Germany.)

two aromatic rings and the remaining vinyl group. As discussed previously, even with a fourfold excess of MDDPE, the addition of poly(styryl)lithium in benzene at room temperature produced the dimer in 87% yield [76]. In contrast, with only a twofold excess of PDDPE, the corresponding reaction with poly(styryl)lithium produced only 7% dimer.

For the synthesis of three-armed, heteroarm, star-branched polymers, the first step involves the addition of a polymeric organolithium compound (e.g., poly(styryl)lithium) with the macromonomer, 16, to form the corresponding coupled product, 17, a diphenylalkyllithium. For stoichiometric amounts of polymer lithium and macromonomer, it was found that the efficiency of this coupling reaction is > 96%. Finally, the third arm is formed by addition of monomer (e.g., styrene) in the presence of THF to promote the crossover reaction to form the third arm. SEC analyses of the hetereoarm star polymer product, 18, formed after alcohol termination, showed that each of these steps proceeds efficiently to give the expected products; only relatively small amounts of nonstar product were observed, which corresponds in retention volume to the small amount of unreacted macromonomer and a small amount of polystyrene corresponding to the second



arm as shown in Figure 13.11. A narrow-molecular-weight distribution star product (Mw/Mn = 1.02) was easily obtained by one fractionation step.

The limitations of this star polymer synthesis are analogous to those discussed for the hetero four-armed star polymers. Only reactive polymeric organolithiums will react with the 1,1-diphenylethylene-functionalized macromonomer. However, any monomer that undergoes living anionic polymerization can be added to the coupled lithium adduct (17) to grow an arm and form heteroarm star-branched polymers with well-defined structures and low degrees of compositional heterogeneity. With respect to the macromonomer synthesis, although the meth-

Figure 13.12 SEC chromatograms of first, second and third

products during preparation of hetero, three-arm star polymer using macromonomer methodology

(see Scheme 13.11). —.—.—.—., base polystyrene (PSLi); ———————, coupling

product (24) of PSLi with macromonomer (23); ———, hetero three-armed star

polymer (25) [polystyrene, polydimethylsiloxane, poly(t-butyl methacrylate)].

(From Ref. 88; reprinted by permission of Elsevier Science Ltd., Oxford, UK.)



odology shown in Scheme 13.9 is limited to reactive chain ends, any synthetic methodology that generates a 1,1-diphenylethylene functionality at the terminus of a polymer chain can be used to synthesize the macromonomer.

For example, a 1,1-diphenylethylene functionalized polydimethylsiloxane macromonomer, 23, was prepared as shown in Scheme 13.10 [88]. After frac-

Scheme 13.10

tionation to obtain a narrower molecular weight distribution, this macromonomer was reacted first with poly(styryl)lithium followed by crossover to t-butyl methacrylate to form the corresponding ABC hetero, three-armed star polymer as



Scheme 13.11

shown in Scheme 13.11. An excess of macromonomer was used and the resulting star polymers, 25, exhibited broad, multimodal molecular weight distributions; fractionation provided “fairly narrow molecular weight distributions” (see Fig. 13.12). It is highly probable that poly(styryl)lithium will also attack the siloxane bonds in the macromonomer, which would cause multiple linking of the poly(styryl)lithium with the macromonomer. In spite of these shortcomings, these results demonstrate the potential versatility of the macromonomer method based on 1,1-diphenylethylene functionality.

III. Conclusions

A variety of useful methods are available for the synthesis of regular star-branched polymers. The end-linking reactions of polymeric organolithium compounds with multifunctional chlorosilanes are a versatile and efficient method for preparing a wide variety of well-defined star-branched polymers. However, the syntheses of multiarm stars generally require the use of dienyllithium chain ends and an excess of living arms to minimize the formation of stars with intermediate degrees of branching. As a result, fractionation is generally required to obtain a pure star-branched polymer. Both multistep syntheses with silyl halides and living linking reactions have been utilized to prepare heteroarm, star-branched polymers with low degrees of compositional heterogeneity. In all cases, adequate characteriza-



tion of the products from each reaction step, as well as the final star-branched polymer, is essential.

References

1. J. Roovers, in Encyclopedia of Polymer Science and Engineering, 2nd ed., J. I. Kroschwitz, Ed., Wiley-Interscience, New York, 1985, Vol. 2, p. 478.

2. Source-based nomenclature for copolymers, Pure Appl. Chem., 57, 1427 (1985).


4. B. J. Bauer and L. J. Fetters, Rubber Chem. Technol., 51, 406 (1978).

5. K. W. Doak, in Encyclopedia of Polymer Science and Engineering, 2nd ed., J. I. Kroschwitz, Ed., Wiley-Interscience, New York, 1986, Vol. 6, p. 383, 386.

6. D. L. Beach and Y. V. Kissin, in Encyclopedia of Polymer Science and Engineering, 2nd ed., J. I. Kroschwitz, Ed., Wiley-Interscience, New York, 1986, Vol. 6, p. 454.

7. P. A. Small, Adv. Polym. Sci., 18, 1 (1975).

8. J. E. Mark, A. Eisenberg, W. W. Graessley, L. Mandelkern, E. T. Samulski, J. L. Koenig, and G. D. Wignall, Physical Properties of Polymers, 2nd ed., American Chemical Society, Washington, D.C., 1993.

9. L. J. Fetters and E. L. Thomas, in Material Science & Technology, Vol. 12, VCH Verlagsgesellschaft, Weinheim, Germany, 1993, p 1.

10. P. Rempp and J. E. Herz, in Encyclopedia of Polymer Science and Engineering, 2nd ed., J. I. Kroschwitz, Ed., Wiley-Interscience, New York, 1989, Supplement Volume, p. 493.

11. R. P. Quirk, B. Lee, and L. E. Schock, Makromol. Chem., Macromol, Symp., 53, 201 (1992).

12. R. P. Quirk, T. Yoo, and B. Lee, J. Macromol. Sci. Pure Appl. Chem., A31, 911 (1994).

13. H. Iatrou and N. Hadjichristidis, Macromolecules, 25, 4649 (1992).





18. S. Bywater, in Encyclopedia of Polymer Science and Engineering, 2nd ed., J. I. Kroschwitz, Ed., Wiley-Interscience, New York, 1985, Vol.2, p. 1.

19. H. L. Hsieh, Rubber Chem. Technol., 49, 1305 (1976).

20. R. P. Quirk, D. J. Kinning, and L J. Fetters, in Comprehensive Polymer Science, G. Allen and J. C. Bevington, Eds., Pergamon Press, Vol. 7, 1989, p. 1.

21. R. N. Young and L. J. Fetters, Macromolecules, 11, 899 (1978).




27. C. Tsitsilianis, P. Chaumont, and P. Rempp, Makromol. Chem., 191, 2319 (1990).

28. C. Tsitsilianis, S. Graff, and P. Rempp, Eur. Polym. J., 27, 243 (1991).

29. T. Masuda, Y. Ohta, T. Yamauchi, and S. Onogi, Polym. J., 16, 273 (1984).

30. L.-K. Bi and L. J. Fetters, Macromolecules, 9, 732 (1976).

31. J.-M. Widmaier, Makromol. Chem., 186, 2079 (1985).

32. G. D. Karles, W. H. Christiansen, J. G. Ekerdt, I. Trachtenberg, and J. W. Barlow, Ind. Eng. Chem. Res., 30, 646 (1991).

33. M. Schmidt and W. Burchard, Macromolecules, 14, 370 (1981).

34. G. Odian, Principles of Polymerization, 3rd ed., Wiley-interscience, New York, 1991, p. 460.

35. H. Eschwey, M. L. Hallensleber, and W. Burchard, Makromol. Chem., 173, 235 (1973).

36. P. Lutz and P. Rempp, Makromol. Chem., 189, 1051 (1988).

37. C. Tsitsilianis, P. Lutz, S. Graff, J.-P. Lamps, and P. Rempp, Macromolecules, 24, 5897 (1991).


39. S. Varshney, C. Jacobs, J.-P. Hautekeer, P. Bayard, R. Jerome, R. Fayt, and Ph. Teyssie, Macromolecules, 24, 4997 (1991).

40. H. Eschwey and W. Burchard, Polymer, 16, 180 (1975).

41. R. P. Quirk and F. Ignatz-Hoover, in Recent Advances in Anionic Polymerization, T. E. Hogen-Esch and J. Smid, Eds., Elsevier, New York, 1987. p. 393.


43. Y. Tsai, unpublished work at the University of Akron.

44. B. J. Wakefield, Organolithium Methods, Academic Press, New York, 1988, p. 107.

45. I. P. Beletskaya, G. A. Artamkina, and O. A. Reutov, Russ. Chem. Rev., 45, 330 (1976).

46. H. E. Zieger and D. Mathisen, J. Am. Chem. Soc., 101, 2207 (1979).

47. T. Altares, D. P. Wyman, V. R. Allen, and K. Meyerson, J. Polym. Sci., A3, 4131 (1965).

48. A. M. Belu, M. O. Hunt, Jr., J. M. DeSimone, and R. W. Linton, Macromolecules, 27, 1905 (1994).

49. D. Wittenberg and H. Gilman, Quart. Rev., 13, 116 (1959).

50. M. Morton, T. E. Helminiak, S. D. Gadkary, and F. Bueche, J. Polym. Sci., 57, 471 (1962).

51. T. Masuda, Y. Ohta, and S. Onogi, Macromolecules, 4, 763 (1971).



SecLover

Highlight

SecLover

Highlight

SecLover

Highlight

SecLover

Highlight

SecLover

Highlight

SecLover

Highlight

59. P. J. Toporowski and J. Roovers, J. Polym. Sci., Polym. Chem. Ed., 24, 3009 (1986).

60. J. Roovers, N. Hadjichristidis, and L. J. Fetters, Macromolecules, 16, 214 (1983).

61. A. B. Nguyen, N. Hadjichristidis, and L. J. Fetters, Macromolecules, 19, 768 (1986).

62. L.-L. Zhou and J. Roovers, Macromolecules, 26, 963 (1993).

63. L.-L. Zhou, N. Hadjichristidis, P. M. Toporowski, and J. Roovers, Rubber Chem. Technol., 65, 303 (1992).

64. J. Roovers, L.-L. Zhou, P. M. Toporowski, M. van der Zwan, H. Iatrou, and N. Hadjichristidis, Macromolecules, 26, 4324 (1993).

65. J. Roovers, P. Toporowski, and J. Martin, Macromolecules, 22, 1897 (1989).

66. R. W. Pennisi and L. J. Fetters, Macromolecules, 21, 1094 (1988).

67. N. Khasat, R. W. Pennisi, N. Hadjichristidis, and L. J. Fetters, Macromolecules, 21, 1100 (1988).

68. J. W. Mays, Polym. Bull., 23, 247 (1990).

69. H. Iatrou and N. Hadjichristidis, Macromolecules, 26, 2479 (1993).

70. S. J. Wright, R. N. Young, and T. G. Croucher, Polym. Int., 33, 121 (1994).

71. J. Roovers and P. M. Toporowski, Macromolecules, 14, 1174 (1981).

72. R. P. Quirk, S.-H. Guo, M. Alsamarraie, and F. Ignatz-Hoover, Polym. Prepr., Am. Chem. Soc., Div. Polym. Chem., 29(2), 298 (1988).

73. R. P. Quirk and T. Yoo, Polym. Bull., 31, 29 (1993).

74. R. P. Quirk, in Comprehensive Polymer Science. First Supplement, S. L. Agarwal and S. Russo, Eds., Pergamon Press, Oxford, 1992, p. 83.

75. L. H. Tung and G. Y.-S. Lo, Macromolecules, 27, 1680 (1994).

76. R. P. Quirk and T. Yoo, Polym. Prepr., Am. Chem. Soc., Div. Polym. Chem., 34(2), 578 (1993).

77. A. Ikker and M. Möller, New Polymeric Mater., 4, 35 (1993).



80. R. P. Quirk, L. E. Schock, and B. Lee, Polym. Prepr., Am. Chem. Soc., Div. Polym. Chem., 30(1), 113 (1989).

81. A. Yamagishi, M. Szwarc, L. Tung, and G. Y-S. Lo, Macromolecules, 11, 607 (1978).

82. L. J. Fetters and R. N. Young, in Anionic Polymerization. Kinetics, Mechanisms, and Synthesis, J. E. McGrath, Ed., ACS Symposium Series No. 166, American Chemical Society, Washington, D.C., 1981, p. 95.



SecLover

Highlight

14 Graft Copolymers

I. Introduction

A graft copolymer is a polymer comprising molecules with one or more species of block connected to the main chain as side chains, having constitutional or configurational features that differ from the main chain, exclusive of branch points [1]. In a graft copolymer, the distinguishing feature of the side chains is constitutional: the side chains comprise units derived from at least one species of monomer different from those that supply the units of the main chain [1]. The recommended IUPAC nomenclature for graft copolymers states that the name of the backbone polymer should be given first, and the name of the grafted branch second, the word graft indicating the structure of the macromolecules [1]. Thus, poly(styrene-graft-isoprene) or poly(styrene-g-isoprene) refers to a graft copolymer containing a polystyrene backbone with polyisoprene grafts or branches [1].

Graft polymerization is an attractive method for the modification of polymer properties [2–4]. Since the main chains and the branch chains are generally thermodynamically incompatible, most graft copolymers are multiphase materials in the solid state; other examples of multiphase materials are polymer blends, block copolymers, and interpenetrating networks [4–6]. Because the two immiscible phases are joined by covalent bonds, analogous to block copolymers, a limited range of composition-sensitive phase and morphological behavior is expected [7–13]. Microphase-separated graft copolymers can exhibit many of the



Administrator

Underline

unique thermal and mechanical properties observed for block copolymers, including thermoplastic elasticity. Since the morphology of heterophase polymers can be affected by the casting solvent and the nature of its interaction with the polymer blocks [5,8,14], their physical properties are also expected to depend on the casting solvent. These factors are responsible for technologically important materials, such as high-impact polystyrene (HIPS), ABS resins (acrylonitrile-butadiene-styrene graft copolymers), and MBS resins (methacrylate-butadiene-styrene graft copolymers) [15]. Graft copolymers are also used as emulsifiers, surface-modifying agents, coating materials, adhesives, and compatibilizing agents for polymer blends [16].

Free radical polymerization methods are the simplest, oldest, and most widely used procedures for the synthesis of graft copolymers [2]. Unfortunately, these methods generally lead to materials difficult to characterize and that contain varying amounts of the corresponding homopolymers and gel fractions. Methods have recently been developed to produce graft polymers with relatively well-defined structures [16,17].

II. Structure and Characterization

A graft polymer will have one or more graft branches per molecule. One of the important variables to be defined and controlled is the average number of graft branches per macromolecule and the spacing distribution of graft branches along the backbone chain (e.g., random, block, etc.). The amount of crosslinking should also be determined. Other important structural features to control and define are the average length of the branches and their length distribution. Without such basic structural information, a rather Edisonian approach to structure-property relationships is all that is possible.

With structural definition such as the average number, distribution, and size of branches, the morphological features of the graft copolymers can be investigated in a rational manner. One would anticipate that the morphology and properties of graft copolymers, similarly to block copolymers [7–13], would be dependent on the molecular weights of the segments and their volume fractions. In general, the level of structural definition required to explore these variables has not been available.

Preparative methods are being developed for the synthesis of model graft polymers that have the well-defined structural features outlined above [4,16,17]. With the availability of such materials, the relationships between the structure of the graft polymer, morphology, and the properties of the material can be explored. Furthermore, an understanding of what the optimum structural features are for a given system will provide insight into the best synthetic method to achieve or approach the desired structure.

The subject of compositional heterogeneity among grafted polymer chains is



Administrator

Text Box

乳化剂,乳化器

Administrator

Pencil

only a point of conjecture because of the rather primitive state of reliable characterization methods [18–20]. However, compositional heterogeneity would be expected to be an issue in systems that undergo phase separation during the graft polymerization reaction. Compositional heterogeneity can be manifested as variations in the number of branches formed vs. time, as changes in the average graft chain length distribution, and as alterations in the predominant termination or transfer modes. Compositional heterogeneity can be monitored by analyzing the grafting products of the grafting reaction at various stages of the reaction.

III. General Aspects of Graft Copolymer Synthesis

There are three general methods for the synthesis of graft copolymers: (a) “grafting from” (Scheme 14.1), in which active sites are generated on a backbone

Scheme 14.1 Grafting From

polymer chain and are used to initiate the polymerization of a second monomer, M2, to form a graft branch; (b) “grafting onto” (Scheme 14.2), in which a

Scheme 14.2 Grafting Onto

backbone polymer chain contains randomly distributed reactive functional groups, X, which can react with another macromolecular chain that carries antagonistic, reactive functional groups, Y, at the chain ends; and (c) “grafting through” (Scheme 14.3), in which a growing polymer chain with reactive chain end sites

Scheme 14.3 Grafting Through



incorporates a branch chain by addition to another polymer chain that possesses reactive unsaturated groups and copolymerizes with the backbone-forming monomer (M1) [16].

IV. Anionic Synthesis

Anionic polymerization is particularly suitable for the synthesis of graft copolymers with low degrees of compositional heterogeneity [4,16,17]. The absence of termination and chain transfer reactions allows the preparation of polymers with narrow molecular weight distributions and predictable molecular weights [21–28]. When applied to graft copolymer synthesis, anionic nonterminating systems have the advantage that branches with uniform length can be prepared. In principle, anionic systems can provide graft polymers with well-defined structures to delineate the relationships between structure, morphology, and properties. However, there is a major limitation for all of the available anionic methods for graft copolymer synthesis: there is generally a random distribution of branch points along the backbone chain since there is no control over the generation of the loci for branch formation.

There are three general methods for the anionic synthesis of graft copolymers: (a) metalation of the backbone chain followed by anionic graft branch growth is an example of a “grafting from” process; (b) linking of chains with anionic chain ends with functional groups on the backbone chain is an example of a “grafting onto” process; and (c) copolymerization of monomers with anionically synthesized macromonomers that have polymerizable chain-end functionality is an example of a “grafting through” process.

A. Metalation Grafting

The “grafting from” method (Scheme 14.1) generally involves generation of carbanionic (organometallic) initiator sites on the backbone chain by (1) metalation of carbon-hydrogen bonds using an organometallic compound, RMt (Eq. 14.1); (2) metal-halogen exchange reactions with carbon-halogen bonds using an organometallic compound or an alkali metal (M) (Eqs. 14.2, 14.3); and (3) by addition of an organometallic compound to a reactive vinyl group on the polymer (Eq. 14.4) [29–32]. Subsequently, monomer is added and anionic polymerization

(14.1)

(14.2)

(14.3)

(14.4)

takes place to form grafted branches at the sites of metalation. By water [23] has reviewed earlier work on these grafting techniques and has discussed their limita-



Administrator

Text Box

地点，所在地, [数]轨迹

Administrator

Pencil

Administrator

Highlight

tions (see also [25,33–35]). To obtain grafts with uniform chain length, the rate of initiation of anionic chain growth must be at least of the same order of magnitude as the propagation rate [21,36]. This requirement is readily fulfilled for styrene and diene monomers using homogeneous organometallic initiators such as radical anions and alkyllithiums, especially sec-butyllithium [37]. However, since many of the metalation sites for grafting involve stabilized species such as allylic, benzylic, or aryl anions in the presence of coordinating ligands, it is not necessarily correct to assume that chains of uniform length will be produced, especially from more reactive, polar monomers. Initiation may be slow relative to propagation, thereby producing chains with nonuniform lengths.

One of the most generally studied metalation procedures for graft polymerization utilizes an alkyllithium compound complexed with N,N,N',N'-tetra-methylethylenediamine (TMEDA) [35]. These reactive complexes can metalate allylic, benzylic, and even aromatic C-H bonds. Active sites for initiation are generated in a random process. Limitations of this method include aggregation phenomena, which can render the metalated backbone insoluble, lack of quantitative metalation, and inhomogeneous initiation. The lack of quantitative initiation not only precludes estimation of the number of grafts formed (i.e., lack of control of stoichiometry, and therefore, Mn), but can also generate homopolymer of the monomer to be grafted. As discussed previously, the lack of homogeneous chain initiation leads to branches with nonuniform length.

Polybutadiene (see Scheme 14.4) and polyisoprene have been metalated with

Scheme 14.4

the complex of n-butyllithium and TMEDA, followed by “grafting from” reactions with styrene [38,39]. It was reported that the intrinsic viscosity of the polymer decreased after hydrolysis of the metalated polymer. Apparently, some degradation of the backbone chain occurs during the polydiene metalation reaction at 50°C. The amount of degradation reportedly increased with the relative amount of the n-BuLi/TMEDA complex. Grafting efficiencies of 65–97% (Eq. 14.5) were determined by acetone extraction of homopolystyrene. Catalyst effi



Administrator

Text Box

烯丙基的

Grafting Efficiency = %G.E. = {[wt of styrene grafted]/ [total wt of styrene polymerized]} × 100 (14.5)

ciencies of 75–95% were determined by measuring the Mn for the polystyrene branches after OsO4/t-butyl hydroperoxide oxidative degradation of the polybutadiene backbone and comparing this with Mn calculated from the charge ratio of the grams of monomer per mole of metalating agent. The molecular weight distribution of the grafted polystyrene was described as being “rather wide” [38].

It has been reported that the use of either t-butyllithium/alkali metal alkoxides at 50°C [39] or sec-butyllithium/TMEDA at room temperature [40] eliminates backbone chain degradation. Catalyst efficiencies of 40–58% were reported using the alkoxide-activated system [39]. This corresponds to branches with Mn = 15–22 × 103 g/mol vs. a calculated Mn = 8.7 × 103 g/mol. In addition to the problem of homogeneous chain initiation, double titration analysis indicated that 42–60% of the active chain ends were lost after 1–2 h at 50°C [39]. Falk and co-workers [41] have described a simple procedure for the preparation of poly-(butadiene-g-styrene) using the sec-butyllithium/TMEDA complex. Grafting efficiencies of 96% were reported for this procedure. Earlier work by this group indicated that these graft copolymers behave like thermoplastic elastomers if the polystyrene content is between 34 and 42%, with an average of 10–20 graft sites per polybutadiene chain [40]. Lower melt viscosities than those of the corresponding ABA triblock polymers were also noted. The hydrogenated graft copolymer derivatives of these poly(diene-g-styrene) have also been described [40,41]. Only weak materials were obtained by analogous grafting of polystyrene from EPDM rubbers [43]. These graft copolymers were characterized merely by extraction methods (with acetone) and tensile property measurements.

Hadjichristidis and Roovers [44] have utilized the sec-butyllithium/TMEDA metalation procedure developed by Falk et al. [40] to prepare model poly-(isoprene-g-styrene) for polymer conformational analysis in dilute solution. In contrast to the observations of Falk et al. [40], 20–28% homopolystyrene was observed by SEC. From SEC, the uncorrected polydispersity (Mw/Mn) was 1.29. The molecular weight of the polystyrene branches isolated after m-chloroper-benzoic acid/HIO4 degradation of the polyisoprene backbone was identical with that of the ungrafted homopolystyrene. On the basis of the observed extinction coefficients, it was concluded that not all of the sec-butyllithium is converted to allylic, backbone organolithium species. It was suggested that a side reaction occurs that forms a lithium compound that does not metalate the backbone, but that can still initiate styrene homopolymerization. From light-scattering measurements, the radius of gyration of the polyisoprene backbone was reported to be smaller than that of the polystyrene branches. These results were interpreted in terms of a core-shell model for the graft copolymer conformation in dilute solution, with the backbone preferentially occupying the core.



Administrator

Underline

Administrator

Underline

Administrator

Highlight

Administrator

Rectangle

The dependence of grafting efficiency on the microstructure of polyisoprene has been investigated [45]. Metalation of polyisoprenes with various microstructures was effected using the sec-butyllithium/TMEDA complex in cyclohexane at 30°C for 24 h followed by addition of styrene monomer. It was reported that the grafting efficiency was improved and the molecular weight distribution was narrowed from > 3 to 1.2–1.5 by adding 5% THF to the metalation solution. The results are shown in Table 14.1.

As expected from the number of allylic C-H bonds for each type of microstructure, the maximum grafting efficiency was obtained with the highest 1,4-microstructure (7 allylic H per repeat unit). Lower grafting efficiencies result with an increase in 3,4- (4 allylic H per repeat unit) and 1,2-content (no allylic H per repeat unit).

The use of 1,2-dipiperidinoethane (DIPIP) in place of TMEDA with sec-butyllithium in cyclohexane for metalation of 1,4-polybutadiene has been investigated to prepare graft copolymers with styrene and butadiene [46]. In general, the rate of metalation was slower than with TMEDA, but the grafting efficiencies were comparable. In contrast to sec-BuLi/TMEDA metalations, the use of DIPIP generates highly delocalized species absorbing at 445 and 540 nm that are poor grafting initiators. The fraction of active lithium for grafting was estimated from the molecular weight of the homopolymer produced during the grafting reactions using Eq. 14.6, where Ms and Mobs are the stoichiometric and observed molecular

fraction active C - Li = Ms/Mobs (14.6)

weights, respectively; the values varied from 0.28 (7.5 h) and 0.43 (8.75 h) to 0.38 (24 h) for DIPIP, while a value of 0.61 (2 h) was found for TMEDA. The grafting efficiencies (Eq. 14.5) using the sec-BuLi- DIPIP complex were as high as 82%

Table 14.1 Influence of Polyisoprene Microstructure on Grafting Efficiencies

Polyisoprene Microstructure (%) PS G.E. Mw/Mn

Cis-1,4 Trans-1,4 3,4 1,2 (wt %) (%) (SEC)

95 — 2 — 35 98 1.3

62 91 1.2

68 25 7 — 31 84 1.5

62 78 1.5

2 38 50 10 30 50 1.4

60 42 1.3

— 11 61 28 35 44 1.5

65 39 1.4

Source: Ref. 45.



YFP

Underline

YFP

Underline

YFP

Rectangle

YFP

Highlight

YFP

Highlight

YFP

Highlight

YFP

Highlight

YFP

Text Box

烯丙基氢

YFP

Rectangle

after a 24 h lithiation time, which is comparable to the grafting efficiencies using the corresponding TMEDA complex after a 2 h lithiation time.

In conclusion, although the metalation grafting procedure is relatively simple, and in principle could produce well-defined graft polymers, present results indicate that this method has certain limitations. The observation of broad molecular weight distributions indicates that inhomogeneous initiation of graft branches may be occurring, which results in lack of control of graft molecular weight and molecular weight distribution. The formation of the homopolymer of the grafting monomer also indicates that backbone initiation sites are not formed quantitatively. Thus, these methods are not completely suitable for the preparation of structurally well defined graft polymers.

Earlier work on the use of the lithium-halogen exchange reaction to generate metalated backbone polymers for grafting reactions has been reviewed elsewhere [23,33,34]. In general, one of the major problems with metalation by metal-halogen exchange is the problem of removing all of the excess organometallic compound, which can initiate the homopolymerization of the grafting monomer.

The anionic graft polymerization of polar monomers can be initiated with less reactive anionic groups than those required for diene and styrene polymerization. However, side reactions with the polar functional groups can complicate these polymerizations [23]. Pivalolactone is one polar monomer that can be polymerized anionically without chain termination or chain transfer, using carboxylate salts as initiating species [47–49]. Since the tetrabutylammonium carboxylate initiation rate is comparable to the propagation rate and the resulting poly(pivalolactone) chain end is stable in the absence of impurities, homopolymers with relatively narrow molecular weight distributions have been prepared [50–52]. A relatively simple procedure for the grafting of crystalline poly(pivalolactone) blocks from various elastomers has been described as shown in Scheme 14.5 [50,51,53–56]. For example, a combined block-graft copolymer

Scheme 14.5

was prepared by further metalation of α,ω-dilithio-cis-1,4-polyisoprene (Mn = 35–90 × 103 g/mol) with n-butyllithium/TMEDA at 55–60°C or sec-butyllithium/TMEDA at 25°C followed by carboxylation, conversion to the tetrabutylammo-



YFP

Highlight

YFP

Underline

YFP

Underline

YFP

Pencil

YFP

Highlight

YFP

Highlight

YFP

Underline

YFP

Text Box

新戊内酯

YFP

Line

nium salts, and subsequent polymerization of pivalolactone[53]. The characterization of the carboxylated polymer was not reported, although side reactions are known to occur in these carboxylations[57–59]. A variety of polymers with carboxyl functionality undergo controlled “grafting from” reactions with pivalolactone[50,51,53,55,56]. Pivalolactone conversions of 93% and 3.8–12 wt % homopolymer formation were reported[51]. No experimental characterization of the number, distribution, molecular weight, or polydispersity of the pivalolactone graft branches was reported. These results suggest that analogous procedures can be used to prepare graft polymers with well defined structures, provided that appropriate backbone chain functional groups and monomers that polymerize anionically in a controlled fashion are chosen.

B. Addition Reactions

The “grafting from” method (Scheme 14.1) involving the generation of organometallic active centers on a polymer backbone using addition reactions has only been investigated for a few specialized systems. For example, it has been reported that both m- and p-diisopropenylbenzene can be polymerized or copolymerized with α-methylstyrene anionically to form linear polymers (Mw/Mn < 1.3), under well defined conditions (i.e., using 1-phenylethylpotassium as initiator in THF at -30°C and at < 50% conversion) [60]. In principle, each repeating unit bears an unsaturated side group. Addition of 1.5 molar equivalents of sec-butyllithium (relative to double bonds) in benzene at temperatures above the ceiling temperature of the α-methylstyryl-type units (35°C) results in an insoluble species because of chain end association. A homogeneous polylithiated polymeric initiator was obtained by addition of THF and this species was used to initiate the anionic graft copolymerization of styrene as shown in Scheme 14.6. The addition of

Scheme 14.6

excess sec-butyllithium generated polystyrene homopolymer that was separated by fractional precipitation. Evidence for intermolecular branching was reported.

A similar procedure has been reported using the controlled butyllithium-initiated copolymerization of m- or p-divinylbenzene (DVB) with 1,1-diphenyl-



YFP

Rectangle

YFP

Underline

YFP

Underline

YFP

Highlight

ethylene (DPE) in THF at -78°C to produce a linear polymer[61,62]. The monomer reactivity ratios at -78°C for m- and p-divinylbenzene were 1.2 and 2.8, respectively. Since the monomer reactivity ratio for DPE = 0, a highly alternating structure was formed. Reaction of this unsaturated, alternating copolymer (ca. 1/1 DVB/DPE) with one equivalent of sec-butyllithium per double bond at -78°C in THF formed a polyanionic initiator for the graft copolymerization of methyl methacrylate (see Scheme 14.7). However, it was necessary to deactivate

Scheme 14.7

many of the backbone organometallic initiating sites (80%) with methanol prior to addition of methyl methacrylate to prevent gel formation. No analogous gelation reaction was observed for t-butyl methacrylate. High grafting efficiencies (95–97%) were obtained when methanol deactivation was employed.

Jenkins and co-workers[63] have recently prepared a backbone from the radical-initiated copolymerization of p-isopropenylstyrene (PIPS) with styrene; gel formation was observed with > 5 mole % PIPS. The resulting copolymers were reacted with butyllithium in stoichiometric amounts relative to the isopropenyl groups followed by addition of methyl methacrylate at -78°C. It was deduced that little reaction between initiating sites and unreacted isopropenyl



YFP

Highlight

YFP

Highlight

YFP

Highlight

YFP

Highlight

YFP

Rectangle

YFP

Line

groups on other backbone chains occurred, based on the fact that the products were always soluble; however, significant branching could occur before gelation is observed. No characterization of the metalated polymer by quenching prior to grafting monomer addition was reported; this type of characterization would have provided information regarding the extent of branching reactions.

C. Coupling Procedures

This “grafting onto” method (Scheme 14.2) involves coupling of an electrophilic functional group on the backbone polymer chain with a preformed polymer chain containing a carbanionic chain end. Since the carbanionic branch chain can be prepared using standard, well-controlled anionic polymerization techniques, this method overcomes one of the common limitations of the metalation procedure: lack of uniform branch length. In addition, the backbone chain can be characterized separately; thus, because coupling procedures are reactions between preformed polymers, they offer the potential of providing well-defined graft copolymers. Backbone functionalities that have been used for “grafting onto” reactions include ester, anhydride, benzylic halides, nitrile, chlorosilane, epoxide, and pyridine groups [16,23]. In general, these electrophilic functionalities are distributed randomly along the polymer backbone chain.

Rather disappointing results were obtained from attempts to graft poly(styryl)lithium onto lightly brominated polymers derived from polyisobutylene (with 1.6% copolymerized isoprene), cis-1,4-polybutadiene, or an ethylene-propylene copolymer (1.2 mol% copolymerized 1,7-octadiene) [64]. Inefficient grafting and gelation were observed, presumably resulting from competing side reactions such as elimination, lithium-halogen exchange, and subsequent Wurtz-type coupling. Similar results were reported for reactions of α,ω-disodium polystyrene with chlorinated ethylene-propylene-terpolymer, poly(chloroprene), chlorinated butyl rubber, poly(vinyl chloride), poly(epichlorohydrin), and epichlorohydrin-ethylene oxide copolymer [65]. The mechanistic complexities of the reactions of organolithium compounds with alkyl halides are well documented in the literature [29,66–70].

The reactions of living anionic polymers with chloromethylated polystyrenes and related compounds (see Scheme 14.8) are among the most studied types of

Scheme 14.8



YFP

Line

YFP

Rectangle

YFP

Underline

YFP

Text Box

亲电性官能团

YFP

Pencil

YFP

Rectangle

YFP

Underline

YFP

Underline

YFP

Underline

YFP

Underline

YFP

Underline

YFP

Underline

YFP

Underline

YFP

Underline

YFP

Underline

YFP

Underline

YFP

Underline

YFP

Underline

YFP

Underline

YFP

Underline

YFP

Underline

YFP

Underline

YFP

Underline

YFP

Underline

YFP

Text Box

氯丁二烯

YFP

Line

YFP

Text Box

环氧氯丙烷,表氯醇,氯甲代氧丙环

YFP

Line

YFP

Line

YFP

Highlight

polymer coupling reactions [71–78]. In general, these grafting reactions with benzylic halides are accompanied by formation of dimers of the grafting anionic chain end as well as terminated chains. For chloromethylated polystyrene, the dimerization (38–45%) and termination (6–9%) are significant for poly(styryl)-lithium at 23°C in benzene, but only 2–4% and 20–23%, respectively, for poly(styryl)potassium at 0°C in THF [72a]. Although the amount of dimerization increased to 67% in the presence of small amounts of THF for PSLi, increasing amounts of THF to 44 vol% increased the grafting efficiency to > 90% [72]. It was noted that side reactions that must be considered include metal-halogen interchange, α-hydrogen abstraction, one-electron transfer, and carbene generation [72a]. Since the dimer formation occurred even for grafting reactions with fluoro-methyl groups [74], it was concluded that dimer formation may be occurring by one-electron transfer rather by lithium-halogen exchange followed by Wurtz coupling. However, the extent of dimer formation increased in the order C1 < Br < I, which is consistent with lithium-halogen exchange also [72]. Quite surprising was that the coupling reaction of poly(styryl)lithium with poly(p-fluoro-methylstyrene) (PFMS) and poly(p-fluoromethylstyrene-co-styrene) in a THF/benzene mixture (50 vol %) proceeded in high yield (> 90% grafting efficiency of polymeric anions) at low extents of grafting ([PSLi]/[CH2F] < 0.5) or for the copolymer with five styrenes per fluoromethylstyrene [74]. It was estimated that because of the hindering effect of branches, the maximum number of graft branches for PFMS is approximately 75% of the available CH2F groups. Similar results were obtained for grafting reactions of poly(styryl) anions with poly(p-vinylstyrene oxide) and the corresponding styrene copolymer [79].

Highly branched (arborescent) graft polystyrenes have been prepared by a repetitive sequence of polystyrene chloromethylation followed by anionic polymer coupling reactions [73]. The efficiency of the coupling reaction was improved significantly by end-capping poly(styryl)lithium with 1,1-diphenylethylene in the presence of THF prior to “titration” of the resulting polymeric diphenylalkyl-lithium with chloromethylated polystyrene at -30°C.

Evidence for formation of graft copolymers from reaction of polystyrylpotassium with poly(1,3-diisopropenyl)benzene via addition to the isopropenyl groups has been presented, although the reaction is quite slow [60]. It is interesting to note that soluble polymers of 1,3-diisopropenylbenzene were formed using 1-phenylethylpotassium in THF at -30°C using conversions < 50% to prevent broadening of the molecular weight distribution and ultimately branching.

An alternative procedure utilized coupling reactions with the dimethyl-chlorosilyloxymethylated polystyrenes [80,81], although significant amounts of dimer formation were still observed [80]. The problems of graft polymer chain coupling have been circumvented by Roovers [82], who has coupled chloromethylated polystyrenes from anionically prepared polystyrene with the anionically synthesized, potassium salt of carboxyl-terminated polystyrenes in the presence of crown ether (dicyclohexyl-18-crown-6) in a mixture of benzene and acetonitrile.



YFP

Highlight

YFP

Line

YFP

Line

YFP

Line

YFP

Line

YFP

Highlight

YFP

Underline

YFP

Underline

YFP

Underline

YFP

Underline

YFP

Underline

YFP

Highlight

YFP

Highlight

YFP

Highlight

YFP

Rectangle

YFP

Text Box

乙腈,氰甲烷,乙腈

YFP

Line

Another method for circumventing the side reactions often encountered in the direct anionic coupling reactions was reported by Cameron and Qureshi [83], which involves functionalization of polybutadiene with chlorosilane groups, followed by coupling with either poly(styryl)lithium or poly (α-methylstyryl)lithium as shown in Scheme 14.9. With the use of an excess of chlorosilane groups relative

Scheme 14.9

to polymer lithium grafting chains, no homopolystyrene was detected in the product by fractional precipitation. These preliminary results suggest that this may be a useful method for preparing graft copolymers with well defined structure.

The simple coupling reaction of the potassium salt of living poly(ethylene oxide) (Mn = 38.5 × 103 g/mol) onto partially chloromethylated polystyrene (Mn = 18.4 × 103 g/mol; anionically prepared) in THF at 50°C is a rather inefficient procedure for the preparation of these amphiphilic graft copolymers [84].

Morton [25] has recently reviewed the literature for anionic synthesis of graft and comb-type polymers with both elastomeric and nonelastomeric backbone chains.

D. Macromonomer Procedures

As discussed in Chapter 11, a macromonomer is an oligomer with a polymerizable end group that can copolymerize with various backbone-forming monomers to form comb-type graft copolymers (“grafting through” method; see Scheme 14.3) with pendent, preformed polymer chains as shown in Scheme 14.10 [85–93].

Scheme 14.10



YFP

Highlight

YFP

Highlight

YFP

Rectangle

YFP

Highlight

YFP

Underline

Macromonomers can be prepared by a variety of polymerization methods. However, living anionic methods provide unique control of average chain length, chain length distribution, and chain-end functionality [21–28]. Once a macromonomer has been prepared, free radical, cationic, Ziegler-Natta (coordination), or anionic polymerization methods can be used to copolymerize the macromonomer with the backbone-forming monomer. Thus, a wide choice of comonomer structures is possible. The macromonomer procedure provides a powerful complement to the metalation and coupling procedures. However, control of backbone chain length and chain length distribution is lost if methods other than living anionic or cationic polymerizations are used to form the comb-type graft polymers.

The instantaneous composition for copolymerization of macromonomer M1 with monomer M2 can be described by the Mayo-Lewis equation (Eq. 14.7) [94]:

d[M2]/d[M1] = (1 + r2[M2]/[M1]/(1 + r1[M1]/[M2]) (14.7)

where r1 and r2 are the respective monomer reactivity ratios (see Chapter 10). In this type of copolymerization the molar feed ratio [M2]/[M1] is usually much greater than 1 because the molar concentration of the macromonomer is so small. Therefore, the instantaneous copolymerization composition equation is reduced to Equation 14.8. Under these conditions, the monomer reactivity ratio for the

d[M2]/d[M1] = r2 [M2]/[M1] (14.8)

macromonomer does not significantly affect the instantaneous copolymer composition. A resulting Bernoullian distribution of grafts along the chain would be expected with little intermolecular compositional heterogeneity, provided that conversion is limited to about 70% [90,95]. However, these conclusions are based on the assumption that the copolymerization is carried out under homogeneous reaction conditions, which may not prevail throughout the course of the copolymerization since phase separation can occur in these systems [88,89]. It is important to note that the monomer reactivity ratios for comonomers (M2), determined from analysis of copolymer compositions from anionically prepared macromonomers (M1), usually agree with literature values [89,96].

Schulz and Milkovich [88,89] have presented a detailed study of the preparation of graft copolymers using the macromonomer approach. A methacrylate-terminated polystyrene was copolymerized with various vinyl monomers, using free radical initiators, to form graft copolymers. The functionalized polystyrenes showed “very high monofunctionality” (> 95%), although the chain end functionality was not independently characterized. Only size exclusion chromatography data were presented to verify this description of the efficiency of this functionalization reactions. No estimate of the amount of non-macromonomer-containing polymer was provided, although it was stated that it is possible to prepare “pure graft copolymer comtaminated with little or no Macromer homopolymer” [88].

In principle, this method produces polymers with a random distribution of grafts with controlled, narrow, molecular weight distributions. However, all of the



YFP

Underline

YFP

Highlight

YFP

Rectangle

YFP

Highlight

vagaries of free radical copolymerization (e.g., compositional heterogeneity among and along chains, chain transfer, and gelation) can complicate the final polymer architecture. A further problem is the fact that phase separation of the components can occur during the copolymerization, and this will obviously result in compositional heterogeneity among the polymer chains formed.

An indication of the actual structural complexity of these graft copolymers is the presence of high gel contents (68–89% by suspension polymerization) [88]. The morphological characteristics of these graft copolymers were not reported. The generation of hazy or opaque products in certain cases was attributed to phase separation during the copolymerization, which generated large amounts of unreacted macromonomer. This problem increased with increasing macromonomer molecular weights [88]. On the positive side, the product of copolymerization of the macromonomer (25 g) with ethyl acrylate (37.5 g) and butyl acrylate (37.5 g) was a transparent elastomer with a tensile strength of 9.72 MPa and 790% ultimate elongation [88].

E. Homopolymerization of Macromonomers

Macromonomers should be polymerizable by a variety of mechanisms, depending on the chain-end functional group. However, simply because of the low concentration of reactive groups per gram of macromonomer, one would expect slow rates of homopolymerization. Thus, for polymerization mechanisms in which chain termination reactions and chain transfer reactions can compete with propagation, these side reactions can limit conversion and molecular weight. The homopolymerization of macromonomers is of considerable interest from the standpoint of structure-property relationships, since a polymacromonomer can be considered as the ultimate branched polymer with one branch (or graft) per repeating unit as shown in Equation 14.9 (P = polymer chain) [93,97,98].

(14.9)

The anionic polymerization of styrene-terminated macromonomers provides a useful system for considering the scope and limitations of homopolymerization of macromonomers, since these should be living polymerizations free of chain termination and chain transfer reactions. The anionic homopolymerization of a macromonomer is analogous to attempts to prepare polymers with molecular weight > 106 (i.e., impurities can be at concentrations comparable to chain ends) [99,100]. The butyllithium-initiated homopolymerization of a p-vinylbenzyl-functionalized polystyrene macromonomer (Mn = 3 × 103 g/mol) in benzene at 40°C provided a polymer with Mw(light-scattering) = 97 × 103 g/mol [97]; the



YFP

Highlight

YFP

Underline

YFP

Underline

YFP

Line

YFP

Underline

YFP

Rectangle

YFP

Rectangle

YFP

Highlight

YFP

Highlight

YFP

Highlight

YFP

Underline

YFP

Highlight

calculated number average molecular weight was 21 × 103 g/mol. Only a small amount of residual macromonomer was evident in the SEC trace of the polymacromonomer that attests to the high functionality of the macromonomer and also to the ability to obtain high-molecular-weight polymacromonomers using alkyllithium-initiated polymerization. Analogous results were obtained for the low-temperature anionic polymerization of a methacrylate-functionalized poly-styrene macromonomer [97]. No homopolymerizations have been investigated for diblock macromonomers that could form thermoplastic elastomers of the type —[(hard phase)-block-(soft phase)]n—.

Anionic Copolymerization of Macromonomers

Although the range of monomers that can be polymerized anionically is limited relative to free radical polymerization, anionic copolymerization of a structurally well-defined macromonomer with another monomer has the advantage of controlled polymerization with respect to the molecular weight and molecular weight distribution of the polymer backbone. Few anionic copolymerizations of macro-monomers have been reported. The poly(styryl)lithium-initiated copolymerizations of p-vinylbenzyl-functionalized polystyrene macromonomers (Mn = 2.7 × 103, 5.6 × 103, and 12.7 × 103 g/mol) with both styrene and p-methylstyrene in benzene at 25°C were investigated [101]. For the macromonomer molecular weights investigated, the monomer reactivity ratios were not dependent on the molecular weight of the macromonomer in accord with the Flory equal reactivity principle [102]. The observed monomer reactivity ratios (r2) were 1.98 for styrene and 0.997 for p-methylstyrene. These results may not be general, however, because of the expected compatibility of the monomers with the macromonomers and the comb-type graft copolymer product. Thus, phase separation during the copolymerization would not be a problem for this system. In other systems of interest for preparation of phase-separated thermoplastic elastomers in which the backbone chain and the graft chain are not miscible, phase separation is often a problem. This can be dependent on molecular weight of the macromonomer and can occur during the copolymerization [4].

The butyllithium-initiated copolymerization of p-vinylbenzyl-functionalized polystyrene macromonomers with butadiene has been examined in cyclohexane at 25°C [103]. Sodium tert-butoxide ([Na]/[Li] = 1) was added to these copolymerization mixtures to promote randomization in the copolymerization (see Chapter 10) [104]. The proposed synthesis of polybutadiene-graft-polystyrene was investigated to explore the effects of the molecular weight of the graft chains and their distribution along the elastomeric backbone on the properties of these heterophase systems. The macromonomer molecular weights varied from Mn = 1.6 × 103 to 38 × 103 g/mol and the predicted polybutadiene backbone molecular weight was designed to be Mn = 250 × 103 g/mol.

The copolymerization data were analyzed using an integrated form of the macromonomer copolymerization equation (Eq. 14.10; cB and cM represent the



YFP

Underline

YFP

Highlight

YFP

Highlight

(14.10)

concentrations of butadiene and macromonomer, respectively) and the results are shown in Table 14.2 in terms of 1/rB, which is equal to kBM/kBB, i.e., the rate of crossover of poly(butadienyl)lithium chain end to macromonomer compare to the rate of addition of butadiene. This has been described as a measure of the reactivity of the macromonomers, since it would be expected that the rate constant for homopolymerization of butadiene would be constant. It is noteworthy that for low conversions and low-molecular-weight macromonomers (< 10,000 g/mol) a “random” copolymerization behavior (rB = 1) for the poly(butadienyl)lithium chain end is observed below conversions of 25–30%. For a given macromonomer, the macromonomer reactivity (i.e., 1/rB) decreases with increasing butadiene conversion. In addition, this measure of macromonomer reactivity decreases with increasing molecular weight of the macromonomer for a given degree of conversion. This is the type of behavior that would be expected for a system that exhibits phase separation during the copolymerization. Based on data for polystyrene—block—polybutadiene—block—polystyrene copolymers and theoretical calculations, a critical minimum molecular weight for phase separation between 5000 and 10,000 g/mol would be expected [105]. To the extent that phase separation occurs during formation of the graft copolymer, the macromonomer would tend to be located in a polystyrene-rich phase and the butadienyllithium chain end would be in a polybutadiene-rich phase. This effect would limit the accessibility of the

Table 14.2 Values of 1/rB as a Function of Conversion for the Copolymerization of p-Vinylbenzylpolystyrenes with Butadienes

1/rB for Different Macromonomer Molecular Weights and Macromonomer Feeds

Conversion of Butadiene (%)

Mn = 5000a

F = 25–33%bMn = 9000a

F = 20–30%bMn = 9000a

F = 40–50%bMn = 30,000a

F = 25–33%b

30 1.02 1.00 0.92 0.44

40 0.98 0.93 0.88 0.43

50 0.91 0.83 0.70 0.42

60 0.84 0.74 0.60 0.41

70 0.81 0.65 0.51 0.38

80 0.76 0.59 0.46 0.37

Conversionc 98.8 97.7 98.3 89.9

aMacromonomer molecular weights

bAmount of macromonomer in feed (wt %).

cMaximum conversion of macromonomer in the anionic copolymerization.

Source: Ref. 103.



YFP

Highlight

YFP

Rectangle

YFP

Underline

YFP

Underline

macromonomer to the polymeric organolithium chain end. Another factor that will tend to promote phase separation with increasing molecular weight of the macromonomer is the quality of the solvent for polystyrene. Since cyclohexane is a theta solvent for polystyrene (Tθ = 34–35°C) [106], this would also tend to limit access of the macromonomer to the chain end with increasing molecular weight of the macromonomer.

Model Graft Copolymer Synthesis

The macromonomer-based copolymerization method for the synthesis of combtype graft copolymers solves some of the classic problems in traditional grafting reactions: (a) the lack of control of the grafted branch molecular weight and molecular weight distribution; and (b) the contamination of the graft copolymer with the homopolymers of the backbone and the grafting monomer [4]. However, this method does not control the number of graft branches per molecule or the distribution of graft branches along the polymer backbone. In principle, living anionic polymerization with a nonhomopolymerizable macromonomer can provide a method of preparing branched and graft polymers with control of all of the

Scheme 14.11



YFP

Underline

YFP

Underline

YFP

Highlight

structural parameters and low degrees of compositional heterogeneity, as illustrated in Scheme 14.11. Thus, living polymerization will form the first (A) backbone segment, 1, which will react with a nonhomopolymerizable macromonomer to only add one macromonomer unit to the chain end to form 2 and maintain the living nature of the polymerization. Addition of more backbone-forming monomer (nM1) will then generate a new (B) backbone segment whose length will be defined by the ratio of the grams of monomer added to moles of active chain end. The living polymer 3 can then react with another equivalent of nonhomopolymerizable macromonomer to place another graft branch at the precise point on the backbone defined by the B segment length. Addition of more backbone-forming monomer (mM1) to the adduct 4 will then generate a new (C) backbone segment whose length will be defined by the ratio of the grams of monomer added to moles of active chain end. At this point the whole sequence of macromonomer and comonomer addition could be repeated to generate more graft branches at specific locations along the backbone, or the living polymer can be terminated to form the precisely defined graft copolymer with two grafted branches located at segment distances A and A + B along the polymer backbone. Although such a model graft copolymer synthesis has not been accomplished, a demonstration of the feasibility of this approach has been performed using a 1,1-diphenylethylene-functionalized polystyrene macromonomer (see Scheme 11.20, Chap. 11) to form a hetero three-armed, star-branched copolymer as shown in Scheme 14.12 [107].

Scheme 14.12



YFP

Underline

For the synthesis of three-armed, heteroarm, star-branched polymers, the first step involves the addition of a polymeric organolithium compound (e.g., poly(styryl)lithium) with the 1,1-diphenylethylene-functionalized polystyrene macromonomer, 7, to form the corresponding coupled product, 8, a diphenylalkyllithium. For stoichiometric amounts of poly(styryl)lithium and macromonomer, it was found that the efficiency of this coupling reaction is > 96%. This result also shows that the vinyl functionality of the macromonomer is > 96%. Finally, the third arm was formed by addition of monomer (e.g., styrene) in the presence of THF to promote the crossover reaction. Size exclusion chromatographic (SEC) analyses of the heteroarm star polymer product showed that each of these steps proceeded efficiently to give the expected products; only relatively small amounts of nonstar product were observed, which corresponds in retention volume to the small amount of unreacted macromonomer and a small amount of polystyrene corresponding to the second arm. A narrow-molecular-weight distribution star product (Mw/Mn = 1.02) was easily obtained by one fractionation step. This methodology can be extended to other backbone-forming monomers, especially polydienes, to form model graft thermoplastic elastomers.

V. Conclusions

A variety of anionic methods are available for the synthesis of graft copolymers with control of the grafted branch structure. Using a well-defined, preformed polydiene backbone, metalation-grafting reactions, in principle, provide procedures for the synthesis of graft copolymers with controllable branch molecular weight and random distribution. Coupling of polymeric anions with a variety of complimentary functional groups on well-defined backbone polymers has been developed into a reliable method for graft copolymer synthesis. The copolymerization of well-defined macromonomers with backbone-forming monomers provides the most versatile method for the synthesis of a wide variety of graft copolymer compositions, since almost all polymerization mechanisms can be used for the backbone-forming reaction and therefore a wide variety of monomers can be copolymerized to form the backbone.

References

1. Basic definitions of terms relating to polymers, Pure Appl. Chem., 40, 479 (1974); Copolymers, Pure Appl. Chem., 57, 1427 (1985).

2. H. A. J. Battaerd and G. W. Tregear, Graft Copolymers, Interscience, New York, 1967.

3. R. J. Ceresa, Ed., Block and Graft Copolymerization, Vol. 1 and 2, Wiley-Interscience, New York, 1973.



4. P. Dreyfuss and R. P. Quirk, in Encyclopedia of Polymer Science and Engineering, J. I. Kroschwitz, Ed., Wiley, New York, Vol. 7, 1987, p. 551.

5. A. Noshay and J. E. McGrath, Block Copolymers: Overview and Critical Survey, Academic Press, New York, 1977.

6. E. Molau, Colloidal and Morphological Behavior of Block and Graft Copolymers, Plenum Press, New York, 1971.

7. D. J. Meier, J. Polym. Sci., Part C, 26, 81 (1969).

8. D. J. Meier, in Block and Graft Copolymers, J. J. Burke and V. Weiss, Eds., Syracuse University Press, Syracuse, NY, 1973, Ch. 6, p. 105.

9. D. J. Meier, Polym. Prepr., Am. Chem. Soc., Div. Polym. Chem., 15(1), 171 (1974).

10. D. J. Meier, Ed., Block Copolymers: Science and Technology, MMI Symposium Series, Harwood Academic Publishers, New York, 1983.

11. E. Helfand, Macromolecules, 8, 552 (1975).

12. E. Helfand and Z. R. Wasserman, Macromolecules, 9, 879 (1976).

13. E. Helfand, Acc. Chem. Res., 8, 295 (1975).

14. D. J. Meier, Polym. Prepr., Am. Chem. Soc., Div. Polym. Chem., 11(2), 400 (1970).

15. C. B. Buchnall, Toughened Plastics, Applied Science Publishers, London, 1977.

16. P. F. Rempp and P.J. Lutz, in Comprehensive Polymer Science, Vol. 6, Polymer Reactions, G. C. Eastmond, A. Ledwith, S. Russo, and P. Sigwalt, Eds., Pergamon Press, Elmsford, NY, 1989, p. 403.

17. R. P. Quirk, Rubber Chem. Technol., 57, 557 (1984).

18. I. Sakurada, Y. Ikada, and F. Horii, Makromol Chem., 139, 171 (1970).

19. H. Inagaki and T. Tanaka, in Developments in Polymer Characterization—3, J. V. Dawkins, Ed., Applied Science Publishers, London, 1982, p. 1.

20. Y. Ikada, Adv. Polym. Sci., 29, 48 (1978).


22. S. Bywater, in Encyclopedia of Polymer Science and Engineering, 2nd ed., J. I. Kroschwitz, Ed., Wiley-Intercience, New York, 1985, Vol. 2, p. 1.

23. S. Bywater, Prog. Polym. Sci., 4, 27 (1974).



26. M. Szwarc and M. Van Beylen, Ionic Polymerization and Living Polymers, Chapman & Hall, New York, 1993.



34. A. F. Halasa, D. N. Schulz, D. P. Tate, and V. D. Mochel, Adv. Organometal. Chem., 18, 55 (1980).

35. A. W. Langer, Ed., Polyamine Chelated Alkali Metal Compounds, Adv. Chem. Ser. 130, Am. Chem. Soc., Washington, D.C., 1974.


37. H. L. Hsieh and O. F. McKinney, Polym. Lett., 4, 843 (1966).

38. D. P. Tate, A. F. Halasa, F. J. Webb, R. W. Koch, and A. E. Oberster, J. Polym. Sci., Part A-1, 9, 139 (1971).

39. A. F. Halasa, G. B. Mitchell, M. Stayer, D. P. Tate, A. E. Oberster, and R. W. Koch, J. Polym. Sci., Polym. Chem. Ed., 14, 497 (1976).

40. J. C. Falk, R. J. Schlott, D. F. Hoeg, and J. F. Pendleton, Rubber Chem. Technol., 46, 1044 (1973).

41. J. C. Falk, J. Van Fleet, D. F. Hoeg, J. F. Pendleton, and R. J. Schlott, Macromolecular Synthesis, 8, 57 (1982).

42. J. C. Falk, D. F. Hoeg, R. J. Schlott, and J. F. Pendleton, J. Macromol. Sci. Chem, A7, 1669 (1973).

43. J. C. Falk and R. J. Schlott, J. Macromol. Sci. Chem, A7, 1663 (1973).

44. N. Hadjichristidis and J. Roovers, J. Polym. Sci., Polym. Phys. Ed., 16, 851 (1978).

45. M. M. F. Al-Jarrah, J. D. H. Al-Kafaji, and R. L. Apikian, Brit. Polym. J., 18, 256 (1986).

46. J. Roovers and P. M. Toporowski, Polym. Commun., 27, 266 (1986).

47. T. Tsuruta and Y. Kawakami, in Comprehensive Polymer Science, Vol. 6, Polymer Reactions, G. C. Eastmond, A. Ledwith, S. Russo, and P. Sigwalt, Eds., Pergamon Press, Elmsford, NY, 1989, p. 457.

48. N. Spassky, Makromol. Chem., Macromol. Symp., 42/43, 15 (1991).

49. D. J. Brunelle, Ed., Ring-Opening Polymerization. Mechanisms, Catalysis, Structure, Utility, Hanser Publishers, New York, 1993.

50. S. W. Caywood, Rubber Chem. Technol., 50, 127 (1977).

51. W. H. Buck, Rubber Chem. Technol., 50, 109 (1977).

52. H. K. Hall, Macromolecules, 2, 488 (1969).

53. R. P. Foss, H. W. Jacobsen, H. N. Cripps, and W. H. Sharkey, Macromolecules, 9, 373 (1976).

54. J. F. Harris, Jr., and W. H. Sharkey, Macromolecules, 10, 503 (1977).

55. R. C. Thamm and W. H. Buck, Polym. Prepr., Am. Chem. Soc., Div. Polym. Chem., 17(1), 205 (1976).

56. S. A. Sundet, R. C. Thamm, J. M. Meyer, W. H. Buck, S. W. Caywood, P. M. Subramanian, and



63. A. D. Jenkins, E. Tsartolia, D. R. M. Walton, J. Horska-Jenkins, P. Kratochvil, and J. Stejskal, Makromol. Chem., 191, 2501 (1990).

64. B. W. Brooks, J. Polym. Sci., Part B, 5, 641 (1967).

65. Y. Minoura, H. Hironaka, T. Kasabo, and Y. Ueno, J. Polym. Sci., Part A-1, 6, 2773 (1968).

66. J. L. Wardell, in The Chemistry of the Metal-Carbon Bond, Vol. 4, F. R. Hartley, Ed., Wiley, New York, 1987, p. 1.

67. G. A. Russell and D. W. Lamson, J. Am. Chem. Soc., 91, 3967 (1969).

68. H. R. Ward, R. G. Lawler, and R. A. Cooper, J. Am. Chem. Soc., 91, 746 (1969).

69. G. C. East and H. A. Ellis, J. Polym. Sci., Polym. Chem. Ed., 19. 869 (1981).

70. A. F. Halasa and H. E. Adams, J. Polym. Sci., Part C., 30, 169 (1970).

71. T. Altares, Jr., D. P. Wyman, V. R. Allen, and K. Meyerson, J. Polym. Sci., Part A, 3, 4131 (1965).

72. (a) M. Takaki, R. Asami, and Y. Kuwata, Macromolecules, 12, 378 (1979); (b) M. Takaki, R. Asami, and M. Ichikawa, Macromolecules, 10, 850 (1977).

73. M. Gauthier and M. Möller, Macromolecules, 24, 4548 (1991).

74. M. Takaki, R. Asami, H. Inukai, and T. Inenaga, Macromolecules, 12, 383 (1979).

75. T. Fujimoto, H. Narukawa, and M. Nagasawa, Macromolecules, 3, 57 (1970).

76. J. Pannel, Polymer, 13, 2 (1972).

77. F. Candau and E. Franta, Makromol. Chem., 149, 41 (1971).

78. C. Price and D. Woods, Polymer, 14, 82 (1973).

79. M. Takaki, R. Asami, and M. Mizuno, Macromolecules, 10, 845 (1977).

80. D. Rahlwes, J. E. L. Roovers, and S. Bywater, Macromolecules, 10, 604 (1977).

81. J. Roovers, Polymer, 20, 843 (1979).

82. J. Roovers, Polymer, 17, 1107 (1976); 20, 843 (1979).

83. G. G. Cameron and M. Y. Qureshi, Makromol. Chem., Rapid Commun., 2, 287 (1981).

84. M. H. George, M. A. Majid, J. A. Barrie, and I. Rezaian, Polymer, 28, 1217 (1987).

85. Y. Yamashita, J. Appl. Polym. Sci., Appl Polym. Symp., 36, 193 (1981).

86. D. J. Walbridge, in Dispersion Polymerization in Organic Media, K. E. J. Barrett, Ed., Wiley, London, 1975, p 45.

87. R. Milkovich, in Anionic Polymerization: Kinetics, Mechanism and Synthesis, J. E. McGrath, Ed., ACS Symp. Ser. 166, American Chemical Society, Washington, D.C.,. 1981, p. 41.

88. G. O. Schulz and R. Milkovich, J. Appl. Polym. Sci., 27, 4773 (1982).



96. K. Ito, Y. Masuda, T. Shintani, T. Kitano, and Y. Yamashita, Polym. J.15,443 (1983).

97. P. Rempp, P. Lutz, P. Masson, P. Chaumont, and E. Franta, Makromol. Chem., Suppl., 13, 47 (1985).

98. P. Rempp and E. Franta, in Recent Advances in Anionic Polymerization, T. E. Hogen-Esch and J. Smid, Eds., Elsevier, New York, 1987, p. 353.

99. D. McIntyre, L. J. Fetters, and E. Slagowski, Science, 176, 1041 (1972).

100. E. L. Slagowski, L. J. Fetters, and D. McIntyre, Macromolecules, 7, 394 (1974).

101. Y. Gnanou and P. Lutz, Makromol. Chem., 190, 577 (1989).

102. P. J. Flory, Principles of Polymer Chemistry, Cornell University Press, Ithaca, NY, 1953, p. 75.

103. M. Arnold, W. Frank, and G. Reinhold, Makromol. Chem., 192, 285 (1991).


105. D. J. Meier, J. Polym. Sci., Part C, Polym. Symp., 26, 81 (1969).

106. H. G. Elias, in Polymer Handbook, 3rd ed., J. Brandrup and E. H. Immergut, Eds., Wiley, New York 1989, p. VII/205.

107. R. P. Quirk and T. Yoo, Polym. Bull., 31, 29 (1993).



V COMMERCIAL APPLICATIONS OF ANIONICALLY PREPARED POLYMERS



15 Commercial Applications of Anionically Polymerized Products*

I. Introduction

Natural rubbers have been used for a long time. As early as 500 B.C., there were indications that the Indians in Mexico played with rubber balls. Spanish explorers in the 15th and 16th centuries wrote of bouncing balls and waterproof shoes from Brazil and Mexico. Today, at least well over 400 different shrubs, vines, and trees have been identified that produce natural rubber latex. Hevea brasilliensis (natural rubber trees) are cultivated in Central America, Mexico, Southeast Asia, Africa, and South America.

Synthetic rubbers are manufactured polymeric materials that have the general properties of natural rubbers, such as high deformability, rapid recovery from deformation, good mechanical strength, and others. The term elastomers was coined to describe this class of polymers: polymers having elastic behaviors regardless of the chemical compositions of the origin of the materials. One can define elastomers or rubbers technically as polymers that can undergo very large, reversible deformations at relatively low stress. In this book, the terms rubber and elastomer are used interchangeably.

Synthetic rubbers can be classified into two general types: general-purpose and specialty. General-purpose synthetic rubbers, such as polybutadienes (BR),

*Text used from Ref. 92, reproduced with permission from American Chemical Society.



YFP

Underline

YFP

Text Box

三叶胶树

YFP

Line

YFP

Highlight

YFP

Highlight

YFP

Underline

YFP

Highlight

YFP

Highlight

YFP

Underline

YFP

Highlight

YFP

Rectangle

YFP

Rectangle

polyisoprenes (IR), and styrene/butadiene copolymers (SBR) are most often used as a direct replacement for natural rubber (NR). In contrast, the specialty rubbers, such as polychloroprenes (CR), nitrile/butadiene copolymers (NBR), polysiloxanes (Q), fluoroelastomers (CFM, FKM, or FFKM), epichlorohydrin elastomers (CO, ECO), polyacrylics (ACM), polyurethane (AU, EU, or U), polysulfide (T, PTR), and others, offer one or more definite outstanding performance attributes, such as heat resistance, low-temperature flexibility, weatherability, fuel resistance, or chemical resistance. Specialty rubbers are generally high priced and are used only in applications where their specific performance attributes are needed. Overall, the total volume is small comparing with the general-purpose rubbers used for tires, shoes, mechanical goods, and other items. However, the specialty rubbers fill an ever-growing demand for components for use in hostile environments.

Somewhere in between the general-purpose and specialty rubbers are elastomeric products such as ethylene/propylene (EP) rubbers, ethylene/propylene/diene (EPDM) rubbers and thermoplastic rubbers (TPR) of styrene-diene-styrene triblock type, copolyester-ether type, and olefinic type. These rubbers are generally priced in between the two classifications.

Anionic polymerization occupies a key position in the industrial production of polydiene rubbers, solution styrene/butadiene rubbers (SBR), thermoplastic elastomers of styrenic type, and other nonrubber products, such as clear impact-resistant polystyrene resins, binders for solid rocket fuel, and others.

For readers who wish to know more about the historical events and technical details of various synthetic rubbers, see the recommended readings at the end of the references section of this chapter.

The polymerization of vinyl monomers with alkali metals was reported as early as 1910–1914 by Mathews and Strange [1], Harris [2,3], and Shlenk [4]. A review of the early work on sodium polymerization of dienes has been given by Taft and Tiger [5]. A recounting of all these early publications and the controversy that surrounded the claims to priority has been presented by Tornquist [6]. Ziegler [7] in 1929 first described the polymerization of butadiene in the presence of n-butyllithium (n-BuLi) and in 1936 in a review article Ziegler [8] proposed mechanisms that describe the polymerizations by metallic lithium and butyllithium. The mechanisms consisted of initiation and propagation steps, and it was recognized that termination and transfer reactions may not contribute significantly. Ziegler reported that they had obtained resinous and rubberlike substances from butadiene using butyllithium in ether solution. The early German workers created methyl rubber in World War I (1914–1918) using 2,3-dimethylbutadiene, sodium metal, and carbon dioxide (a chain modifier). Later some polybutadiene rubbers were produced with sodium or potassium in World War I by the Germans. However, with the exception of Russians who continued to produce polybutadiene rubbers by using sodium or potassium metals in World War II (1939–1945), every other country decided to pursue emulsion processes. Emulsion polymerization



YFP

Text Box

聚氯丁烯,氯丁橡胶

YFP

Line

YFP

Text Box

环氧氯丙烷,表氯醇,氯甲代氧丙环

YFP

Line

YFP

Line

YFP

Text Box

多硫化物,聚硫化物,多硫化合物

YFP

Line

YFP

Line

YFP

Highlight

YFP

Underline

was an easier reaction to control and yielded a greater output of a most useful synthetic rubber styrene-butadiene copolymer [9].

In 1952, Firestone research chemists, under the direction of F. W. Stavely, explored the polymerization of diolefin monomers with lithium metal. In their investigations, Firestone chemists discovered that the polyisoprene produced with the lithium metal in hydrocarbon solution exhibits the structural features of the natural rubber hydrocarbon. The process for the synthesis of an elastomer closely resembling natural rubber (cis-polyisoprene) and methods for the technological use of that elastomer in tire compounds were disclosed by Firestone [10–12]. Hsieh and Tobolsky [13] reported that high cis-polyisoprene could also be prepared with alkyllithium in heptane or benzene. The stereospecificity of isoprene polymerization stimulated and regenerated interest in alkyllithium-initiated polymerizations. Intensive research in this area has been done both in industrial laboratories and in academic institutions. While the stereospecificity created the initial excitement, it is the unprecedented control over polymer properties provided by anionic polymerization that led to the development of several important commercial products. New products and new applications continue to appear in the marketplace based on anionic polymerization. This control is most easily exerted when polymerization is initiated by hydrocarbon soluble organolithium and includes:

Polymer composition (see Chap. 10)

Microstructure (see Chap. 9)

Molecular weight (see Chap. 4)

Molecular weight distribution (see Chap. 4)

Monomer sequence distribution in copolymers (see Chap. 10)

Choice of functional end groups (see Chap. 11)

Branching (see Chap. 13)

In alkyllithium-initiated solution polymerization of dienes, some polymerization conditions affect the configurations more than others. In general, the stereo-chemistry of polybutadiene and polyisoprene responds to the same variables. Thus, solvent profoundly influences the stereochemistry of polydienes when initiated with alkyllithium. In nonpolar solvents, polymerization of isoprene results largely in cis unsaturation (70–90%) polymerization of butadiene gives about equal amounts of cis and trans unsaturation. Aromatic solvents such as toluene tend to increase the 1,2 or 3,4 linkages. Polymers prepared in the presence of active polar compounds such as ethers, tertiary amines, or sulfides show increased 1,2 (or 3,4 in the case of isoprene) and trans unsaturation [14–17]. It appears that the solvent influences the ionic character of the propagating ion pair, which in turn determines the stereochemistry.

Figures 15.1 and 15.2 show the dependence of polymer microstructure on the molecular weight of the polymer and therefore on the original initiator concentration. Polymerization temperature also affects the microstructure, as can be seen in



YFP

Highlight

YFP

Highlight

YFP

Highlight

YFP

Rectangle

YFP

Rectangle

YFP

Rectangle

Figure 15.1 Microstructure of polybutadiene initiated with alkyllithium in

cyclohexane at 50°C. (From Refs. 17, 18 used with permission from John Wiley & Sons.)

Figure 15.2 Microstructure of polyisoprene initiated with alkyllithum

in cyclohexane at 50°C. (From Refs. 17, 18 used with permission from John Wiley & Sons.)



Figure 15.3. The overall heat activation energy leading to 1,2 addition is greater than that leading to 1,4 addition [15,18]. Initiator structure (i.e., organic moiety of the initiator), monomer concentration, and conversation have essentially no effect on polymer microstructure.

II. Polydienes

Firestone and Shell started commercial production of cis-polyisoprene by the anionic process in the 1950s, but these plants are no longer in operation. Firestone technology employed lithium metal, while Shell used alkyllithium initiator. The anionic produced polyisoprene rubber has 92–94% cis and is linear, very high molecular weight, with narrow molecular weight distribution, and is free from traces of transition metals, which could be troublesome in certain applications. The other technology using trialkylaluminum/titanium tetrachloride and initiator produces polyisoprene rubber of 96–98% cis and is virtually indistinguishable from natural rubber by infrared spectroscopy, ensuring more rapid crystallization and hence higher tear strength in vulcanizates, which are close in mechanical properties to those from natural rubber [19–21]. About the same time, Phillips started manufacturing polybutadienes by the anionic route and ever since their use

Figure 15.3 The effect of temperature on microstructure of polybutadiene

prepared in cyclohexane. (From Refs. 17, 18 used with permission from John Wiley & Sons.)



has grown steadily, particularly in the tire-tread and tire-carcass formulations. These solution polybutadienes generally have low vinyl contents, but some interesting applications have been found for medium vinyl polybutadienes as well [22]. Polybutadienes with 50–55% vinyl contents behave like emulsion-polymerized SBR in tire-tread formulations and exhibit similar tread wear, wet skid resistance, low heat buildup, and other characteristics. A three-way blend of 45% vinyl polybutadiene with SBR and cis-polybutadiene (30/35/35) is even better in that respect than a 65/35 blend of SBR and cis-polybutadiene. Medium vinyl polybutadienes are also effective as partial or complete replacement for SBR in a variety of nontire applications.

Polydienes prepared with enough initiator to give a molecular weight of 250,000 will have a desirable combination of vulcanizate properties, but will also have high cold flow, high compounded Mooney viscosity, and poor processability. Increasing the molecular weight does lower the cold flow, but greatly decreases the processability. Lowering the molecular weight improves the processability, but leads to intolerable cold flow problems. These problems can usually be solved by altering the molecular weight distribution and/or introducing branching in the polymer molecule.

There are several ways to broaden the molecular weight distribution. One can pick an alkyllithium that will give a slow rate of initiation compared to the rate of polymerization [23–25]. Generally, this results in a relatively minor broadening, as shown by the gas permeation chromatography curve in Figures 15.4a and b. One can use an initiator of limited solubility so that it slowly dissolves and initiates polymerization throughout the process. However, a more controllable and practical procedure is continuous initiator addition to a batch process [26]. By programming the rate of addition, a variety of molecular weight distributions can be obtained. Broadening the molecular weight distribution by continuous polymerization is a common technique (Fig. 15.5). Here, all reagents are added continuously to the reactor and the product is constantly withdrawn at the same rate. The molecular weight distribution of polymer from such a system is quite broad, with both very low and very high molecular weight species. Processability of these broad molecular weight products is improved by the presence of low-molecular-weight material, but cold flow still can be unacceptably high.

Cold flow can be most effectively decreased and performance of the final product improved by using a variety of techniques to branch the polymer molecule. For example, branching comonomers such as divinylbenzene can be used [27]. The amount needed to branch the polymer adequately is generally so low that the reagent can hardly be detected in the final product. Additives that cause metallation of the polymer chains can be included in the polymerization mixture or added after polymerization to create new growth sites along the polymer chain [28]. As an alternative, polymer can be heat-soaked by increasing the residence time. While these methods result in random branching with very little control over



Figure 15.4 (a) Molecular weight distribution of essentially monodisperse

polymer. (b) Molecular weight distribution of polymer from slow initiation or programmed continuous initiator addition. (From Ref. 92, used with permission from American

Chemical Society.)



Figure 15.5 Molecular weight distribution of polymer from a continuous

polymerization. (From Ref. 92, used with permission from American Chemical Society.)

placement or extent of branching in any particular molecule, the products can be quite useful commercially.

Linear polymers can be branched by employing a polyfunctional coupling agent at the end of polymerization. This technique has been used to prepare model polymers for rheological studies [29,30] of specific types of branching and also commercially for some rubbers. By controlling the functionality of the reagent, one can prepare nearly pure trichain, tetrachain, star- and comb-shaped branched polymers (Fig. 15.6). Use of a polyfunctional initiator and branching comonomer in conjunction with terminal coupling will result in both branching and molecular weight broadening.

Introduction of one or two long-chain branches into a polydiene molecule to form tri- or tetrachain molecules leads to profound changes in rheological behavior. For instance, in polybutadiene, at low molecular weights, the Newtonian viscosity is decreased relative to a linear polymer of same molecular weight [31] (Fig. 15.7). At molecular weights exceeding 60,000 (trichain) or 100,000 (tetrachain), the Newtonian viscosity rises rapidly above the corresponding value for a linear polybutadiene. Figure 15.8 shows the relationship between viscosity (η) and molecular weight (mol wt) at higher shear rate (s = 20 s-1). At this shear rate, the viscosity of the branched polymers is uniformly lower than that of the linear samples of identical molecular weight. In other words, the non-Newtonian behavior of branched polymers becomes rapidly more pronounced at higher molecular weight. Long-chain branched polymer has higher resistance to flow at low shear rate (i.e., low cold flow) and has more flow at moderate and high shear rate (i.e., better processing) than the corresponding linear polymer.

Polydiene rubbers are comprehensively reviewed in Chapter 16.



Figure 15.6 Molecular weight distribution of polymer terminally

coupled with polyhalide coupling agents. (From Ref. 92, used with permission from American Chemical Society.)

III. Diene-Styrene Copolymers

In the copolymerization of butadiene or isoprene and styrene, the reactivity ratios are influenced by the solvent [32–34]. Figure 15.9 shows typical conversion curves of a 75/25 butadiene-styrene copolymerization in various hydrocarbon solvents. Note that the solvents change only the overall rates while the general shapes of the curves remain similar. While analyzing samples at various conversions (Fig. 15.10), one can see that the styrene contents are initially lower than in the monomer charge, that they gradually increase until the inflection points of the conversion curves of Figure 15.9 are attained, and that they increase rapidly thereafter. Furthermore, in the analysis of the samples by oxidative degradation [35], polystyrene segments are recovered only after the inflection points are reached (Fig. 15.10). Since the final polymers are essentially homogeneous in composition and in molecular weight, it follows that the process has resulted in a



Figure 15.7 Dependence of Newtonian viscosity on

molecular weight. (From Ref. 31, used with permission from John

Wiley & Sons.)

Figure 15.8 Viscosity vs. molecular weight at shear

rate 20 s-1. (From Ref. 31, used with permission from John

Wiley & Sons.)



Figure 15.9 Polymerization of butadiene-styrene in different solvents at 50°C.

(From Refs. 24,52, used with permission from Rubber Div., A.C.S. and Syracuse University Press.)

Figure 15.10 Copolymerization of styrene from butadiene-styrene (75/25) at

50°C. (From Refs. 24,52, used with permission from Rubber Div., A.C.S. and Syracuse University Press.)



tapered (or graded) block copolymer, one segment being a butadiene-styrene copolymer and the other a polystyrene block with an overall B/S-S structure. These results seem to contrast with what is generally known about anionic homopolymerization of styrene: styrene homopolymerizes faster than butadiene and yet, in a mixture, butadiene polymerizes faster. However, if one examines the cross-propagation rates in hydrocarbon solvents [36–40], the reason for this behavior is evident (Equations 15.1–15.5):

(15.1)

(15.2)

(15.3)

(15.4)

(15.5)

where rb is about 50 times larger than rs. The inversion phenomenon, as one would expect from a kinetic point of view, is independent of polymerization temperature and is shown in Figure 15.11.

Figure 15.11 Styrene incorporation from butadiene-styrene (75/25) in

cyclohexane solutions at 30–121°C. (From Refs. 24,52, used with permission from Rubber Div., A.C.S. and Syracuse University Press.)



Only a limited number of monomer pairs form block copolymers in this manner. Examples are conjugated dienes and vinyl aromatics that have similar reactivity-polarity (Q-e) values. The nature of the anionic initiator (i.e., the ionic character of the carbon-metal bond) plays an important role in both the amount and sequence of block formation. For instance, when potassium or cesium initiators are used, styrene polymerizes first (Fig. 15.12).

A tapered block copolymer containing 75% butadiene and 25% styrene, marketed as Solprene 1205, was the first solution copolymer produced commercially by Phillips in 1962. This polymer has outstanding extrusion characteristics, low water absorption, low ash, and good electrical properties. As a result, it is useful in such diverse applications as wire and cable coverings, shoe soles, and floor tiles.

Zelinski at Phillips also discovered [33] that the four rate constants discussed above can be altered by adding small amounts of ether or tertiary amine, which results in reduction or elimination of the block formation. Figures 15.13 and 15.14 illustrate the effect of diethyl ether on the rate of copolymerization and on the incorporation of styrene in the copolymer. Indeed, we can prepare random copolymers of butadiene and styrene or isoprene and styrene by using alkyllithium as initiator in the presence of small amounts of ether or tertiary amine. Use of these

Figure 15.12 Styrene incorporation. (From Refs. 24,52, used with permission

from Rubber Div., A.C.S. and Syracuse University Press.)



Figure 15.13 Effect of amount of diethyl ether on rate of

copolymerization at 50°C.

Figure 15.14 Effect of amount of diethyl ether styrene incorporation

at 50°C.



polar randomizers also increases the vinyl unsaturation in the copolymer. Butadiene—styrene random copolymers can also be prepared by slow and continuous addition of monomers [41] or by incremental addition of butadiene to a styrene-rich monomer mixture during polymerization. These two methods, unlike the ether or tertiary amine-complexing systems, would produce copolymers of low vinyl unsaturation (˜<10%). One can also produce constant composition copolymers by continuous polymerization.

There is yet another general method to prepare random copolymer. As stated earlier, when one uses potassium, rubidium, or cesium initiator, styrene polymerizes first and gives an S/B-B type of tapered block polymer. When one mixes an alkyllithium with a potassium compound such as potassium tert-butoxide, we obtain quite a different system [42–44].

Alkyllithium compounds as well as polymer—lithiums associate not only with themselves but also with other alkalimetal alkyls and alkoxides. In a polymerization initiated with combinations of alkyllithiums and alkalimetal alkoxides, dynamic tautomeric equilibria between carbon—metal bonds and oxygen—metal bonds exist and lead to propagation centers having characteristics of both metals, usually somewhere in between. In this way, one can prepare copolymers of various randomness and various vinyl unsaturation. This reaction is quite general, since one can also use sodium, rubidium, or cesium compounds to get different effects.

The solution random copolymer prepared with an ether as randomizer generally contains about 32% cis, 41% trans, and 27% vinyl unsaturation compared to 8% cis, 74% trans, and 18% vinyl unsaturation in emulsion copolymer of the same monomer composition. The principal effect of slightly higher vinyl unsaturation in solution copolymer is a small increase in the glass transition temperature (-58°C vs. -62°C for the emulsion copolymer). However, both solution and emulsion-polymerized copolymers exhibit satisfactory low-temperature performance for general uses.

Solution copolymers have inherently very narrow molecular weight distributions, while those of emulsion copolymers are broad (Fig. 15.15). The low-molecular-weight component in the emulsion copolymer helps it to process better, but has an undesirable influence on hysteresis. The solution copolymer, on the other hand, is characterized by a sharp peak with only a small amount of higher molecular weight fraction. This feature leads to improved hysteresis properties and better abrasion resistance, but also results in more difficult processing. Solution copolymers, like polydienes, also exhibit cold flow and, as in the case of dienes, the solution to these two problems is found by altering the molecular weight distribution and/or branching of the polymer molecule. Phillips developed a series of multifunctional initiators that are ether-free, hydrocarbon-soluble, multiple-lithium initiators to produce both branching and molecular weight broadening in many of our random copolymers, both low and medium vinyl. These initiators are based upon the reaction of an alkyllithium, such as sec-butyllithium, and a multifunctional vinyl compound, such as divinylbenzene [45], diiso-



Figure 15.15 Molecular weight distribution of anionically polymerized

styrene-butadiene random copolymer and emulsion polymerized SBR.

propenylbenzene [46], trivinylphosphine [47,48], or tetravinylsilane [47,48]. Functionality of these initiators can be varied to provide more or less branching and/or molecular weight broadening.

Solution random copolymers prepared by the above procedures have performed well in tire-tread formulations [49]. They require about 20% less accelerator than an emulsion SBR and give higher compounded Mooney, lower heat buildup, increased resilience, and better retread abrasion index.

In view of the “living” nature oforganolithium polymerization, one can also synthesize block polymers in which the sequence and length of the blocks are controlled by incremental (or sequential) addition of monomers [50–55]. This general method of preparing block polymers is readily adaptable to commercial production; indeed, a number of block copolymers are manufactured this way. For example, the diene—styrene two-phase systems with two distinctly different glass transition temperatures have three or more blocks and are characterized by high raw strength, complete solubility in common organic solvents, and thermoplasticity. They are referred to as “thermoplastic elastomers” because they behave as vulcanized elastomers at room temperature and yet can be processed as thermoplastics at elevated temperatures. The simplest form of this diene-based elastomer is the linear triblock polymer SBS or SIS, where S represents polystyrene; B, polybutadiene; and I, polyisoprene. Other variations are known as radial block polymers and have a controlled number of branches of equal length and composition, (SB)n-X, where X is a coupler residue and n can be 3, 4, 5, or higher (for details of radial block polymers and their preparation, see [56]). Production of



elastomeric radial block polymers began in October, 1967, and, at present, several on the market belong to the radial block polymer family.

Elastomeric block polymers of styrene and butadiene or isoprene and their products of hydrogenation are finding increasing use. Linear and radial block polymers are used extensively in injection-molded rubber goods, footwear, pressure-sensitive and hot-melt adhesives, and in mechanical rubber goods, such as hose, tubing, cove base, toys, drug sundries, rubber bands, stoppers, and erasers.

A more detailed discussion on styrene-diene random and block copolymers can be found in Chapter 17. Chapter 18 is devoted to the styrenic thermoplastic elastomers.

The outstanding feature of diene—styrene block polymers in pressure-sensitive adhesive formulations is their excellent resistance to adhesive creep. These adhesives are said to have established new performance highs for shear holding when applied to selected metal and plastic surfaces, while maintaining a high degree of tack [57]. At equal molecular weights, radial polymers exhibit lower melt and solution viscosities than linear polymers [58,59]. This is important because it allows the use of radial polymers of higher molecular weights with a corresponding improvement in shear resistance. Isoprene—styrene block polymers have an advantage over butadiene—styrene polymers since, in an oxidative atmosphere, polyisoprene degrades by chain scission rather than gelling by cross-linking so that they exhibit better tack retention. Adhesive applications are discussed in Chapter 19.

Styrene—butadiene block polymers also find applications in blends with polystyrene and ABS plastics. When minor amounts of rubbery SBS or (SB)n-X block polymer are blended with high-molecular-weight polystyrene such that the polystyrene forms the continuum, impact resistance improves with increasing styrene block length. For good results, styrene block lengths should be of the order of 20,000 as shown by Childers et al. [60], who blended butadiene—styrene diblock polymers (25% styrene) with polystyrene (1:3) and cured the blend with 0.1% dicumyl peroxide. When triblock or branched multiblock polymers are used in blends with polystyrene, it is not necessary to cross-link the rubbery domains [61,62], although peroxide treatment does produce some additional improvement in impact and tensile strength. Another use of butadiene—styrene block co-polymers is in blends with ABS. When ABS scrap is reprocessed, the resulting loss in toughness may be restored by the addition of small amounts (˜5%) of a block polymer [62]. It is interesting that SB diblocks are as effective as SBS or (SB)n-X multiblock polymers in this application. It must be remembered that the butadiene—styrene diblock polymers are also used as the source of elastomeric components in the polymerization of certain types of ABS resins. Although little information is available on this application, it is conceivable that diblock polymers are sometimes preferred over polybutadienes because of the advantages in the processes and/or properties of the final products.



Blends of butadiene—styrene block polymers with polyolefins (particularly polypropylene) improve the impact strength of the polyolefins [63]. Similar improvements can be realized from the use of polyolefin block polymers, but the blends have not gained much recognition. However, butadiene—styrene radial triblock polymers are blended into polyethylene film to increase the tear resistance and tensile impact [64].

Butadiene—styrene block polymers with functional end groups, such as carboxy terminated polymers, have been developed [65] as impact modifiers in sheet molding compounds based on fiber-reinforced unsaturated polyester and styrene. On curing, these blends show a good balance of mechanical properties, improved impact strength, low shrinkage, no sink, excellent surface, and improved pigmentability and paintability. Thermoplastic modifications with styrenic thermoplastic elastomers are discussed in Chapter 19.

A relatively new development is the modification of asphalt by butadiene—styrene block polymers [66]. The block polymers help reduce the low-temperature brittleness and impart resistance to flow at elevated temperatures. One can foresee applications in mastics, automobile body undercoatings, and waterproofing materials such as high-quality roofing membranes. More information on asphalt modification can be found in Chapter 19.

IV. Telechelic Polymers

Anionic polymerization has also been used to make telechelic polymers (formed from the Greek telos, for end, and chele, for claw), that is, polymers with reactive terminal groups [67,68]. Uraneck at Phillips coined the term telechelic in 1957 and it has been accepted in technical and patent literature. Liquid carboxy- and hydroxy telechelic polybutadienes initiated with difunctional organolithium initiators have been produced commercially since 1962. Some of the physical properties [69–71], production details [72], and uses of these polymers (e.g., in solid rockets) have been described [73]. Solid, telechelic elastomers, the polymer chain ends of which incorporate into the vulcanizate network and thus obtain superior physical properties, have been developed [68]. Mercapto-, hydroxy-, and aziridinyl telechelic elastomers, when cured with peroxide or sulfur accelerator recipes, exhibited improved stress-strain and dynamic properties in comparison of those of the controls. Mercapto- and aziridinyl telechelic butadiene—styrene copolymers also show outstanding properties in tread formulations. However, these polymers are not produced commercially. Functionalization is discussed in Chapter 11 and telechelic elastomers and prepolymers are discussed in Chapter 22.

V. Viscosity Index-Improving Polymers

Selectively hydrogenated random and block copolymers of vinyl aromatic monomers and dienes are used as viscosity index (VI) improvers in multigrade lubricat



ing oils [74–77]. The block copolymers are of the SB type, where S is most commonly styrene and B is butadiene or isoprene. Hydrogenation of the diene component imparts resistance to oxidative degradation. Crystallinity is undesirable in these applications and is prevented by control of microstructure and a random incorporation of styrene. Since the viscosity of a heavy lubricating oil stock decreases with increasing temperature more rapidly than that of a lighter stock, such a VI-improving polymer is added to the light oil to match the viscosity of a heavier oil at high temperature. The modified oil, although thickened, must still be more fluid than the heavy oil at low temperatures. In other words, the desired property of a VI improver is a relative viscosity that remains constant or decreases as the temperature falls.

An outstanding property of these polymers is their shear stability. Sonic shear stability tests [78] indicate that these polymers are superior to some of the currently used polymers of ethylene—propylene or methacrylate. The excellent stability of hydrogenated diene—styrene polymers is attributed to their relatively low molecular weight and narrow distribution, which is consistent with the established theory of shear degradation of polymers [79]. The most recent developments in this field are block polymer VI improvers with dispersant properties built into the molecule by chemical modification of the rubbery block [80,81]. Readers can find a more detailed discussion of these hydrogenated polymers as VI improver in Chapter 17.

VI. Styrene-Butadiene Resin

Styrene-butadiene block copolymers containing over 60% styrene constitute a family of transparent resinous thermoplastics with good impact strength [82]. The lower diene content makes these copolymers nonelastomeric. These resins are generally crystal clear, have moderate to good toughness, good impact strength, excellent to good rigidity, excellent printability and gloss, and excellent organoleptic properties. These resins supplied in free-flowing pellet form can be injection molded, blow molded, or thermoformed. Ease of processing allows for a variety of end uses including packaging, medical devices, toys, bottles, office articles, disposable cups, and films. The synthesis, properties, and applications of these clear resins are described in Chapter 20.

VII. Graft Copolymers and Macromers

Polybutadiene rubber made with alkyllithium initiation is commonly used commercially to produce impact polystyrene. Impact polystyrene is a two-phase system consisting of a rigid matrix of styrene homopolymer in which soft rubber particles are imbedded. Improved toughness is the result of stretching of the matrix and compression of the rubber (see Chap. 16). A medium cis-polybutadiene consisting of about 35% cis, 52% trans, and 12% vinyl configuration is



most widely used. The vinyl configuration (1,2-addition) is more prone to cross-linking and grafting than the internal double bonds formed by 1,4-addition.

Graft copolymers can also be produced via anionic reactions (see Chap. 14). One approach can be described as “graft from” (Fig. 15.16). Halasa [83,84] prepared anionic graft copolymers by polylithiating polymer backbones with n-BuLi/N,N,N',N'-teramethylethylene diamine (TMEDA) complex. Complexation with TMEDA dramatically increased the reactivity of n-BuLi [85]. Graft copolymers of EPDM by the reaction of (dialkylamino) alkylmethacrylates with polylithiated (n-BuLi/TMEDA) polymer was reported [86]. Similar techniques can also be used to prepare graft copolymers by using n-alkyllithium/potassium or sodium alkoxide (Superbases). The Superbases arepowerful metallation agents [87,88]. Another approach can be described as a “graft on” (Fig. 15.17). In this approach one reacts the “living” polymers with a backbone polymer containing functional groups along the chain leading to the addition reactions. As far as we know, no commercial products are being produced by these two approaches.

Macromer [89] is a trademark by CPC International of a family of macromonomers (a macromolecular monomer). The macromonomers are synthesized via anionic polymerization and therefore their molecular weights can be easily controlled (generally in 5000–30,000 g/mole) and have narrow molecular weight distributions. The typical polymeric portions of the macromonomer that have been investigated are polystyrene, polybutradiene, and blocks of the two [89]. A wide variety of functional groups that are useful for various polymerization mechanisms have been synthesized [89]. The macromonomer with a functional group on one chain end can be copolymerized with the appropriate comonomers such as

Figure 15.16 Graft copolymers “graft from.”



Figure 15.17 Graft copolymers “graft onto.”

acrylates, vinyl chloride, styrenes, ethylene, ethylene/propylene, acrylonitrile, N,N-dimethylacrylamide, etc. The copolymerizations that lead to graft copolymers with controlled polymer structures can be carried out with a variety of polymerization mechanisms (i.e., free-radical, ionic, condensation, or coordination systems). Figure 15.18 shows the simple scheme of formation of graft

Figure 15.18 Copolymers from monomer A and “macromer.”



copolymers via macromonomers. More detailed discussions can be found in two excellent review articles[90,91] and in Chapter 11.

VIII. Conclusions

In conclusion, anionic polymerization has emerged from laboratory curiosity to an important industrial process. Millions of tons of elastomers, thermoplastic elastomers, thermoplastic resins, specialty polymers such as viscosity index improvers for lubricants, solid fuel rocket binder, and others are manufactured in plants around the world. Dedicated and skillful industrial scientists, engineers, and technologists, often working behind the public scene, should be congratulated for their part in harnessing this technology for commercial applications.

References

1. F. E. Mathews and E. H. Strange, Br. patent 24,790 (1910).

2. C. H. Harris, US patent 1,058,056 (1913).

3. C. H. Harris, Ann, 383, 157 (1911).

4. W. Schlenk, W. J. Appenrodit, A. Michael, and A. Thal, Ber., 47, 473 (1914).

5. W. K. Taft and G. J. Tiger, in Synthetic Rubber, G. S. Whitby, Ed., Wiley, New York, 1954, Ch. 21, pp. 734–747.

6. E. G. M. Tornquist, in Polymer Chemistry of Synthetic Elastomers, J. P. Kennedy and E. G. M. Tornquist, Eds., Part 1, Wiley-Interscience, New York, 1968, pp. 46–50.

7. K. Ziegler, Ann, 473, 1 (1929).

8. K. Ziegler, Angew Chem., 49, 499 (1936).

9. G. S. Whitby, Ed., Synthetic Rubber, Wiley, New York, 1954.

10. Private communication to the Office of Synthetic Rubber of the Federal Facilities Corporation; Chem. Eng. News, 33, 3553 (Aug. 29, 1955).

11. F. W. Stavely and co-workers, Ind. Eng. Chem., 48, 778 (1956). Presented before the Div. Rubber Chem., ACS, Philadelphia, PA, November, 1955.

12. F. C. Foster and J. L. Binder, Advances in Chemistry Series No. 19, American Chemical Society, 1957, p. 26.

13. H. Hsieh and A. V. Tobolsky, J. Polym. Sci., 25, 245 (1957).

14. C. E. Rogers and A. V. Tobolsky, J. Polym. Sci., 40, 73, 1959; Rubber Chem. & Tech., 33, 655 (1960).

15. R. S. Stearns and J. E. Forman, J. Polym. Sci., 41, 381 (1959).

16. I. Kuntz and A. Gerber, J. Polym. Sci., 42, 299 (1960).

17. H. L. Hsieh, Rubber Plastics Age, 46, 394 (1965); Rubber Chem. & Tech., 39(3), 491 (1966).




22. H. E. Railsback and N. A. Stumpe, Rubber Age, 107(12), 27 (1965).


24. H. L. Hsieh and W. H. Glaze, Rubber Chem. & Tech., 43, No. 1, 22–73 (1970).

25. H. L. Hsieh and O. F. McKinney, Polym. Letters, 4, 843 (1966).

26. R. C. Farrar, IUPAC Meeting, Rio de Janeiro, Brazil, 1974.

27. R. P. Zelinski and H. L. Hsieh, US patent 3,280,084.

28. A. F. Halasa, ACS Polymer Preprints, 13(2), 678 (1972).

29. R. P. Zelinski and C. F. Wofford, J. Polym. Sci., A3, 93 (1975).

30. C. A. Uraneck and J. N. Short, Rubber Chem. & Tech., 41, 1375 (1968).

31. G. Kraus and J. T. Gruver, J. Polym. Sci., A3, 105 (1965).

32. Phillips Petroleum Company, Br. patent 895,980.

33. R. P. Zelinski, US patent 2,975,160.

34. R. N. Cooper, US patent 3,030,346.

35. I. M. Kolthoff, T. S. Lee, and C. W. Carr, J. Polym. Sci., 61, 25 (1962).

36. M. Morton and F. R. Ells, J. Polym. Sci., 61, 25 (1962).

37. J. Smid and M. Szwarc, J. Polym. Sci., 61, 31 (1962).

38. A. F. Johnson and D. J. Worsford, J. Makromol. Chem., 85, 273 (1965).

39. I. Kuntz, J. Polym. Sci., 54, 569 (1961).

40. A. A. Korotkov and N. N. Chesnokova, Polym. Sci., USSR, 2, 284 (1960).

41. J. N. Short, US patent 3,094,512.

42. H. L. Hsieh and C. F. Wofford, J. Polym. Sci., A1(7), 461 (1969).

43. H. L. Hsieh, J. Polym. Sci., +iA1(8), 533 (1970).

44. C. F. Wofford, US patent 3,294,768.

45. R. C. Farrar, US patent 3,652,516.




49. R. S. Hanmer and H. E. Railsback, Paper presented before Division of Rubber Chem., ACS, Detroit, May 1964.



60. C. W. Childers, et al., in Colloidal and Morphological Behavior of Block and Graft Copolymers, G. Molau, Ed., Plenum Press, New York, 1971, pp. 193–207.

61. R. R. Durst, et al., ACS Org. Coatings Plast. Division Preprint, 34(2), 320 (1974).

62. R. H. Burr, Phillips Petroleum Company, private communication.

63. A. L. Bull and G. Holden, Rubber Chem. & Tech., 49, 1351 (1976).

64. R. H. Burr, Phillips Petroleum Company, private communication.

65. A. South, Paper presented before Soc. of Plastic Engs. Meeting, Detroit, MI, Nov. 6–8, 1979.

66. G. Kraus and D. S. Hall, Paper presented before Midland Macromolecular Inst., Midland, MI, August 20–24, 1979.

67. C. A. Uraneck, H. L. Hsieh, and O. G. Buck, J. Polym. Sci., 46, 535 (1960).

68. C. A. Uraneck, H. L. Hsieh, and R. J. Sonnenfeld, J. Appl. Polym. Sci., 13, 149 (1969).

69. W. W. Crouch, Rubber Plast. Age, 42, 276 (1961).

70. D. M. French, et al., ACS Meeting Polymer Prepr., 7, 447 (1966).

71. R. M. Screaton and R. W. Seemann, ACS Meeting Polymer Prepr., 8, 1379 (1969).

72. C. A. Wentz and E. E. Hopper, Ind. Eng. Chem. Res. Dev., 6, 209 (1967).

73. D. C. Sayles, Rubber World, 153(2), 89 (1965).

74. S. Schiff, M. M. Johnson, and W. L. Streets, US patent 3,554,911.

75. D. J. St. Clair and D. D. Evans, US patent 3,772,196.

76. W. S. Anderson, US patent 3,763,044.

77. R. J. Eckert and J. Heemskerk, US patent 3,752,767.

78. T. W. Johnson and M. T. O'Shaughnessy, in The Relationship Between Engine Oil Viscosity and Engine Performance, Publication SAE-SP-Y16, R. M. Stewart and T. W. Selby, Eds., Soc. Auto Engrs., Warrendale, PA, 1977, pp. 57–69.

79. F. J. Bueche, J. Appl. Polym. Sci., 4, 101 (1960).

80. W. J. Trepka, US patent 4,145,298.

81. T. W. Kiovski, US patent 4,033,388.

82. L. M. Fodor, A. G. Kitchen, and C. C. Biard, in New Industrial Polymers, R. F. Gould, Ed., ACS Symposium Series 4, ACS, Washington, D.C., 1974, Chapter 4, pp. 37–48.

83. A. F. Halasa, in Polyamine-Chelated Alkali Metal Compounds, A. W. Langer, Ed., ACS, Washington, D.C., Chapter 8.

84. A. F. Halasa and D. P. Tate, US patent 3,976,682.

85. A. W. Langer, Polyamine-Chelated Alkali Metal Compounds, ACS, Washington, D.C., Adv.



92. H. L. Hsieh, presented at the Symposium on Anionic Polymerization, Div. of Polym. Chem., ACS at the 179th Nat. Meeting, March 24–28, 1980; H. L. Hsieh, R. C. Farrar, and K. Vdipi, in Anionic Polymerization: Kinetics, Mechanisms, and Synthesis, J. E. McGrath, Ed., ACS Symposium Series 166, ACS, Washington, D.C., 1981, Chapter 25, pp. 353–366; CHEMTECH, Oct. 1981, ACS, Washington, D. C.

Further Reading: Books, Monographs, and Reviews on Synthetic Rubbers

G. S. Whitby, C. C. Davis, and R. F. Dunbrook, Synthetic Rubber, John Wiley & Sons, Inc., New York, and Chapman Hall, London, 1954.

J. P. Kennedy and E. G. Tornquist, Polymer Chemistry of Synthetic Elastomers, Part I and Part II, Interscience Publishers, New York, 1969.

W. M. Saltman, The Stereo Rubbers, John Wiley & Sons, New York, 1977.

N. R. Legge, G. Holden, and H. E. Schroeder, Thermoplastic Elastomers, A Comprehensive Review, Hanser Publishers, Munich/Vienna/New York, 1987.

L. R. Treloar, The Physics of Rubber Elasticity, Oxford University Press, 1975.

F. R. Eirich, Science and Technology of Rubber, Academic Press, New York, 1978.

J. Lal and J. E. Mark, Advances in Elastomers and Rubber Elasticity, Plenum Press, New York, 1986.



16 Polydiene Rubbers

I. Introduction

Polybutadiene rubbers are generally classified by their microstructure (Table 16.1).

High-cis polybutadiene

Medium-cis polybutadiene

Medium-vinyl polybutadiene

High-vinyl polybutadiene

High-trans polybutadiene is nonelastomeric and can be made by using a coordination catalyst of Ziegler-Natta type. Catalysts with titanium, vanadium, chromium, rhodium, iridium, and nickel have been reported [1]. High-transpolybutadiene resembles balata and gutta-percha [2]. Detailed discussions of the effects of cis/trans/vinyl ratios on the physical properties of polybutadiene can be found in the two excellent articles by Kraus and others [3,4]. Emulsion process, which was the process of choice to produce styrene-butadiene copolymer rubber (SBR) until the development of solution process based on anionic polymerization, can also be utilized to produce polybutadiene rubber. However, due to the lack of control over the polymer properties relative to the anionic process, emulsion process is not favored for the production of polybutadiene rubbers (BR).



Table 16.1 Typical Microstructure of Polybutadiene Group

Structure (%)

Name cis (1,4-) trans (1,4-) Vinyl (1,2-)

High-cis 93–98 1–3 1–4

Medium-cis 42 48 10

Medium-vinyl 27 31 42

High-vinyl 15–18 10–12 70–75

Very high-vinyl 1 1 98

cis-BR cannot be produced by anionic initiator. However, in view of its commercial importance and its parallel history of developments, a brief description is included. A more detailed discussion can be found in Saltman's The Stereo Rubbers [1].

II. cis-Polybutadiene Rubbers

A. Process Overview

Development of solution processes for manufacturing high (>90%) cis-polybutadiene during the last half of the 1950s paralleled that for the chemically related solution SBR and cis-polyisoprene processes. Operation of the first commercial facility utilizing this organometallic initiator technology started in 1960. Unlike the multifaceted array of products that result from the solution SBR process, the relatively small group of cis-polybutadiene products are largely used in automotive and truck tire applications.

The high cis-polybutadiene processes utilize initiator systems that are very sensitive to various impurities (water, oxygen, sulfur, nitrogen, and/or acetylenic compounds) and require careful selection of a hydrocarbon solvent. The polymerization is well suited to continuous operation although batchwise polymerization is also satisfactory. The process flow diagram in Figure 16.1 shows a typical arrangement of continuous processing steps used to manufacture high cis-polybutadiene for tire rubber applications.

B. Monomer and Chemical Preparation

Polymerization-grade butadiene is flashed to remove inhibitor and treated over a fixed absorbent bed. Fresh, polymerization-quality solvent (e.g., n-hexane, cyclohexane, benzene, toluene, or selected mixtures) is joined with recycle solvent from the solvent removal step and fractionated to remove water, light impurities, and



Figure 16.1 Solution cis-polybutadiene process.



heavy impurities. The dry, deoiled solvent may also be absorbent-treated before it is fed to the polymerization step.

The several initiator constituents, together with shortstop and antioxidant chemicals, are diluted or dissolved in dry solvent and stored under nitrogen pressure maintenance to exclude atmospheric oxygen and moisture. The initiator constituents, which typically include metal alkyls, metal alkyl halides, and/or metal halides (Tables 16.2–16.4) deserve special storage and handling considerations to accommodate their pyrophoric or adverse reactivity characteristics.

Table 16.2 cis-Polybutadiene Rubbers

Stucture (%)

Catalyst Type cis (1,4-) trans (1,4-) Vinyl (1,2-)

Ti 94 2 4

Co, Ni 97 2 1

U, Nd 99 <1 <0.5

Table 16.3 Typical Catalyst System for the Production of cis-Polybutadiene

Titaniuma Nickelb

TiI4 + R3Al Ni(COOR)2 + BF3 + R3Al

TiCl4 + AlI3 + R3Al Ni(COOR) 2 + TiCl4 + R2AlF

TiCl4 + I2 + R3Al Uraniumb

TiCl4 + R2AlI U (Alkyl) 3X + RAlX2

Cobalta U (OR) 4 + RAlX2 + R3Al

CoCl2 pyridine + R2AlCl Neodyniumb

Co(acac) 2 + R2AlCl Nd (OR) 3 + RAlX2 + R3Al

Nd (COOR) 3 + RAlX2 + R3Al

Table 16.4 Nature of Halogen Ligand vs. cis-Contenta in Polybutadiene

Halogen

Catalyst Metal F Cl Br I

Ti (5) 35 75 87 93

Co (5) 93 98 91 50

Ni (5) 98 85 80 10



C. Polymerization

Purified solvent, recycle solvent containing unreacted butadiene from the flash step, and purified butadiene are individually metered, then mixed and precooled to reaction temperature. The ratio of solvent to butadiene monomer in this continuous charge stream may be controlled within the range of 8–12. Carefully controlled rates of the two or more initiator constituents are added to the charge stream as it enters the first in a series of four to eight stirred polymerization reactors. Reactor cooling coils are used to remove the heat of polymerization and control temperature. The targeted control temperature (broadly within the range of 10–65°C) will depend on the initiator system and desired polymer characteristics. Average residence time in the reactor train ranges between 2 and 6 h. Butadiene conversion is frequently 80–90%.

Polymer solution from the final reactor is mixed with a shortstop chemical to terminate the reaction and antioxidant to protect the polymer. The stabilized solution is concentrated in one, or more, flash stages and collected in blend tanks to homogenize any product variations. The flash vapor is condensed and returned to the polymerization step. Extender oil (when used) may be added continuously to the concentrated solution stream, or batchwise to the blend tank.

D. Recovery of Solvent

Blended solution containing approximately 10–20% polymer and extender oil (when used) is continuously metered to a stripping vessel. The solvent is vaporized overhead with steam and the contained polymer (with oil) agglomerates into small rice-like particles that are carried from the process as a slurry in water. A surfactant may be added to the process to control polymer particle size. The condensed overhead solvent phase is transferred to solvent purification for recycle to polymerization. The rubber-in-water slurry is collected in a surge tank.

E. Polymer Recovery

The rubber-water slurry containing approximately 5–10% rubber solids is continuously transferred to a screw press or similar device to remove most of the water for recycling to the solvent recovery step. Wet rubber from the press is extruder dried and compressed into 75 lb rectangular bales. Other systems, including apron dryers, are used to dry the dewatered polybutadiene agglomerate. Direct evaporative drying of solution polymers on rotary drum dryers and extrusion desolventing has also been practiced on a commercial scale in place of the steam stripping, rubber dewatering, and drying steps.

F. Environmental Considerations

Airborne emissions can include small quantities of solvent vapors resulting from handling rubber solution in low-pressure tankage and water vapor that contains



small amounts of process-related hydrocarbons from polymer drying. Release of excess water, approximately equal to the weight of product, from the solvent removal system may contain as much as 0.5% rubber fines and other process related hydrocarbons. Other liquid emissions, including heavies from the solvent purification column and desorbate from the regeneration of absorbent beds, can be used as fuel or incinerated. Solid wastes, including spent absorbents and rubber product spillage, should be very small from a plant that is properly designed and operated.

Processes that employ n-hexane or benzene solvent will require handling and disposal techniques appropriate for these hydrocarbons.

G. Compounding

The linear nature of the polymer, narrow molecular weight distribution, and resistance to polymer breakdown during mixing all contribute to the poor processing characteristics of high cis content polybutadiene rubbers (BR). However, processing is satisfactory when the BR is blended with other polymers such as emulsion or solution SBR, and natural or synthetic polyisoprene rubber.

H. Applications

The polybutadienes have outstanding abrasion resistance, high resilience, low heat generation, low Tg, and are superior to many polymers in resistance to cut growth and flex cracking. The cis-BRs have poor wet traction however. Most polybutadiene is used in blends with other rubbers. The vast majority of the BR is used in tire treads, retread rubber, tire carcasses, and tire sidewalls. Other uses capitalize on the high resilience, good heat stability, low Tg, and excellent abrasion resistance of the polybutadiene.

III. Anionic Solution Process

A. Introduction

The solution process utilizing anionic polymerization initiator such as butyllithium offers the simplest of polymerization systems, but its greatest asset lies in its versatility. It is capable of producing solution SBR, low-cis, medium-vinyl, or high-vinyl polybutadiene rubbers, cis-polyisoprene rubber, block copolymers, and thermoplastic rubbers. A single plant can supply a variety of polymers suitable for tires, electrical and mechanical goods, footwear, adhesives, and plastic modification.

B. Process Description

The hydrocarbon solvent, diene, and styrene (if required to produce copolymers) are dried to a very low moisture level before charging into a reactor. If other



chemicals are required for microstructure modification or randomization, they should be added before introduction of the initiator. Polymerization is essentially adiabatic, but in some cases peak temperature is controlled by cooling. All of the monomers charged are polymerized to quantitative conversion. After deactivation of the active chain ends, a stabilizer is added and a portion of the solvent is ecovered for direct recycle to the reactor. The concentrated cement is blended to the desired properties before separation of solvent and rubber. Once recovered, the solvent is purified and recycled, while the rubber is dried, baled, and packaged.

C. Monomer and Solvent

Water, alcohol, and alikes react with alkyllithium rapidly and destroy the initiator. Even at very low levels, in some cases compensation of the initiator level must be applied to get the targeted molecular weight of the product in commercial operation. However, the slow-reacting impurities often are more troublesome. For example, 1,2-butadiene, a frequent contaminant of 1,3-butadiene, reacts with alkyllithium only very slowly [8,9] serving as a terminator. External acetylenes such as 1-propyne and 1-butyne react with alkyllithium quickly to form lithium acetylide, and then metallation reaction occurs, in stepwise fashion, at the sites of beta-hydrogen of the acetylide, eventually replacing all the beta-hydrogen atoms with lithium [10,11]. This sequence of reactions resulted in destruction of the initiator (rapid formation of the acetylide), termination of the propagating chain (slow metallation of the beta-hydrogen atoms), and even chain transfer reaction (initiations from the metallated acetylide). Allen can be isomerized to 1-propyne causing problems. Vinylacetylene, another common impurity in butadiene, also forms acetylide and then undergoes not only metallation of the beta-hydrogen atoms but also a polymerization reaction [10]. These 1-alkynes are often the source of yellow color in the products as well.

Typical composition of C4 cut from naphtha cracker and typical specifications of butadiene and cyclohexane, a polymerization solvent, are shown in Tables 16.5–16.7.

D. Polymerization Process

The simplified flow diagram for solution polymerization rubber process is shown in Figure 16.2. This diagram is applicable to polydienes production as well as styrene-diene copolymer production.

E. Medium-cis Polybutadiene Rubbers

The first generation of polybutadienes (e.g., Phillips Solprene 200 and Firestone's Diene 55) manufactured by alkyllithium initiation in hydrocarbon solvent were introduced to the marketplace in the early 1960s. These are of medium cis content,



Table 16.5 Typical Composition of C4 Cut From Naphtha Cracker

Ingredient Weight (%)

Ethane 0.02

Propane 1.48

Propylene 0.53

Cyclopropane 0.04

Allene 0.08

Propyne (ppm) <10

Isobutane 1.08

n-Butane 3.74

I-Butene 18.20

Isobutylene 5.20

cis-2-Butene 4.14

trans-2-Butene 5.15

1,2-Butadiene 0.16

1,3-Butadiene 59.95

Vinylacetylene (ppm) <10

Heavies 0.20

Nitrogen (ppm) 1

Sulfur (ppm) 3

Oxygen (ppm) 60

generally around 50% cis, 40% trans, and 10% vinyl, and vary in molecular weight and other structure characteristics such as long-chain branching. In the solution process, polybutadienes can be made very light in color and odor-free and, in addition, contain a high percentage of actual rubber hydrocarbon. This allows use of the polymer in light-colored rubber stocks and as an additive for plastics. Some grades are, in fact, manufactured specifically for the plastics trade and are used primarily in the preparation of high-impact resins. It is especially useful for the production of high-impact polystyrene where the polybutadiene is dissolved in styrene and the solution polymerized to produce a graft polymer.

In view of the fact that medium-cis polybutadiene rubbers are extensively used to produce high-impact polystyrene resin, some more detailed discussion here seems appropriate. In these uses, impact polystyrene is a two-phase system consisting of a rigid matrix of styrene homopolymer in which soft rubber particles are imbedded. Increased toughness is the result of stretching of the matrix and compression of the rubber.



Table 16.6 Typical Properties of 1,3-Butadiene (99% Minumum)

Typical Properties

Conjugated diene content (min. wt. %) 99.0

Peroxides (as hydrogen peroxide; max. ppm) 5

Total acetylenes (as vinyl acetylene; max. ppm) 100

Methyl acetylene (max. ppm) 10

Ethyl acetylene (max. ppm) 50

1,2-Butadiene (max. ppm) 150

Carbonyl (as acetaldehyde; max. ppm) 50

Sulfur (as hydrogen sulfide; max. ppm) 5

Butadiene dimer (max. wt. %) 0.1

Nonvolatile residue (max. wt. %) 0.1

4-Tertiary Butylcatechol (min. ppm as shipped) 100

Methanol (max. ppm) 15

Allene (max. ppm) 10

Oxygen content of vapor over liquid in filled tank car (max. vol. %) 0.3

C5 Hydrocarbon (difference in boiling point between 2.0 and 0.5 ml residue volumes in Cottrell boiler; max. °C)

0.4

Table 16.7 Typical Solvent Specifications: 98% Cyclohexane

Specifications

Property Typical Minimum Maximum

Cyclohexane content, wt. % 98.5 98.0 —

mol. % 98.9 — —

Sulfur content (ppm) 1 — 5

Benzene plus toluene (ppm) 25 — 100

Distillation evap. (°C at 760 mm)

UBP 80.7 79.5 —

DP 81.0 — 82.0

Specific gravity (60°/60°F) 0.7821 0.777 0.787

Freezing point (°C) 3.7 — —

Saybolt color +30 +30 —



Copper corrosion (2 h at 212°F) 1 — 1

Doctor test Negative Negative —

Nonvolatile matter (g/100 ml) 0.0005 — 0.001



Figure 16.2. Flow diagram shows solution polymerization rubber process.



Four general requirements should be considered when selecting an elastomer for toughening polystyrene.

1. The rubber should remain elastic down to low temperatures. The lower the glass transition temperature, the less elastomer required to obtain equal impact strength.

2. The elastomer should be completely or at least partially soluble in styrene monomer, but not in polymer.

3. The elastomer should have large particles. Large rubber particles 0.5–5 =305m provide higher impact strength, whereas smaller rubber particles, 0.08–0.25 µm, give better gloss.

4. The elastomer should have graftable sites. Rubber particles must be cross-linked to develop toughness.

Impact polystyrene is made in three steps. First, the polybutadiene is cut up and dissolved in styrene monomer. The next step is prepolymerization. As the styrene starts to polymerize, the rubber particles precipitate and become dispersed in the syrup. Droplets of polystyrene form with phase separation. When nearly equal phase volumes are reached, phase inversion occurs and the polystyrene droplets become the continuous phase. Particle size of the dispersed rubber can be adjusted by varying the amount of shear after the phase inversion.

The polymerization is completed in mass, solution, or aqueous suspension. During the final polymerization the particle size of the rubber is affected by a number of variables:

Viscosity of the matrix (hence the rubber particle size) depends on the amount of chain transfer agent used.

High-viscosity polybutadiene leads to larger particles than low-molecular-weight rubber.

Azo initiators provide larger particles than peroxides.

Highly crosslinked rubber leads to a fine distribution of the graft polymer in the rubber particles.

Grafting and crosslinking reactions begin at the vinyl sites and then occur later at the internal 1,4 sites. It is desirable to keep the temperature low to avoid an undesirable increase in melt viscosity due to crosslinking of the rubber.

Other properties that characterize this type of polybutadiene are high resilience, excellent resistance to abrasion, and ability to utilize high loadings of block and oil. These properties make it valuable for blending with SBRs or natural rubber for use in tires or footwear. The solution-polymerized polybutadiene provides greatly improved resistance to cracking. Added to natural rubber, this polymer also imparts better resistance to thermal degradation and reversion. Another feature of these polybutadienes is a very low brittle point (around -74°C).



In Figure 16.3 the horizontal axis is time and vertical is torque (shear modulus). The natural rubber carcass stock reaches a peak in torque after 4 min at 356°F and shows sharp reversion (16 units) at 20 min cure time. The 40/60 “Solprene 200” polybutadiene rubber/natural rubber blend requires slightly longer to reach a maximum on torque, and display a decrease of only 7 units. The blend increased the resistance to heat degradation in both tread and carcass compounds, in which long cure times, retreading operations, or high temperatures developed in service can degrade the vulcanizate [12].

Changes in the performance of “Solprene 200” polybutadiene/SBR 1712 blends with the level of polybutadiene and carbon block-oil levels are shown in Table 16.8 [12]. It is apparent that a significant improvement in abrasion resistance over oil-extended SBR is obtained with 75 parts carbon black and 50 parts oil at either 25 or 50 part levels of polybutadiene. At black-oil levels of 95–75 the great tolerance of extension inherent in the polybutadiene permits improved abrasion resistance to be realized if a 50/50 ratio of polybutadiene rubber is used with SBR

Figure 16.3. Reversion of natural rubber and polybutadiene blend at

-356°F, preheat 60 s. (From Ref. 12, used with permission from MCM Publishing Ltd.,

London, England.)



Table 16.8 Effect of Solprene 200 Polybutadiene/SBR 1712 Ratio and Extension of Treadwear

Solprene 200/SBR 1712 SBR 1712

Rubber ratio 25/75 50/50 25/75 50/50 100

HAF carbon black (phr)

75 75 95 95 70

Highly aromatic oil (phr)

50 50 75 75 40

Abrasion index 109 112 96 113 100

Cost/quality indexa 114 117 110 130 100

Source: Ref. 12.

1712. All of the blend compounds show a much better cost-quality index than the SBR 1712 control stock [12].

The medium-cis polybutadiene rubbers are manufactured in large quantity and consumed all over the world.

F. Medium-Vinyl Polybutadiene Rubbers

In the 1970s the large-scale development of medium-vinyl polybutadienes began [13–18]. Medium- to high-vinyl polybutadiene can be useful in fatigue-resistant uses, but when styrene became more expensive and less plentiful than medium-vinyl, polybutadiene came into focus as a general purpose rubber.

It was shown that there is a linear relationship between Tg of the polymer and vinyl content, as shown in Figure 16.4 [16,20]. Tg is closely related to many properties of a polymer. Work reported by Railsback and Zelinski [13] shows decreased abrasion resistance and increased skid resistance as the vinyl configuration of the polymer is increased; these are accompanied by improved processability as evidenced by tack, milling, and extrusion. The styrene content in solution SBR is similar to vinyl in that abrasion resistance increases and skid resistance decreases as styrene or vinyl content is reduced. Processability suffers as styrene content increases as well [13].

Duck and Locke compared higher-vinyl materials with the low-vinyl (˜10%) polybutadienes and showed some interesting differences [16,20] (Figs. 16.5, 16.6). It is known that hysteresis and skid resistance improve as the second-order transition temperature (Tg) increases. Good abrasion resistance and good tread wear are found in low-Tg rubber such as the high 1,4-polybutadienes (low-vinyl), but these have not been satisfactory tread rubbers by themselves because of inadequate road holding and wet skid deficiencies. These commercial tire rubbers have always had compositions that represented best-balance compromises of wear and traction properties. The Tg of the rubber blends used in tires generally falls in



Figure 16.4. Relationship of glass transition temperature to percentage

1,2 configuration in polybutadiene. (From Refs. 16,20,40, used with permissions from The Plastics Rubber Inst., London,

England and John Wiley & Sons.)

Figure 16.5. Brabender processing characteristics of high- and

low-vinyl linear polybutadienes. (From Refs. 16,20,40, used with permissions from The Plastics Rubber Inst.,

London, England and John Wiley & Sons.)



Figure 16.6. Processing of tread stock compounds showing the effect

of vinyl content of polybutadiene used. (From Refs. 16,20,40, used with permissions from The Plastics Rubber Inst.,

London, England and John Wiley & Sons.)

the range of -50 to -75°C. SBR achieves this Tg range by incorporating styrene along with butadiene in the polymer chain. One can approach the same Tg range of polybutadiene rubbers with the vinyl configurations in the range of 35–55%.

The results of tests conducted with master batches of 35–55% vinyl polybutadiene with 37.5 phr highly aromatic oil as compared with SBR 1712 indicate that polybutadiene homopolymer of selected vinyl structure (35–55% vinyl) is similar to SBR in many ways and can be substituted for SBR: perhaps complete substitution or perhaps in blends as shown in Table 16.9. Blowout resistance is excellent for the polybutadienes. Abrasion resistance decreases fairly rapidly and wet traction improves rather slowly with higher vinyl content. If the latter properties are given considerable weight, 55% vinyl polybutadiene may be the nearest replacement for SBR 1712 (Fig. 16.7). If better tread wear is desired, or for higher extension with carbon black and oil, a lower vinyl configuration may be desirable. Blends of 45% vinyl polybutadiene and SBR 1712 have slightly lower tensile and tear but most vulcanizate properties improve. Skid and traction on wet surface remain almost constant and abrasion resistance and cut growth resistance show a significant improvement with 50% or more of polybutadiene in the compound [21]. Medium-vinyl polybutadiene can also be used satisfactorily for whole or partial replacement of other rubbers for applications in high-quality shoe soling, conveyor belts, V-belt components, weather stripping, and dock fenders [21].

For commercial production of medium- or high-vinyl polybutadiene rubbers,



Table 16.9 Medium-Vinyl Polybutadienes in Passenger Tire Treads

Polybutadiene

SBR 1712 35% Vinyl 45% Vinyl 55% Vinyl

Rubber 100 100 100 100

Total oil (phr)a 45 45 45 45

N339 Black (phr)b 75 75 75 75

Masterbatch ML-4 53 35–47 40–45 41–54

Compounded ML-4 58 65–80 70–80 72

Extrusion rate (g/min) 79 83 86 88

Extrusion appearance (3–12, 12 best)

10+ 12 12 12-

Die swell (%) 96 72 68 58

Scorch time (min) 14 12 12 13

Dispersion (0–10, 10 best) 8 7 7 7

300% Modulus (MPa) 8.7 8.8 9.1 8.9

Tensile (MPa) 21.2 17.5 18.4 17.9

Elongation (%) 600 520 530 530

Shore A hardness 58 60 59 59

Heat build-up (°C) 45 42 41 41

Resilience (%) 52 60 59 58

Blowout time (min) 11 >60 >60 35->60

Gehman freeze point (°C) -46 -59 -50 -46

Abrasion indexc 100 (145)d 120 100

Skid and traction index (wet)c 100 (90)d 90–100 95

Plus typical amounts of zinc oxide, stearic acid, and antioxidants.

Source: Ref. 21.

donor additive or an alkali-metal alkoxide additive other than lithium is commonly used (Figs. 16.8, 16.9; Table 16.10; see also Chap. 9). There is a report [23] claiming 100% vinyl polybutadiene can be made by using additive bis-piperidinoethane.

G. Long-Chain Branched Polydiene Rubbers

Polybutadiene rubbers made with alkyllithium initiators are linear and in general have very narrow molecular weight distribution. Cold flow and poor processability were two serious problems for commercial applications in early productions. The introduction of one or two long-chain branches



Figure 16.7. Blends of medium-vinyl polybutadiene and SBR 1712.

(From Ref. 21, used with permission from MCM Publishing Ltd., London, England.)

100,000 (tetrachain), the Newtonian viscosity rises rapidly above the corresponding value for a linear polybutadiene. However, non-Newtonian behavior of the branched polymers becomes more pronounced the higher the molecular weight, so that at moderate to high rates the viscosity of the branched polymers is uniformly lower than that of linear polymers of identical molecular weight. Many commercial polybutadiene rubbers are made in long chain branched form [27] to reduce cold flow (low shear) and improve processability (high shear).

Divinylbenzene, a difunctional monomer, is also commonly used in commercial productions of anionically polymerized rubbers either as a coupling (linking) agent at the end of polymerization to form multichain branched molecules or as a comonomer introducing random long-chain branching during polymerization [28]. Rubber molecules that are long chain branched are often referred to as “star” or “radial” (see Chap. 13).



Figure 16.8. Effect of donor additives on vinyl content of polybutadiene.

(From Refs. 16,20, used with permission from The Plastics Rubber Inst., London, England and John Wiley & Sons.)

Natural rubber breaks down readily during mechanical mixing, and rubber processors have traditionally made good use of this property. Emulsion SBR breaks down with great difficulty and some polybutadiene rubbers do not break down at all.

One unique class of solution elastomers, both polybutadiene and butadiene-styrene copolymers, was designed to undergo controlled breakdown during processing [29]. They are made by introducing carbon-tin bonds into the main chain. When such a product is mixed with an organic acid, for example, stearic, acid, some of the carbon-tin bonds are ruptured and breakdown occurs. The extent can be controlled readily for the mixing conditions and the amount of organic acid employed [30].

IV. Polyisoprene Rubber

Sodium or potassium metals were used in large scale in Russia and Germany to produce polydiene rubbers earlier, particularly during the period between the World Wars [31].



Figure 16.9. Effect of tetrahydrofuran and temperature on

polybutadiene vinyl content.

Table 16.10 Relative Vinyl Unsaturation of Polybutadienes

Vinyl Unsaturation (%)a

M RM or M BuLi/t-BuOMb

Li 10 7

Na 65 67

K 45 48

Rb 62 55

Ca 59 53

aPolymerization is hydrocarbon solvent.

bOptimum ratio for maximum vinyl unsaturation.

Source: Ref. 22.



According to Forman [32], in 1952, Firestone's director of research, Dr. F. W. Stavely, instituted research programs for the purpose of exploring synthetic routes that were hoped to lead to the synthesis of cis-polyisoprene (IR). During those investigations, a polymer exhibiting the structural features of the natural rubber hydrocarbon was obtained from the reaction of lithium metal with isoprene.

Firestone disclosed the findings, including synthesis and technological use, to the Office of Synthetic Rubber of the Federal Facilities Corporation in 1955 [33]. Later that year extensive details were presented at a meeting of the American Chemical Society and then published [34]. At a subsequent meeting of the Society, Foster and Binder [35] reported some relationships between microstructure and key physical properties of both polyisoprene and polybutadienes prepared by using lithium, sodium, potassium, rubidium, and cesium metals. It is clear that among these alkali metals, only lithium and organolithium compounds were able to produce predominantly cis configuration in polyisoprene.

Hsieh and Tobolsky first reported the synthesis of cis-polyisoprene by the way of n-BuLi in hydrocarbon solution [36]. This work was done in 1955 after the announcement that natural rubber was synthesized was made and before the methods were disclosed or known to them. This prompted Tobolsky to make the remark of “too late, alas” [37].

At about the same time and shortly after Ziegler's discovery that the suspension obtained by treating TiCl4 with Et3Al can polymerize ethylene, Hore and colleagues discovered the synthesis of cis-polyisoprene with TiCl4 and R3Al [38]. With this Ziegler catalyst, a cis content of 96% could be obtained, while the lithium catalysts produced lower cis content, in the range of 85–92%. Since, in the case of polyisoprenes, the processing characteristic of crystallinity under stress and green strength depend mainly on the stearic purity and also on molecular weight distribution, the differences between these two types of synthetic polyisoprenes are more significant in relation to their applications than was the case for the polybutadiene rubbers. Although the very high-cis-polyisoprene prepared with the Ziegler catalysts can be used as a total replacement for natural rubber (97–100% cis), the lithium metal or alkyllithium-initiated polyisoprene rubbers should not be treated as if they were natural rubber [39]. Figure 16.10 [40] illustrates the differences in stress—strain properties of unvulcanized tread stocks of natural rubber and synthetic polyisoprene rubber made with alkyllithium initiator. High green strength is important in that it contributes to the ability to pull itself over the mill rolls during processing. It is generally agreed that both green strength and tack improve significantly as the cis content increases toward 96%.

Figure 16.11 [40] shows the effect of masticating in a Brabender natural rubber (I), Ziegler type, very high-cis polyisoprene (II), and the polyisoprene prepared using alkyllithium catalyst (III).

Numerous reports relating microstructure to the physical properties of polyisoprene have been published [41–44]. In addition, Brock and Hachthorn [45]



Figure 16.10. Green strength: stress-strain properties of unvulcanized

tread stocks (Extension values 0.2 in/min). (From Ref. 40, used with permission from

John Wiley & Sons.)

pointed out that the insertion of a few isopropenyl units (from 3,4-addition) into the main chain changed head-to-tail, tail-to-head point and this made the segment in polyisoprene crystal lattice lose the possibility of the tight arrangement and led to slow crystallization rate and low crystallinity. Furukawa [46] reported that the melting point (Tm) is an important parameter to characterize the rubber. The strength of the rubber increases with increasing Tm. However, for various polyisoprene rubbers, if Tm is too high the low temperature property will be sacrificed. Both Tm [45] and Tg [42] depend upon the regularity of the chain element structure and they are the function of the cis content.

Firestone Tire & Rubber Co. produced cis-polyisoprene rubber with lithium metal in the late 1950s. Shell Chemicals and Shell Netherland Chemie NV started production using alkyllithium in the early 1960s. Asahi of Japan and Negromex SA of Mexico also had production facilities using alkyllithium. However, by the mid 1970s, Firestone, Shell, Asahi, and Negromex all stopped the productions. Shell Group in Europe stayed in production a little while longer and then also discontinued the activity. The inability of the lithium system to produce cis content in the 96% range made it significantly less desirable than the titanium system for the direct replacement of natural rubber, particularly in the applications.

Shell did produce a latex from their cis-polyisoprene solution [47]. These



Figure 16.11. Brabender mastication of polyisoprenes. (From Ref. 40, used

with permission from John Wiley & Sons.)

lattices have been used to make surgical gloves and other dipped goods. The lower modulus of the alkyllithium polyisoprene gum vulcanizate gives a softer, more pleasant glove. Tear strength and other properties are excellent [48].

V. Homopolymers of Piperylene and Butadiene/Isoprene Copolymers

1-Methylbutadiene or 1,3(pentadiene-1,3), more commonly known as piperylene, is a byproduct in the synthesis of isoprene from C5 petroleum source. As such, it has an effect on the economics of the synthesis of isoprene by a dehydrogenation process. In spite of many efforts to convert piperylene into a commercially valuable polymer, little success has been reported [32].

Piperylene consists of a mixture of the cis and the trans isomers in varying proportions. The trans isomer was found to polymerize more rapidly than the cis monomer with lithium type of catalysts [49]. A group of scientists at Strasbourg [50–55] reported a preponderant tendency to form 1,4 polymer and no 3,4 additions in piperylene polymers obtained from monomer mixed isomers either in bulk or in hydrocarbon solution and catalyzed either by lithium metal or butyllithium. Jenner [54] found higher 1,2 content from polymerization in very high pressure, in tetrahydrofuran (THF), or using a sodium metal as catalyst.



The known resistance of 1,4-polypiperylene to attack by ozone [56] has led to consideration of this polymer as a desirable material for sidewall stocks of tires [32]. Polypiperylenes have the advantage over other rubbers resistant to oxygen in being completely compatible with existing polymers in commercial use. However, no large-scale production of these polymers was undertaken.

Rakova and Korotkov [57] studied the copolymerization of the two most common dienes, butadiene and isporene, in hexane by butyllithium. The results are shown in Table 16.11. Using these data, the sequence distribution was analyzed according to the method of Harwood and Ritchey [58] as shown in Table 16.12.

A more comprehensive study of anionic copolymerization of butadiene and isoprene with sec-butyllithium in hexane was carried out in the laboratories of Ward and colleagues [59]. These authors made the following conclusions.

1. Isoprene is a more active monomer than butadiene in homopolymerization, but the apparent activation energy of the propagation reaction is 19.2 kcal/mole for both monomers.

2. In copolymerization butadiene reacts preferentially; however, a significant concentration of isoprene units is also incorporated in a random manner during the early stage of reaction. Butadiene has higher reactivity ratio than

Table 16.11 Copolymerization of Butadiene and Isoprene in Hexane with Butyllithiuma

Initial Mixture (mole %)

Copolymer Composition (mole %)

Butadiene

Isoprene

Conversion (%)

Butadiene

Isoprene

21.1 78.9 42.3 42.5 57.5

22.0 78.0 50.8 37.0 63.0

30.3 69.7 16.8 45.6 54.4

47.8 52.2 13.4 71.3 28.7

48.8 51.2 35.4 68.1 31.9

48.9 51.1 84.6 56.3 43.7

51.0 49.0 14.5 73.0 27.0

51.0 49.0 65.5 65.5 34.5

51.3 48.7 25.2 71.7 28.3

79.1 20.9 11.9 92.0 8.0

79.3 20.7 40.2 93.4 6.6

80.2 19.8 64.5 90.5 9.5

aData of Rakova and Korotkov [57]; r1 (butadiene) =3.38 ± 0.14; r2 (isoprene) = 0.47 ± 0.03.



Table 16.12 Sequence Distribution in Butadiene-Isoprene Copolymers

Analysis Calculated from Run Number

Random, AA, BB, AB + BA,

FBAA

A B % % % % FAAA and fAAB fBAB

42.5 57.5 48.9 21.1 36.3 42.9 0.245 0.500 0.255

37.0 63.0 46.6 15.3 41.3 43.4 0.171 0.485 0.344

45.6 54.4 49.6 23.6 32.4 44.0 0.268 0.500 0.232

71.3 28.7 40.9 53.5 10.9 35.7 0.562 0.375 0.063

68.1 31.9 43.5 50.6 14.4 35.0 0.552 0.382 0.066

56.3 43.7 49.2 38.9 26.3 34.9 0.476 0.428 0.096

73.0 27.0 39.4 55.8 9.8 34.5 0.583 0.361 0.056

65.5 34.5 45.2 48.8 17.8 33.5 0.554 0.381 0.065

71.7 28.3 40.6 55.1 11.7 33.3 0.589 0.357 0.054

92.0 8.0 14.7 85.3 1.3 13.4 0.860 0.135 0.005

93.4 6.6 12.3 86.8 0.0 13.3 0.863 0.132 0.005

90.5 9.5 17.2 84.2 3.2 12.7 0.865 0.131 0.005

A, % butadiene species in copolymer by analysis; B, % isoprene species in copolymer by analysis; AA and BB are respective percentage probabilities of diads; AB + BA, run number calculated by method of Harwood (339) and is the percentage probability that an A unit is either preceded or followed by a B unit; fBAA (or fAAB), fraction of A units centered in triad. Copolymer analysis from data of Rakova and Korotkov [57]. A and B are units derived from butadiene and isoprene, respectively.

Source: Ref. 32.

isoprene. The reversal of reactivity is caused by the less stearic and polar factors in the interaction of butadiene molecules with an active anionic center.

3. Kinetic results show that essentially pure isoprene blocks are formed in the latter stage of reaction after “inversion” point. The phenomenon is somewhat dependent on the relative molar concentration and temperature.

4. Typical reactivity ratios at 20°C are rb = 2.64 and rs = 0.404. Preliminary evidence was reported that suggests the copolymerization is more selective at lower temperature.

Although we have no knowledge that butadiene-isoprene copolymers are produced commercially, it is conceivable that copolymers of this type are produced and may be used “in-house.” Copolymerization of isoprene and piperylene was also extensively studied in the laboratory [60]. No commercial production of these diene copolymers is reported.

References



2. H. E. Railsback, J. R. Haws, and C. R. Wilder, Rubber World, 142, 67 (1960).

3. G. Kraus, J. N. Short, and V. Thornton, Lecture No. 7, International Synthetic Rubber Symposium, London (1957); Rubber & Plast., 38, 880 (1957).

4. J. N. Short, G. Kraus, R. P. Zelinski, and F. E. Naylor, Rubber Chem. & Tech., 32, 614 (1959).

5. C. M. Throckmorton, Kaut. Gummi Kunstst., 22, 293 (1969).

6. M. Bruzzone, et al., Rubber Chem. & Tech., 47, 1175 (1974).

7. H. L. Hsieh and H. C. Yeh, in Advances in Polymer Synthesis, W. M. Culbertson and J. E. McGrath, Eds., Plenum Publishing Corp., New York, 1985.

8. K. C. Eberly and H. E. Adams, J. Organometal. Chem., 3, 165 (1965).

9. H. E. Adams, R. L. Bebb, K. C. Eberly, B. J. Johnson, and E. L. Kay, Kaut. Gummi, 18, 709 (1965).

10. H. L. Hsieh and J. A. Favre, US patent 3,303,225.

11. R. West, P. A. Carney, and I. C. Mineo, J. Am. Chem. Soc., 87, 3788 (1965).

12. H. E. Railsback and J. R. Haws, Rubber & Plast. Age, October, 1967.

13. H. E. Railsback and R. P. Zelinski, Kaut. Gummi Kunst., 25, 254 (1972).

14. E. F. Engel, Rubber Age, 105, 25 (1973).

15. K. H. Nordsieh, Kaut. Gummi Kunst., 25, 87 (1972).

16. E. W. Duck, Eur. Rubber J., 155(12), 38 (1973).

17. H. E. Railsback and N. E. Stumpe, Jr., Rubber Age, 107, 12, 27 (1975).

18. J. Platner, Rubber World, 169, No. 5, 43 (1974).

19. A. Ledwith and A. M. North, in Molecular Behavior and the Development of Polymeric Materials, Chapman and Hall, London, 1974, p. 125.

20. E. W. Duck and J. M. Locke, J. Inst. Rubber Ind., 2, 223 (1968).

21. J. R. Haws, L. L. Nash, and M. S. Wilt, Rubber & Plast. Age, 107, June 1975.

22. H. L. Hsieh and C. F. Wofford, J. Polym. Sci., A-1, Vol. 7, 449 (1969).

23. A. F. Halasa, D. F. Lohr, and J. E. Hall, J. Polym. Sci., Chem. Ed., 19, 1357 (1981).

24. R. P. Zelinski and C. F. Wofford, J. Polym. Sci., A3, 93 (1965).

25. J. T. Gruver and G. Kraus, J. Polym. Sci., A2, 797 (1964).

26. G. Kraus and J. T. Gruver, J. Polym. Sci., A3, 105 (1965).





38. Br. patent 827,365.

39. G. W. Atkinson and J. M. Gopperi, Third TNO Conference on New Commercial Technical Developments, 55, Rotterdam, February 26–27, 1970.

40. E. W. Duck and J. M. Locke, in The Stereo Rubbers, W. M. Saltman, Ed., Wiley Interscience, New York, 1977.

41. M. Bruzzone, G. Corrodini, and F. Amato, Rubber Plast. Age, 46, 278 (1965); Rubber Chem. & Tech., 39, 1593 (1966).

42. K. W. Scott, G. S. Trick, R. H. Mayor, W. M. Saltman, and R. M. Pierson, Rubber Plast. Age, 42, 175 (1961).

43. E. Butta and V. Frosini, Chim. Ind. (Milan), 45, 703 (1963); Chem. Abstr., 60, 10898 (1964).

44. E. G. Kent and F. B. Swinney, Ind. Eng. Chem. Prod. Res. Dev., 5(2), 134 (1966).

45. M. J. Brock and M. J. Hachathorn, Rubber Chem. & Tech., 45, 1303 (1972).

46. J. Furukawa, Plast. Rubber Materials Appl., 2, 33 (1977).

47. Br. patents 957,967; 1,016,235; 1,016,236.

48. A.R. Bean, et al., Encyclopedia of Polymer Science, Vol. 17, Wiley, New York, 1967, p. 829.

49. R. S. Stearns, US patent 3,147,242.

50. F. Schue, Bull. Soc. Chim., 980 (1965).

51. F. Schue, Bull. Soc. Chim., 973 (1963).

52. F. Schue, C. Ortlieb, A. Mailand, and A. Deluzarche, Bull. Soc. Chim., 982 (1965).

53. G. Friedman, F. Schue, M. Brini, A. Deluzarche, and A. Mailard, Bull. Soc. Chim., 1343; 1728; 3636 (1965).

54. G. Jenner, Bull. Soc. Chim., 2851 (1965).

55. G. Jenner, Bull. Soc. Chim., 1127 (1966).

56. W. K. Taft and G. L. Tiger, in Synthetic Rubbers, G. S. Whitby, Ed., Wiley, New York, 1954, p. 688.

57. R. V. Rakova and A. A. Korotkov, Rubber Chem. & Tech., 33, 623 (1960).

58. H. J. Harwood and W. M. Ritchey, J. Polym. Sci., B, 2, 601 (1964).

59. I. C. Wang, Y. Mohajer, T. C. Ward, G. L. Wilkes, and J. E. McGrath, in Anionic Polymerization, Kinetics, Mechanisms, and Synthesis, ACS Symposium Series 166, J. E. McGrath, Ed., Am. Chem. Soc., Washington, D.C., 1981, Chapter 34.

60. G. V. Rakova, A. A. Korotkov, and T.-C. Li, Proc. Acad. Sci., USSR, 126,377 (1959).



17 Styrene-Diene Rubbers

I. Styrene-Butadiene Diblock Copolymers

A. Introduction

Zelinski and his co-workers in the Phillips laboratory have conducted an extensive program on copolymerization of butadiene and styrene in solution with alkyl-lithium from 1957 on. In this program, these researchers discovered the synthesis of tapered block copolymers and random copolymers using polar compounds as adjuvent to cause the randomization [1,2]. In 1960, Crouch of Phillips Petroleum Company reported briefly at the Second International Synthetic Rubbery Symposium held in London on a copolymer produced in pilot plant reactors for market development study [3]. This copolymer, Solprene 1205 rubber, was the first solution block copolymer of butadiene and styrene produced commercially [4]. It is a block copolymer containing 75 parts butadiene and 25 parts styrene. It is useful in many applications and has unique properties not found in styrene-butadiene rubber (SBR) types [5]. Another solution copolymer, Solprene 1204 rubber, of butadiene and styrene was offered in commercial quantity by Phillips 1 year later [6]. Solprene 1204 copolymer is random copolymer. The polymerization chemistry and properties of these two types of copolymers were reported by Hsieh at the Fourth International Synthetic Rubber Symposium held in London in 1964 [7]. Both Solprene 1205 and 1204 copolymers are no longer produced in the



United States. These rubbers are still being produced by former Phillips licensees elsewhere.

B. Tapered Block Copolymers

Tapered block copolymers such as Solprene 1205 are prepared directly from monomer mixtures in hydrocarbon solvent. Typical polymerization results are shown in Figure 17.1 and Table 17.1.

In addition to the tapered block copolymers produced by copolymerization of mixtures of monomers in the absence of a randomizer such as an ether, pure block and random block copolymers of diene and styrene (Table 17.2) can be produced by incremental addition of monomers.

C. Properties of Diblock Copolymers

Diblock copolymers can have very special properties by virtue of the particular arrangement, for example, glossy block of polystyrene and rubbery block of essentially polybutadiene. The polystyrene block at room temperature is some 75°C below its glass transition temperature, while the polybutadiene block is

Figure 17.1 Polymerization of butadiene-styrene in cyclohexane at 50°C.



Table 17.1 Formation of Tapered Block Copolymers of Butadiene and Styrene

Butadiene/Styrene charge Total styrene (%) Block styrene (%)

85/15 15 12

75/25 25 20

50/50 50 42

25/75 75 66

some 110°C above its glass transition temperature. The block copolymers exhibits both glass transitions when taken through the -100–150°C range.

The glassy polystyrene block is not soluble to any extent in the rubber portion of the polymer and a simple mixture of the two homopolymers would tend to separate to produce an opaque mixture of resins. However, when the blocks are tied together on a molecular level, as in a pure diblock or tapered block copolymer, they cannot separate on a macro scale and, therefore, tend to separate on a micro scale.

Simple styrene-butadiene diblock copolymers appear to be homogeneous and indeed exhibit great clarity when molded into sheet. Nonetheless, the two blocks are still basically incompatible and while massive separation may be prevented by the carbon-carbon links between the rubbery and glossy blocks, separation will still occur. On a micro scale, the glossy polystyrene block tends to separate from the rubbery block and associate with similar blocks from other polymer chains. The polystyrene blocks concentrate to form into domains of glassy polystyrene in a rubbery block matrix. The domain size is in the 100–400 Å size, which cannot be seen by the naked eye. Therefore, the molded polymers appear to be homogeneous and transparent. The products are rubbery and yet exhibit two glass transitions typical of the two homopolymers. Such products will exhibit no green tensile strength and for nearly all the applications will have to

Table 17.2 Diblock Copolymers of Butadiene (B) and Styrene (S)

Type Monomer sequence

Pure block BBBBBBBBBBBBSSSSSSSS

(B-S)

Tapered block BBBBBBSBBBBSBBSBSBSSBSSSSS

(B S-S)

Random block BSBBBSBBSSBBSBSBBSSSSS

(B/S-S)



be vulcanized to develop the strength needed in molded articles. Typical properties of some tapered diblock copolymers are shown in Table 17.3. This is quite different from those thermoplastic elastomers with linear triblock structure S-B-S or radial structure (S-B-)-2,3 S made with anionic initiators. In these block copolymers, the polystyrene blocks associate in domains, but since there is one at each end of a long rubbery block, a single polymer molecule may be connected to two different domains. As long as the polystyrenes are in their glassy, rigid state, the product will appear as a network structure with the rubbery chain anchored between two fixed domains. This phenomenon is known as “physical crosslinking” and the products require no vulcanization (chemical crosslinking) to obtain the strength. These thermoplastic elastomers are discussed in Chapter 18.

D. Applications of Diblock Copolymers

Features of the tapered block copolymer are high modulus, high hardness, low shrinkage, good extrusion, high resistance to abrasion, and a low brittle point. Hardness, shrinkage, and abrasion qualities of these products are well suited for shoe soles, floor tile, and cove base. The transparency, high hardness, and light color allow production of very attractive floor tile. High extrusion rates and glossy appearance are useful in cove base. Solprene 1205 rubber can be used in blends with other rubbers to control shrinkage and has also been used to replace low

Table 17.3 Typical Properties of Some Diblock Copolymers of Butadiene (B) and Styrene (S)

Solprene rubbers

Property 1205 303a 410 476

Oil content (phr) — — — 37.5

B/S Ratio 75/25 52/48 52/48 52/48

Block polystyrene (%) 18 11 32 32

Color White/ Noncontamination

Very light/ Noncontamination

Mooney, M1–4, 100°C 47 45 47 35

Hydrocarbon content (%; SBR 1502 is 92.93%)

99 99 99 99

Specific gravity 0.932 0.972 0.925 0.959

Refractive index 1.536 1.556 1.533 1.553

Vulcanization, 145°C 50 mins

Tensile strength (MPa) 15.7 15.7 16.7 14.7

Elongation in break (%) 650 450 500 450

aRandom block copolymer.

bNaphthenic oil.



Table 17.4 Styrene-Butadiene Block Copolymer in Floor Tile

Total filler (parts) 445

Percentage rubber 16.7

Block Copolymer

Random copolymer(emulsion)

Compounded ML-4 at 212°F 54 78

Mill shrinkage (%) 0 4.5

Cured 12 min at 158°C (320°F)

Shore D hardness 48 46

Abrasion index 184 100

temperature plasticizers. The polymer is more thermoplastic than others and has excellent mold flow characteristics.

Other applications of these styrene-butadiene block copolymers are wire and cable insulation, sponges, conveyor belts, rubber-covered rolls, hoses, and molded or extruded goods requiring good abrasion resistance.

In floor tile application, the block copolymer has lower mill shrinkage and much improved wear resistance, as shown in Table 17.4. In Table 17.5 the comparison of the block copolymer and an oil-extended emulsion SBR in a cove base stock indicates the excellent appearance of the product made from the block copolymer.

The improvement in extrusion characteristics is shown in the mechanical rubber goods formulation (Table 17.6). The four columns indicate the increased amount of block copolymer replacing the emulsion SBR. The extrusion rate in both centimeters per minute and grams per minute is increased by increasing the

Table 17.5 Block Copolymer in Cove Base Stock

Total filler (parts) 400

Total plasticizer (parts) 42

Block copolymer

Oil extended emulsion copolymer

Extrusion at 82°C (180°F) (cm/min) 270 112

Appearance Excellent Good

Cured 30 min at 153°C 307°F)

Tensile strength (MPa) 4.4 6.1

Elongation (%) 310 350

Shore A hardness 84 83



Table 17.6 Improvement of Extrusion Characteristics

SBR random copolymer 100 50 25 —

SB block (25% S, 17% SB) — 50 75 100

Extrusion rate (cm/min) 167 193 218 258

(g/min) 138 148 159 160

Appearance rating 6 7 11 11

σ300 (kg/cm2) 66 70 76 73

σb (kg/cm2) 119 119 122 105

εb (%) 530 620 640 670

level of solution block copolymers. The surface appearance of the mechanical rubber goods is improved by increasing levels of solution block copolymers.

The reduction in mill shrinkage in a mechanical rubber goods formulation in which natural rubber is replaced by a solution block copolymer is shown in Table 17.7. Mill shrinkage is reduced from 35% to 9.5% by the replacement of 20% of the natural rubber with styrene-butadiene block copolymer.

An integral molded sole is one in which the sole and heel are formed as one unit. The advantage for the Solprene 1205 rubber, a tapered block copolymer of styrene and butadiene, compared to the two emulsion random copolymers, is the much improved resistance to wear (Table 17.8).

Microcellular soling in a blown shoe sole in which the pores are small enough has to be viewed only through a microscope. The advantages of the block copolymers include shorter cure time and increased Shore hardness (Table 17.9).

In electric wire insulation compound the use of a block copolymer offers high extrusion rate, improved extrusion appearance, lower freeze point, and improved tear strength, as shown in Table 17.10.

In automobile sponge compounds, the block copolymer offers lower compression set and higher resilience than emulsion SBR. The sponge rug underlay

Table 17.7 Reduction of Mill Shrinkage in Natural Rubber Mechanical Goods Formulation

Percent of NR replaced by SB blocka

0 5 10 20

Mill shrinkage (%) 35 16 11 9.5

σ300 (kg/cm2 43 388 41 40

σb (kg/cm2) 102 115 107 112

εb (%) 490 550 540 550

Shore A hardness 57 5999 62 63



Table 17.8 Integral Molded Soling

Solprene 1205 SBR 1502 SBR 1506

(Cured 10 min at 300°F)

300% Modulus (MPa) 6.6 5.4 4.5

Tensile (MPa) 7.6 8.3 7.9

Shore A hardness 72 71 66

Flex Crack (cm) a 0.44 0.78 0.34

Abrasion index 143 100 100

a Growth at 200,000 flexures.

Table 17.9 Microcellular Soling

SBR 1509 75 —

SB block (48% total styrene, 11% block styrene)

— 75

Rubber dust 25 25

High-styrene resin 6 10

Filler 75 110

Blowing agent 5.5 3

Oil — 3

Cure at 153°C

Time (min) 12 8

Density (g/cc) 0.55 0.56

Shrinkage (%) 3.5 0.8


Table 17.10 Electrical Wire Insulation

SBR 1503 25 25

SBR 1008 75 —

SB block (25% total styrene, 17% block styrene) — 75

Mooney viscosity 63 41

Extrusion rate (g/min) 79 119

Appearance rating 9 12



Brittle point (°C) -51 -63

Tear strength (kg/cm) 26 34



made from the block copolymer has an extremely smooth surface, and a much better appearance than those made from random SBR or natural rubber.

II. Styrene-Butadiene Random Copolymers

A. Introduction

Random copolymers made in solution with anionic initiators (lithium-based compounds) are used in both tire and nontire products. Merits of these polymers include light color, low nonrubber content, good dimensional stability in extruded products, good mold flow, high resilience, good tear resistance, and improved resistance to cracking.

The comparison of styrene-butadiene rubber made in emulsion process (emulsion SBR) and styrene-butadiene rubber made in solution process (solution SBR) is inevitable. The former was developed and mass produced in the United States during the period of World War II, while the solution SBR was not introduced until some 20 years later. These two processes employ the same pair of monomers and both give essentially random distribution of the two comonomers. In some degree the random copolymers made in either system are alike. However, a number of differences do exist. The high rubber hydrocarbon content and light color of the solution products have been mentioned. Other key differences are found in microstructure, branching, and molecular weight distribution (Table 17.11).

Table 17.11 Comparison of Solution and Emulsion Butadiene/Styrene Copolymers

Emulsion SBR

Typical solution SBR

1500

1502

1503

Styrene (wt. %) 25 23.5 23.5 23.5

Fatty acid (wt. %) 0.5 — 2.9 5.8

Rosin acid (wt. %) — 6.1 2.9 —

Ash (wt. %) 0.1 0.7 0.6 0.15

Water absorption (mg/sq.in.) 4 — — 9

Rubber hydrocarbon (wt. %) 97.5 92 92 92

Color White Dark Light Light

Mooney viscosity (ML-4) 55 52 52 52

Molecular weight distribution Narrow Broad Broad Broad

Microstructure (diene portion)

% cis 32a, 44b 8 8 8

% trans 41, 47 74 74 74

% Vinyl 27, 9 18 18 18

aWith ether as randomizer.



The solution system inherently gives polymers of narrow molecular weight distribution. Usually there are less low- and high-molecular-weight fractions in the solution rubbers.

These differences in molecular weight distribution probably account for the improved hysteresis properties of the solution copolymers. These differences in molecular weight distribution and the greater linearity of the polymer molecule [8] give solution copolymer stocks with a minimum nerve and that are more difficult to process than the emulsion products. For tire applications, the solution SBRs have lower heat build-up, higher resilience, and higher abrasion resistance [9–11].

Moore and Day [12] reported that some of the more important differences between the two types of SBR involve the cure rate, hysteresis, modulus, and tensile and green strength. These researchers found that solution SBR (10–12% vinyl type) is faster curing than emulsion SBR, but the scorch time varies with respect to the individual polymer. Solution SBR usually provides lower hysteresis and higher modulus while emulsion SBR generally provides a higher tensile strength.

With the same bound styrene level, the solution SBR provides lower rolling resistance, improved dry slide traction, and superior wear resistance [12]. The emulsion SBR provides better wet slide traction [12]. With the same Young's Modulus Index (here one is making the assumption that this index is a reasonable method for comparing the compounded glass transition temperature of the polymer), the solution SBR is better for rolling resistance, dry slide, and wet peak traction, while both polymers are equal for wear [12]. This is significant because the tire compounder now can design a compound that improves both wet traction and rolling resistance with no sacrifice in wear.

B. Commercial Solution Styrene-Butadiene Rubbers

In the early productions, a 60-Mooney product was marketed as a general-purpose rubber for use in tires, footwear, and other products. Several rubbers are produced for more specialized use. A slightly higher Mooney rubber has proved useful in physical blends with polystyrene to improve impact resistance. Another version was made with controlled breakdown characteristics [13]. This has been useful to make sponge items such as rug underlay and has been found to display better mold flow, color, resilience, load-bearing ability, and permanent set than other rubbers used in sponge. An additional copolymer with a higher styrene content is introduced for use in sponge and footwear, especially for microcellular sponge.

Oil-extended random copolymers possess many of the same features as their non-oil-extended counterparts, such as good resilience and resistance to abrasion and also the cost and mixing advantages associated with oil extended rubber. Processing characteristics are good.

As is the case with emulsion rubbers, the oil-extended, random solution rubbers are particularly well suited for passenger car tires, either in the tread or



carcass, and for small truck tires. In tires the solution copolymers impart improvements in resistance to wear and cracking; further improvements can be obtained by blending these with solution-polymerized polybutadiene.

The amount of oil, as well as the type of oil, in these master batches depends upon the end use. Highly aromatic hydrocarbon is often used for tire and industrial goods applications. Light colored, non-staining, naphthenic oil is used in master batches that are useful for some types of wire and cable coverings, footwear, adhesives, various molded or extruded goods, and certain passenger tire carcass stocks in which color and nonstaining qualities are important.

In the early and mid-1970s, intensive efforts were made to improve further the performance of the random copolymers. This effort was focused in particular upon increasing abrasion resistance or decreasing rolling resistance while maintaining or improving processing characteristics and wet traction [11,14–20]. Variables that showed the greatest effects on polymer properties are the butadiene-styrene ratio (Fig. 17.2), degree and type of long-chain branching, microstructure (Tables 17.12 and 17.13), molecular weight, and molecular weight distribution. Although these variables are often treated as independent, components or polymerization conditions used to vary one property can affect other properties (e.g., use of modifiers to increase long-chain branching can result in a broadened molecular weight distribution). To devise a polymer with the best balance of properties, many polymers must be prepared and evaluated. Synthetic chemists have many tools available to them to tailor-make the macromolecule in an anionic polymerization process. One can change molecular architecture by selecting modes of initiator addition, polyfunctional initiators, difunctional monomers,

Figure 17.2 Effect of styrene of vinyl solution SBR.



Table 17.12 Effect of Butadiene Microstructure and Copolymer Composition on Tread Properties

Polymer type

SBR

Copolymerwith

medium vinyl

Copolymerwith low

vinyl

BR

Catalyst Free radical

Alkyllithium plus ether

Alkyllithium Alkyllithium

Medium Emulsion Solution Solution Solution

Molecular weight Very broad

Moderate Moderate Moderate

Distribution

% Styrene 25 25 21 0

Microstructure (butadiene portion)

% cis (1,4) 10 23 33 36

% trans (1,4) 70 49 58 55

% Vinyl (1,2) 20 28 9 9

% Steel ball rebound

24°C (72°F) 31 33 42 52

100°C (212°F) 54 61 62 63

Young's modulus index (°C)

-39 -42 -55 -70

Coefficient of friction on wet concrete

100 100 98 85

Wear rating 100 100 135 145

Source: Ref. 17.

various coupling agents, various randomization modifiers, monomer addition sequences, and other variables.

It was shown [14] abrasion resistance decreases and skid resistance increases as styrene or vinyl content is increased. Processability is also improved by the increase of styrene or vinyl content in the rubber (Fig. 17.2). The relationship of Tg and tan δ to styrene content in solution SBRs of a constant vinyl configuration (10%) are shown in Figure 17.3.

Rolling loss of tires using tread polymers of variable characteristics was studied [15]. Aggrawal et al. [18] reported that rolling resistance (i.e., fractional energy dissatisfied in a rolling tire) is predominantly related to loss tangent (i.e., the tan δ at comparatively low frequency and at appropriate temperatures above the glass transition temperature, [Tg]). Rolling resistance is related



Table 17.13 Effect of Butadiene Microstructure on Properties of Solution SBR

Microstructure (butadiene portion)

cis content (%) 32 45

trans content (%) 40 46

Vinyl content (%) 28 9

Processing

Mix time (min) 3.0 3.7

Extrusion at 12°C, Garvey die (g/min) 140 102

Rating (12 = best) 12 11

Cure 30 min at 153°C

300% modulus (MPa) 7.2 7.8

Tensile stength (MPa) 21.6 19

Heat build-up (°C) 41 43

Resilience (%) 57 59


Abrasion resistance index 100 127

Wet-skid resistance asphalt 100 91

Polished concrete 100 96

Recipe for butadiene–styrene rubber (85/15) is:

Rubber 100

Carbon black ISAF (N 220) 70

Oil 40

Zinc oxide 3

Stearic acid 2

Antioxidant/antiozonant 3

Paraffin wax 2

Sulfur 2.1

Accelerator 1.3



Figure 17.3 Relationship of Tg and tan δ to styrene content in solution SBRs.

(From Ref. 18, used with permission from Plenum Press.)

link density, and chain architecture strongly influence loss tangent in the rubber plateau region.

One quite surprising and unique finding of this study [18] is that the tan δ remains constant with vinyl content up to about 50% of polybutadiene rubbers. At higher vinyl content, the tan δ increases sharply. The glass transition temperature (Tg) of the polybutadienes increases nearly linearly with increasing vinyl content (Fig. 17.4). Thus, in the low- and medium-vinyl region of butadiene-based rubbers, traction can be increased without changing rolling resistance.

In conclusion, today the tire builders have a large number of choices of tailor-made rubbers and combinations of rubbers with which to produce tire compounds with a superior and desired balance of performances.

C. Solution SBR Manufacturing Process Overview

The solution SBR polymers are typically formed by copolymerization of styrene and butadiene with an alkyllithium initiator. The polymerization can be continuous or batchwise, although the latter seems to be more versatile to produce a mix of random, block, and triblock copolymers while minimizing polymer gel formation and equipment fouling tendencies associated with this “living polymer” system. If suitably equipped, a solution SBR facility can also produce medium cis-polybutadiene, medium- and high-vinyl polybutadiene, and isoprene—styrene



Figure 17.4 Relationship of Tg and tan δ to vinyl configuration in

polybutadienes. (From Ref. 18, used with permission from Plenum Press.)

copolymers. The broad range of polymers from such a plant will require a variety of polymer recovery equipment to furnish the product forms (bales, pellets, granules, etc.) best suited to customer needs. The solution SBR process diagram shows typical processing steps involved for a facility based upon batch polymerization (see Fig. 16.2 in Chapter 16). Practical considerations are likely to limit the processing options in an actual, commercial processing line.

Fresh monomers with typical purities above 99% and with exacting specifications to limit selected sulfur, oxygenated, and acetylenic compounds are required. Treatment of butadiene and isoprene (optional) can include fractionation to remove water, flashing to remove inhibitor, and, possibly, final treatment over fixed bed absorbent. Styrene is treated in one or more absorbent beds to accomplish the same purposes. Refrigerated storage of “polymerization-ready” monomers is advisable in warm climates to reduce dimer/oligomer formation rates and minimize risk of autopolymerization. Butadiene (and perhaps isoprene) containing substantial amounts of close-boiling paraffins and olefins can be utilized if appropriate equipment is provided to remove these impurities from the solvent circuit.

Fresh makeup solvent (cyclohexane, for example) joins recycle solvent and is fractionated to remove water, light impurities, and heavy impurities (butadiene dimer, etc.). The dry, deoiled solvent stream may also be absorbent-treated before being returned to the polymerization reactor.



The alkyllithium initiator (n-butyllithium, for example) and other reaction-modifying, coupling, shortstop, and antioxidant chemicals are diluted or dissolved in dry solvent and stored under nitrogen to exclude atmospheric oxygen and moisture. Special storage and handling procedures are used to accommodate the highly reactive or pyrophoric nature of the initiator and some of the process chemicals.

The principal raw materials, butadiene and styrene, must be of high purity and almost free of certain possible contaminants. Typical properties of commercial grade butadiene and styrene used in the process are shown in Tables 17.14 and 17.15. Table 17.16 shows typical properties for 98% cyclohexane solvent, which is a satisfactory solvent.

Carefully controlled amounts of purified solvent, butadiene and styrene monomers, and reaction-modifying chemicals (when used) are transferred into the well-mixed batch reactor. A simple charging recipe might include (wt. parts): butadiene 75, styrene 25, and solvent 600–800. Initial batch temperature is established by heating or cooling the charge ingredients. Polymerization proceeds rapidly, with addition of a specific amount of alkyllithium initiator that is related to the desired polymer molecular weight. The batch temperature rises adiabatically as monomer conversion proceeds to essentially 100%. If reactor cooling coils are used, the heat removal strategy is mainly concerned with tempering the temperature rise, perhaps to permit a relatively smaller amount of solvent in the charging recipe. When total conversion is achieved (i.e., temperature peak) a

Table 17.14 1,3 Butadiene (99% Minimum): Typical Properties

Conjugated diene content (min. wt. %) 99.0

Peroxides (as hydrogen peroxide) (max. ppm) 5

Total acetylenes (as vinyl acetylene) (max. ppm) 100

Methyl acetylene (max. ppm) 10

Ethyl acetylene (max. ppm) 50

1,2 Butadiene (max. ppm) 150

Carbonyl (as acetaldehyde) (max. ppm) 50

Sulfur (as hydrogen sulfide) (max. ppm) 5

Butadiene dimer (max. wt. %) 0.1

Nonvolatile residue (max. wt. %) 0.1

4-Tertiary butylcatechol (min. ppm as shipped) 100

Methanol (max. ppm) 15

Allene (max. ppm) 10

Oxygen content of vapor over liquid in filled tank car (max. vol. %) 0.3

C5 Hydrocarbon (difference in boiling point between 2.0 and 0.5 ml residue volumes in Cottrell boiler) (max. °C)

0.4



Bottle polymerization test (after drying) Pass



Table 17.15 Styrene (Synthetic Rubber Grade)

Specifications Typical properties

Purity 99.2 (wt.% min.) 99.6 (wt.%)

Aldehydes (as benzaldehyde) 0.01 (wt.% max.)

Peroxides (as hydrogen peroxide) 0.001 (wt.% max.) 99.6 (wt.%)

Sulfur (as S) 0.002 (wt.% max.) 0.001 (wt.%)

Chlorides (as C1) 0.002 (wt.% max.) 0.0005 (wt.%)

Color Pt-Co scale (APHA) 20 (max.) 0.001 (wt.%)

Density 0.9127

4-Tert-Butylcatechol (ppm) 10–20

15 for 30 day storage at max. of 90°F

Polymerization activity test Pass

Bottle polymerization test Pass

shortstop is added to terminate the polymer lithium. The hot solution is transferred to a concentration step to flash part of the solvent for direct recycle to the reactor. The concentrated solution flows to one of several large blend tanks. Antioxidant solution may be added directly to the reactor after the shortstop, to the solution transfer line, or into the blend tank. Batch cycle time is typically near 2 h and may vary somewhat with the polymer type.

Table 17.16 98% Cyclohexane

Specifications

Property Typical Minimum Maximum

Cyclohexane content (wt.%) 98.5 98.0 —

(mol%) 98.9 — —

Sulfur content (ppm) 1 — 5

Benzene plus toluene (ppm) 25 — 100

Distillation evap. (°C at 760 mm)

IBP 80.7 79.5 —

DP 81.0 — 82.0

Specific gravity (60°/60°F) 0.7821 0.777 0.787

Freezing point (°C) 3.7 — —

Saybolt color +30 +30 —



Copper corrosion (2 h at 212°F) 1 — 1

Doctor test Negative Negative —

Nonvolatile matter (g/100 ml) 0.0005 — 0.001



The reactor-charging recipe and the order in which the monomers, initiator, and various modifying chemicals are charged can be changed to permit the production of a very broad variety of polymer types. Some examples follow:

1. The ratio of styrene and butadiene monomers can be changed to give copolymers with widely differing properties.

2. Isoprene can be substituted for butadiene to give styrene—isoprene copolymers. Terpolymers can also be produced.

3. Coupling agents such as SiCl4 can be substituted for the shortstop or a divinyl monomer, such as divinylbenzenes, can be added in the polymerization recipe to introduce long-chain branching. Multilithuim initiatory can be used to form star polymers.

4. Reaction-modifying chemicals can be used to control the diene addition microstructure and/or randomize the distribution of monomer units in the copolymer chain.

5. Incremental or continuous additions of initiator and monomer(s) are possible.

This array of possible polymers with widely differing rheological behaviors, neat as well as in solutions, pose challenging process design situations. The most important polymerization area design criteria will provide precise charging controls to permit batching repeatability and thorough mixing in the polymerization reactor throughout the change in solution viscosity as the monomers are converted to polymer.

Solution polymers made with alkyllithium initiation are inherently linear and have very narrow molecular weight distributions, while emulsion polymers contain branchings and are broad in molecular weight distributions. These differences are reflected by the better processing characteristics of the emulsion SBR and improved hysteresis properties and better abrasion resistance of the solution counterpart. In commercial productions of solution rubbers of this type (solution SBR and polybutadiene rubbers), several methods are employed to broaden the molecular weight distribution and/or introduce long-chain branching. One can add alkyllithium initiator in increments to spread out molecular weights. One can add a reactive multifunctional compound as coupling agent at the end of polymerization to couple three or more polymer chain ends together to form branched macromolecules. Divinylbenzenes are often used in the polymerization recipe to introduce random long-chain branching. These two chemical methods also tend to broaden the molecular weight distribution of the final product. The coupling reactions are much less effective for the copolymers than butadiene homopolymers. In copolymers, the last unit is often the styrene and the living end is much more sterically hindered. By the inclusion of a conjugated diene, such as butadiene in a coupling procedure, the efficiency is greatly increased [21].

In the copolymerization of butadiene (or isoprene) and styrene, the reactivity ratios are influenced by the solvent. In hydrocarbon solvent, the alkyllithium-



initiated copolymerization leads to a tapered block copolymer, one segment being a butadiene-rich tapered copolymer and the other, a polystyrene block with an overall B/S-S structure. However, addition of a small amount of ether or tertiary amine results in the elimination of block formation. Thus a random copolymer can be produced. The use of ether or tertiary amine also increases the vinyl unsaturation from ˜10% to 25% or higher (based on butadiene units) in the copolymer. Vinyl unsaturation comes from the 1,2-addition of butadiene unit. Tetrahydrofuran is the most commonly used reagent to randomize the copolymerization. By using a compound such as potassium tert-butoxide, random copolymers are produced with little change of microstructure. Butadiene/styrene random copolymers can also be prepared by slow and continuous addition of monomers or by incremental addition of butadiene to a styrene-rich monomer mixture during polymerization. These two methods would produce copolymers of low vinyl unsaturation, ˜10%. Emulsion copolymer contains around 18% vinyl unsaturation.

Blended solution containing approximately 10–20% polymer and extender oil (when used) is continuously metered to a stripping vessel. The solvent is vaporized overhead with steam and the contained polymer (with oil) agglomerates into small rice-like particles that are carried from the process as a slurry in water. A surfactant may be added to the process to control polymer particle size. The condensed overhead solvent phase is transferred to solvent purification for recycle to polymerization. The rubber-in-water slurry is collected in a surge tank.

The rubber—water slurry containing approximately 5–10% rubber solids is continuously transferred to a screw press or similar device to remove most of the water for recycling to the solvent recovery step. Final drying of wet rubber from the press can be accomplished by several means. An extruder dryer can produce pelleted products, a form that is favored for plastics modification. As an alternative, the extruder-dried polymers can be compressed into 75 pound rectangular bales if the product is destined for classic, compounded rubber applications. Furthermore, an optional system is available commercially to produce a fine, granular product. In this system, the energy of grinding is used to dry the rubber evaporatively.

The pelleted and granular product forms will often be surface treated to minimize agglomeration to a package-sized mass. Other drying systems, including apron dryers, are applicable in solution SBR production.

D. Solution SBR Processing Alternatives

Important commercial examples of solution SBR and related processes exist that represent significant departures from the process described above. In the polymerization area, continuous polymerization is practiced to produce solution SBR and polybutadiene polymers with relatively broader molecular weight distributions.



Continuous process to produce random copolymers was extensively practiced by the Firestone Tire & Rubber Company in the United States and Asahi Chemical Industry in Japan. By omitting the use of a polar compound as the randomizer, the process produces a low vinyl content of about 9% [22,23]. Phillips also developed a process to produce random copolymers by continuous addition of monomers without a chemical modifier [24]. However, it was never put in commercial operation. Instead, low-vinyl copolymers are produced commercially with the use of alkyllithium-potassium alkoxide initiator [26]. These low-vinyl solution SBRs produced by the continuous process were thoroughly studied regarding their structure/property relationships [19,20,25]. Tables 17.12 and 17.13 illustrate some of the differences in properties between copolymers with different vinyl contents.

All three processes, namely continuous, ether randomization, and potassium alkoxide randomization, are practiced commercially around the world. Likewise, extrusion desolventizing is commercially practiced to replace the steam stripping, rubber dewatering, and drying steps. Direct evaporative drying of solution polymers on rotary drum dryers has also been practiced on a commercial scale.

E. Rubber Processing and Compounding

In general, solution SBRs have less branchings and narrower molecular weight distribution. As a consequence, stocks containing these polymers display reduced nerve and shrinkage. To take advantage of this property in internal mixers, it is necessary to employ larger batches than would be used when mixing emulsion rubbers [27]. Larger batch sizes result in better mixing efficiency (larger output per unit of time), better dispersion of fillers, and faster take-up of plasticizers or extender oils. The low shrinkage and nerve exhibited by the solution rubbers give a tendency to band loosely on the mill roll and in some cases to stick to the back roll. This can be prevented by maintaining low stock temperatures and by maintaining the front roll at a lower temperature than the back roll (20–30°C difference). Loose banding or bagging can often be eliminated by changing the distance between the two rolls (mill gauge). The gauge is increased or decreased depending on the type and condition of roll mill and the type compound. Some improvement in mill banding can oftenbe obtained by reducing the speed ratio between the two rolls.

Because of the low “green strength” of solution SBRs it is advisable to feed the extruder with a thicker but narrower band (strip) than would normally with emulsion SBR. A roller feeder at the throat of the extruder is helpful. Benefits can be gained by undercutting the screw and/or blocking a portion of the feeder entrance. It is helpful to maintain approximately 30°C differential between the extruder barrel and screw. Whether to heat the barrel and cool the screw or cool the barrel and heat the screw depends on the equipment and the compound.



Table 17.17 Typical Solution Butadiene/Styrene Random Copolymers and Uses

Butadiene/styrene ratio

Oil (phr)

Typical Mooney (ML-4)

Uses

75/25 0 58 General purpose: tires, footwear, mechanical goods

75/25 0 75 Plastics modification

75/25 0 32 Sponge products

52/48 0 45 Footwear, sponge products

75/25 37.5a 50 Footwear, mechanical goods

75/25 50a 50 Footwear, mechanical goods

75/25 37.5b 50 Tires

phr, Parts per hundred rubber.

aNaphthenic oil.

bHighly aromatic oil

Source: Ref. 24.

The different mill banding and extrusion characteristics of the solution and emulsion tire polymers can be eliminated by blending the two polymers.

In summary, the differences in mill banding and extrusion characteristics between emulsion and solution SBR tire rubbers can be offset by blending polymers or changes in equipment or procedures. Furthermore, the new versions of solution rubbers have much improved processing characteristics. The solution SBR may require a slightly higher level of curatives due to the higher percentage rubber hydrocarbon level of the solution polymer. Solution rubber can also tolerate higher levels of carbon black and processing oil.

F. Applications

The solution SBRs tend to be used in the same applications as emulsion copolymers. In some applications, the better hysteresis properties and low color of the solution rubbers are at advantage. Typical applications are summarized in Tables 17.17 and 17.18.

III. Styrene-Isoprene-Butadiene-Rubbers

The latest example of harnessing the versitilities of anionic polymerization to produce a tailor-made product is the production of styrene-isoprene-butadiene rubber (SIBR). Halasa and his co-workers at Goodyear [28] employed a chain of single-stirred reactors (CSTRs) wherein the choice of varying the respective steady-state concentration of any individual monomer and modifier level at any degree of conversion is attainable by multiple feedings of monomer and modifier



Table 17.18 Uses of Styrene-Butadiene Rubber

Pneumatic Tires Hard Rubber Molded Goods (cont.)

Tread cap Steering wheels Seals

Tread base Caster wheels O-rings

Ply coats Battery boxes Motor mounts

Inner liner Bowling balls Stoppers

Bead insulation Tank lining Gaskets

Retreads (treads) Rubber-Covered Rolls Bumpers

White sidewalls Textile finishing Wiper blades

Undertread Typewriter Brake boot

Flat Belting Printing Brake cup

Skim coat Paper mill Brake pedal cover

Cover Hard rubber Pipe gasket

Friction Sporting Goods Bumpers

V-belt Protective equipment Trank tread pads

Jacket Padding Hose

Base Swim fins Washing machine

Footwear Rafts Sand blast tubing

Unit soles Tennis balls Milk tubing

Heels Basket balls Flooring

Calendered out soles Bladders for balls Floor tile

Injection molded soles Grips Cove base

Cellular soling Molded Goods Wire and Cable

Slab soling Floor mats Insulation

Integral soling Mud flaps Jacket

Sports soling Sink mats Rubber-Covered Fabrics

Direct molded soles Spark plug boots Rain coats

Microcellular soling Diaphragms Rain boots

Sponge Rubber Products

Bathtub mats Bicycle Tires

Door seals

Window seals

Armrest



(TMEDA) along the reactor chain to achieve simultaneous control of composition, microstructure, and sequence distribution. These elastomers are said to have diverse visoelastic response to suit specific tire applications.

In one of their patents [29], an example of the the SIBR is described as a terpolymer rubber comprised of about 10–35 weight % bound styrene, about 30–50 weight % bound isoprene, and about 30–40 weight % bound butadiene. The terpolymer is characterized by having a Tg of about -40°C to about -10°C. Furthermore, the bound butadiene structure contains about 30–40 weight % vinyl units, the bound isoprene structures contains about 10–30 weight % isopropenyl



units, and the sum of the percentage of vinyl units and isopenyl units is in the range of about 40–70%. This terpolymer rubber can be blended with cis-PI rubber, cis-PB rubber and/or natural rubber to give a tread compound that is claimed [29,30] to provide a balanced rolling resistance, traction, and wear.

Another patent [31] discloses a segmented polymer comprised of a first segment of polyisoprene having 75–98 weight % 1,4-linkages and a second segment of an “essentially random” copolymer of styrene and butadiene. The second segment is comprised of 30–95% butadiene and from 5–70% styrene, wherein (%S × 1.7) + (% vinyl) = 50–92%. The segmented SIBR can typically be prepared by initiating the polymerization is isoprene with alkyllithium in hydro-carbon solvent in the first reactor and then copolymerizing the butadiene/styrene monomer mixture onto the living polyisoprene segments in the presence of a modifier such as N,N,N'N'-tetramethylethylene diamine in the second reactor.

Another example of an SIBR that has multiple Tgs and exhibits multiple viscoelastic responses is said to be prepared in one single reactor by terpolymerization of the three monomers in an organic solvent at a temperature of no more than about 40°C in the presence of an alkyllithium compound plus tripiperdino phosphine oxide or alkali metal alkoxide such as potassium amylate [32]. In this process styrene and butadiene are copolymerized rapidly first, and then isoprene is polymerized at much slower rate to form an essentially diblock copolymer with two glass transition temperatures. It is claimed that by utilizing such terpolymers in tire treads, tires having improved wet skid resistance can be built without sacrificing rolling resistance or tread wear chracteristics.

Tire tread compounds can be made using segmented SIBR without the need to blend with additional rubbers. However, it is claimed [29–33] that, in many cases, it will be desirable to blend the segmented SIBR with one or more additional rubbers to attain the desired performance characteristics for the tire tread compound.

It is reported [28–32] that unique morphologies are attainable for SIBR ranging from lamellar to spherical. Some of the enhanced physical properties of the SIBR vulcanizates compared with the blends of SBR and polyisoprene having the same monomer ratios are attributed, at least in part, to the unique morphology.

It is publicized by The Goodyear Tire & Rubber Company that some high-performance tires and all-season tires introduced in the recent years contain some form of SIBR in the tread compound.

One can employ similar technique [28,33] to produce diene block copolymers having diverse viscoelastic response to suit specific tire applications. One patent [33] described the use of two reactors in series to produce diene block polymers having multiple glass transition temperatures. One example showed a copolymer of butadiene and isoprene having glass transition temperatures of -47°C, 24°C, and -10°C. It is a segmented copolymer consisting of a polybutadiene segment with medium vinyl configuration, a copolymer segment, and a



polyisoprene segment with very high isopropenyl configuration. In another example a block polymer of butadiene consisting of low-vinyl segment, intermediate-vinyl segment, and high-vinyl segment was illustrated. Diene and styrene copolymers with multiple segments can also be produced in this two-reactor process.

IV. Polymers from Barium-Based Catalysts

Several catalysts based on organobarium compounds and barium salts were developed by Gencorp (formerly General Tire & Rubber Corp.) [34,35]. In one system, compounds such as dixanthenyl barium and diphenylmethyl barium were prepared by a metallation and electron transfer reactions of barium with the appropriate hydrocarbon. Monomers such as butadiene and styrene and polar monomers such as methyl methacrylate are polymerized with these compounds. In another system, an alkyllithium such as butyllithium is combined with a barium salt such as barium t-butoxide/hydroxide to form the catalyst. Styrene—butadiene rubbers can be prepared having a unique structure with this catalyst. At vinyl contents of ˜10%, the trans content can be as high as 78%. The molecular weight distributions of the rubbers are broad and bimodal. These rubbers can undergo strain-induced crystallization analogous to natural rubber. They have good building tack and green strength. In a third system, an organomagnesium compound with an organoaluminum compound and an alkaline earth salt such as barium t-butoxide form an effective polymerization catalyst for butadiene and styrene. It was reported that high trans butadiene rubbers and styrene—butadiene rubbers are prepared with this catalyst, which have outstanding building tack and green strength for a synthetic rubber and equal or better than natural rubber. These synthetic rubbers are claimed to be useful as a replacement for natural rubber in vulcanizates requiring improved oxidative and fatigue resistance [18,36]. The trans-polybutadiene segment of these rubbers can contain as high as 90% trans with vinyl contents as low as 2–4%. The polymerizations with this catalyst system were reported to be “living” and block copolymers, functional group addition, and molecular weight and molecular weight distribution control can be readily attained. Rubber blends containing high trans SBRs have given advantages as tread rubbers, having low rolling resistance and high wet traction. These anionic polymerization processes by which crystallizing butadiene rubbers and styrene—butadiene copolymers can be produced are not believed to be in use commercially.

V. Lubricating Oil Additives

Selectively hydrogenated copolymers of vinyl aromatic monomers and dienes are used as viscosity index (VI) improvers in multigrade lubricating oils. Both random and block copolymers are described in the patent literature [37–40]. The block



polymers are of the AB-type, where B is most commonly styrene and A is butadiene, isoprene, or a random copolymer of styrene and the appropriate diene. Hydrogenation is essential to impart resistance to oxidative degradation. Crystallinity is undesirable and is prevented by control of microstructure and randomly incorporated styrene. Molecular weights range typically from 50,000 to 150,000, with the characteristically narrow molecular weight distribution of organolithium-initiated solution polymers.

An outstanding property of these polymers is their shear stability. Johnson and O'Shaughnessy [41] have published sonic shear stability test results (ASTM Method 2603) for several commercial and experimental oils containing various VI improvers. Their results are summarized in Table 17.19. To a considerable degree, the excellent stability of the hydrogenated diene—styrene polymers is attributable to their relatively low molecular weight and narrow molecular weight distribution, consistent with the established theory of shear degradation of polymers [42].

The viscosity of a heavy lubricating oil stock decreases with temperature more rapidly than that of a lighter stock. When a VI-improving polymer is added to a light oil to match the viscosity of a heavier oil at high temperature, the modified oil, although thickened, must still be more fluid than the heavy oil at low temperatures. Therefore, the desired property of a VI improver is a relative viscosity that remains constant or decreases as the temperature falls. Decreases in relative viscosity are difficult to achieve in practice and they are usually of modest magnitude. Figure 17.5 shows the relative viscosity of a high-quality methacrylate-type VI improver (Mw = 540,000, Mn = 150,000, c = 2.6% by wt.) in a typical lubricating oil as a function of temperature and shear rate. Similar data for a hydrogenated butadiene-styrene block polymer (Mw = 79,000, Mn = 60,000, total styrene = 59%, block styrene = 20%, c = 2.35%) are shown in Figure 17.6. In both instances the relative viscosity is of the order of two over the entire range of conditions. However, in the block polymer the shear rate dependence is minimized at the higher temperatures, even though accentuated at the lower temperatures.

Table 17.19 Shear Stability of Lubricating Oils

VI improver type

% Decrease in viscosity(sonic shear test)

Ethylene-propylene (EP) 2.1–9.7

Styrene-ester (SE) 12.9–17.2

Methacrylate (MA) 4.2–16.8

EP/MA 8.7

Hydrogenated diene-styrene 0.5–2.3

Source: Ref. 41.



Figure 17.5 Relative viscosity of lubricating oil containing methacrylate-type viscosity index improver.

(From Ref. 41, with permission from Soc. of Auto. Eng.)

Figure 17.6 Relative viscosity of

lubricating oil containing hydrogenated butadiene-styrene

block polymer. (From Ref. 41, with permission

from soc. of Auto. Eng.)



The first of these effects is desirable and is a consequence of lower molecular weight and narrower molecular weight distribution. The second has been shown to be the result of micelle formation, the oil being a selective solvent for the block polymer at the lower temperatures. The micelles consist of a polystyrene core with a highly solvated copolymer shell and are apparently not shear-stable, giving rise to increasingly non-Newtonian behavior. Although a practical manifestation of a most interesting effect, which has been described in some detail in the scientific literature [43], the micellar character of these solutions appears to play no decisive role in VI improvement; at the shear rates actually encountered in service, the relative viscosity is much the same as with other types of VI improvers. Moreover, Johnson [44] has found that at yet lower temperatures, the zero-shear relative viscosity again decreases. The flow of micellar block polymer solution is not well understood [45].

The most recent developments in this field are VI improvers with dispersancy properties built into the molecule via chemical modification of the hydrogenated block copolymer [46,47]. VI improvers with high thickening power [48] and VI improver soluble in synthetic poly(alpha-olefin) lubricants [49] have also been described. In one patent [50] the use of a small amount of hydrogenated SB random or block copolymer in a hydrocarbon liquid or petroleum distillate to improve its pour-point and/or render it more resistant to thermal degradation is claimed.

References

1. Br. patent 895,980.


3. W. W. Crouch and J. N. Short, 2nd International Synthetic Rubber Symposium, London, October 1960; Rubber & Plastics Age, 42, No. 3, 276 (1961).

4. Announcement, Rubber Age, 92, No. 6, 870 (1963).

5. H. E. Railsback, C. C. Baird, J. R. Haws, and R. C. Wheat, Rubber Age, 94,583 (1964).

6. Announcement, Chem. & Eng. News, 43, No. 11, 54 (1964).

7. H. L. Hsieh, 3rd International Synthetic Rubber Symposium, London, October 1964; Rubber & Plastics Age, 46, No. 4, 394 (1965).

8. J. T.. Gruver and G. Kraus, J. Polym. Sci., A-2, 797 (1964).

9. R. S. Hanmer and H. E. Railsback, Rubber Age, 73, October 1964, based on a paper presented at the 85th Meeting of the Division of Rubber Chemistry, Am. Chem. Soc., Detroit, MI, 1964.

10. H. E. Railsback and J. R. Haws, Rubber and Plastics Age, October 1967.

11. R. N. Cooper and L. L. Nash, Rubber Age, 55, May 1972.

12. D. G. Moore and G. L. Day, Comparison of Emulsion vs. Solution SBR On Tire Performance, presented at the Akron Rubber Group Meeting, January 24, 1985.

13. C. A. Uraneck and J. N. Short, J. Polym. Sci., 14, 1421 (1970).



14. J. R. Haws, L. L. Nash, and M. S. Wilt, presented at the Third Australian Rubber Technology Convention, September 1974; Rubber Industry, 107, June 1975.

15. C. R. Wilder, J. R. Haws, and T. C. Middlebrook, presented at the 124th Meeting of The Rubber Division, ACS, Houston, October 1983; Kautsch. Gumi. Kunstst., 39, 683, August 1984.

16. K. H. Nordsich, Kautsch, Gumi Kunstst. 38, 178, (1982).

17. R. Bond, Proc. Royal Soc., London, A399, 1 (1985).

18. S. L. Aggarwal, I. G. Hargis, R. A. Livigni, H. J. Fabris, and L. F. Marker, in Advances in Elastomers and Rubber Elasticity, J. Lal and J. E. Mark, Eds., Plenum Press, New York, 1986, p. 17.

19. F. C. Weissent and B. L. Johnson, Rubber Chem. & Tech., 40, 590 (1967).

20. H. E. Railsback, W. S. Howard, and N. A. Stumpe, Jr., Rubber Age, April (1974).

21. R. C. Farrar and C. F. Wofford, US patent 3,692,874.

22. Br. patent 994,726.

23. N. F. Keckler, US patent 3,558,575; Canadian patent 769,096.

24. J. N. Short, US patent 3,094,512.

25. J. M. Willis and W. W. Barbin, Rubber Age, 100, 53 (1968).

26. C. F. Wofford and H. L. Hsieh, J. Polym. Sci., A-1, 7, 461 (1969); C. F. Wofford, US patents 3,294,768, 3,496,154, and 3,498,960.

27. J. R. Haws, Rubber & Plastics Age, 46, No. 16, 1144 (1965).

28. A. F. Halasa, B. Gross, B. Hsu, and C. C. Chang, Eur. Rubber J., 36, June (1990); Rubber & Plastics News, 63, May 26, 1990.

29. A. F. Halasa, J. Bergh, and F. A. J. Fourgon US patent 5,047,483.

30. A. F. Halasa, J. Bergh, and F. A. J. Fourgon, US patents 5,159,020 and 5,191,021.

31. W.-L. Hsu and A. F. Halasa, US patent 5,070,148.

32. W.-L. Hsu and A. F. Halasa, US patent 5,137,998.

33. A. F. Halasa, B. B. Gross, and J. L. Cox, US patent 4,843,120.

34. I. G. Hargis and R. A. Livigni, US patents 3,846,385; 3,907,933; 3,928,302; 3,965,080; 3,966,638; and 44,012,336.

35. I. G. Hargis, R. A. Livigni, and S. L. Aggarwal, US patents 3,992,501 and 4,026,115.

36. I. G. Hargis, R. A. Livigni, and S. L. Aggarwal, in ACS Symposium Series 193, J. Mask and J. Lal, Eds., American Chemical Society, Washington, D.C., 1982.

37. S. Schiff, M. M. Johnson, and W. L. Streets, US patent 3,554,911.



46. W. J. Trepka, US patents 4,402,843; 4,402,844; 4,145,298; also Macromolecules, 17, 497 (1984).

47. T. E. Kiovski, US patent 4,033,888.


49. S. Schiff and W. J. Trepka, US patent 4,418,234.

50. W. L. Street, US patent 3,419,365.



18 Styrenic Thermoplastic Elastomers

I. Introduction

Thermoplastic elastomers (TPE) combine the flexibility and impact resistance of rubbers with the strength and easy processability of thermoplastics. In addition, they have frictional properties and hardness that are generally intermediate between those of conventional rubbers and those of thermoplastics. The TPE can be repeatedly processed and molded, and retains its elastomeric properties.

Anionic polystyrene-polydiene-polystyrene triblock copolymers and their branched (radial) versions are one of the several important industrial TPEs. Others are thermoplastic urethane elastomers, thermoplastic polyester elastomers, ionomeric thermoplastic elastomers, poly-α-olefin-based thermoplastic elastomers, and some other block copolymers such as polyesteramides, polyetheresteramides, polyetheramides, and others [1]. In addition, some TPEs are blends of rubber (e.g., EPDM rubber) and thermoplastics (e.g., polypropylene). The readers can find detailed discussions on various TPEs in a comprehensive review edited by Legge and colleagues [1].



II. Historical Developments of Diene/styrene Block Copolymers as TPE

A. Introduction

In the late 1950s, Szwarc [2,3] coined the term “living polymer,” while several major industrial laboratories were having an active research program on anionic polymerizations of dienes and vinyl aromatic monomers. In the following few years, scientists at the Shell and Phillips laboratories, independently and at almost the same time, discovered the diene/styrene thermoplastic elastomers by polymerizing the two monomers in sequence (or sequences) to form the triblock copolymers. These new discoveries literally started a new industry.

It is interesting to note that Tobolsky [4] in 1959 predicted that new block polymers might be synthesized that have high crystalline regions and amorphous regions of low Tg. The general macrostructure of styrenic TPE is two glassy end blocks connected by the amorphous, elastomeric polydiene block.

B. Block Copolymers

Copolymers are polymer products formed by polymerizing two monomers, M1 and M2, into the same polymer chains. The copolymer chains may be a mixture of units of the two monomers with both randomly distributed throughout the polymer chain. As an alternative, individual segments of the chain may contain only one monomer unit followed by a segment containing only the other monomer unit. The first is a random copolymer, whereas the second is a type of block copolymer. Molecules of these two types of polymers may be visualized as follows:

-[A-A-B-A-A-B-A-B] x - -[A-A-A-A-A] x -[B-B-B] x -

A/B Random Copolymer Molecul A-B Block Copolymer Molecule

The difference between the two can be seen by examination of a section of the polymer chain. In truly random copolymer molecules, any section of the chain containing > 10 monomer units would show a constant composition of monomer unit A and B reflecting the ratio of the two monomers in the original charge. By contrast, if the A and B units of the polymer chain are arranged in segments of all A units connected to segments of all B units, we have a simple A-B block copolymer molecule. An examination of any section of the chain (except that containing the A-B joining bond) would show either all A or all B units depending on which part was examined. An example of this would be a copolymer consisting of a block of polystyrene (A-A-A-A-A) connected to a block of polybutadiene (B-B-B). For simplicity, this polymer would be designated as an A-B block copolymer, and A and B would now refer to blocks of monomer units rather than to single monomer units. The dash between the A and B says the two



segments are connected to each block, A—B structure or a more complex A—B—A triblock or even more complex A-B-A-B-A multiblock types. Block copolymers also can contain, for example, molecules that contain a block of A and a block of a mixture of copolymerized A/B monomer. Such a block copolymer molecule can be represented by A-(A/B).

Simple block copolymers, since they contain segments of two or more totally different polymer types, will have some properties common to each type of homopolymer. However, when the two polymer segments are connected, some new characteristics will be exhibited. For example, polystyrene is not appreciably soluble in polybutadiene. A mixture of the two polymers will be translucent at best and perhaps opaque depending on the percentage polystyrene in the mixture as well as its molecular weight. By contrast, a block copolymer of the same overall composition will be transparent. The appeal of block copolymers and their commercial importance is usually due to that rare combination of properties created by the hybridization of the homopolymers. The morphology or phase-separation plays a key role in the properties of the block polymers.

Block copolymers can have very special properties by virtue of the particular



arrangement of, for example, glassy blocks of polystyrene and rubbery blocks of polybutadiene. The polystyrene blocks at room temperature are some 75°C below their glass transition temperature while the polybutadiene blocks are some 110°C above theirs. The block copolymer exhibits both glass transitions when taken through the -100 to 150°C range.

The glassy polystyrene block is not soluble to any extent in the rubber portion of the polymer and a simple mixture of the two homopolymers would tend to separate to produce an opaque mixture of resins. However, when the blocks are tied together on a molecular level, as in an A-B block copolymer molecule, they cannot separate on a macroscale and, therefore, tend to separate on a microscale.

Simple butadiene-styrene (B-S) or A-B block copolymers appear to be homogeneous and indeed exhibit great clarity when molded into sheet. Nonetheless, the two blocks are still basically incompatible and while massive separation may be prevented by the carbon-carbon links between the blocks, separation will still occur. On a microscale, the polystyrene blocks tend to separate from the rubbery polybutadiene blocks and associate with similar blocks from other polymer chains. The polystyrene blocks concentrate to form into domains of glassy polystyrene in a rubbery block matrix. The domain size is in the 100–400A size, which cannot be seen by the naked eye. Therefore, the molded polymers appear to be homogeneous and transparent. The products are rubbery and yet exhibit two glass transitions typical of the two homopolymers. Such products will exhibit no green strength (raw strength) and for most applications will have to be vulcanized to develop the strength needed in molded articles.

B-S-B block copolymers also exhibit phase separation and again the polystyrene center blocks associate to form the domain structure. The only difference is that each polystyrene block has two rubbery blocks hanging into the rubbery matrix. Again, the polymer must be crosslinked or vulcanized to develop tensile

strength. For more detailed discussions on copolymerization, block copolymers and star-branched polymers, see Chapters 10, 12, and 13, respectively.



Styrenic Thermoplastic Elastomer

If the blocks are arranged with the rubbery (soft) block in the middle of two polystyrene (hard) blocks, S-B-S type, an unusual, commercially useful property results, namely green tensile strength. Again, as with the other block copolymers, the polystyrene blocks associate in domains, but since there is one at each end of a long rubbery block, a single molecule may be connected to two different domains. Indeed most of the blocks appear to be arranged in this fashion. As long as the polystyrenes are in their glassy, rigid state, the product will appear as a network structure with the rubbery chains anchored between two fixed domains.

S-D-S Triblock Copolymer (Thermoplastic Elastomer)

The terms S-D-S, TPE, and TPR are used interchangeably. These copolymers are characterized as follows:

Two phases

Two Tgs

Characterized by:

High raw strength

Complete solubility

Reversible thermoplasticity

In this form, as long as the polystyrenes are below their glass transition temperature, the polymer has good tensile strength without chemical crosslinking or vulcanization. When the temperature of the polymer is raised above the glass transition temperature of the polystyrene segments (usually 100–110°C), the glassy domains soften and the resin loses its strength because the chains are no longer anchored at each end of the rubbery blocks. When the resin temperature falls below 100°C, the block structure reforms and green strength returns.

Elastomeric block copolymers TPE or TPR are useful in a variety of fields. The unique behavior of A-B-A and (AB)x star block copolymers combines the features that were previously considered to be mutually exclusive, namely, thermoplasticity together with elastic behavior.



C. Kraton TPE and Shell Chemical Company

The discovery and development of Kraton thermoplastic elastomers at Shell were described in detail by Legge, the medalist, in his 1987 Charles Goodyear Medal address entitled, Thermoplastic Elastomers [5].

Legge's address is briefly summarized here:

The discovery was made while Shell's researchers were engaged in means to impart some green strength to cis-polyisoprene rubber made with alkyllithium, a Shell product at that time.

At the time of the discovery of S-I-S copolymer possessing green tensile strength, Legge was the Director of Elastomer Research Laboratory (Torrance, CA). He transferred from Shell Development in Emeryville to the Director position at Torrance in January, 1961.

Legge and his senior staff believed the discovery was a major scientific break-through and received immediate and strong support from the managers of Synthetic Rubber Division and Shell Development.

Obtaining broad coverage of the triblock copolymers through patent applications was considered to be the major task. The basic US Patent on the triblock copolymers of styrene and dienes was filed in January, 1962, and issued in 1964 [6]. G. Holden and R. Milkovich were the inventors. Milkovich received his M.S. degree in 1957 from the University of Syracuse, and with M. Levy developed the all-glass, high-vacuum technique and established the living electron-transfer polymerization in Prof. Szwarc's laboratory [3]. Subsequently he moved to Prof. M. Morton's laboratory at the University of Akron and received his Ph.D. degree in 1959. He joined Shell following his graduation.

The thermoplastic elastomers based on block copolymers of styrene and butadiene were announced in 1965 [7,8]. A series of papers were presented in 1967–1968 describing the details of the triblock structure, the theory of the thermoplastic elastomer behavior, and some structure limitations.

The trade name, Kraton, was coined by Tom Baron, then Manager, R&D, in the Synthetic Rubber Division, later President of Shell Development Company. From a book on greek mythology he was reading, he selected the name of a god of strength, Kratos, and changed the last letter to give a technological sound.

In 1972, Shell announced the Kraton G, an S-EB-S triblock copolymer. It is a product of hydrogenated S-B-S copolymer.

Legge considered the successful innovation of TPR came primarily from technological “push,” with very little market demand “pull.”

Legge concluded that Shell's success was due to several favorable elements:

Patentable technical breakthrough.

Ample pilot plant capacity at the time of discovery.



By happy change, there was thermoplastic processing and testing equipment on hand.

A small and very capable research and development staff.

Management was oriented toward new products and overall company management was willing to support new ventures in polymers.

D. Solprene Rubbers and Phillips Petroleum Company

The discovery and development of Solprene TPE was reviewed by Hsieh [9]. Here is the summary of his presentation:

Hsieh, in 1957, joined Phillips and explored the preparations of dilithium initiators with main objectives of synthesizing polymers with reactive functional end groups (telechelic polymers, see Chap. 22).

Zelinski, in the same laboratory, had prepared tapered diene-styrene block copolymers in hydrocarbon solvent and diene/styrene random copolymers in the presence of a polar adjuvant in early 1957.

By 1958 several dilithium initiators were successfully prepared in ether and carboxy- and hydroxy-telechelic polybutadienes were prepared. Zelinski used the same initiators to prepare S-B-S and B-S-B triblock copolymers by incremental additions of monomers. The term teleblock copolymer was coined within Phillips to describe the triblock copolymers. The word of teleblock (distance blocks) was coined shortly after the term telechelic (distance functional groups) was used internally at Phillips.

Kraus, in 1958, at Phillips carried out a study of the stress-strain behavior of a series of S-B-S and B-S-B “teleblock copolymers” having S/B ratios from 30/70 to 70/30. Kraus observed [10] that S-B-S types of copolymers exhibited high elasticity and tensile strength without the benefits of crosslinking or filler reinforcement and that S-B-S teleblock copolymers exhibited “a type of physical crosslinking by association of styrene blocks.”

Encouraged by the earlier results, Phillips researchers used alkyllithium initiator to synthesize systematically four basic types of ordered copolymers of styrene and dienes, namely random copolymers (S/D), simple diblock copolymers (S-D), teleblock (triblock) copolymers (S-D-S), and multiblock copolymers (S-(-D-S-)n-D). These experimental polymers were extensively examined for their mechanical and rheological properties.

During the period of 1960–1962, the priority at Phillips was to bring the B-S tapered block copolymer (Solprene 1205), B/S random copolymer or solution SBRs (e.g., Solprene 1204 and also Solprene 300 series), cispolybutadiene rubber (Cis-4), medium-cis polybutadiene rubber (Solprene 200 series), and carboxy- and hydroxy-telechelic liquid polybutadiene (Butarez CTL and Butarez HT) into commercial production. Teleblock copolymers remained in the laboratory.



S-B-S teleblock thermoplastic elastomers in radial (three- or four-armed star branched) form were produced in development quality in the pilot plant in 1964 and were not put in commercial production until 1973 (Solprene 400 series).

By 1977, no fewer than a dozen types of Solprene 400 series products were commercially produced for various applications. Several hydrogenated versions were semicommercially produced as Solprene 500 series.

The TPE plant in Borger, Texas, was sold to Synthetic Rubber Corp., Taiwan, in April, 1986. It was moved to Taiwan and has been in full production since January, 1988. The plant has since expanded its capacity and product types.

E. Past and Future

Shell introduced its Kraton commercially in 1965 as the first producer of styrenic TPE. It started at a modest 2000 metric ton production in 1965 and reached around 30,000 metric tons in 1972 (Chem System estimates). Phillips started the commercial production of its radial copolymers in 1973. From 1973 to 1982, Shell and Phillips provided the bulk of styrenic TPE to the marketplace. The total production in 1982 reached nearly 80,000 metric tons. After Phillips stopped its production in 1982, Shell became the dominant producer, with a small portion of the market being served by Firestone since 1981. Firestone had an active research group (Tate, Halasa, Schulz, and others) in the 1960s and 1970s in anionic polymerization.

Several major rubber and chemical companies in the United States, Western Europe, Japan, and China also had active R&D programs focusing on styrenic block copolymers during this period.

According to a recent article, Royal Dutch/Shell will spend $170 million through 1991 to boost worldwide capacity 39% to 320,000 metric tons (m.t.)/year [11]. According to another article, new plants in the United States are planned by leading importer Enichem, in a joint venture with Arco (since then Arco has withdrawn from the joint venture), and newcomers Dow and Exxon in their Dexco joint venture [12]. Since that publication, Dexco has begun production (see Table 18.1). Enichem, the second-ranking producer worldwide with around 70,000 m.t./year capacity, Petrofina, and Repsol Quimica (all once licensees of Phillips) have added capacity or are doing so. Taiwan Synthetic Rubber Corp., as mentioned earlier, has restarted the ex-Phillips plant it shipped from Borger, Texas, with an annual capacity of 25,000 m.t./year. Japan has a production capacity of over 50,000 m.t./year at Asahi, Nippon Zeon, Japan Elastomer Company (joint venture of Asahi and Showa Denko), and JSSR. Forecasts by Chem. Systems (Tarrytown, NY) show a U.S. demand of 145,000 m.t./year and exports of 30,000 m.t./year in 1991. Europe's and Japan's demands and exports should reach 163,000 m.t./year and 34,000 m.t./year, respectively (Table 18.1).



Table 18.1 Producers of Styrenic Thermoplastic Elastomers

Producer

Trade name

Plant capacitya,b

(1000 tons/yr)

Comments

United States

Shell Chemicals Kraton 120

Dexco Polymers Vector 32 Dow Chemicals and Exxon Chemicals

Firestone Synthetic Rubber & Latex

Stereon 19

Europe

Enichem Elastomer Europrene 68

Petrofina NV Finaprene 40 Philips licensee

Rheinische Olefinwerke Ceriflex 35 Deutsch Shell and BASF

Repsol Quimica Calprene 40 Philips licensee

Shell Chimie, S.A. Cariflex 35

Japan

Asahi Chemical Industry Tufprene 11 Firestone licensee

Tuftee

Nippon Elastomer Solprene 10 Philipps licensee

Asaprene

Nippon Geon Quintac 10

Shell JSR Elastomer JSRTR 20

Kuraray — —

Asia (other than Japan)

Taiwan Synthetic Rubber 25 ex-Phillips plant

China SINOPEC — c

aEuropean and Japanese capacity data include extended oils.

bIt is reported that Enichem is building a 37,000 ton/yr plant at Baytown, Texas [13], and Shell plans to build a 20,000 ton/yr plant in Brazil [14].

cCapacity unknown.



III. Synthesis

A. Method 1: The Use of a Dilithium Initiator and Sequential Addition of Monomers

These three types are described here.

For simplicity, R is a very minor moiety of the polymeric chain (Dn) and is not included for this formulation.



B. Method 2: The Use of an Alkyllithium Initiator and Sequential Additions of Monomers

These four types are described here.

C. Method 3: The Use of an Alkyllithium Initiator and Utilizing Coupling (Linking) Reactions

These types are described here.



D. Initiation and Polymerization

Method 1, using a dilithium initiator, may very well be the earliest example of synthesizing triblock copolymers [15]. The initiators used in these examples at Phillips contained some diethyl ether or THF, and therefore, the microstructure of the polybutadiene or polyisoprene chain had somewhat higher vinyl (from 1,2-addition) or isopropenyl (from 3,4-addition) configuration, respectively, than the similar block copolymers prepared in the absence of a polar compound. However, a truly hydrocarbon soluble dilithium compound was not available until some 20 years later.

Dow Chemicals [16,17] disclosed a family of dilithium initiators by reacting soluble difunctional 1,1-diphenylethylenes where the reactive double bonds are separated by a variety of moieties such as p-phenyl, biphenyl, p,p-biphenylether, and m-phenyl with sec-butyllithium. In one case, the initiator was prepared in cyclohexane/toluene solution by reacting sec-butyllithium with 1,3-bis(1-phenylethenyl) benzene in the presence of pentamethyldiethylenetriamine.

Triblock copolymers of S-D-S types were prepared by using this initiator, which is said to be soluble in hydrocarbon solvents.

Hocker and Latterman [18,19] and McGrath and his co-workers [20] confirmed that this metaisomer will react with sec-butyllithium in toluene to form a hydrocarbon-soluble dilithium initiator suitable for the preparation of triblock copolymers.

In Method 2, in which an alkyllithium initiator is used, styrene is polymerized first, followed by a diene monomer and then another increment of styrene monomer. In a series of patents assigned to Shell Chemicals [6,21], sec-butyllithium was the preferred initiator to prepare the linear triblock copolymers. The rate of initiation of styrene with sec-butyllithium is much higher than with n-butyllithium as reported by Hsieh [22]. In fact, the same author had shown that with n-butyllithium, in hydrocarbon polymerization solvent such as cyclohexane, by the time all the styrene was polymerized, substantial amounts of the initiator remained. This would lead to the formation of diblock copolymer (B-S) in the subsequent steps. The presence of diblock polymer in S-D-S TPE is undesirable



1,3-phenylene-bis(3-methyl-1-phenylpentyllidene)bis (lithium)

because of the diblock polymer's inability to participate in the physical network formation.

sec-Butyllithium is more costly and less stable in storage than n-butyllithium. Hsieh [23] discovered that when a trace of THF or Et3N (0.01–0.4 ppm in solvent) is used with n-butyllithium, the initiation rate of styrene is increased severalfold more than the increase in propagation rate. Thus, triblock polymers with excellent stress-strain properties can be produced from the use of n-butyllithium.

In Method 3, styrene is polymerized first also. Either the use of fast initiator such as sec-butyllithium or the use of trace THF or Et2N with n-butyllithium is essential to produce good quality TPE.

Many di- and polyfunctional coupling agents are known and used in the commercial productions of linear and radial block polymers. Patented examples of some more common coupling agents are active dihalo compounds [24], polyepoxides [25], dianhydrides [25], chlorosilanes [26], carbon dioxide or disulfide [26], halides and mixed halides [27], monoesters such as benzoates or acetates [28,29], adipates or terephthalates [30], divinylbenzene [31–33], vinylhalomethylarene [34], and others.

Hsieh [35] studied the coupling efficiencies and degrees of coupling of various coupling agents for the preparations of linear S-B-S and radial (S-B-)-nX block polymers. Laboratory bottle polymerization technique was used in these studies and the results more closely resembled those one could obtain in commercial operations.

Coupling efficiencies were calculated in two ways from gel-permeation chromatographs. For polyfunctional agents that gave good shifts in coupled peak



position from the precursor peak, efficiencies were calculated by measuring the height of the precursor before and after coupling:

Coupling efficiency = 100(Pp - Pc)/Pp

where Pp is the height of peak in the uncoupled precursor and Pc the height of precursor peak in coupled product.

Dichain coupling presented a special problem. This shift in position of the peak from precursor to product was so small that accurate measurement of the residual precursor in the product was difficult because of peak overlap. It is fortunate that dichain products allow convenient calculation of efficiency from the number-average molecular weights by the following formula:

Coupling efficiency = 200(Mc - Mp)/Mc'

where Mc and Mp are the number-average molecular weights of product and precursor, respectively.

The degree of coupling was calculated from the hydrodynamic volume of the peak positions of the gel peurmeation chromatographic (GPC) curves for the precursor and coupled product [36].

The results of these studies are shown in Tables 18.2 and 18.3 and Figures 18.1–18.3.

Figure 18.4 shows the GPC curves from samples after styrene polymerization(s), after butadiene polymerization (S-B), and after coupling with SiCl4((S-B-)-nSi). The significant point needs to be made here is that these samples were taken directly from a Phillips commercial reactor in Borger, Texas.

One can see that in a typical commercial SiCl4-coupled radial block copoly

Table 18.2 Difunctional Coupling Agents

Coupling

Reagent Efficiency (%) Degree

a,a'-Dibromo-p-xylene 94 2.0

a,a'-Dichloro-p-xylene 94 2.0

Bis(chloromethyl)ether 95 2.0

Methylene iodide 94 2.0

Iodine 93 2.0

1,4-Dibromo-2-butene 91 2.0

1,4-Diiodo-2-butene 90 2.0

Ethyl acetate 90 2.0

Terephthalaldehyde 83 2.0

Anthraquinone 81 2.0



Table 18.3 Polyfunctional Coupling Agents

Coupling

Reagent Efficiency (%) Degree

Dimethyl phthalate 90 2.9

Phosphorus trichloride 85 2.9

Methyltrichlorosilane 94 2.9

Silicon tetrachloride 94 3.8

Hexachlorodisilane 93 5.5

Divinylbenzene 94 7–14

Divinylbenzene + butadiene 90 4

mer, no more than 2–3% of the styrene homopolymer and about equal amount of the diblock copolymers are present. The latter have been found [37,38] to have a dramatic effect on the strength of the material. The diblock copolymers are free chain ends in the network. The presence of a certain amount of styrene homopolymer, however, can be tolerated [37,38] since it is apparently incorporated into polystyrene domains.

In large-scale commercial productions, certain amounts of termination occur.

Figure 18.1 GPC curves of the precursor and coupled dichain

linear block polymers. (From Ref. 35, used with permission from Rubber Div., A.C.S.)



Figure 18.2 GPC curves of tri- and tetra-chain radial block polymers.

(From Ref. 35, used with permission from Rubber Div., A.C.S.)

Terminations can occur during the first step. Fast-reacting impurities from solvent or monomer such as water or alcohol are less of a problem overall. These impurities generally destroy the active initiator immediately. In a modern commercial plant equipped with in-line analyzers and computer-controlled automatic initiator level compensations, the predicted (desired) molecular weight of the final product still can be consistently obtained. Molecular weight distribution is also not much affected due to the rapid reaction in the initial stage of the process. Of course, this discussion is assumed only with very low level of the impurities. Compensation of impurities with initiator is economically unwise in a commercial operation. Furthermore, the side reaction products such as lithium alkoxides,

Figure 18.3 GPC curves of divinylbenzene-coupled radial block polymers.

(From Ref. 35, used with permission from Rubber Div., A.C.S.)



Figure 18.4 Intermediate and final products of a typical commercial

butadiene—styrene radial plastomer (PZN, polymerization). (From Ref. 35, used with

permission from Rubber Div., A.C.S.)

hydroxide, or sulfides in any appreciate amounts do affect the rates of initiation and polymerization [39].

Terminations caused by slow-reacting impurities such as 1-alkynes are more troublesome. In the first increment, these side reactions lead to homopolymer of various chain length. In the second increment, they lead to the terminated diblock copolymers of which can neither participate in further polymerization to form the triblock copolymer nor participate in the coupling reaction to form radial copolymers. The fast-reacting impurities in the monomer of the second increment lead to terminated homopolymer formed in the first stage. These processes go on with each incremental addition of monomer. In Method 3 where coupling reaction is involved, the efficiency of the coupling is rarely 100%. This again leads to the uncoupled diblock copolymers.

Morton and his co-workers [40,41] using high vacuum techniques studied systematically the synthesis, properties, and morphology of the triblock copoly-



mers. Morton [42] concluded that the requirements for making high-quality copolymers are as follows.

1. Any possible chain termination must be reduced to a negligible level.

2. The polymerization system must be capable of producing a polydiene block of high 1,4 chain structure (i.e., a soft elastomer).

3. The initiation rate for each block must be much faster than the propagation rate.

Also using high vacuum techniques, Fetters and associates examined the formations and properties of radial block copolymers with coupling agents of divinylbenzene [43] and polychlorosilanes [44,45].

IV. Triblock Copolymers Utilizing Vinylaromatic Monomers Other Than Styrene

The stress—strain properties of S-D-S triblock copolymers are limited by the glass transition temperature (Tg) of the polystyrene. These materials show serious losses of strength even at 60°C. The Tg for polystyrene is around 105°C. Significant efforts were made to use vinylaromatic monomers of higher Tgs to produce triblock copolymers by anionic polymerization. Most of these efforts were centered around the use of α-methylstyrene, and to a lesser extent, tert-butylstyrene.

Poly-α-methylstyrene has a Tg of about 165°C and poly-tert-butylstyrene has a Tg of about 130°C. α-Methylstyrene has low “ceiling” temperature and the polymerization of the monomer had to be carried out at reduced temperature, using polar solvents to accelerate the rate [46–50]. Also, the use of a dilithium initiator, Method 1, is advantageous for the preparation of α-MeS-D-α-MeS copolymers. With this method, one can polymerize the diene first under normal conditions, then polymerize the α-methylstyrene at low temperature and high polar adjuvant or cosolvent. In some studies, mixtures of styrene and α-methyl-styrene were used to form a copolymer as end blocks of a triblock copolymer [46,51,52]. Triblock copolymers with poly-tert-butylstyrene end blocks were reported by Cunningham [53].

V. Hydrogenated Triblock Copolymers

For many years, chemical modifications of unsaturated polymers, particularly the hydrogenation of polydienes, have received a great deal of attention [54–61]. Hydrogenation of the unsaturated polymers can provide an alternative route to produce a variety of unique and useful elastomers and thermoplastics with specific structures and properties. Many of the polymers with novel monomer sequence distribution and composition that are difficult or impossible to prepare directly through conventional polymerization methods can be easily obtained by the



hydrogenated polymers. Another important feature of the hydrogenation of polydienes is that hydrogenated products normally would have improved resistance to oxidation and thermal degradation compared to the parent polymers containing unsaturated double bonds.

The hydrogenation catalysts [62] frequently used for the hydrogenation of polydienes are (a) insoluble metal and nobel catalysts, (b) homogeneous organometallic catalysts, and (c) diimide generated in-situ from p-toluene sulfonyl hydrazide. As illustrations, several examples of organometallic hydrogenation catalysts for polydienes are listed in Table 18.4.

Diimide generated in-situ is very convenient to use in the laboratory. It is not used in commercial operation mainly because its high cost.

Shell [69] announced in 1972, a “second-generation” TPE based on anionic polymerization. It was described as S-EB-S triblock copolymers that have higher cohesive strength, retain structural integrity at higher temperatures, and have excellent resistance to degradation by oxygen, ozone, and ultraviolet (UV) light. The S-EB-S copolymers can be extended with paraffinic or naphthanic oils up to 200 parts without bleed-out.

The effect of hydrogenation of polyisoprene and polybutadiene is illustrated in Table 18.5.

When one polymerizes butadiene to form the center soft (elastomeric) segment of the triblock copolymers, one can control the vinyl (1,2-addition segment) configuration by adjusting appropriate amount of polar adjuvant in the polymerization medium. Thus, one can control the ratio of ethylene (E) and butylene (B) units in the soft segment after hydrogenation. This, in turn, effects the Tg and elastomer crystallinity of the segment as shown in Figure 18.5 [5]. The butylene concentrations of the commercial products are generally in the range of 30–60%, depending on the applications.

Although several industrial laboratories worked actively on the hydrogenation catalysts and processes to produce the “second-generation” TPE in the 1960s

Table 18.4 Examples of Organometallic Hydrogenation Catalysts for Diene Polymers

Catalyst Reference

Organic salt of Ni or Co + BF3 + RnM (e.g. R3Al) 63

Ni, Co or Fe naphthanate acetylacetonate, 2-ethylhexanoate or 3,5-diisopropylsalicylate + RLi or R2Mg

64

Cobaltous bis-lactam or CoCl2 .6–8 lactam + R3Al 65

Ni. 3,5-diisopropylsalicylate or Ni 2-ethyl-hexoate + R3Al 66

Ni acetylacetonate, p-nonylphenol + BuLi 67

Isopropyltitanate + R3Al 68



Table 18.5 Effect of Hydrogenation on Polydiene Structures

Precursor After [H] Tg After [H]

1,4-Polyisoprene segment Ethylene/propylene (1:1) (E/P) segment ˜-60°C

1,4-Polybutadiene segment Linear polyethylene (PE) segment 136°C

1,2-Polybutadiene segment Polybutylene (B) segment ˜ -18°C

and 1970s, Shell seemed to have a leading position as witnessed by the number of patents [70–79] granted and their strong market position. No intent is made by the authors to cover all the pertinent patents issued and assigned to Shell.

Finely divided metal catalyst residue is insoluble in hydrocarbon solvent and most of it can be removed after hydrogenation by filtration. Hassell described the removal of nickel catalyst from hydrogenated polymer by volatilizing as nickel carbonyl [80] and by complexing and extracting [81]. In the latter process, the solution is sparged with oxygen, air, or aqueous H2O2, or tert-butylhydroperoxide is added. Aqueous citric acid is added to complex the metal.

Figure 18.5 Effect of butylene concentration on the crystallinity of Tg of

ethylene/butylene copolymer in the elastomer phase. (From Ref. 5, used with

permission from Rubber Div., A.C.S.)



VI. Properties

A. Tensile Properties

Block copolymers made up of incompatible sequences can form network with high rubberlike elasticity without the introduction of chemical crosslinks, provided they contain at least two long sequences of a monomer with a higher Tg than the monomer or monomers making up the continuous phase, usually the monomer present in excess [82–90]. The triblock copolymers of S-D-S type, in which S is styrene and D is a diene or a random copolymer of dienes, are outstanding examples of this network formation. When phase separation occurs, S blocks from different molecules share in making up domains of S dispersed in a continuum of D, provided that D is present in sufficient excess. Between the transitions of S and D, the polymer then consists of rubbery chains (the D blocks) starting and terminating in glassy domains (the segregated S blocks). This structure (Fig. 18.6) resembles a crosslinked filler-reinforced vulcanizate.

Mier [91] predicted with his theoretical calculations that the critical end block molecular weight for domain formation in a styrene—butadiene block copolymer would be between 5000 and 10,000 when the polybutadiene block was of the order of 50,000 molecular weight. Bishop and Davison [92] calculated

Figure 18.6 Schematic of a styrene—butadiene—styrene triblock copolymer.

(From Ref. 42, used with permission from Hanser Publishers.)



that a 15S-100B-15S block copolymer (i.e., a triblock copolymer with polybutadiene center block of 100,000 molecular weight and polystyrene end blocks of 15,000 molecular weight each), with a domain size of about 30 nm, would have approximately 200 chains emanating from each domain. Hence the domains serve as physical crosslinking sites of very high functionality.

Holden and colleagues [93] compared the stress—strain properties of the unvulcanized S-B-S triblock copolymers with vulcanized natural rubber (NR) and vulcanized styrene—butadiene rubber (SBR) as shown in Figure 18.7. The stress—strain values, particularly tensile strength of S-B-S, are much higher than those obtained from unreinforced vulcanizate of SBR or polybutadiene. Holden and co-workers explained by postulating that (a) the domains act as reinforced filler, and (b) the increased tensile strength resulting from the slippage of entangled chains. It

Figure 18.7 Stress—strain properties of elastomers. (From Ref. 93,

used with permission from John Wiley & Sons.)



Table 18.6 Effect of Molecular Weight on Mechanical Properties of S-B-S

Segment MW 6S-81B-6S 10S-52B-10S 14S-74B-14S 20S-110B-20S

Styrene content (%) 13 27.5 27.5 26.5

Total MW (× 10-3) 93 73 102 1500

Tensile strength (MPa) 1.0 27 27.9 24.8

Elongation at break (%) 1000 860 820 850

Shore hardness 41 65 71 61

Source: Ref. 93.

is quite possible that both apply [93,94]. These authors also confirmed Meier's prediction about the critical end block molecular weight for domain formation (Table 18.6).

The sample of 6S-81B-6S with polystyrene end blocks with 6000 molecular weight is in the lower end of Meier's calculated range and, indeed, the sample had only very modest strength. The other three samples, all of which exceeded the calculated critical end block molecular weight for domain formation, had very good tensile strength without vulcanization.

Childers and Kraus [89] reported that at polystyrene levels below 30–35%, interaction between different polystyrene regions is apparently small. This is illustrated by elastomeric stress-strain curves shown for 10–80–10 and 15–70–15 (weight ratios) S-B-S block polymers. They resemble stress-strain curves of filler-reinforced vulcanizates. At higher polystyrene level, 25–50–25 S-B-S, an initial high modulus region at low strain apparently results from interaction between polystyrene regions. As elongation proceeds, irreversible softening occurs during which stress is constant with increasing elongation. This process is followed by elastomeric behavior at higher elongation.

Childers and Kraus [89] also showed that at constant composition, increasing polystyrene block length increases both tensile strength and storage modulus of S-B-S block polymers. The effect of molecular weight on tensile strength of 15–70–15 S-B-S block polymers is illustrated in the tabulation below:

Approximate relative molecular weight (M)

Polystyrene (%)

Tensile strength(MPa)

Elongation (%)

M 27.9 8.8 900

2 28.4 15.9 1200

4 29.0 33 1000

This effect has been attributed to increased Tg of polystyrene by increased chain length [95]. Other factors that may be involved are size and shape of



polystyrene domains, effect of stress on cold drawing of polystyrene [96], and thermal history. In one instance [40], slow cooling increased modulus and reduced transparency, indicating that the size of the polystyrene regions was increased.

It should be pointed out that the data tabulated above may very well be another illustration that the polystyrene molecular weight in S-B-S and S-I-S polymers must be high enough to cause the formation of strong, well-separated domains under the condition of the test to obtain optimum strength. Once this requirement is met, the tensile moduli and tensile strengths of these materials with constant polystyrene content are not molecular weight-dependent [93,97].

Tensile strength falls off at elevated temperatures [89] and the maximum useful temperature for this class of materials has been suggested as 60°C [7].

Plasticizers (e.g., naphthanic and highly aromatic oils) sharply reduced tensile strength of S-B-S block polymers, although both preferentially associate with the elastomer phase [84]. Since extender oils seem to have low affinity for polystyrene regions, their strong depressing effect on tensile strength is not understood at present.

Morton and his associates systematically studied and compared the tensile properties of S-I-S and S-B-S block copolymers. Their findings and comments have been recently summarized by Morton [42]. Highlights of this summary are listed below.

S-I-S Block Copolymers

For S-I-S copolymers, the tensile modulus appears to be mainly dependent on the polystyrene content and independent of the molecular weight of the center elastomeric block. Hence the latter cannot be considered as representing the molecular weight between crosslinks (Mc) of the network. A number of chain entanglements between the polystyrene “crosslinks” and the “network chain” can really be considered as equivalent to the Mc value.

The tensile strength appears to be largely independent of either polystyrene content or molecular weight, with the exception of the sample having very low polystyrene molecular weight. It is believed the low polystyrene molecular weight prevented a good phase separation (substantial plasticization of the polystyrene domains by polyisoprene).

The retraction curves exhibits a large hysteresis loop and considerable uncovered deformation (set), which increases with degree of strain and of polystyrene content. This phenomenon is apparently largely due to a distortion of the polystyrene domains. The distorted domains can be restored to their original condition by heating the sample at or above the Tg of the polystyrene or by swelling in a specific solvent for the polystyrene, namely tetrahydrofurfuryl alcohol. This type of treatment also restores the tensile behavior of the pre-stretched sample.

The polystyrene domains are responsible for the integrity and strength of the



network. Tensile failure involves rupture of the domains. The exceptionally high strength shown by these polymers is a result of both the regularity of the network, which helps to distribute the stress more evenly, and the fact that the polystyrene domains can act as an “energy sink” to absorb the elastic energy and this delay failure.

Polymers containing 40% styrene show an initial yield point in the stress-strain curves. This occurs only on the first stretch and not on any subsequent stretch unless the sample is heated. This phenomenon is believed to be due to high concentration of polystyrene domains approaching the critical fraction for volume packing. Hence there would be some “connections” between the domains that would be broken on the first stretch.

S-B-S Block Copolymers

S-B-S block copolymers show some similarity in behavior to the S-I-S polymer in that the styrene content does control the modulus and the latter is independent of the molecular weights.

Unlike S-I-S polymers, the tensile strength of S-B-S polymers depends greatly on the styrene content. The increase in tensile strength of S-B-S polymers with increase in styrene content is most likely related to a better phase separation of the polystyrene domains at higher volume fractions of this component. In the case of the S-I-S polymers, it appears the incompatibility of the two phases is already high enough not to be affected very much, at 20–40% level, by the styrene content.

Morton [42] also compared the stress-strain properties of polymers with α-methylstyrene (mS) and styrene (S) end blocks. Thus, when two samples of mS-I-mS and S-I-S, with nearly identical molecular weight and composition were compared, the former sample had substantially higher tensile modulus and tensile strength. Morton took this as a strong confirmation of the idea that the plastic domains are the principal stress bearers, since poly-α-methylstyrene has higher tensile modulus than polystyrene. Also, the data showed the marked superiority of the mS-I-mS polymer at elevated temperatures. The dependence of tensile strength on the temperature also points to the plastic domains as the key factor in the strength of the network.

In addition to the mechanism of ductile failure in the styrene (plastic) domains, there are two other possible mechanisms to explain the tensile failure in styrenic block copolymers. These are brittle fracture in the styrenic domains and elastic failure in the polydiene (elastomeric) center segments [94]. Holden and Legge [94] suggested that it is possible that which of the mechanisms will apply in a particular case depends on the conditions. At high temperatures, as the domains soften, ductile failure will predominate. The same effect will apply when the time scale of the test is long. At low temperatures or shorter time one of the other two mechanisms will take over.



Polymers of the S-EB-S type are described and characterized in a paper by Gergen[98]. One area of particular interest is the comparison of tensile properties exhibited by the S-B-S and S-EB-S polymers. Of the two polymers. the S-EB-S exhibits higher modulus. The higher modulus is attributable to the higher association energy and the absence of substantial interface volume in the S-EB-S polymer. The elongation of the S-EB-S polymer is lower than that of the S-B-S polymer because the S-EB-S polymer has the lower contour length. The rate of tensile loss with increase in temperature is much less for the S-EB-S polymer when compared to the S-B-S polymer. This difference in temperature-tensile relationship allows the S-EB-S polymer to be used in applications where it would not be possible to utilize the S-B-S type polymers.

B. Structure/Morphology

S-D-S block copolymers are two-phase systems. The two phases, polystyrene and polydiene, retain many of the properties of the respective homopolymers. Block polymers of B-S type prepared by sequential addition of monomers clearly display two glass transitions, a result of microphase separation [85]. This effect [83,86–90] plays a crucial part in the characteristic behavior of block polymers of the S-D-S type in the unvulcanized state.

The fundamental parameters governing the physical behavior of S-D-S type block copolymers are total composition; total molecular weight; and the number, length, and composition of the individual block sequence [99]. These in turn govern the morphology of the domain structure. If the glassy polystyrene block totals more than 50% in volume, the glassy phase then becomes continuous and load bearing. These materials are not likely to be rubbery.

Even in polymers containing 50% styrene in volume, there may be appreciable interconnection of the polystyrene domains. These are broken on stretching, giving rise to a strong stress-softening effect [40,89] which is generally reversible only on heating. This effect has been demonstrated with stress birefringence measurements [100,101].

Deposition from solvents varying in quality with respect to the two types of block sequences leads to films of different morphologies and physical properties [102–105]. There are five types of fundamental domain structures as revealed by electron microscopy, which may be demonstrated by casting, from a common (good) solvent, block polymers of widely varying composition: spherical domains of S in a matrix of D, rodlike domains of S in D, alternating lamellae of S and D, rods of D in S, and spheres of D in S [103]. Compression-molded samples also exhibit these basic morphologies [90,105,106]. A generalized scheme of the solid state morphology of block copolymers is shown in Figure 18.8 [107]. Orientation and degree of order are strongly dependent on the method of sample preparation, particularly with cylindrical and lamellar morphologies [108]. Extruded samples



Figure 18.8 A generalized scheme of the solid state morphology of

ABA block copolymers. (From Ref. 107, used with permission from Plenum Press.)

Figure 18.9 Effect of solvent solubility

parameter on the solution viscosity of an S-B-S block copolymer.

(From Ref. 111, used with permission from Soc. of

Plastics Engineers.)



of extremely high degree of order have been prepared from a S-B-S block polymer [109]. These resemble single crystals consisting of hexagonally spaced long rods of polystyrene in the polybutadiene matrix. From electron microscopic examination and small-angle x-ray diffraction studies, domain sizes and spacings are known to range from approximately 50 to 1000 Å, depending greatly on polymer composition molecular weight and method of sample preparation [90,110].

C. Solution Properties

The properties of dilute solutions of these polymers in relatively good solvent are quite normal. By measuring the dilute solution properties as intrinsic viscosity in a range of solvents, Paul and his associates [111] showed that the viscosity is at maximum when solubility parameter of the solvent is about 8.6 (cal/cc)1/2 (Fig. 18.9). In contrast, the viscosity of the concentrated (24.8% W/V) solution is at minimum in the same region (Fig. 18.10).

Figure 18.10 Effect of solvent solubility parameter on the intrinsic

viscosity of an S-B-S block copolymer. (From Ref. 111, used with permission from Soc. of Plastics Engineers.)



As the solution became more concentrated, phase separation begins and evidence of ordered structure is observed [112] and domain sizes, interdomain distance, and morphologies were examined by the use of small-angle x-ray scattering [113–117]. Stacy and Kraus [118] reported that the domain sizes depend on the polystyrene molecular weights and on the thermal histories of the solutions.

D. Viscous and Viscoelastic Properties

Under low shear conditions, the melt viscosities of S-B-S and S-I-S block copolymers are much higher than those of polybutadiene [119], polyisoprene [120], or random styrene-butadiene copolymers [121] of equivalent molecular weights. Moreover, these block copolymers show non-Newtonian behavior both under steady-state [89,93] and dynamic conditions [82,122]. This is rationalized in terms of two-phase domain structure of these polymers, which persists to a significant degree in the melt [82,93]. In such a structure, flow can only take place by the polystyrene segments at the ends of the elastomer chains being pulled out of the domains.

The melt rheology of star-branched block polymers of butadiene and styrene is also very different from that applying to similarly branched homopolymers or random copolymers [123]. The steady-flow Newtonian melt viscosity of narrow distribution, star-branched polybutadienes is smaller than that of linear polybutadiene below a certain molecular weight, but the reverse is true at high molecular weights [124]. Below the glass transition of the polystyrene blocks of the star-branched block copolymers on the other hand, the effects of branching were masked by differences in the morphology of the domain structure unrelated to branching. The effects of branching are minor when polymers are compared at equal lengths of the terminal block sequences.

E. Hardness

Hardness of a S-B-S polymer is strongly dependent on polystyrene content. The Shore hardness increases with an increase in polystyrene content, as shown in Table 18.7.

Table 18.7 Effect of Styrene Content on Hardness

Wt.% Styrene Molecular weight Shore A hardness

20 160,000 47

30 140,000 65

40 130,000 91



VII. Commercial Styrenic Block Copolymers

A. Availability of Polymers

Some trade names and types of thermoplastic elastomers based on styrenic block polymers are listed in Table 18.8.

Block polymers are commercially available as clear nonextended (neat), oil-extended, or fully compounded products with either saturated or unsaturated center blocks. Partial lists of commercial polymers and compounds are provided in Tables 18.9–18.11. As can be seen, a wide range of properties is possible.

Commercial products are produced in a variety of forms including bale, crumb, pellet, and ground particles. Optimal choice of product and product form is dependent on the consumer's process equipment and intended application. Typical properties of some commercially available compounds designed for various applications are shown in Table 18.12.

Table 18.8 Some Trade Names of Thermoplastic Elastomers Based on Polystyrene/Elastomer Block Copolymers

Trade name (manufacturer)

Type

Hard segment

Soft segment

Notes

Kraton D and CariflexTR (Shell)

S-B-S, S-I-S and (S-B)nX

S B or I

Solprene 400a

(Phillips)(S-B)nX

Finaprene (Fina) (S-B)nX

Stereon (Firestone) S-B-S S B General purpose,soluble

Tufprene & Asaprene (Asahi)

S-B-S

Europrene Sol T(Enichem)

S-B-S or S-I-S S B or I

Kraton G (Shell) S-EB-S S EB Improved stability,soluble when uncompounded

Elaxar (Shell) S-EB-S S EB Wire and cable

C-Flex (Concept) S-EB-S and silicone oil S EB Medical applications

aNo longer made in the United States.

Source: Ref. 125.



Table 18.9 Partial Listing of Nonextended Polymers with Saturated Center Block and Typical Properties

Kraton G Rubber (Shell)

1650 1652

Styrene/olefin ratio 30/70 30/70

Propertya

Specific gravity 0.91 0.91

Melt flow, 190°C/21.6 kgb 0 2.5

300% modulus (MPa) 5.5 4.8

Tensile (MPa)c 34.5 31.0

Elongation (%) 560 520


aShell Chemical Company, SC:68–88.

bASTM D1238

cTypical values determined on film cast from a toluene solution.

B. Typical Mechanical Properties

The nonextended polymers do not require vulcanization, but have excellent green strength and elastomeric properties (Table 18.13). Unlike plastics, these materials exhibit low permanent set after extension and are highly flexible and resilient. Surface friction is high, with coefficient of friction values similar to those of conventional vulcanized rubbers. Due to their thermoplastic nature, block polymers exhibit a loss in tensile strength and hardness as temperature increases. Thermoplasticity also contributes to increased flow and ease of processing at elevated temperatures.

C. Typical Thermal Characteristics

Block polymers remain flexible at very low temperatures, as indicated by the data in Table 18.14. Indicative of a two-phase nature, two glass transition points are exhibited by the elastomers. The upper transition temperature is dependent on the styrenic block and the lower transition temperature on the elastomeric block. Distortion characteristics can be altered somewhat by varying structure (molecular weight, branching, etc.) or more so by compounding. Maximum service temperature can vary widely, depending on formulation and specific requirements.



Table 18.10 Partial Listing of Oil-Extended Polymers and Typical Properties

Solprene and Finaprene Rubber Kraton D R

475 478 480 481 4113 4124 4122

Structure (SB)n (SB)n (SB)n (SB)n SBS SBS SBS

Styrene/butadiene ratio 40/60 40/60 30/70 48/52 33/67 33/67 48/52

Extractable (% wt: oil) 33 44 33 37 26 16 22

Property

Specific gravity 0.94 0.93 0.93 0.95 0.94 0.93 0.9

Melt flow, 200°C/5 kga 5 35 3 5 26 16 22

300% modulus (MPa) 2.1 1.1 1.4 1.5 2.1 1.4 1.6

Tensile (MPa)b 18.6 13.8 9.0 17.2 11.7 11.7 15.5

Elongation (%) 1000 1300 1100 1100 1150 1350 1150

Shore A hardness 67 48 45 68 46 43 62

aASTM 1238.

bTypical values on polymer compression molded at 149°C (300°F).



Table 18.11 Partial Listing of Nonextended Polymers with Unsaturated Center Block and Typical Properties

Solprene and Finaprene Rubber Kraton D

406 414 411 416 418 1101 1102 1107

Structure (SB)n (SB)n (SB)n (SB)n (SI)n SBS SBS SIS

Styrene/diene ratio 40/60 40/60 30/70 30/70 15/85 30/70 28/72 14/8

Propertya

Specific gravity 0.95 0.95 0.94 0.94 0.92 0.94 0.94 0.

Melt flow, 200°C/5 kgb

0 4 0 3 3 <1 6 9

300% modulus (MPa) 4.1 4.1 2.1 3.0 1.0 2.8 2.8 0.

Tensile (MPa) 26.9c 27.6c 19.3c 20.0c 20.0c 31.7d 31.7d 21.

Elongation (%) 700 750 700 720 1050 880 880 1300

Shore A hardness 93 90 78 68 34 71 62 37

Set at break (%) 10 10 10 10 10 10 10 10

aPhillips Chemical Company Solprene Rubbers (no longer made in United States) bulletins and Shell Chemical Compan(SC:68–88).

bASTM 1238.

cTypical values on polymer compression molded at 149°C (300°F).

dTypical values determined on film cast from a toluene solution.

Source: Ref. 126.



Table 18.12 Typical Propertiesa of Commercially Available Compounds

Compounds designed for Specific gravity Tensile (MPa) Elongation (%) Shore A

Footwear: Slipper soling 1.10 6.7 410 81

Sneaker soling 1.14 5.9 520 44

Unit soling 1.02 6.3 450 67

“Earth” shoe 1.02 5.9 480 59

General-purpose extruded products

Cove base 1.68 5.2 210 86

Rubber band 1.02 20.0 960 59

Garden hose 1.26 7.6 800 65

Gasket (appliance) 1.27 13.9 910 69

Milk tubing 1.26 11.0 1040 49

General purpose molded products

Flexible toys 1.00 5.6 870 46

Swim fins 0.98 12.4 780 76

Crutch tips 0.99 7.2 630 48

Golf club grip 0.98 8.6 880 62

Pharmaceutical, medical Tubing, packaging

0.93 6.2–13.1 800–1300 45–55

Automotive

Sight shields 1.16 13.8 900 84

Bumper fillers

aRepresentative data for some of Phillips Petroleum Co. (Solprene LR) compounds (no longer made in United States) orproducts.

Source: Ref. 126.



Table 18.13 Typical Mechanical Properties

Type center block

Polymer variables

Butadiene

Butadiene

Saturated-olefinic

Isoprene

Styrene content, % 30 40 30 ca. 15

Melt flow range (200 C/5 kg) b

2–10 <1 2–10 <1 <1 5

Compression moldeda

300% modulus (MPa) 2.8 4.1 3.8–5.5 1.0

Tensile (MPa) 17.2–24.1 27.6 27.6 17.2–20.7

Elongation (%) 750 750 550 1000

Shore D 27 41 30 —

Share A at 25°C 70–75 92 72–85 35

at 70°C 42 70 65 79 68 —

at 100°C 15 55 25 55 48 —

Tensile (MPa) at 50°C 3.8 7.6 12.4 17.2 12.4 4.8–6.9

at 60°C 2.4 — 4.1 13.1 8.3 1.4–2.8

at 70°C 1.4 55.5 3.1 11.7 4.8 —

Set (after 300% elongation, %)

5–10 15–20 10–30 <5

Yerzley resilience (%) 70 — 55 80 75–80 85

aTest values may be varied to some extent by method of processing (injection molded, extruded, solvent case, etc.).

bASTM D1238.

Source: Ref. 126.

D. Environmental and Chemical Resistance

Unsaturated block polymers are similar to SBR rubber in resistance to ozone, oxidation, and ultraviolet radiation; these can be used in many applications without special stabilization. Additives can markedly improve aging characteristics under more demanding conditions. Polymers with saturated rubbery sections are inherently resistant to effects of ozone, oxidation, and ultraviolet radiation. Block polymers can be used in contact with water, alcohols, weak acids, or bases with little change in properties. However, many hydrocarbons, esters, and ketones dissolve or cause excessive swelling of the rubbers. Solubility characteristics of block polymers make them well suited for solvent-based formulations (adhesives, sealants, caulks, etc.), but limit their use in applications requiring resistance to many chemicals (Tables 18.15 and 18.16).



Table 18.14 Thermal Characteristics

Type center block

Polymer variables Butadiene Saturated-olefinic Isoprene

Styrene content (%) 30-40 30 15

Melt flow range (200°C/5 Kg)a

2-10 <1 <1 5

Properties, °C

Tg, lower -90 -52 -55

upper 100 100 100

Gehman freeze point -90 -60 -60

Brittleness temperature <-60 <-60 <-60

Vicat softening point 60-75 6-70 -

Distortion temperatureb 50-55 45-55 -

% distortion at 100°Cc 80 0 0 100

Thermal conductivityd 3.6e - 3.6e

Specific heat, cal/° C/gm. 0.45-0.50e - 0.45-0.50e

Thermal expansion, 10-5 in/in°C

13-13.7e - 13-13.7e

aASTM D1238.

bASTM D648, 66 psi.

cASTM D2633.

d10-4 cal/s/sq cm/1 (°C/cm).

eValue from Modern Plastics Encyclopedia, 1974-1975.

Source Ref. 126.

Table 18.15 Chemical Resistance of Styrenic TPR

Type center block

Polymer variables Butadiene Butadiene Saturated-olefinic

Styrene content (%) 30 40 30

Volume swell, % after 22 h at 25°C in:

10% NaOH 0 1 0.4



Table 18.16 Chemical Resistance of Kraton D1101 Rubber

Wt.% gain twoweeks at r.t

% decrease in tensile strength

Wine (20% V alcohol) 0.3 none

Beer (3.2% V alcohol) 0.5 none

Milk (fresh daily, tested @ 40°F) 0.8 none

Acetic acid (5% W and 10% W) 2.0 <10

Sulfuric acid (3% W and 10% W) 0.3 <10

Phosphoric acid (30% W and 60% W as P2O5) 0 <10

Boric acid (3.1% W) 0 none

Oxalic acid (3.1% W) 1.0 none

Lactic acid (3.8% W) 0 none

Distilled H2O 0.7 none

Sodium chloride (10% W) 0.1 <10

Sodium carbonate (2.7% W) 1.0 none

Potassium hydroxide (5% W) 0.1 none

Ammonium hydroxide (3.4% W) 1.0 <10

Ammonium nitrate (50% W) 0 none

Nitric acid (10% W and 20% W) 4.5 790

Source: Shell Chemical Company, SC: 198–87.

F. Permeability

The highly permeable nature of block polymers (Table 18.18) suggests their use in breathable-type packaging. Permeability is much greater than that of Saran Wrap or polyethylene.

G. Viscosity and Rheological Properties

Solution viscosity of plastomers depends on a number of factors. The block structure in itself contributes to low solution viscosity. At 10% solids, solutions of block polymers may display a viscosity about one-tenth that of SBR or natural rubber. Other factors exerting strong effects are monomer ratio, block styrene content, molecular weight and structure of the elastomer molecule, and, of course, solids concentration. Some typical solution viscosities for branched polymers are noted in Table 18.19.

Trends observed in solution viscosity with changes in molecular structure (linear vs. branched) and in molecular weight are illustrated in Figure 18.11. These are generalized data for solutions in solvents such as toluene or naphthatoluene blends.

Melt viscosity of block polymers depends on monomer type and ratio,



Table 18.17 Electrical Properties

Type center block

Polymer variables Butadiene Butadiene Saturated-olefinic

Styrene content (%)

30 40 30

Dielectric constant

1 kHz 2.51 2.53 2.30

1 MHz 2.50 2.53 2.30

Dissipation factor 1 kHz 4 × 10-4 1 × 10-4 13 × 10-5

1 MHz 8 × 10-4 7 × 10-4 23 × 10-5

Volume resistivity (ohm-cm) 1 min 3 × 1016 2 × 1016 9 × 1016

5 min 1 × 1017 2 × 1016 2 × 1017

Source: Ref. 126.

molecular structure, and molecular weight as well as temperature and shear rate. Overall, it would appear that one of the most important considerations in melt behavior is the length of the polystyrene terminal blocks. Butadiene leads to higher viscosity than isoprene. Increasing styrene in copolymers with butadiene appears to increase viscosity up to approximately 30% styrene followed by a decrease in viscosity at higher styrene levels. Branching reduces viscosity if total molecular weight is constant.

As noted in Table 18.19, some neat polymers are low in melt flow. Response of these polymers to changes in pressure and temperature may vary, but the

Table 18.18 Gas Permeability

Type center block

Polymer variables

Butadiene

Butadiene

Saturated-olefinic

Sarana

High-density polyethylene

Styrene content (%) 30 40 30 — —

Permeation (cc/1200 sq. in./mil thickness/24 h) Oxygen

3600 700 2000 0.1–0.2 110

Carbon dioxide 13,000 2800 8500 0.3 350

Nitrogen 1300 180 700 0.025 50

Moisture–vapor transmissionb

36 17 8 1.8 —

aVinylidene chloride.



Table 18.19 Typical Solution and Melt Viscosities

Solprene rubber

Polymer description 417 416 411 414 406 418

Type center block Butadiene Isoprene

Styrene content (%) 20 30 30 40 40 15

Mol. Wt. Mw/Mn (× 10-3)

200/150 140/110 300/220 130/100 280/200 300/230

Type structure Radial Radial Radial Radial Radial Radial

Toluene solution viscosity

5 wt.% (cs) 10 30 7 18 15

25 wt.% (cp) 3500 1400 17,000 2300

Melt Flow, g/10 minutesa

180°C, 5 Kg 1 1 0 2 0 2

200°C, 5 Kg 2 2 0 5 0 —

ASTM D1238.

Source: Ref. 126.

apparent viscosity generally decreases with an increase in shear rate. Often the subject polymers are blended with other polymers or compounded with oils and resins that may alter flow properties. Melt viscosities and response to shear rate of compounds and blends will often be similar to that displayed by thermoplastics such as polyethylene or polystyrene.

Figure 18.11 Molecular weight and structure effects. (From Ref. 126, used

with permission from Van Nostrand Reinhold Co.)



VIII. Compounding (126)

A. General Comments

The thermoplastic elastomers (TPE) of styrene-diene types play an interesting balance of properties and can be used as manufactured in various applications, including products formed by heat and pressure as well as those deposited from solvents. These polymers can also be used to modify properties of other polymers (polyolefins, polystyrene, etc.) without the need for other added ingredients. For many uses, however, such as adhesives, hot melts, extrusions, and moldings, the addition of minor or major amounts of other materials may prove advantageous. Thermoplastic elastomers display good strength and can be compounded to provide a very wide range of properties [126]. The polymers are easily mixed with compounding ingredients either as a melt or as a solution. These rubbers respond well to compounding, which can be useful to adjust cost, viscosity and flow, hardness, flexibility, tack, deformation resistance, oxidation and ozone resistance, flammability, and other properties.

Various materials are useful in compounding block polymers: fillers, plasticizers, resins, and antidegradants. Choice of ingredients will often depend on the composition of the neat polymer as well as the effect desired. Another factor to be kept in mind is the two-phase nature of these polymers. Ingredients may be miscible or compatible with the hard segments, the soft segment, both, or neither. The morphology can be affected: addition of a sizable amount of some polymer or resin may cause inversion of the continuous phase.

B. Plasticizers

Various oils and waxes serve as plasticizers for block polymers. In most cases, it is desired to use a material that will soften and plasticize the elastomeric portion and not the polystyrene segment. Oils that generally appear most useful are those classified as paraffinic, naphthenic, or white mineral oils. Aromatic content of the oil should be low to avoid softening of polystyrene blocks and resultant excessive decreases in strength and hardness. Of course, all oils will reduce tensile and abrasion resistance but serve to regulate hardness and modulus as well as increasing melt flow. Oils may provide benefits in resistance to crack growth during flexing.

In summary, the use of oils and waxes as plasticizers in styrenic TPE decreases hardness and modulus, eliminates drawing, reduces melt and solution viscosity, improves compounding processability, and decreases cohesive strength or increases plasticity. Materials other than oils may serve as plasticizers, especially at elevated processing temperatures. Some of the resins and process aids discussed in following sections serve in this capacity.

Plasticizers or other compounding ingredients may reduce the stability of



compounds compared to the neat polymer upon exposure to ultraviolet radiation. In such cases, addition of pigments or other ultraviolet (UV) stabilizers may be necessary.

C. Fillers

Large quantities of inexpensive fillers can be added to block polymers, usually in combination with plasticizers and/or resins to reduce cost and modify properties. The effect of a few fillers at several loadings is shown in Table 18.20.

In general, fillers decrease melt flow and tensile strength; however, strength at elevated temperatures may be increased. Other changes to be expected include an increase in tear strength, flex life, and abrasion resistance imported by fine particle size silicas, carbon blacks, and hard clays. These fillers also increase modulus and hardness. Calcium carbonates have less effect on properties, but improve flex life with only a slight increase in hardness or stiffness. Fillers such as titanium dioxide, zinc oxide, or carbon black are sometimes added to improve resistance to ultraviolet radiation. Glass beads, talc or other materials may also be useful as fillers for some applications.

D. Resins and Plastic Polymer Additives

A number of resins and plastic polymers can be used to advantage in developing specific properties in block copolymer adhesive and thermoformed compounds. These materials tend to decrease elasticity or “rubbery feel” and are generally used in combination with plasticizers (and fillers). In adhesives, resins are used to improve tack, adhesion, peel strength, and shear strength. In extruded or molded goods, resins and plastics can be added to adjust hardness and improve a variety of physical properties including distortion resistance. These materials also often reduce viscosity at elevated temperatures, permitting easier processing. Different resins affect the ratio of properties in different ways: some increase melt flow with small effect on hardness while others have the reverse effect. Resins that are more compatible with the styrenic end-block are likely to produce a hard, nontacky material; a sticky, soft, flexible composition is likely to result from resins that are more compatible with the rubbery central block. Some useful resins or plastics and effects of primary interest include the following:

Polystyrene: increases tear strength, abrasion resistance, flex life, and hardness

Polyethylene: increases abrasion resistance, hardness, and distortion resistance

Ethylene vinyl acetate copolymer: improves ozone resistance

Polypropylene: increases modulus, hardness, and distortion resistance

Vinyl toluene copolymer: increases melt flow and adhesion

Polyindene: increases melt flow, hardness, tensile strength, and adhesion



Table 18.20 Effect of Fillers

Polymer (60/40 butadiene/styrene) 100 100 100 100 100 100 100

Calcium carbonate — 40 80 120 — — —

Hard clay — — — — 40 80 120

Fine silica — — — — — — —

Melt flow (180°C/5 Kga) 3 1.5 0.9 0.5 1.4 0.5 0

300% Modulus (MPa) 4.1 4.0 3.9 4.0 9.0 11.9 13.8

Tensile (MPa) 27.5 19.3 13.4 9.7 17.9 14.8 13.8

Elongation (%) 780 750 700 630 700 470 300 7

Crescent tear (kg/cm) 24.9 23.7 24.9 21.5 39.6 42.9 47.5

Shore D hardness 36 38 42 46 42 48 52

NBS abrasion index (% of RMA standard)

52 39 34 30 67 64 63

Ross flex (M flexures to 0.5'' cut growth)

3.5 23 40 13 >200 >200 30

aASTM D1238.

Source: Ref. 126.



Coumarone-indene copolymer: increases distortion resistance, melt flow, tensile and tear strength

Pentaerythritol ester of hydrogenated rosin: increases melt flow, adhesion, and flex life

Care should be exercised to disperse the resin or plastic in block polymer rubber goods adequately by fluxing at sufficiently high temperature (see discussion on processing). One should also be aware that the morphology of blends can be altered to some extent by various conditions of molding or extrusion, causing some differences in physical properties. This latter effect is especially evident in blends with polystyrene. For example, compression molding below the softening point of polystyrene (< 220°F) can result in considerably higher hardness than molding at higher temperature (see Fig. 18.12). Extruded or injection-molded specimens containing polystyrene usually have properties more similar to those obtained by compression molding at the lower temperature.

Figure 18.12 Effect of mold temperature on hardness of blends with

polystyrene. (From Ref. 126, used with permission from Van Nostrand Reinhold Co.)



E. Blends with Other Rubbers

Block polymers are compatible with most conventional rubbers and many types of blends are possible. Green strength and processing characteristics of conventional rubbers may be improved by the addition of block polymer. Conversely, block polymer properties such as solvent resistance or ozone resistance can be upgraded by blending with minor amounts of some rubbers including neoprene, nitrile, polyurethane, propylene oxide, or ethylene—propylene—diene (EPDM). High green-strength rubbers are best suited for these blends in order to maintain optimum physical properties. Probably of most commercial significance at present are blends composed of unsaturated block polymers with 15–30% of a high green-strength EPDM. Such compositions exhibit good thermoplastic properties and a high degree of ozone resistance. Examples of such blends are shown in Table 18.21. Process conditions can affect the degree of ozone resistance obtained, with high shear operations (such as injection molding) or relatively low temperature in compression molding (< 290°F) thought to be most favorable.

F. Stabilizers

A number of factors should be considered in choosing stabilizers for block polymers and their compounds. Of primary importance are the type of central block in the polymer, compounding ingredients to be added, and process and application requirements. In general, block polymers with a butadiene or isoprene segment are similar in stability to SBR rubber or natural rubber, respectively. Polymer with a saturated elastomeric block is inherently more stable.

Sufficient stabilizer is added by the manufacturer to prevent degradation during normal finishing operations and storage of polymer. For many applications, additional stabilizer may not be required. However, under some conditions deterioration can occur due to oxidation, ultraviolet radiation, or ozone attack. Non-staining stabilizers are preferred for most applications utilizing these elastomers. Food and Drug Administration (FDA) acceptance of stabilizers is also required for many uses.

Table 18.21 Ozone Resistance of Blends

Polymer (60/40 butadiene/styrene)

100 80 75 —

High green strength EPDM — 20 25 100

300% Modulus (MPa) 4.5 4.3 4.0 —

Tensile (MPa) 27.6 23.1 17.9 7.10

Elongation (%) 750 670 630 1010

Ozone resistance Poor Fair-Good Good Good

Source: Ref. 126.



Oxidative conditions such as are encountered in air at elevated temperatures may cause polymer with a butadiene block to crosslink, increasing in hardness and viscosity; an isoprene segment will undergo scission when oxidized, becoming softer and less viscous. Polymer with a saturated center segment can also be affected by high-temperature exposure, usually resulting in reduced viscosity. Combinations of phenolic antioxidants or dithiocarbamates with dilauryl thiodipropionate effectively increase oxidative stability.

Ultraviolet radiation can cause surface embrittlement of neat, unsaturated polymers and, depending on formulation ingredients, can cause discoloration and embrittlement of compounded stocks regardless of the chemical composition of the polymer center block. Only stable plasticizers and resins should be included in compounds intended for outdoor use, where direct exposure to ultraviolet light might cause objectionable property changes. Certain chemical additives such as nickel dibutyldithiocarbamate, benzophenones, or benzotriazoles will increase resistance to ultraviolet light. Also, fillers that absorb or reflect ultraviolet radiation (carbon black, zinc oxide, titanium dioxide) can be added to increase UV stability. Typical stabilizers used in the rubber industry are shown in Table 18.22.

When under stress, unsaturated block polymers are susceptible to attack by ozone. However, ozone protection is easily attained by blending with minor amounts of EPDM or ethylene vinyl acetate polymers (see Sec. VIII.E). Also, chemical antiozonants including nickel dibutyldithiocarbamate, dibutyl thiourea, or certain waxes offer some ozone protection.

G. Miscellaneous Additives

Process aids or lubricants may be added to reduce viscosity, decrease tackiness, increase flow, and/or improve surface gloss of finished articles. Stearic acid (0.5–2 phr) and some metallic stearates are excellent process aids. Certain waxes, polyamides, polyethylene glycols, and low-density polyethylenes are also effective.

Pigments, dyes, or color concentrates are often added to obtain various shades of color. Colorants may be dispersed during initial mixing, or dry blended and subsequently dispersed during fabrication. Stability (durability) and ease of incorporation should be considered in choosing colorants. Inorganic pigments

Table 18.22 Typical Stabilizers Used in the Rubber Industry

Typical antioxidants Typical antiozonato UV inhibitors

Hindered phenolic Nickel dibutyldithio- Substituted benzotriazole

Cresol carbamate Hindered amine

Phosphite Dibutylthiourea Benzophenone

Zinc dibutyldithiocarbamate Monobenzoate



generally possess good stability and are easy to disperse. Organic pigments often produce brighter colors and are more soluble in organic solvents and polymers. Dyes also produce bright colors, but many are relatively poor in stability and tend to bleed or migrate.

Other additives that are sometimes utilized include antistatic or antiblocking agents, flame retardants, fungicides, and blowing agents. Although added for specific effects, additives often cause some change in general characteristics; their effect on all properties of interest should be determined. A summary of the effects of compounding ingredients is provided in Table 18.23.

IX. Processing

A. Mixing

As noted, the high-strength plastomers may be used alone or in admixture with other ingredients [126]. When mixing or compounding is desired, several methods may be considered: solution (or liquid) mixing (including emulsions and plastisol-type mixes), melt processing, and dry blending.

B. Solution Mixing

This method is useful for the preparation of solvent-based adhesives, sealants, and coatings. Triblock copolymers are soluble in a wide range of common, inexpensive solvents. The polymers dissolve quickly and display fast solvent release. Since two phases are present, a hard segment of polystyrene and an elastomeric segment, both must be considered in the selection of solvents and additives. Both segments must be dissolved if the polymer is to be considered truly in solution. Dissolution of the polystyrene domains temporarily destroys the network, which reforms on subsequent solvent release, thus restoring strength to the polymer or compound.

The morphology of the polymer can differ if deposited from different sol-



vents and this can affect properties of coatings or adhesives [88]. Triblock copolymers often display lower solution viscosity than SBR random copolymers or natural rubber (depending on molecular weight). The structure of the polymer (linear or branched) must be considered since this will affect viscosity; branched polymers give lower viscosity than linear at the same molecular weight (see earlier discussion under Properties).

Good solvents for elastomeric block copolymers of styrene with polybuta-diene or polyisoprene include cyclohexane, toluene, methyl ethyl ketone, diethyl ether, and styrene. Mixtures ofsolvents are often found practical, for example, (a) naphtha-toluene, (b) hexane-toluene, or (c) hexane-toluene-ketone. The viscosity is sensitive to the solvent blend composition as broadly indicated in Figure 18.13.

The thermoplastic elastomers can be converted to emulsions by suitable methods. Films and coatings can then be prepared by evaporation of water. Since the polystyrene will tend to be deposited in discrete particles, higher strength will be developed if the film or coating is heated after drying.

Block copolymers based on styrene and butadiene can absorb large quantities of oil and still provide some useful properties. Another useful processing method involves the combination of oil and ground polymer to make fluid mixture, which can then be formed into a product by compression molding, injection molding, rotational molding, casting, and perhaps other methods. As the mixture is heated, the oil is incorporated into the polymer forming a solid (usually soft) product.

Figure 18.13 Solution viscosities in hexane-toluene blands. (From Ref. 126,

used with permission from Van Nostrand Reinhold Co.)



C. Melt Mixing

Block copolymers soften on heating and can be processed easily using equipment designed either for rubber or for plastics. Internal mixers of various types are preferred for addition of compound ingredients, but conventional roll mills are also suitable. Proper temperature control is necessary in order to achieve maximum efficiency in melt mixing. Block polymers can be fluxed at 200–250°F; however, higher temperatures are often required to disperse high-softening-point resins, fillers, or other additives. Banbury dump temperatures of 280–320°F are usually adequate; a few additives, such as polypropylene, may increase requirements to 350°F or more.

Dispersion of resins and fillers is best accomplished by early addition in the mix cycle and fluxing with rubber before addition of plasticizers or softeners. Block polymers will build up heat during mastication, readily softening and mixing with moderate amounts of added materials. If a high plasticizer level is required, addition should be in several increments to prevent the mix from becoming lubricated, with resultant slipping on rotors. Compounds to be mixed with high levels of filler and oil may require addition of part of the oil initially in order to flux the dry mixture.

A typical Banbury mix schedule follows:

Formulation

Polymer (60/40 butadiene/styrene) 100

Calcium carbonate 150

Polystyrene 40

Naphthenic oil 60

“B” Banbury, 158°F, 118 rpm rotor speed

0 min: add polymer, polystyrene, and calcium carbonate

280°F (approximately 2 min): add ½oil

Power-up: add remaining oil

280–320°F (approximately 4½ min): dump. Sheet off on roll mill at 240°F.

A roll mill temperature of 200–250°F is generally satisfactory for banding stocks. If a temperature differential exists between rolls, compounds will tend to band on the hotter roll; for this reason, it may be advantageous to operate with the front (preferred) roll 15–25°F hotter than the back roll.

Hot melt adhesive formulations containing plastomers are readily mixed in either a sigma-blade-equipped mixer or by simple stirring in a heated vessel. Mixer temperatures of 330–380°F are common.

D. Dry Blending

Dry blends can be prepared by mechanically mixing ground or powdered block polymer with other components at a temperature less than that required to flux. In



most formulations, added oil tends to bind an even coating of filler and other ingredients to polymer particles, resulting in a homogeneous, free-flowing mixture. This mixture can be fed directly to fabrication equipment capable of handling powder forms. Here the material is fluxed prior to formation of finished articles. Energy requirements may be reduced by this method of mixing compared to melt mixing and subsequent finishing operations.

Intensive mixers (of the Henschel type) or ribbon blenders are suitable for dry blending. Typical mix procedures are shown below. Dump temperature should be kept below the point (depending on formulation) at which ingredients tend to stick together, losing their free-flowing characteristics. Components should be fairly uniform in particle size. If large differences exist, the mixture may tend to segregate or layer during storage or finishing operations.

High-Intensity Mix Procedure: Dry Blending

Use cooling water on jacket; 625 rpm rotor speed (rotor speed required will vary with capacity of mixer; generally reduced with increasing mixer capacity).

0 min: Add all dry ingredients.

0.5 min: Begin addition of oil. If oil used is of high viscosity, it should be heated to about 120°F. Addition time is dependent on the batch size and polymer particle size. Mix until compound appears free flowing.

2–5 min: Dump. Temperature is usually less than 150°F.

Ribbon Blender Mix Procedure: Dry Blending

0 min: Add all ingredients except oil, other liquids, and small amount of filler.

0.5 min: Add liquids slowly, preferably through a spray head. Mix until ingredients are dispersed; add remainder of filler.

5–15 min: Dump.

E. Extrusion

Extrusion techniques similar to those established with other thermoplastics can be used to form film, tubing, and other cross-sections from block polymers and their compounds. Low die swell is exhibited by these materials, permitting the use of relatively simple dies to produce complex sections.

A temperature gradient should be maintained between feed zone and die, with the following ranges typical:

Barrel temperature (°F)

Feed zone 240–270

Intermediate zone 250–300

Final zone 270–340



Adjustments may be required to optimize rates and prevent excessive melt temperatures, depending on formulation. Excessive heat increases the possibility of degradation and also weakens the extrudate, increasing the chance for distortion before cooling.

Extruder size can vary considerably; satisfactory results have been demonstrated with L:D ratios of 10:1–25:1. The longer barrel may be preferred for cold feed, a shorter barrel for hot feed (direct from internal mixer or sheet-off mill). In either case, low-compression screws are recommended.

F. Molding

Thermoplastic block polymers can be molded by most conventional methods including injection, compression, rotation, or blow molding. A wide selection of equipment exists. A brief discussion and general guidelines applicable to the subject rubbers follow.

Injection molding block polymer compositions can generally be accomplished easily with melt temperature in the 300–400°F range and mold temperatures of 60–150°F. Melt fracture or excessive orientation can occur if melt temperature is too low (producing insufficient flow); degradation is possible if temperature is too high. Injection time should be kept to a minimum to avoid appreciable cooling before the mold is completely filled, possibly causing skin formation or layering. Relatively large gates, short sprues, and short runners are desirable. Due to low shrinkage (usually less than 1%) and high friction characteristics, ejection of molded parts may be difficult compared to that encountered with plastics. This problem can be alleviated by tapering and/or rounding mold cavity corners and edges or by use of additives.

To obtain satisfactory specimens by compression molding, it is recommended that copolymers or their compounds be placed in a hot mold and allowed to soften before enough pressure is applied to initiate flow into the cavity. Mold temperatures are usually 250–350°F. The ideal temperature should provide good flow under pressure, but cause little deformation of specimens during removal from the mold. In practice it is sometimes necessary to cool the mold (below softening point of the composition) before opening, to prevent distortion of the sample. The use of mold release agents and/or molding between sheets of foil is also helpful.

Blow molding techniques are similar to those established for polyethylene. In general, imperfections in mold finish are less critical with block polymer compositions than with polyethylene or other crystalline thermoplastics. However, tearing of the parison could be more of a problem since block polymers are usually lower in melt strength. If tearing occurs, corrective steps suggested are to (a) reduce melt temperature as low as possible (near 300°F), (b) reduce mold closure rate, and (c) round off corners at point of “pinch-off.”



G. Sealing and Bonding

Articles prepared from block polymers and their compositions can be sealed or bonded by a number of procedures. Methods include sealing by heat, solvent, microwave, ultrasonic treatment, adhesive bonding, or thermoplastic welding. These processes can also be employed to form laminates with plastic or metallic materials. Equipment can range from very simple to highly sophisticated systems, permitting quick and easy assembly of parts and efficient sealing of films. Block polymers remain elastomeric at point of sealing or welding.

X. Applications

The properties or combination of properties of elastomeric materials determine, to a great extent the ultimate uses [125,127]. For the thermoplastic elastomers or plastomers, three general areas of use have developed: adhesives, sealants, and coatings; plastics modification; and molded and extruded elastomeric items.

Some of the more notable features of thermoplastic elastomers based on polystyrene blocks connected with polydiene or polyolefin type segments are listed in Table 18.24 and the general areas where these properties should be of interest are indicated.

Characteristics of the neat polymers considered with the effects of compounding ingredients serve as a guide to the properties that can be achieved in compounded stocks. As an additional aid, the following list shows some of the variations in important properties that can be obtained.

Hardness (shore A) 25–95

Flexural modulus (MPa) 34–138

300% modulus (MPa) 0.7-12.4

Elongation at break (%) 100–1200

Tensile strength, 26.5°C (80°F) 3.4–34.5

Resilience (%) 20–85

Specific gravity 0.9–1.2a

Deformation temperature (°C, °F in parentheses) 38–132 (100–270)

Tear strength (kg/cm) Up to 56 (500 lb/in)

Brittleness temperature Down to <–100°F

Volume resistivity (ohm-cm) Up to 1016



Table 18.24 Features of Block Polymers

Application areas

Adhesives, sealants, coatings

Molded anextruded

items

Plastic modification

Soluble in many organic solvents X

Fast solvent release X

High strength X X X

Compatible with numerous resins X X

Compatible with many other elastomers

X

Compatible with many plastics X X

Good tear resistance X X X

Good abrasion resistance X

Resistant to crack growth X

Restistant to shear X

Low brittle point X X X

High elongation X X X

High coefficient of friction X

Good resilience X

Hardness can be varied over wide range

X

Low specific gravity X X X

High purity X X X

Light color X X X

Thermoplastic X X X

Available in various forms X X X

Does not need vulcanization X X

Easy mixing X X

Fast, smooth extrusion X X

Easy molding X X

Low mold shrinkage X



Scrap can be recycled X

High permeability X

High oil levels can be used X X

Good electrical properties X X

Resistant to bases, weak acids, water, alcohol

X

Ozone and oxidation resistant (Special grades or compounds)

X X X

X, Denotes applications where features are especially attractive.



Table 18.25 Modification of Polystyrene

Melt index

Tensile(MPa)

Elongation(%)

Notched izod impact (J/m)

General-purpose polystyrene 3.5 63.2 9 21.3

(0.4 ft-lb/in)

Above polystyrene blended with a 4.0 38.3 38 96.0

80/20 butadiene/styrene radial block

(1.8 ft-lb/in)

polymer

sealants. Thermoplastic block polymers described in this chapter can be formulated to provide excellent quality of tack, quick stick, peel strength, and resistance to shear [128–132]. More detailed discussions on the use of TPRs in adhesive industry can be found in Chapter 19.

Block polymers based on styrene and dienes are used to modify properties of plastics in both polymerization and mechanical melt blending systems. Polystyrene can be toughened considerably by the addition of block copolymers. An example of impact improvement in polystyrene is provided in Table 18.25.

Resistance to both impact and stress cracking has been improved in polyolefins by incorporation of thermoplastic block polymer. More detailed discussions on the use of TPRs for the modifications of plastics can be found in Chapter 19.

Among molded and extruded elastomeric items, footwear is the major application for styrenic TPRs. Discussions can be found in Chapter 19 of this book. Several specific applications for styrenic type thermoplastic rubbers or where there is now consideration of their use are listed below.

Adhesives or hot melt pressure-sensitive, contact construction, product assembly)

Sporting goodsToys

Coatings (for chemical milling, paper, fabrics) Wire covering

Sealants Flooring products

Sound insulators Hose and tubing

Modifier for asphalt Gaskets, seals, weatherstrip

High-impact plastics Pharmaceutical items

Plastics resistant to stress cracks Auto parts



Soluble in a wide range of common solvents

Fast dissolution and solvent release

Processible as a melt

Easily formulated with many common compounding materials

Their scrap can be recycled

Low die swell

Benefits in performance include the following:

High strength at moderate temperature

Highly elastic and resilient

Flexible at low temperatures

Resistant to most acids and bases

Can be formulated to produce a wide range of property combinations for many different end-uses

No vulcanization residues

Good color and transparency

Additional benefits of the saturated rubbers (Kraton G series):

Resistance to ozone, oxygen, and UV light

Resistance to degradation at elevated temperatures

Major limitation: Limited upper useful temperature. Strength and hardness become unacceptable as the softening point of the polystyrene is approached.

References

1. N.R. Legge, G. Holden, and H. E. Schroeder, Thermoplastic Elastomers, A Comprehensive Review, Hanser Publishers, Munich/Vienna/New York, 1987.

2. 2. M. Szwarc, Nature, 178, 1168 (1956).


4. A. V. Tobolsky, Rubber World, 138, 857 (1959).

5. N. R. Legge, Rubber Chem. & Tech., 60, No. 3, G83 (1987).

6. G. Holden and R. Milkovich, US patent 3,265,765.

7. J. T. Bailey, E. T. Bishop, W. R. Hendricks, G. Holden, and N. R. Legge, Rubber Age, 98, 69 (1966).

8. M. A. Luftglass, W.R. Hendricks, G. Holden, and J. T. Bailey, Machine Design, 38, 1914 (1966).



15. R. P. Zelinski, US patents 3,251,905 and 3,287,333; and Phillips Petroleum Company, Br. patent 895,980.

16. L. H. Tung, et al., US patents 4,172,100; 4,172,190; 4,182,818; 4,196,153; 4,196,154; 4,200,718; 4,201,729; and 4,205,016.

17. L. H. Tung, G. Y. S. Lo, and D. E. Beyer, Macromolecules, 11, 616 (1978).

18. H. Hooker and G. Latterman, Macromol. Chem., 158, 191 (1972).

19. H. Hooker and G. Latterman, J. Polym. Sci., Polym. Symp., 54, 361 (1976).

20. A. D. Broske, T. L. Huang, R. D. Allen, J. M. Hoover, and J. E. McGrath, in Recent Advances in Anionic Polymerization, T. E. Hogen-Esch and J. Smid, Eds., Elsevier, Amsterdam-London, 1987, p. 363.

21. Shell Chemical Company, Br. patent 1,035,873 (a composition patent); L. M. Porter, US patent 3,149,182; G. Holden, et al., US patent 3,231,635; and F. Dudley, et al., US patent 3,390,207.


23. H. L. Hsieh, US patents 3,639,521 and 3,778,490.



26. J. Darcy, US patent 3,598,887.

27. J. Darcy, US patent 3,632,682.

28. M. M. Wald, US patent 3,700,633.

29. H. E. DeLaMare, US patent 3,766,301.

30. H. E. DeLaMare, US patent 3,594,452.


32. Shell Chemical Company, Br. patent 1,025,295.

33. R. T. Prudence, US patent 3,949,020.


35. H. L. Hsieh, Rubber Chem. & Tech., 49, No. 5, 1305 (1976).

36. G. Kraus and C. J. Stacy, J. Polym. Sci., A-2, 10, 657 (1972).

37. M. Morton, in Block Polymers, S. L. Aggarwal, Ed., Plenum Press, New York-London, 1970, Ch. 1.

38. M. Morton, Encyclopedia of Polymer Science and Technology, John Wiley and Sons, New York, 1971, Vol. 15, p. 508.

39. H. L. Hsieh, J. Polym. Sci., A-1, 8, 533 (1970).



50. V. P. Skatalov, et al., (Voronezh Synthetic Rubber), USSR patents 401,153 and 401,155; Br. patents 1,409,956; 1,414,924; Fr. patent 2,245,686.

51. R. E. Cunningham, Fr. patent 2,229,710; Br. patent 1,264,741.

52. L. H. Tung and G. Y. Lo, in Advances in Elastomers and Rubber Elasticity, J. Lal and J. E. Mark, Eds., Plenum Press, New York/London, 1986, p. 129.

53. R. E. Cunningham, J. Appl. Polym. Sci., 22, 2907 (1978).

54. C. W. Moberly, in Encyclopedia of Polymer Science and Technology, Interscience Publ., New York, 1964, Vol. 7, p. 557.

55. R. V. Jones, C. W. Moberly, and W. B. Reynolds, Ind. Eng. Chem., 45, 1117 (1953).

56. L. A. Mango and R. W. Lenz, Macromol. Chem., 163, 13 (1973).

57. H. J. Harwood, D. B. Russell, J. A. Vertke, and J. Zymonos, Makromol. Chem., 163, 1 (1973).

58. J. C. Falk and R. J. Schlott, Agnew Makromol. Chem., 21, 17 (1972).

59. Y. Mohajer, G. L. Wilkes, I. C. Wang, and J. E. McGrath, Polymer, 23, 1523 (1982).

60. W. P. Gergen, Kaut. Gummi Kunst. 37, 284 (1982).

61. H. L. Hsieh and H. C. Yeh, in Advances in Elastomers and Rubber Elasticity, J. Lal and J. E. Mark, Eds., Plenum Press, New York-London, 1987, p. 197.

62. H. Rachapudy, G. G. Smith, U. R. Raju, and W. W. Graessley, J. Polym. Sci., Polym. Phys. Ed., 17, 1211 (1979).

63. Bridgestone Tire, Br. patent 1,198,195.

64. T. Yoshimoto, et al., US patent 3,541,064.

65. A. F. Halasa, US patents 3,872,072 and 3,868,354.

66. Shell Chemical Company, Br. patent 1,030,306.

67. F. C. Loveless, et al., US patent 3,855,185.

68. A. I. Yakubchik, et al., USSR patent 165,883.

69. N. R. Legge, et al., in Applied Polymer Science, J. C. Craver and R. W. Tess, Eds., Organic Coatings and Plastics Chem. Div., ACS, Washington, D.C. 1975, p. 349.

70. W. R. Haefele, et al., US patent 3,333,024.

71. M. M. Wald, et al., US patents 3,595,942; 3,700,633; and 3,706,817.

72. R. C. Jones, US patent reissue 27,145.

73. O. Johnson, US patent 3,415,759.

74. H. B. Minor, US patent 3,507,934.

75. H. L. Hassell, US patent 3,644,588.



88. J. F. Beecher, L. Marker, R. D. Bradford, and S. L. Aggarwal, J. Polym. Sci., C, 26, 117 (1969).

89. C. W. Childers and G. Kraus, Rubber Chem. & Tech., 40, 1183 (1967).

90. H. Hendus, et al., Kolloid-Z, 216–217, 110 (1967).

91. D. J. Meier, J. Polym. Sci., C, 26, 54 (1969).

92. E. T. Bishop and S. Davison, J. Polym. Sci., C, 26, 54 (1969).

93. G. Holden, E. T. Bishop, and N. R. Legge, Proc. Int. Rubber Conf., 1967, Maclaren and Sons, London (1968), p. 287; J. Polym. Sci., C, 26, 37 (1969).

94. G. Holden and J. R. Legge, in Thermoplastic Elastomers, A Comprehensive Review, N. R. Legge, G. Holden, and H. E. Schroeder, Eds., Hanser Publishers, Munich/Vienna/New York, 1967.

95. R. F. Boyer, Rubber Chem. & Tech., 36, 1303 (1963).

96. D. H. Ender and R. D. Andrews, J. Appl. Phys., 36, 3057 (1965).

97. M. Morton, Rubber Chem. & Tech., 56, 1069 (1983).

98. W. P. Gergen, Hydrogenated block copolymers, paper presented at the 124th Meeting of the ACS Rubber Division, Houston, Oct. 25–28, 1985.

99. G. Kraus, in The Stereo Rubbers, W. M. Saltman, Ed., John Wiley and Sons, New York, 1977, Ch. 11.

100. J. F. Henderson, et al., J. Polym. Sci., C, 16, 3121 (1968).

101. E. Fischer and J. F. Henderson, J. Polym. Sci., C, 26, 149 (1969); 30, 459 (1970).

102. G. L. Wilkes and R. S. Stein, J. Polym. Sci., A-2, 7, 1525 (1969).

103. G. Uchida, et al., J. Polym. Sci., A-2, 10, 101 (1972).

104. T. Miyamoto, et al., J. Polym. Sci., A-2, 8, 2095 (1970).

105. G. Kraus, et al., J. Polym. Sci., A-2, 10, 2061 (1972).

106. M. Matsuo, et al., Polymer, 9, 415 (1968).

107. G. E. Molau, in Block Polymers, S. L. Aggarwal, Ed., Plenum Press, New York London, 1970, Ch. 1.

108. R. P. Lewis and C. Price, Polymer, 13, 20 (1972).

109. J. Dlugosz, et al., Kolloid-Z, Z. Polymer, 242, 1125 (1970).

110. A. Keller, et al., Kolloid-Z, Z. Polymer, 238, 385 (1970).

111. D. R. Paul, J. E. St Lawrence, and J. H. Troell, Polym. Eng. Sci., 10, 70 (1970).

112. E. Vanzo, J. Polym. Sci., A-1, 4, 1727 (1966).

113. M. Shibayama, T. Hashimoto, and H. Kawai, Macromolecules, 16, 16 (1983).



124. G. Kraus and J. T. Gruver, J. Polym. Sci., A, 3, 105 (1965).

125. G. Holden, in Thermoplastic Elastomers, A Comprehensive Review, N. R. Legge, G. Holden, and H. E. Schroeder, Eds., Hanser Publishers, Munich/Vienna/New York, 1987.

126. J. R. Haws and R. F. Wright, in Handbook of Thermoplastic Elastomers, B. M. Walker, Ed., VanNostrand Reinhold Company, 1979, Ch. 3.

127. G. Kraus and D. S. Hall, in Block Copolymers, Science and Technology, D. J. Meier, Ed., MMI Press Symposium Series Vol. 3, Harwood Academic Publishers, London/New York, 1983.

128. O. L. Marrs, F. E. Naylor, and O. L. Edmonds, J. Adhesion, 4, 211 (1972).

129. O. L. Marrs, R. P. Zelinski, and R. C. Doss, J. Elastomers Plastics, 6, 246 (1974).

130. O. L. Marrs and O. L. Edmonds, Adhesive Age, 14, 15 (1971).

131. L. D. Jurrens and O. L. Marrs, Adhesive Age, 18, 31 (1975).

132. R. A. Gray and O. L. Marrs, Adhesive Age, 19, 51 (1976).



19 Applications of Styrenic Thermoplastic Rubbers in Plastics Modifications, Adhesives, and Footwears

I. Thermoplastics Modifications With Styrenic Thermoplastic Rubbers

A. Introduction

Most plastics are susceptible to brittle fracture. Although energy-absorbing processes, such as crazing or shear yielding, operate in these plastics, they do so only in highly localized areas around the crack tip. In order to increase the toughness, energy-dissipating mechanisms must occur over a large volume of the plastic. And they must simultaneously limit the growth and breakdown of voids and cracks to prevent catastrophic failure of the plastic 1–3.

The incorporation of dispersed rubber particles in plastics has been found to achieve this goal. Rubber particles act as stress concentrators. When rubber particles are close together, the stress field between them is enhanced. This in turn initiates localized energy-absorbing mechanisms from many sites, allowing a much greater volume of the matrix polymer to be involved. The result is a tough system that can absorb large amounts of energy without catastrophic failure.

Optimum stress field enhancement depends on the following conditions: high rubber loading, small particle size, and uniform dispersion. In order to obtain these conditions, it is necessary to have the right degree of compatibility between the rubber and the plastic. Incompatibility leads to large rubber particles with poor toughening. The other extreme, miscibility of the rubber and the plastic also leads



to poor results — significantly reduced stiffness related properties, HDT and flexural modulus, of the plastic.

The most efficient rubbers for impact moderation will also have a low glass transition temperature (Tg) and an internal strength mechanism. The low Tg improves the low-temperature toughness of the plastic. A rubber with an internal strength mechanism should be able to absorb energy not only upon initiation of crazing or shear yielding in the matrix but also by the work necessary to rupture the rubber particles themselves.

Other requirements for the rubber depend on the application for the final material. One requirement might be processability of the rubber, which will influence the melt flow of the final blend, while others may include the rubber's heat aging and weatherability characteristics, its transparency within the plastic, and its effect on the surface appearance of the final blend.

Styrene-diene (S-D) thermoplastic elastomers are used in the modification of plastic resins (Table 19.1). These block copolymers are discussed in Chapter 18. They are readily processed in typical plastic equipment and frequently allow the user to reduce the melt temperature by 11–22°C (20–40°F), thus reducing the power requirements and cost. In many cases the modifier can be mixed with the desired resin in a simple blend during fabrication.

This enables the fabricator to reduce the inventory by varying the amount of one modifier to meet requirements in several resins. The proper modifier for each application can be chosen by considering a few generalizations that can be made for these polymers.

1. Higher butadiene levels offer greater impact improvement and flexibility.

2. Higher-molecular-weight polymers are more efficient modifiers.

3. Lower-molecular-weight polymers disperse more easily in a blend.

By knowing the mixing capability of the processing equipment to be used, one can make the best modifier choice for the application.

B. Polystyrene Modification

Blends with Polystyrene

Blends of crystal polystyrene (PS) with elastomeric block polymers of S-D-S and (S-D)n-X types show an increase in impact strength and elongation (Table 19.1). Tensile strength and flexural modulus remain at a good level for most applications. In certain applications, these blends have the additional feature of high gloss that can be used to good advantage. The polymer modifiers do not reduce the gloss level, as is expected with other modifiers.

Polystyrene is the most widely used plastic resin. Its versatility allows it to be used in many ways. There are, however, applications for polystyrene that need improvement in impact strength and elongation. These include film and sheet



Table 19.1 Properties of Blends of PS and S/B Block Copolymer

Values for various compounds

Composite/properties

Crystal PS with S/B radial triblocka

Crystal PS with S/B diblockb

Polystyrene (%) 100 95 85 100 70 50 100

Block polymer (%) 0 5 20 0 30 50 0

Melt flow (5 kg., 200°C)(g/10 min)

2.5 2.5 1.7 12.5 7.8 5.8 3.6

Flexural modulus (103 psi) 448 434 356 428 326 249 260

Tensile strength (psi) 7520 7670 6220 4990 4120 3980 5800

Elongation (%) 4 5 6 3 25 50 75

Izod impact (ft-lb/in)

Notched 0.3 0.7 2.8 0.4 0.6 0.6 2.1

Unnotched — — — 1.6 2.4 3.7 —

Heat-deflection temperature under 450 kPa. load (°C)

84 82 78 88 87 86 —

Shore D hardness — — — 88 80 75 75

abutadiene/styrene radial-structure triblock polymer (Phillips Solprene 414).

b70/30 SB diblock polymer, of which 40% is block styrene (Solprene 314D).

Source: Ref. 1.



extrusion, injection molding, and thermoforming. Diene/styrene block polymers, which can be added during fabrication, are recommended.

The use of a diblock polymer (by definition, it is not a thermoplastic elastomer) as a low-haze modifier for crystal polystyrene will serve as an example for our discussion of sheet and film extrusion. The modifier can be used to produce low-haze sheet that processes like impact PS and has good toughness.

The properties can, of course, be varied considerably by extrusion conditions and rates, as well as by screw configuration or the polystyrene molecular weight. However, a number of trials have been made with extruders and extrusion conditions virtually unchanged from impact PS settings. A melt temperature range of 420–440°F is recommended for best results in sheet extrusion, with a maximum melt temperature of 450°F. The use of higher temperatures may result in gelling of the butadiene/styrene copolymers.

Clear disposable products are a rapidly growing segment of the polystyrene market. Unless a modifier is used, it is necessary to use oriented polystyrene or some specialty resin to achieve the combination of clarity, processability, and toughness needed. Since the costs of these products is high, there has been considerable interest in ways to achieve economy. One means is to use a blend (40–45%) of a 70/30 styrene/butadiene diblock polymer containing 40% block styrene with general-purpose PS. By definition, the diblock polymer is not a TPR.

Table 19.1 shows typical data for blends of 30% and 50% diblock polymer with polystyrene. The block polymer used in this example contains 70% styrene, of which 40% is block styrene, the remaining 30% being randomly incorporated in copolymer block. In addition to increasing flexibility and toughness in the final product, the addition of the block polymer causes increased melt strength, resulting in improved sheet extrusion processability and permitting a deep draw in thermoforming. Clarity is excellent up to a thickness of about 0.5 mm.

The sheet can be easily thermoformed into various individual-portion containers: cups for ice-cream sundaes, cocktail glasses, sandwich packages, a piewedge package, or lids. At 45% addition, the cost of materials is only 16% higher than for crystal PS; oriented sheet cost is at least 60% higher. Since the blend can be made during sheet fabrication, the manufacturer also has control of the sheet supply and can reuse the regrind. Cups can also be made by injection molding to reduce scrap and breakage during shipping.

One further application in the polystyrene area is in film. Polystyrene has long been used in rubber bale wrap, produce wrap, window envelopes, stationery supplies, and others, but unless oriented, its toughness and elongation have been inadequate. The use of modifiers such as a 60/40 butadiene/styrene radial triblock polymer will improve toughness and elongation in a simple cast film or blown tubing while maintaining clarity.

Blends of butadiene/styrene radial (star) triblock polymers with impact polystyrene are shown in Table 19.1. These polymers are considerably more efficient as modifiers when used with impact PS, as can be seen from the impact strength



improvement. This holds true for all impact PS, allowing the development of impact strength to near ABS levels.

The size of the styrene block in S-D-S or (S-D)n-X copolymers is the main determining factor of its effectiveness as the impact modifier. The copolymer with the longest styrene block is the most efficient impact modifier, but also the most difficult to process. Table 19.2 [2] compares the properties of three triblock copolymers with different styrene block sizes in HIPS at the same loading.

In many cases, increased impact strength is accompanied by loss of stiffness. However, ternary blends of crystal polystyrene, HIPS, and block copolymers can be made that not only retain the stiffness of the original HIPS but also show large increase in impact strength. As an alternative, products with the same impact strength, but increased gloss, softening point, and stiffness are possible. Table 19.3 [2] illustrates the properties of several ternary blends.

Polystyrene/styrenic block copolymer blends with higher impact strength and elongation find uses in the fabricating areas of film, injection molding, sheet, and thermoforming. Specific uses are sheet and thermoforming, television (TV) and radio cabinets, clear disposables, decorative moldings for furniture, light diffuser panels, foamed door and window facings, cove base, and others, and scrap reprocessing.

C. Polyethylene Modifications

High-density polyethylene (HDPE) is available in a wide range of grades, suitable for many applications. Even so, a number of uses can benefit from modification. One of these is film. Although HDPE film has good tensile strength and modulus,

Table 19.2 Binary Blends of Various Kraton D Polymers with HIPS

Composition (%w)

HIPS 100 90 90 90

SBS Aa — 10 — —

SBS Bb — — 10 —

SBS Cc — — — 10

Properties

notched izod (ft-lb/in @ 73°F) 1.6 2.3 2.9 3.2

Flex modulus (Mpsi) 310 281 256 255

aSBS A is a linear triblock with 28/72 stryrene/rubber ratio and medium/low molecular weight (Kraton D 1102).

bSBS B is a linear triblock with 31/69 styrene/rubber ratio and medium molecular weight (Kraton D 1101).

cSBS C is a radial triblock with 30/70 styrene/rubber ratio and high molecular weight (Kraton D 1184).

Source: Ref. 2.



Table 19.3 Ternary Blends of Kraton D, HIPS, and Crystal Polystyrene

Composition (parts by weight)

HIPS 100 70 40 40

Crystal PS — 30 60 60

SBS Bb — 5 15 —

SBS Aa — — — 15

Properties

Notched Izod (ft-lb/in) 1.1 1.15 1.75 0.95


Gardner gloss (45/45) (%)

14 14 33 38

Melt index (G) (g/10 min)

15 14 12 17

Softening point (°C) 84 87 94 93

aSBS A is a linear triblock with 28/72 styrene/rubber ratio and medium/low molecular weight (Kraton D 1102).

bSBS B is a linear triblock with 31/69 styrene/rubber ratio and medium molecular weight (Kraton D 1101).

SBS C is a radial triblock with 30/70 styrene/rubber ratio and high molecular weight (Kraton D 1184).

Source: Ref. 2.

it is often found wanting in tear strength and impact resistance. This is true for such uses as merchant bags, heavy-duty bags, and industrial film. Modification can result in considerable improvement in Elmendorf tear, dart drop impact, and bag drop.

When blended into polyethylene film, butadiene-styrene block polymers increase resistance to tearing and tensile impact even though these two polymers are immiscible. Typical block polymer concentrations are 10–30%. Data in Table 19.4 illustrate the effect of a (SB)n-X block polymer on HDPE and low-density polyethylene (LDPE) film properties. In both cases, 10% block polymer caused only minor changes in properties; however, 15–20% resulted in very significant improvement in tear and dart drop. The block polymer and PE pellets can be tumble-blended or meter-fed directly to the film extruder. Data in Tables 19.5 and 19.6 illustrate the effect of a SBS block polymer (Kraton D 1102) on HDPE and LDPE film properties.

Injection-molded items can also benefit from modification. This is most applicable for large molded parts that require high-flow resins. The low-molecular-weight HDPE resins have low-impact strength. It is possible to reduce wall thickness in modified HDPE parts from that required for unmodified resin.

In general, blends of high-density polyethylene with styrenic thermoplastic elastomers (TPE) give products with improved impact resistance and flexibility. HDPE/styrenic TPE blends may be used



Table 19.4 Properties of Films Made with PE Modified with (SB)-X Block Polymer

Values for HDPE and LDPE film

HDPE filma: block copolymerb content (%)

LDPE filmc: block copolymerd content (%)

Property 0 10 20 30 0 10 15

Tensile yield strength (psi)

Machine direction 3670 3300 2700 2200 1100 1030 1060

Transverse direction

2720 3200 2400 1820 2390 1090 1010

Ultimate tensile

Strength (psi)

Machine direction — 4590 4300 3790 1300 2340 2140


— 3750 3760 3640 2600 2170 2110

Elongation (%)



550 610 510 490 470 470 410

Elmendorf tear strength (g)



58 106 275 424 325 406 540

Dart drop at 66 cm (g)

75 80 248 No failure

196 304 372

a1.25 mil blown film; polyethylene density 0.950 g/cc, melt index 0.3.

b60/40 butadiene/styrene radial structure triblock polymer, 40% styrene (Solprene 414).

c2 mil block film; polyethylene density 0.917 g/cc, melt index 2.0.

dButadiene/styrene radial structure triblock polymer, 20% styrene (Solprene 422).

Source: Ref. 1.

Table 19.5 Binary Blends of SBS Polymer with LDPE Film

Composition (%w)



Table 19.6 Binary Blends of SBS Polymer with HDPE Film

Composition (%w)

HDPE 100 90 85 80

SBS polymer (Kraton D1102) — 10 15 20

Propertiesa

Dart impact strengthb (gm) 62 133 150 161

Tensile impact strength (ft-lb/in2)

Machine direction 22 25 34 48

Transverse direction 200 280 210 195

Elmendorf tear strength (gm)

Machine direction 380 680 730 665

Transverse direction 33 100 255 230

Melt flow (E) (gm/10 min) 0.44 0.54 0.58 0.69

aMeasured on blown 1 mil film.

bOne and one-half inch diameter dart, 26 in drop height

Source: Ref. 2.

industrial film, housewares, blow molded products, large molded parts (pails, trash cans, etc.) and sheet and thermoforming.

D. Polypropylene Modifications

In many thermoforming operations, the use of regrind can be a problem, resulting in loss of strength and difficulty in vacuum-forming operations. The use of diene/styrene thermoplastic elastomer modifier can help to alleviate the problems with regrind and can often improve the original properties. Table 19.7 shows data on 20-mil talc-filled polypropylene (PP) sheet modified with 5 and 10% of a radial structure block polymer. Table 19.8 shows the effect of adding various levels of a linear SBS polymer to PP homopolymer.

Blend with butadiene/styrene thermoplastic elastomers gives products with improved impact and low-temperature flex cracking resistance. Blends of this type have considerable practical interest since they enable the manufacturer to combine the high temperature resistance of PP with the low temperature flexibility of polybutadiene.

The factors influencing the use of butadiene/styrene thermoplastic elastomers in PP modification are increased impact strength, increased elongation, ability to be mixed during fabrication, reduction of cost by decrease of part gauge, and improved low temperature properties. General use areas for modified PP include a wide variety of common products, such as housewares, hospital items, film, sheet (thermoforming), and molded horticulture containers.



Table 19.7 Properties of 20 mil PP Sheet Made with (Sb)n-X Block Polymer Blends

Composition/properties Compounds

Composition (%) A C C

PP (virgin, 20% talc) (%) 40 37.5 0

PP (regrind, 20% talc) (%) 60 57.5 90

(SB)-X block polymera (%) 0 5.0 10

Properties of 20-mil Sheet

Melt flow (2160 g, 230°C) (g/10 min) 8.2 6.6 2.9

Modulus of elasticity (103 psi)

Machine direction 107 84 167

Transverse direction 92 75 124

Tensile Strength (psi)



Elongation (%)



Dart impact at 26 in (g) 364 412b 518b

Shore D hardness 25 73 73

aButadiene/styrene radial structure triblock polymer, 40% styrene (Solprene 414).

bAt 60 in height.

Source: Ref. 1.

Table 19.8 Binary Blends of SBS Polymer with PP

Composition (%w)

PP homopolymer 100 90 75 50

SBS polymer — 10 25 50

Properties

Notched izod



@ 73°F (ft-lb/in) 0.95 2.1 7.8 NB

@ 32°F (ft-lb/in) 0.6 1.35 3.15 NB

Falling weight impact strength

@ 73°F (ft-lb/in) 4.6 117 160 >180

@ 32°F (ft-lb/in) <4 37 >180 >180

@ 14°F (ft-lb/in) <4 16 >180 >180


Tensile yield stress (psi) 4640 4060 3260 1960

Ultimate stress, psi 3040 3260 3700 4350

Melt index, g/10 min 3.5 2.8 2.5 1.8

aKraton D 1102.

Source: Ref. 2.



and shows less stress or fold whitening. Through modification it is possible to reduce film gauge 20–25% while maintaining a good balance of physical properties.

Since similar improvements may be realized from use of polyolefin block polymers, SBS is not widely used in polypropylene.

E. Styrene Acrylonitrile Resins Modification

Styrene acrylonitrile (SAN) resins offer a combination of properties that make it an excellent choice for many molding applications. Some of the properties are transparency, rigidity, and strength, plus heat and chemical resistance. There is, however, one property associated with rigidity that should be improved: Low-impact strength can be a problem in several market areas.

The addition of an SB diblock copolymer with 70% total styrene and 40% polystyrene block (Solprene 314D), a low-haze thermoplastic polymer, counteracts the brittleness of SAN resins with no significant reduction in clarity or physical properties. Such blends are formidable candidates for applications requiring a high degree of strength and transparency [4].

Blends of Solprene 314D with SAN can be characterized through the following conclusions drawn from a study of the accompanying data:

1. Comparable clarity is maintained. The blends maintain excellent clarity at all use levels. Light transmittance is reduced somewhat, but see-through remains very good.

2. Izod impact strength is improved at all levels of Solprene 314D addition.

3. General physical properties are maintained at a good level. Tensile strength and flexural modulus are not significantly reduced below the 30% Solprene 314D blend.

4. Resistance to edible oil-induced stress cracking (ESCR) is excellent for the blends that could be tested. There is no reason to suspect any reduction in ESCR at higher SAN ratios.

F. Acrylonitrile—Butadiene—Styrene Resins Modification

Burr [5] described another use of butadiene—styrene block copolymers in acrylonitrile—butadiene—styrene (ABS). When ABS scrap is reprocessed, resulting loss in toughness may be restored by the addition of as little as 5% block polymer (Table 19.9). SB diblock polymers are as effective as SBS or (SB)n-X block polymers in this application.

G. Polyphenylene Ether Modification

Polyphenylene ether (PPE), an engineering thermoplastic, is very compatible with styrene/diene thermoplastic elastomers. In the use of conventional SDS types of copolymers with PPE, extreme care must be taken to avoid excessive degradation



Table 19.9 Effect of Diblock Polymer on Reprocessed ABS

ABS Resin 100a 100b 95b 90b

Butadiene/styrene block copolymerc

— — 5 10

Melt flow, 5 kg, 200°C 0.5 3.8 0.5 0.6

Flexural modulus (MPa) 1400 1000 1320 1300

Tensile strength (MPa) 30.6 23.8 30.5 26.6

Elongation (%) 35 10 50 50

Shore D hardness 70 68 72 67

Izod impact, notched (J/m) 440 133 450 460

Heat distortion, °C @ 1820 kPa 72 60 67 69

aVirgin product injection molded at 440°F.

bReprocessed.

c75/25 butadiene/styrene, 18% block styrene (Solprene 1205C).

Source: Ref. 5.

of the unsaturated rubber due to the high processing temperatures required for PPE.

Saturated (by hydrogenation) triblock copolymers (e.g., Kraton G1650, G1651, and G1652) are recommended where weatherability is important or living degradation during processing is critical.

The presence of the rubbers in PPE imparts increased toughness with corresponding loss of stiffness [2].

Blends of SBS block polymers and poly(2,6-dimethylphenylene oxide), which is compatible with polystyrene, have been described by Shultz and Beach [6]. Moreover, rubbery butadiene—styrene block polymers and their hydrogenated versions can be used for impact reinforcement of poly(2,6-dimethylphenylene oxide)—polystyrene blends [7].

H. Polycarbonate Modification

Gilmore and Modic [8] reported that weak interfacial attraction between polycarbonate and styrenic block copolymers requires careful control of both the structure and use level of the block copolymer for optimum toughening. Blending polycarbonate with 5% styrenic block copolymers (e.g., Kraton G1651 rubber, a saturated linear triblock copolymer and Kraton G1702 rubber, a saturated diblock copolymer) improves impact and heat-aging properties and environmental stress crack resistance.

I. Sheet Molding Compound

“Low-profile” or “low-shrink” sheet molding compounds (SMC) based on fiber-reinforced unsaturated polyester resins and styrene are finding increasing use in automotive applications as a means to reduce weight. These compounds usually



Table 19.10 Typical SMC Formulation

Polyester 50

Block polymer solution (30% in styrene) 50

Calcium carbonate 150

Zinc stearate 3

t-Butyl perbenzoate 1

PEP-100 cure promoter 0.25

Magnesium oxide 1

1 in glass fiber 25–28

aButadiene-styrene, carboxy terminated diblock copolymer such as Solprene 312.

Source: Ref. 9.

require a thermoplastic polymer additive (polyvinyl acetate, polymethyl methacrylate, etc.) to reduce mold shrinkage.

The use of carboxy-terminated butadiene/styrene block copolymers in SMC in place of the thermoplastic results, first of all, in an improvement in impact resistance of these inherently brittle compounds. In addition, and just as important, the block copolymers provide a good balance of mechanical properties, low shrinkage, few sink marks, reduced long-term waviness, improved paintability, and excellent pigmentability [9,10].

Table 19.10 shows typical SMC formulation. Note that the amount of block polymers on the total compound is only about 5% or, based on polymer content, 15%. Table 19.11 shows various physical test results in comparison with a conventional thermoplastic modifier at equal concentration. Internal failure in SMC manifests as audible creaks before the appearance of visible cracks when the

Table 19.11 Properties of SMC with 15% Modifier (Based on Polymer Content)

Modifier

Property

Block copolymera

Thermoplastic (PMMA)

Reverse-impact drop height (cm) 15 9

Failure elongation in radial bend test (%) 1.76 1.26

Torsion test (kg at indicated angle)b

First creak 16 at 45° 9 at 20°

First crack 18 at 54° 14 at 40°

aButadiene/styrene, carboxy-terminated diblock copolymer (Solprene 312).

bData on automobile front end (From Eagle-Picher Industries).



Source: Ref. 19.



part is subjected to bending or torsion. The superiority of the block polymer in this test is particularly pronounced, as is the “reverse impact strength” (dart drop height to cause cracking on the bottom of the sheet).

Shell Chemical Company [11] developed a product (Kraton G1855X) that is a 50/50 blend by weight of Kraton D1300X and G1701X rubbers. Kraton D1300X rubber consists of SB diblock and SBS triblock copolymers. In molding compounds, Kraton D1300X rubber functions as a reactive elastomeric low-profile additive. Kraton G1701X rubber is a saturated S-(E/P) diblock copolymer that imparts thixotropic properties to the molding compound in addition to its contribution to reduce shrinkage.

It is claimed [11] that SMC prepared from Kraton G1855X rubber exhibits outstanding surface appearance, controlled expansion upon molding, excellent mechanical properties, and a high affinity for paints and adhesives.

J. Asphalt Modification

Structure of Block Polymer-Modified Bitumens

Bitumens are complex mixtures of hydrocarbon molecules, ranging from low molecular weight oils to true polymeric species. The latter, which are included in the asphaltene fraction, have been found to have molecular weights of 20,000 and greater. When we wish to blend bitumen and a block polymer, we again encounter the tendency for high polymers to be immiscible, which also causes the phase separation in the block polymer (there are few truly “compatible” polymer pairs). The block polymer absorbs a considerable portion of the lower-molecular-weight or maltene fraction, leaving the rest of the bitumen enriched in asphaltenes. The result is a two-phase structure [14] that is clearly observable under an optical microscope. At the very modest polymer concentration of 14%, the rubber-rich phase is already continuous. In fact, connectivity of the rubbery phase usually sets in between 6 and 8% [12,13]. This is possible because the rubbery phase is highly swollen and hence much more abundant than would be the case in a dispersion of the pure polymer. The degree of swelling depends on the bitumen/polymer combination, but is typically about three times the dry polymer volume.

Evidence from dynamic mechanical and other rheological measurements [13] shows that within the rubber-rich phase polystyrene domains survive. The blend, therefore, exhibits a network-within-a-network structure, as illustrated in Figure 19.1. The result is that the bitumen is converted to a rubberlike material showing essentially complete recovery from large deformations, with remarkable economy of added block polymer.

Solution Temperature and Processing

Block polymer-modified bitumen compositions are prepared [13] by stirring the polymer, in powder or crumb form, into molten bitumen at 180–250°C. High shear mixers are useful, but not essential. Solution times are on the order of 20



Figure 19.1 Schematic diagram of morphology. (From Ref. 12, used with permission

from The Gordon & Breach Publishing Group.)

min to 1 h. The hot melts are fairly stable for about 3–5 h. Large increases in viscosity can occur when maintaining batches at the preparation temperature for extended periods of time. If storage of hot solutions cannot be avoided, lowering the temperature and the use of an inert gas blanket are recommended to minimize degradation. Also, high shear mixing should be replaced by more gentle agitation.

For saturated block copolymer such as Kraton G polymers, an upper processing temperature limit of 240f0C can be used. If the finished blend is to be stored for a period of several days, it is suggested that the storage temperature be maintained at 160°C or below.

The temperature at which a blend becomes a homogeneous melt (Ts) has been found to increase with the asphaltene content of the bitumen, the molecular weight and styrene content of the block polymer, and the concentration [13]. Figure 19.2 shows regions of complete and partial miscibility at 204°C for Solprene 411 in six bitumens in the form of a quasiphase diagram. High asphaltene contents and polymer concentrations tend to favor formation of two-phase melts.

Melt viscosities of block polymer-modified composition are, of course, much greater than those of the pure bitumen. Concentration and molecular weight are the principal polymer variables affecting viscosity. In the oil-extended block polymers, the plasticizing action of the oil is designed to compensate for the higher molecular weight of the rubber. When comparing bitumens containing



Figure 19.2 Regions of complete and partial miscibility at 204°C for

Solprene 411 in bitumens varying in asphaltene content. (From Ref. 12, used with permission from The Gordon & Breach

Publishing Group.)

identical concentrations of a given block polymer, there is often little or no correlation between the viscosity of the modified and neat bitumens. The reason for this is that the solvent quality of various bitumens for the polymer varies with their detailed chemical composition. Other factors affecting this relationship are homogeneity of the melt and possible polymer degradation during mixing. As a consequence, it was reported [13] that no reliable mixture rule for predicting viscosities of modified bitumens. Typical viscosity ranges (at 180°C) are listed in Table 19.12. At 200°C, viscosities are very roughly half of what they are at 180°C.

Penetration and Softening Point

Figure 19.3 shows the effects of temperature and block polymer (Solprene 414) concentration on penetration. Whereas the neat bitumen shows the usual semi-logarithmic relationship, the effect of the polymer modifier increases strongly with temperature [13]. (Concentrations of 30% and higher are included in Figure 19.3 for illustrative purposes only. They are of no practical interest in roofing applications.)

The effect on the ring-and-ball softening point (Tr+b) is shown in Figure 19.4. Tr+b is seen to rise rather quickly at first and approaches a value somewhat in excess of Tg of polystyrene. At higher polymer concentrations the test cannot be



Figure 19.4 Effect of block polymer addition on ring-and-ball softening point (95 penetration bitumen). (From Ref. 12, used with

permission from The Gordon & Breach Group.)

performed properly because the bitumen deforms elastically and tears under the weight of the ball instead of flowing.

High-molecular-weight block polymers produce somewhat larger increases in Tr+b than is shown in Figure 19.4. However, unusually large amounts of polar compounds (nitrogen bases) in bitumen can limit the increase. These compounds apparently have the ability to soften the polystyrene domains of the block polymer.

Stress—Strain Behavior

At polymer concentrations providing a continuous rubber-rich phase, the residual asphalt phase acts somewhat like a reinforcing filler. The stress—strain curve exhibits a maximum or yield value, followed by a region of draw [13]. Figure 19.5 shows the yield strength and stress at 300% elongation for 14% Solprene 411 in six bitumens. Note the excellent correlation of both stresses with asphaltene content of the bitumens.

Because the glass transition of polybutadiene (ca. -100°C) lies so far below that of typical bitumens (usually -25–0°C), the mixed polybutadiene-asphalt phase has a Tg substantially lower than that of the pure bitumen (Table 19.13). This circumstance imparts low temperature ductility to the modified compositions.

Oxidative Aging and Weathering

Unsaturated rubbers, such as the homopolymers and copolymers of butadiene, are rather susceptible to degradation by thermal oxidation and exposure to ultraviolet



Figure 19.5 Stress—strain at 23°C; 14% Solprene 414 in various

bitumens. (From Ref. 12, used with permission from The Gordon & Breach Group.)

Table 19.13 Class Transition Temperaturesa of Modified Bitumens

Tg

Base bitumen penetration value

Modifying polymer

Polymer concentration (%)

Base bitumen

Modified

93 (SB-)-na10 -16 -57

(SB-)-nb14 -16 -52

(SB-)-nc14 -16 -50

73 (SB-)-nb14 -8 -42

118 (SB-)-n14 -1 -36

76 (SB-)-n14 -4 -56, -24b

77 (SB)- n 14 +2 -32

84 (SB)- n 14 -9 -46

a 40% Styrene, medium-molecular-weight (Solprene 414).

b 30% Styrene, high-molecular-weight (Solprene 411).

c 30% Styrene, medium-molecular-weight (Solprene 416).



(UV) radiation [13]. The rubber industry has learned to live with this fact remarkably well; antioxidants are available that strongly retard the degradative processes, chain scission, and crosslinking, without however eliminating them completely. At the concentrations employed in bitumen modification, butadiene-styrene block polymers undergo predominantly chain scission. Table 19.14 shows results of some typical experiments, in which the extent of degradation is judged by gel permeation chromatography (GPC). At the surface of a slab of material exposed to air at 85°C for 8 weeks (a very severe condition), oxidative degradation may be very extensive, but the oxidation is diffusion-controlled and the situation in the interior is not nearly so serious. In the Weatherometer test, degradation is less severe. In practice, the presence of a coating of granules would act as a barrier to UV radiation and diffusion of oxygen, further limiting degradative processes.

The block polymers as marketed do, of course, contain an antioxidant. Additional antioxidants added in formulation have been found to limit further the rate of degradation, but their use has not in general been found necessary. One reason is that the bitumen undergoes hardening on oxidation, whereas chain scission of the rubber causes softening. The result is a partial compensation of oxidative effects on physical properties.

Rheology and Microstructure of Polymer/Asphalt Blends

Bouldin and colleagues [15] reported their characterizations of the viscoelastic properties of polymer/asphalt blends. They also addressed the impact of the judicial selection of asphalt type (i.e., crude slate and methods of manufacture) and asphalt grade (e.g., AC-5-AC-40). The influence of these variables on the rheology, compatibility, morphology, and overall performance of modified asphalt cements were reported.

Work carried out at the Shell Chemical Company [16] showed that the physical and rheological properties for modified asphalt blends are highly dependent on the block copolymer (Kraton rubber) use level. Increase in softening point by 66°C (120°F) and simultaneous decrease in brittle point by 19°C (35°F) are

Table 19.14 Polymer Degradation on Aging at Exposed Surfaces

Condition

Relative OPC peak height (H/Ho)

8 weeks, 85°C, air 0.20–0.35

2 weeks Weatherometer, 100 watt Xe-lamp, 60°C,intermittent water spray

0.55–0.90

Ho, height of polymer elution peak before degradation; H, height at same elution volume after degradation. For no degradation H/Ho =1.

Source: Ref. 13.



commonly provided through modification of asphalt with Kraton rubber. Attempts were made to determine the optimum polymer content by the use of penetration index (PI) [16]. It was observed that at low polymer levels (< 3%) do not have much impact on the PI value. However, a rapid change in PI value occurs between the 3% and 8% use levels. Above about 8% Kraton SBS rubber, the rate of PI value change slows considerably with increasing polymer content, which indicates the 6–8% use level to be an efficient level for temperature susceptibility improvements (a major reason for modifying asphalt with rubber).

It was also shown [16] that stiff modulus is not significantly improved until polymer contents near 6% are provided. Increasing polymer content beyond this level provides higher and higher values for the stiffness modulus of the blend.

Overall, it is recommended [16] that at least 6% neat Kraton rubber be used in the asphalt/polymer blend as a starting point for generating meaningful improvements in the performance of hot mix asphalt concrete. Deviations for this polymer use level should be based upon particular pavement performance requirements.

The typical stress-strain properties for blends of Kraton D rubber and asphalt are shown in Table 19.15.

Recently, Diani and co-workers [17] investigated the complex interaction between bitumen and SBS copolymers. Using dynamic-mechanical spectrometry as a tool, these workers made performance assessments allowing the prediction of viscoelastic behavior in the final service conditions. They found that the SBS copolymer modified-bitumen performs better than neat bitumen in all experimental ranges investigated. This is particularly evident at extreme temperature and frequency values.

Commercial Applications

As an illustration, thermoplastic elastomers produced based on Phillips technology for blends with asphalt are shown in Table 19.16 [18].

These polymeric modifiers listed in Table 19.16 may be used alone as an

Table 19.15 Stress-Strain Properties for Blends of Kraton D Rubber and Asphalt

AC-5 + 3% AC-5 + 6%

AC-5 Kraton D Kraton D AC-20

Maximum stressa @ 4°C (psi) 34 170 592 141

Ductility @ 4°C (cm) 10 48 63 1

Toughness (in-lb) 49 165 328 167

Tenacity (in-lb) 18 126 284 22

aAs measured by force-ductility method.

Source: Ref. 16.



Table 19.16 Solprene (or Finaprene) Modifiers and Their Contribution in Asphalt

Percentage

Product

Styrene

Butadiene

Oil

Molecular structure

Molecular weight

Product form

Solprene 411 or Finaprene

30 70 — Radial High Pellet

Powder


30 70 — Radial Medium Pellet

Powder


40 60 — Radial Medium Pellet

Powder


40 60 50 Radial High Pellet

Powder

Bale


25 75 — Linear tapered diblock

Medium Crumb

Bale

Source: Ref. 18.



Table 19.17 Kraton Polymers for Asphalt Modifications

Kraton D (SBS & SIS)

Kraton G (SEBS)

Neat polymer D1100 G1600

Oil extended D4100 G4600

Easy mixing D4400 N/A

Source: Ref. 16.

asphalt modifier and contribute benefits to the compound. However, blends of these polymers combine the unique contributions of individual polymers. For example, one can combine two polymeric modifiers to improve the resistance to flow and low temperature flexibility or combine another pair of modifiers to improve flow resistance, low-temperature flexibility, and viscosity.

Shell Chemical Company offers several series of polymers for asphalt modifications (Table 19.17).

According to Shell[16], the D4400 series polymers are specially formulated for easy mixing without the need for special equipment. The D4100 series requires specialized moderate- to high-shear mixing equipment for incorporation into asphalt. The neat D1100 Series requires very high shear mixing equipment and peripheral hardware/piping, and tankage in an appropriate configuration.

Summary

Styrenic thermoplastic elastomer-modified asphalt binders deliver improved performance over unmodified asphalt binders by providing:

Increased flexibility at low temperatures

Better resistance to flow and deformation at high temperatures

Significantly reduced temperature susceptibility

Improved tensile strength

Increased stiffness modulus at high temperatures

Increased asphalt—aggregate adhesion

Greater resistance to surface abrasion

These characteristics imparted by the elastomers produce improved performance as asphaltic products such as pavement binders, seal coats, highway joint sealants, waterproofing membrances, coatings, pipeline mastics, pipeline wrapping tapes, and others.

K. Blends of Incompatible Thermoplastics

Block and graft copolymers are often used as compatibilizers for the blending of otherwise incompatible thermoplastics. Most polymer pairs are immiscible and,



therefore, form separate phases comprised essentially of the fuel components when mixed [19–21]. For many polymer pairs there is relatively poor adhesion between these phases [22,23] which precludes efficient transfer of stresses across the interface, leading to inferior mechanical behavior of the mixture. This led to the work of Molau [24–27], Riess [28–32], and others [34–38] in blend additives in an attempt to circumvent the interfacial problems encountered in the blends of immiscible polymer mixture. These early works established the fact that properly designed block or graft copolymers act as interfacial agents or emulsifiers for immiscible polymer pairs. A comprehensive review of this subject was published in 1978 [23]. A more focused review on the use of hydrogenated styrenic thermoplastic rubbers (S-EB-S) as compatible agents was published in 1987 [39].

In this review [39], Paul, mostly based on the experimental work developed in the laboratories of Teyssie and others [40–48] and his own [49], concluded that block copolymers containing segments of polystyrene and hydrogenated polybutadiene can cause signifcant changes in the mechanical properties of immiscible blends of polystyrene and various polyolefins or “compatibilize” them. This result is attributed to the interfacial role played by the block polymer in the blend, resulting in improved adhesion between the phases.

Improved properties were also reported [50] on blends of polyethylene and poly(ethylene-terephthalate) by the addition of a low-molecular-weight S-EB-S block polymers.

II. Adhesives

A. Introduction

Adhesives, sealants, and coatings are one of the most important applications for styrenic thermoplastic elastomers. Linear (SDS) and radial (SD-)x block copolymers of styrene (S) and dienes (D = butadiene or isoprene) are extensively used as pressure-sensitive adhesives, spray and contact adhesives, panel and construction mastics, sealants, and coatings [51–60]. The isoprene-containing polymers have the lowest modulus and therefore can be readily tackified. The butadiene-containing polymers are generally preferred for use in fairly stiff adhesives such as construction adhesives or laminating adhesives. The primary reasons for this are economics and the fact that commercially available SBS or (SB)x block copolymers can be readily formulated to have a high modulus [57]. The hydrogenated block copolymers, S-EB-S type, are preferred for use in weather-resistant adhesives and applications such as sealants.

B. Solubility and Solution Properties

A wide range of solvents will dissolve styrenic thermoplastic elastomers. In choosing a solvent, one must consider the individual solubilities of the polydiene and polystyrene segments of the molecule. These polymers will not dissolve in



aliphatic solvents alone, but are soluble in a mixture of aliphatic and other more polar solvents.

A solvent or solvent blend having solubility parameters between 7.6 and 9.3 can reasonably be assumed to dissolve the block copolymers. There may be anomalies if the solvent has a solubility parameter near the extremes or if the solvent is moderately or strongly hydrogen bonded. A blend of solvents and nonsolvents, or even of two nonsolvents, may be used if the solubility parameter is within the proper range.

Table 19.18 shows some solvents and nonsolvents for the block copolymers, listed in order of increasing solubility parameter. The effects are shown in Figures 19.6–19.10. Figure 19.11 shows the low shear viscosity of 15% weight styrene S-B-S block copolymer in solvent blends composed of various ratios of isooctane (a nonsolvent for polystyrene) and toluene. In pure toluene, both blocks of the polymer are well dissolved and the polymer gives relatively low solution viscosity. The viscosity drops slightly as part of toluene is replaced by isooctane. As a large part of toluene is replaced by isooctane, the viscosity begins to increase very rapidly. This rapid increase in viscosity indicates that polystyrene end-blocks are not well dissolved and are beginning to aggregate, forming a network structure in the solvent [57].

Marrs and colleagues [52] compared (SB-)x (radial branched) and (SBS) (linear) block copolymers and found that the radial branched copolymers have

Table 19.18 Some Nonsolvent Type and Solvent Type Organic Compounds and the Corresponding Solubility Parameters

Solubility parameter

Nonsolvent type

Textile spirits 7.2

Hexane 7.3

Rubber solvent 7.4

Acetone 10.0

Solvent type

80 Hexane–20 toluene 7.6

70 Hexane–30 toluene 7.8

75 Hexane–25 8.0

acetone

Cyclohexane 8.2

1.1.1-Trichloroethane 8.5

Xylene 8.5

50 Hexane–50 ketone 8.6

Toluene 9.1



Figure 19.6 Typical solution viscosities of an (S-B-)x

block polymer (S/B = 30/70) in toluene/hexane mixtures. (From Ref. 51.)

Figure 19.7 Typical solution viscosities of an (S-B-)x block polymer

(S/B = 30/70) in toluene. (From Ref. 51.)



Figure 19.8 Typical solution viscosities of an (S-B-)x block polymer

(S/B = 30/70) in 70% hexane and 30% toluene. (From Ref. 51.)

reduced melt viscosity as well as solution viscosity compared to the linear analogues having the same molecular weight.

For polymers of one monomer ratio, increasing branching at constant molecular weight should effect a decrease in molecular volume and increasing branching at constant molecular volume would require increases in molecular weight. Properties of the polymers should then be commensurate with these changes in branching. Solution viscosities confirm these predictions, as illustrated in Figure 19.12 for polymers containing 30% styrene. Increasing branching does markedly reduce solution viscosity for rubbers of equal molecular weight, and, conversely, allows higher molecular weights at the same viscosity.

Figure 19.9 Typical solution viscosities of an (S-B-)x block

polymer (S/B = 30/70) in 75% hexane and 25% acetone. (From Ref. 51.)



Figure 19.10 Typical solution viscosities of an (S-B-)x block polymer (S/B = 30/70) in 1,1,1-trichloroethane.

(From Ref. 51.)

Figure 19.11 Viscosity of 15% with solution of block copolymer

(Kraton D-1102) in isooctane/toluene blends. (From Ref. 57, used with permission from Rubber Div., A.C.S.)



Figure 19.12 Solution viscosity of linear, trichain and tetrachain block

polymers in toluene/naphtha (25/15).

Although branching in polybutadienes can, depending on the molecular weight value, either increase or decrease in melt viscosity [53], in radial branched block polymers [54] the result is a reduction in viscosity (Figs. 19.13, 19.14). There is, however, less discrimination between tri- and tetrachain polymers in the melt flow values.

The lower solution viscosity of the branched block copolymers is advantageous in adhesive applications, particularly pressure-sensitive adhesives. Formulations must be inexpensive, rapid-evaporating solvents, such as naphtha, and be of relatively low viscosity. However, naphtha is a nonsolvent for polystyrene, and the polystyrene segments of any block polymer greatly impair solubility because of aggregation in these types of solvents. The least possible amount of a good solvent, like toluene, is added to obtain a solution in which solvent cost is balanced against viscosity (ease of application).

Therefore, the adhesive properties of the various triblock polymers were compared, as deposited from a typical formulation, as a function of formulation viscosity. The solvent system contained 10% toluene.

In comparison to linear triblock polymers of equal styrene content and molecular weight, tri- and tetrachain branching reduces the viscosity of the adhesive solutions, allowing a substantial increase in shear resistance at equal viscosity or equivalent shear at substantially lower viscosity (Fig. 19.15).

Shear resistance increases with molecular weight of all the block copoly-



Figure 19.13 Mooney viscosity of SBS and (SB-)x block polymers (30%

styrene).

mers, but at equal formulation viscosities the radial forms are of higher molecular weight (Table 19.19). This, and possibly a higher degree of aggregative crosslinking, may be responsible for the improved performance. All formulations of triblock polymers are more fluid than those containing 40% styrene, even though their molecular weights are the same (Table 19.20).

In poor solvents, such as the 10/90 toluene/naphtha, increased styrene content promotes aggregation even in solution. To obtain equivalent shear resistance, the penalty is much higher formulation viscosity or an increased proportion of toluene.

Figure 19.14 Melt flow of SBS and (SB-)x block polymers (30%

styrene).



Figure 19.15 Shear resistance of block polymers as a function

of formulation viscosity.

Pressure-sensitive adhesives must possess tack: the ability to adhere rapidly to a substrate. As is evident in Figure 19.16, radiality hardly affects tack, but increasing the total styrene content of all triblock polymers to 40% substantially reduces this vital property. It was concluded [52] that compared to linear analogs of the same composition and molecular weight, radial triblock polymers of butadiene and styrene with terminal polystyrene segments exhibit decreased solution viscosities and unexpectedly low melt viscosities. The former property is particularly advantageous in formulating pressure-sensitive adhesives with superior shear resistance without loss of tack. The latter property should be advantageous in formulating lower-viscosity hot melt adhesives. Tack is little affected by branching, but is reduced by increased styrene content in both linear and radial triblock polymers.

Table 19.19 Comparison of Triblock Polymers at Equal Formulation Viscosities

30/70 Sty/Bda

polymerMolecular

weightFormulation

viscosity (Cps)Shear resistance

(h to fail at 90°C)

Linear 84,000 1520 1.0

Trichain 136,000 1580 2.4

Tetrachain 182,000 1820 2.8

aStyrene/butadiene.

Source: Ref. 52.



Table 19.20 Properties of Triblock Polymers Containing 30% and 40% Styrene

Styrene Molecular Formulation Shear resistance

(%) Weight viscosity (Cps) (h to fail at 90°C)

Linear 30 77,000 440 0.1

40 77,000 3920 1.1

Trichain 30 115,000 400 0.4

40 115,000 1620 1.2

Tetrachain 30 182,000 1820 2.8

40 182,000 14000 6.4

C. Compounding Additives

The styrenic thermoplastic elastomer itself has practically no inherent adhesive character. The proper balance of adhesive properties must be developed by the formulator. Most adhesives based on the styrenic block copolymers contain combinations of resins, stabilizers, plasticizers, or fillers.

The unique properties of these block copolymers depend upon the two-phase network, so it is important to consider the effect of compounding ingredients on that network. High-melting-point resins compatible with polystyrene reinforce these polymers and improve the high-temperature tensile strength. Materials that soften polystyrene (i.e., plasticize the polystyrene phase) lower the hardness and tensile strength. Materials compatible with elastomeric phase make the product

Figure 19.16 Tack of block polymer adhesives as a function of

formulation viscosity.



softer. Tackifying resins improve the tack with some loss in tensile strength. Table 19.21 is a listing of typical resin types that are useful in formulating styrenic block copolymers. This is not intended as an all-inclusive listing: similar resins would be expected to give similar results.

Typical fillers used in adhesive formulation are clay, talc, and silica. Most plasticizers used with styrenic block copolymers are oils and waxes. In adhesive formulations these materials decrease hardness and modulus, eliminate drawing, reduce melt and solution viscosity, improve compounding processibility, decrease cohesive strength, or increase plasticity [58].

D. Stabilizers

The polydiene center portion of the styrenic block copolymers is unsaturated and susceptible to oxidative crosslinking and hardening. This can be detrimental to adhesive properties, so stabilizers are added to the adhesive formulations.

Table 19.21 Typical Resins Used in Compounding

Name Softening point (°C) Type Supplier

Reinforcing resin

Picco 6000 100, 110, 120, 130, 140 Aromatic resin Hercules

Neville LX-685 100, 110, 144 Heat-reactivehydrocarbon

Neville

Cumar LX-509 160 Cumaron Neville

Dymerex 150 Dimerized rosen Hercules

Schenectady SP-154 71–88 Phenolic SchenectadChem.

Aromoco 18 100, 115, 145 α-Methylstyrene Amoco

Neville 1035 175 Hydrocarbon Neville

Tackifying resin

Akron M 85,125 Saturated hydrocarbon Arakawa

Escorez 2101 95 Modified hydrocarbon Exxon

Super St Tac 80, 100 Hydrocarbon Reichhold

Zonareg 70, 85, 100, 115 Terpene Hercules

Piccolyte Alpha 100, 115 α-Pinene Hercules

Piccolyte Hm110 110 Modified terpene Hercules

Foral 85, 105 Hydrogenated rosin Hercules

Stabelite Ester 10 83 Hydrogenated rosin ester

Hercules

Pentalyn H 104 Hydrogenated rosinester

Hercules

Nevtac 80, 100, 115 Synthetic polyterpene Neville



It is important to know what the aging conditions will be; stabilizers do not afford the same degree of protection under all conditions. The choice will depend upon cost and effectiveness, and the service conditions. Some antioxidants that have been found to be effective with the block copolymers are listed in Table 19.22. Some antiozonants and UV inhibitors are shown in Table 19.23 and 19.24. The synergistic effect with certain blends of stabilizers should also be considered in aging protection.

E. Morphology and Viscoelastic Behavior of Styrenic Block Copolymers in Pressure-Sensitive Adhesives

Kraus et al. [55] studied the morphology of compositions of linear (SDS) and radial (SD-)y block copolymers of various S/D ratios and molecular weights formulated as pressure-sensitive adhesives using electron microscopy of ultrathin sections. These researchers also made dynamic viscoelastic measurements on these compositions at 35 Hz between -90° and +140°C or higher. Pressure-sensitive tack and holding power were determined and interpreted in terms of morphological and rheological properties. The conclusions of this work [55] are that the formulation of a successful butadiene-styrene or isoprene-styrene block polymer pressure sensitive adhesive requires the following:

1. Use of a tackifying resin that is compatible with the rubbery blocks, but incompatible with the polystyrene blocks.

2. A rubber-continuous morphology of the adhesive. Connectivity of the polystyrene domains leads to low creep compliance and loss of tack, but higher shear resistance.

3. Development of tack requires a high creep compliance to permit the establishment of full contact with the substrate during the short bonding time.

4. Whenever full contact with the substrate is established, the probe tack value is governed by dissipation of strain energy, which increases with styrene content of the block polymer.

5. Effective rubber-compatible tackifiers significantly raise Tg of the rubbery phase. By doing so the adhesive moves into a region of higher mechanical loss, resulting in increased dissipation of strain energy.

F. Applications

Pressure Sensitive Adhesives

This is probably the largest single use for styrenic thermoplastic elastomers. The products are usually applied as hot melts or solvent applications. The use of solvent is attracted by the low solution viscosity of these polymers; concerns about air quality in recent years tend to make it less desirable. Hot melts are more commonly practiced in industry.



Table 19.22 Antioxidants for Various Aging Conditions

Name (70°C [158°F] Oven)

Composition

Supplier

AgeRite White Sym(diβ-nephthyl-p-phenylene diamine) R. T. Vanderbilt

AgeRite Resin D Polymerized 1,2-dihydro-2,2,4-trimethylquinonoline

R. T. Vanderbilt

Antioxidant 330 1,3,5-Trimethyl-2,4,6-tris(3,5-ditert-butyl-4-hydroxy-benzyl)benzene

Ethyl Corporation

Vanox 13 Modified hindered phenol R. T. Vanderbilt

Irganox 1010 Tetrakis[methylene-3(3',5'-ditert-butyl-4-hydroxyhydrocinnamate)] methane

Ciba Geigy

Irganox 1076 Octadecyl-3-(3'-5'-tert-butyl-4-hydroxyphenyl)propionate

Ciba Giegy

Antioxidant 2246 2,2'-Methylene-bis(4-methyl-6-tert-butylphenol)

American Cyanamid

Oxygen bomb

AgeRite White Sym(diβ-nephthyl-p-phenylene diamine) R. T. Vanderbilt

AgeRite Resin D Polymerized 1,2-dihydro-2,2,4-trimethylquinonoline

R. T. Vanderbilt

Antioxidant 2246 2,2'-Methylene-bis(4-methyl-6-tert-butylphenol)

American Cyanamid

177°C (350° F) melting

Polygard Tri(nonylated phenol) phosphite Uniroyal Chemical

AgeRite Geltrol Mixed, phenyl phosphite R. T. Vanderbilt

Irganox 1076 Octadecyl-3-(3'-5'-tert-butyl-4-hydroxyphenyl)propionate

Ciba Geigy

Source: Ref. 51.

Table 19.23 List of Antiozonants

Antiozonant Composition Supplier

NBC Nickel dibutyl-dithiocarbamate du Pont

Pennzone B Dibutyl thiourea Pennwalt

Ozone Protector 80 Reinhold

Source: Ref. 58.



Table 19.24 List of UV Inhibitors

Material Composition Supplier

Tinuvin P/300 Series Substituted benzotriazole Ciba-Geigy Co.

Tinuvin 770 Hindered amine Ciba-Geigy Co.

Permasorb MA Benzophenone National Starch and Chemical Co.

Eastman RMB Monobenzoate Eastman Chemical Co.

Source: Ref. 58.

The high temperatures necessary for mixing hot melt adhesives require that the polymers be protected from oxidation. A blanket of carbon dioxide or nitrogen, in combination with antioxidants, is useful in preventing degradation. SIS polymers are preferred over SBS polymers for hot melt adhesives, because the SIS polymers provide a higher level of tack and are less likely to form gel during processing. The end uses include various kinds of tapes, labels, and adhesive fasteners. The mechanism of the formation of tacky products by combining the resins and the thermoplastic elastomer has been described in the literature [51,52] and discussed in the preceding section.

Four typical formulations and adhesive properties are shown in Tables 19.25 to 19.28. Since a soft product is necessary to form the adhesive bond, softer block copolymers are preferred as the materials from which to formulate pressure-sensitive adhesives. Thus, block copolymers of low styrene contents are often

Table 19.25 Pressure-Sensitive Adhesive Using Styrene-Isoprene Radial Block Copolymer

Compounding

(IB-)x type TPE (16% S) a 100

Polyterpene resin 100

Naphthenic oil 30

Antioxidant 3

Property

Batch viscosity, 182°C (Pa.s) 20.3

Polyken probe viscosity (N) 8.4

Rolling ball viscosity (cm) 1.7

Holding power, 58°C (h) >24

180°C peeling (N/m) 870

aSolprene 423 type.

Source: Ref. 63.



Table 19.26 Pressure Sensitive Adhesive Using S-I-S Block Copolymer

Adhesive

Ab Bc

Formulationa

S-I-S block copolymer (Kraton D-1107) 100 100

C5 midblock resin (Wingtack 95) 80

C5 midblock resin (Escorez 1310) 140

Naphthenic process oil (Shellflex 371) 10

Phenolic antioxidant (Irganox 1010) 1 2

Pigment (TiO2) 5

Properties

Adhesiond (N/m) 850 1400

Sheare (h) 75 2

SAFTf (°C) 105 75

aConcentrations are in parts by weight.

bAdhesive was cast from toluene onto 25 µm polyester film at a dry adhesive film thickness of 35 µm.

cAdhesive was hot melt coated on polyethylene prelaminated to cotton scrim duct tape backing at an adhesive film thickness of 125 µm.

dPressure Sensitive Tape Council (PSTC) method number 1.

ePSTC-7, using 13 × 13 mm contact area, 2 kg weight.

fTemperature at which a 25 × 25 mm lap shear bond of tape to polyester fails under a 1 kg weight in a cabinet whose temperature is raised at 22°C/h.

Source: Ref. 57.

used. In Table 19.26, adhesive A is considered to be a high-performance adhesive suitable for use in a wide variety of paper- or film-backed tapes. Adhesive B is considered to be a medium-performance adhesive developed for use in a general-purpose duct tape. In some formulations, as shown in Tables 19.27 and 19.28, combinations of styrenic thermoplastic elastomer and a diblock copolymer of styrene—diene types (S-B, S-I) as well as the hydrogenated versions such as S-EB are used. This diblock content is non-load-bearing. It, and the resins in the elastomer phase, weaken the adhesive. However, this weakening can be tolerated as long as it is not carried to the point where it causes cohesive failure of the adhesive during service. Partial replacement of SIS polymer with SB



Table 19.27 Pressure-Sensitive Adhesives Using Combination of Block Copolymers

Compounding

(SB-)x type TPE (30% S) a 100 100

Tapered diblock copolymer (B/S = 75/25 PS block + 20%)b — 60

Floral 85 100 100

Antioxidant 1 1

Property (30% toluene solution)

Viscosity (poise) 2450 750

Rolling ball viscosity (cm) 5.2 5.2

Polyken probe viscosity (g) 630 930

180°C, peeling (N/m) 670 550

Creep (180°C after 24 h) 0.12 —

Creep (120°C after 24 h) No No

aSolprene 411 type.

bSolprene 1205 type.

Source: Ref. 51.

Construction Adhesives

Construction adhesives (assembly adhesives) are generally formulated with thermoplastic elastomer alone. Both saturated and unsaturated elastomeric blocks are used. Hot melt application is more usual than application from solution. Tack is not important and so harder products are satisfactory. These adhesives are usually

Table 19.28 Hot Melt Pressure-Sensitive Adhesive Using Combination of Block Copolymers

Compounding

(SB-)x type TPE (40% S) a 100 60

Tapered diblock copolymer (B/S = 75/25 PS block + 20%)b — 40

Floral 85 110 100

Antioxidant 5 5

Property

Viscosity, 350°F (poise) 46,000 72,000

Rolling ball viscosity (cm) — 9.6

180°C, peeling (N/m) 820 650



aSolprene 414 type.

bSolprene 1205 type.

Source: Ref. 51.



Table 19.29 A Typical Formulation of Construction Adhesive

S-B-S polymer, 30% styrene 100

Hydrocarbon resin (softening point 100°C) 100

Hydrocarbon resin (softening point 140°C) 100

Hard clay 200

Polymerized 1,2-dihydro-2,2,4-trimethylquinoline 2

Dibetanaphthyl-p-phenylenediamine 0.5

Toluene 25

Textile spirits 350

Ethyl alcohol 12.5

Source: Ref. 51.

formulated to contain both resins that are compatible with the polystyrene phase and resins (and possibly oils) that are compatible with the elastomer phase [59]. The relative proportions of the two types of resin determine the hardness of the adhesive, while the total amount of resin added determines the viscosity of the final product [59].

Table 19.30 Construction Adhesives

Panel masticadhesive

Contact adhesive

Formulationa

S-B-S block copolymer (Kraton D-1107) 100 100

C9 endblock resin (Picco 6140) 100

Rosin Ester midblock resin (Pentalyn H) 100 37.5

Coumarone indene endblock resin (Cumar R-16) 37.5

Filler (CaCO3/silica) 500

Phenolic antioxidant (Plastanox 2246) 2 1.5

Solvent: Hexane 134 210

Cyclohexane 67

Toluene 70

Acetone 70

Properties

10 30



Optimum open time (min)

Lap shear strength 1 day after assemblyb (MPa) 1.7 0.8

Lap shear strength 10 days after assemblyb (MPa) 2.6 —


bShear strength of 25 × 25 mm wood to wood lap shear bond measured at Instron crosshead speed of 0.55 mm/min.

Source: Ref. 57.



A typical formulation for construction adhesive is shown in Table 19.29. Table 19.30 shows formulations for a panel mastic adhesive and a contact adhesive based on a relatively high-molecular-weight S-B-S polymer [57]. Creep resistance is very important for both types of adhesives. Both adhesives contain a combination of elastomer-compatible and polystyrene-compatible resins. Although both resins contribute to adhesion, the primary function of the resin combination is to develop the proper mechanical properties in the cured adhesives [57]. A combination of a relatively low-volatility solvent (toluene) and high-volatility solvents are used here. The small amount of toluene allowed the adhesives a satisfactory open time. The high-volatility solvents ensure that the rate of drying and shear strength build is adequate [57].

Hot melt assembly adhesives are used in a wide range of applications including packaging, laminating, and product assembly. These adhesives are mixed without solvent. After being hot melted, these materials are usually cooled and chopped into convenient form. During application, the adhesive is remelted and applied molten to one of the pieces to be bonded. The other piece is mated under modest pressure while the adhesive is still hot. Bond strength develops as the adhesive cools [57].

Two examples of hot melt assembly adhesives are shown in Table 19.31. In

Table 19.31 Hot Melt Assembly Adhesives

Bookbinding adhesive

Carton-sealing adhesive

Formulationa

S-EB-S block copolymer (Kraton D-1102) 100

S-EB-S block copolymer (Kraton G-1650) 100

Rosin ester midblock resin (Floral 85) 100

C5 midblock resin (Wingtack 95) 100

C9 endblock resin (Picco 6140) 100

C9 endblock resin (Piccotex 120) 50

Aromatic free process oil (Tufflo 6204) 50

60°C MP paraffin wax (Shellwax 234) 100 150

Phenolic antioxidant (Santowhite Crystals) 3 3

Properties

Melt viscosity (Pa.s) 4.6 @ 160°C 5.0 @ 195°C

Hardness (Shore A) 64 61

Tensile yield stressb (MPa) 3.7 1.9


bSamples cut from 1.8 mm thick film, pulled at 23°C at Instron jaw separation rate of 250



Table 19.32 Solvent Cements

S-B-S Polymer (30% styrene) 100 100

Hydrogenated rosin ester 15 40

Coumarone-indene resin 70 70

Acetone 200 200

Hexane 250 250

Toluene 100 100

T-Peel strength on cotton duck (kN/m)

10.7 10.7

Fabric failure Fabric failure

Source: Ref. 51.

the carton-sealing adhesive, a saturated block copolymer was fused so that the adhesive would have better degradation resistance when held hot and exposed to air in the adhesive melt reservoir [57].

Other Applications

Other applications such as sealants, contact cements and coatings have been described [65–68]. Examples of a solvent cement formulation are shown in Table 19.32. Sealants are mostly made with the S-EB-S block copolymers. Both hot melt and solvent-based applications are important. Chemical milling of metals (aluminum, titanium) is the most important application for coating based on either unsaturated or saturated styrenic thermoplastic elastomers [59]. Little is known about the formulations.

III. Styrenic Thermoplastic Rubbers In Footwear Applications

A. Introduction

Along with many other population-dependent industries, the shoe industry is enjoying a rapidly expanding market that requires constant attention to bolster a faltering profit margin. In a situation of this sort, where competition is extremely keen, it is not enough only to choose the most economical raw materials. To keep its share of the market a product must continually evolve into a more desirable item. The manufacturer's dream is to produce an item that more people want at no increase in cost, or, better still, reduced manufacturing cost.

The soles of shoes used to be mainly made from rubber and PVC. The rubber soles have wear-resistance, high strength, good elasticity but their technology of production is complicated. In addition to the higher production cost, the products are often less attractive to the consumers because of odor and dull color. The soles



made from PVC are formed by injection molding. The processing is simple; the cost of production is low, and the production rate is high. However, soles made from PVC lacks elasticity and have poor skid resistance. It is not comfortable to wear.

Solution styrene-butadiene diblock copolymers such as Solprene (or Finaprene) 1205 or random copolymer such as Solprene (or Finaprene) 300 have desirable attributes such as higher abrasion resistance, longer flex life, reduced shrinkage, and increased rates of processing. For shoe and shoe product manufacturers, reduced cost may be realized by the use of these solution styrene-butadiene diblock copolymers. A significant cost factor is their ability to accept greater extension with no loss in performance. In fact, it is often found that the optimum properties of these materials are best attained through high extension. Another feature of solution polymers is their extreme high purity. The result is more rubber hydrocarbon present in each pound of polymer.

Both random copolymers (solution SBRs) and tapered block copolymers (e.g., Solprene [or Finaprene] 1205 with 25% total styrene and 18% block poly-styrene) can be used in footwear applications with advantages over emulsion styrene-butadiene rubbers (emulsion SBRs). The elastomers require vulcanization. In shoe heel compounds, the greatest advantage for the solution polymer is the much improved abrasion resistance.

Block copolymers, because of the presence of block polystyrene, impart excellent flow properties and high hardness to rubber compounds. They are also very beneficial in improving flex and abrasion resistance. Some higher- (> 25%) styrene content tapered block copolymers can replace blend of random SBR, oil-extended or not, and high-polystyrene resin.

The block copolymers have properties important to the footwear industry, such as extendability, abrasion resistance, flex life, resilience, low shrinkage, translucency, and thermoplasticity. Other advantages, more subtle although just as important, are the processing and cost advantages to be gained by the use of these rubbers, such as excellent compatibility with oil and fillers, shorter curing cycles, low calendaring nerve, and improved mold flow.

Commercial compounds have been developed for the following footwear applications using styrene-butadiene tapered block copolymers with advantages [68,69]. Examples of integral molded soling and microcellular soling are shown in Tables 17.8 and 17.9, respectively.

Other examples include the following:

Slab solings (super-high-quality to low-cost compounds)

Translucent compounds

Injection molded shoe soling

Integral molded shoe soling

Low-density microcellular soling



Medium-density microcellular soling

High-density microcellular soling

Translucent microcellular soling

EVA-based microcellular soling

Boot upper

Sponge inner soling

Slipper sponge inner soling

Styrenic thermoplastic rubbers are also extensively used in footwear applications [70,71]. The important difference between the self-reinforcing block copolymers and chemically vulcanized rubbers is the reversibility of the crosslinks formed by the block copolymer. The network structure formed by the association of the polystyrene segments can be disrupted by heating the polymer to a temperature above the softening point of the polystyrene. When the temperature is reduced, the styrene “domains” are reformed and the original properties are restored. For the sole made from styrene-butadiene thermoplastic elastomers (TPE), no vulcanization is required and the scraps can be reused. The low-temperature flexibility and skid resistance are very good. It is comfortable to wear. Its color, brightness, transparency, and hardness can be changed by using different compounding.

Physical properties of the styrenic thermoplastic elastomers approximate those of filled SBR and natural rubber vulcanizates. Higher styrene content increases modulus and hardness; lower-styrene-content polymers are more rubbery and flexible. Both have good low-temperature flexibility and resistance to flex cracking.

B. Compounding

The compounding of styrenic thermoplastic elastomers is similar to that of SBR and natural rubber. The most significant difference, of course, is that no vulcanizing agents are necessary and would, in fact, cancel the reprocessing advantages. Thermoplastic elastomers have inherently high green strength and good low temperature properties, and are considered to be self-reinforcing polymers. The addition of mineral fillers and certain resins modify properties such as tear resistance, abrasion resistance, flex life, and hardness. Polystyrene resins and naphthenic oils influence abrasion resistance, modulus, and melt flow. Balanced levels of these ingredients are used to control the compound properties and cost.

Fillers

The mineral fillers used in soling are whitings and silicas. These are selected to reduce cost, impart hardness, stiffness, abrasion resistance, and flex life. The silicas provide the greatest stiffening to soling compounds, and produce marked improvements in flexing. This filler type is used to produce a translucent soling. In



dry blend mixing, a small amount of silica is helpful in drying up batches formulated with high oil content. Excessive dusting may occur when silicas are used at high loading levels.

Resins

A low-softening-point crystal polystyrene is an excellent additive for soling compounds. Poly-alpha-methyl styrene is another useful resin. These are clear thermoplastic resins with low specific gravities. The addition of polystyrene acts to reinforce a soling compound. Poly-alpha-methyl styrene is less reinforcing, but increases melt flow. Both resin types exhibit excellent compatibility with styrenic thermoplastic elastomers.

To be dispersed adequately by dry blend mix procedures, polystyrene must be usually ground. However, some polystyrene types are available in reactor grades. In this form, polystyrene has the appearance of coarse sand, and does not require further grinding in order to be blended adequately by the dry blend mix procedure.

The poly-alpha-methyl styrene resin in the 990°C (210°F) melt range has been found to produce the best melt flow, and also provides the best translucency in soling. If better resistance to deformation upon demolding is required, higher melting resin provides the optimum performance. Other resins that have found use in soling are the vinyl-toluene alpha-methyl-styrene copolymer types and certain of the polyindene types. The polyindene types, while promoting flow and fabric adhesion, are limited in their use by the odor and the poorer color produced in the compound.

Oils

Oils are used as processing aids and softeners in styrenic TPE-based soling compounds. Used in combination with the mineral fillers and the resins described, oil permits reduction in compound hardness and cost and improves the melt flow. Overall, naphthenic oils have been found to be best for use with the thermoplastic rubbers. Aromatic oils, while compatible, produce poor color and lack good color stability. Paraffinic oils are generally marginal in compatibility, and only low-viscosity grades do not bleed readily from the compounds. This marginal compatibility has been found detrimental to adhesion.

Several factors should be considered before an oil is selected. Among these are:

1. Color.

2. Ultraviolet light absorption (at 260 mµm). For good color stability, this value should be less than 2.

3. Flash Point. This value should be higher than the maximum processing temperature.

4. Viscosity. Range: 420–540 SUS at 37.8°C.



Stabilizers

For general protection, it is recommended that 0.5 phr of dilauryl thiodipropionate and 0.3 phr of a hindered phenolic type antioxidant be incorporated into the compound. In translucent soling, up to 0.5 phr of ultraviolet light absorber should be added for further sunlight protection. In white or colored solings, titanium dioxide or zinc oxide provides some screening of ultraviolet rays. These materials can be used in conjunction with ultraviolet light absorbers for maximum protection, but addition of ultraviolet light stabilizers to white soling is often not cost effective.

Antiblock Agents

The purpose of antiblock agents is to dust pelleted compounds and, occasionally, finished product to prevent them sticking together. Among the most effective of these materials are zinc stearate, silicas, and aluminum hydrate.

Internal Lubricants

Internal lubricants are used to prevent sticking of compounds during mixing, improve extrusion during pelletizing, and to provide good mold release. Zinc stearate, stearic acid, and erucamides provide the best performance.

C. Processing

Most types of equipment in use today by the rubber and plastic industries can be used to mix thermoplastic elastomers. The most common of these methods are dry blending, extruder mixing, internal mixing, and continuous mixing.

Ground forms of thermoplastic elastomers provide all-around mix capability by allowing dry blending. Dry blending is a process by which thermoplastic elastomers, resins, fillers and liquid plasticizers (usually naphthenic oils) are combined to make a dry, free-flowing powder. Frequently, the final blend can be molded as is; more often, it is used as preparation for further mixing in an extruder, internal, or continuous mixer.

The best product forms for dry blending are low-viscosity liquids, friable resin pellets, and powders smaller than 12 mesh. Elastomers with a greater particle size can be used, in some formulations. However, increased elastomer particle size results in a reduced capacity to absorb oil and longer mix times.

Both high- and low-intensity mixers may be used for dry blending. High-intensity mixers allow shorter mix cycles to be obtained. They add frictional heat to the mix, promoting greater oil absorbancy of the elastomer and often resulting in satisfactory mixing of compounds which would not dry out in lower speed mixers. Higher mix temperatures may require addition of cooling equipment to prevent the material from sticking together after discharge.



D. Mix Cycles

Typical mix cycles for dry blending are 3–10 min for high-intensity mixers and 10–45 min for low-intensity mixers. When determining a proper mix cycle, the following points should also be considered:

1. Mixing temperatures should not exceed 91°C (195°F) when cooling equipment is available, or 60°C (140°F) when it is not.

2. Highly absorptive fillers and oil combine to form a deposit on rotor sides. This can be minimized by late filler addition. In high-intensity mixers, withholding the filler until mix temperature has reached 74°C (165°F) is beneficial. Filler levels, particularly of precipitated silica or calcium silicate substantially greater than needed to dry the mix, generate high dust levels and should be avoided.

3. Oil addition, particularly in ribbon blenders, takes place more rapidly as the number of addition points is increased. Heating the oil results in more rapid incorporation, but is not a necessity.

E. Extruders

Extruder mixing provides continuous, intensive mixing. Typical characteristics of an extruder for this purpose are:

1. 24:1 minimum L/D ratio

2. Two stage screw, 2:1–3:1 compression ratio

3. Venting or vacuum between screw stages, particularly for formulations containing high moisture fillers

Either single- or twin-screw machines can be used.

Dry blended compound is the most common form of feed material for extruder mixing, and the most common methods of preparing pellets from the extrudates are die face cutting or strand cutting. Certain fillers (precipitated silica is one) have small particle size and high bound or absorbed moisture content that can cause porosity, rough strands, and die buildup. These adverse effects can be minimized by operating at the lowest possible melt temperature.

A typical temperature profile for extruder mixing plastic rubbers is illustrated as follows:

Feed Zone Front Zones and die Melt

82–121°C 135–160°C 143–154°C

(180–250°F) (275–320°F) (290–310°F)

When establishing operating conditions, the following should be considered:

1. The melt temperature must be high enough to flux the thermoplastic elastomer as well as any resins in the formulation.



2. If strand cutting is used, the melt temperature should be as low as possible consistent with good mixing.

3. A breaker plate should be used. Screens are often desirable to promote back pressure, prevent surging, and to entrap contaminants. A typical screen pack would be a 60 × 20 mesh.

Screw design will strongly influence the optimum temperature profile, particularly with respect to the work done on the material and the amount of frictional heat generated. The recommended screw design for processing thermoplastic rubber includes low-compression ratios and rather deep flighted metering sections. Screws normally have a length to diameter ratio (L:D) greater than 20:1. If long flow paths are encountered after the melt leaves the shear section of the extruder screw, surface roughness can result due to partial reformation of the physical crosslinks that develop among the polymer endblocks as the melt moves to low shear regions. Die heating bands far removed from the screw tip because of long die adaptors or melt piping can also cause this roughness. High-compression-ratio screws (4:1) can result in degradation due to overheating induced by high shear.

Some general considerations for screw design are as follows:

1. The screw pitch is equal to screw diameter.

2. The compression ratio equals depth of feed section/depth of metering section.

3. If vent is used with two stage screw it should be 16–17 flights from beginning of screw feed. The type and levels of ingredients in a compound, as well as its flow characteristics will also have an effect.

F. Internal Mixers

In internal mixing thermoplastic elastomer compounds, an increase in batch volume and a change in mix cycle are needed over those used with most vulcanizable rubbers. A batch volume of 80% of the machine's net chamber capacity is recommended. If a dry blend is charged, the volume should be increased to 85–90% of the chamber capacity. Only for highly filled compounds (20% elastomer) are the 70–75% batch volumes normally used with vulcanizable rubbers appropriate.

Sufficient shear must be obtained in the mix to flux the thermoplastic elastomer with any thermoplastic resin in the compound. This generally requires that part or all of the oil be withheld until a temperature high enough to produce a homogeneous melt is reached. The remaining oil can then be added, usually in two to four increments to prevent an excessive temperature drop and slow incorporation. To ensure good temperature control, an increment should be added only after a 5–10°C (10–15°F) rise from the minimum temperature resulting from the previous increment. Mix times of 9–20 min are typical. Compounds with less than



20% rubber and greater than 50% filler should be mixed by adding all the ingredients except the rubber at the beginning of the mix, with the rubber being the last ingredient added. When fluxing dry blend, a dump temperature 20° higher than the highest melting point among the resins in the formulation should be used. Roll temperatures of 80–105°C are recommended for take-off mills.

G. Continuous Mixers

High mixing rates can be achieved with continuous mixing equipment. Five thousand to 8,000 pounds per hour are possible through a 9-inch machine. The following considerations should be included in the operating conditions:

1. The machine should always be operated in a slightly starved condition.

2. Because of the short stock residence time, the operating temperature should be maintained at 10–23.9°C (50–75°F) higher than needed to flux the highest melting point resin in the formula. The temperature can be controlled by:

a. Steam and/or water on the jacket

b. Rotor speed

c. Discharge orifice

The discharge stock is generally fed to an extruder for subsequent pellet preparation. Die face cutting or strand cutting are both used in pellet preparation. Stair-step dicers have limited application. Compounds having a Shore A hardness in excess of 65 Shore A are required for efficient cutting with these dicers. Underwater pelletizers will handle a wider range of hardness and are therefore most versatile.

Typical operating conditions for an underwater die face cutter include a water temperature of 48.9°C (120°F) and a cutter speed of 900 revs/min (RPM). Die face cutting without water, which can be done with PVC, may also be possible. Because compounds based on thermoplastic elastomers do not cool as rapidly, changes in the cooling system are often needed to prevent pellet sticking.

A rotary pelletizer for strand cutting is also effective. Machines equipped with rotors having a large number of cutting edges are the most satisfactory, especially with low duromoter compounds. A 24-edge rotor has proven particularly effective. Although eight-blade rotors have been used successfully, the condition of the blades, as well as the proper tolerances between knife edge and bed plate, are more critical for the lesser number of blades.

H. Applications

Direct molded and unit solings constitute the major market for the compounded thermoplastic elastomers. Typical formulations for sneaker and unit solings will incorporate the compounding and processing techniques reviewed. Such formulations can be processed on any of the injection machines commonly used in the



footwear industry. These machine types have lower injection pressures than the more sophisticated reciprocating screw machines. The recipes are designed to provide balanced flow properties and set-up times.

Sneaker solings must be economical and exhibit good mechanical adhesion to the canvas uppers to which they are directly molded. The sneaker solings have the necessary balance of flow characteristics and set-up time. By adjusting alpha methyl styrene resin, polystyrene resin, and oil at the indicated levels, adequate melt flow (above 20 g/10 min) adhesion (above 8 pli), and hardness (50–60 Shore A) are obtained. A typical sneaker soling recipe is shown in Table 19.33.

Unit soles combine heel and sole in one molded piece, which is adhered to the lasted upper. All thermoplastic rubber unit soles must be treated with a chlorine solution or other primer to achieve good adhesive bonds. A smooth surface provides the best bond strength. The soles should not be roughed either before or after chlorination or priming. Chlorination of the soles may be accomplished by using any of several commercial solvent-based systems or with an aqueous system. The most common method uses an aqueous chlorination bath consisting of 96% water, 3% household bleach (5.25% sodium hypochlorite solution), and 1% hydrochloric acid (37% by weight HCl).

Optimum dipping time of the soles in the primer solution is 1–2 min. After the aqueous chlorination, soles should be rinsed in plain water and completely dried. The cement margin should not be handled or otherwise contaminated. Properly prepared chlorinated soles have been stored up to 1 year with no loss of

Table 19.33 Sneaker Soling

TPR Aa 100 —

TPR Bb — 100

Polystyrene 19 26.5

Alpha-methylstyrene resin (mp 290°F) 37 26.5

Whiting 31 57

Precipitated silica 22 —

Naphthenic oil 47 40

Stabilizers 0.5 0.5

Titanium dioxide 1.5 3.5


Specific gravity 1.1 1.1

Melt flow, ASTM condition E 25 40

aTPR A is a naphthenic oil extended (50 parts) radial TPR with 40% styrene (e.g., Solprene or Finaprene 475).

bTPR B is a naphthenic oil extended (60 parts) radial TPR with 48% styrene (e.g., Solprene or Finaprene 481).



bonding potential. Nevertheless, it is good policy to rotate warehouse stock to keep all soles relatively fresh.

The aqueous chlorination bath should be tested periodically for chlorine concentration. Test frequency should be approximately every 2 h and perhaps more often, depending on bath size and throughput of unit soles. To ensure proper unit sole adhesion, the acid level in the bath should be controlled to give a pH in the range of 1.0–3.0.

Heat-activated solvent-based polyurethane adhesives are recommended for attaching thermoplastic rubber unit soles to the shoe upper.

Heat-activation temperatures for the cement should be relatively low (e.g., 135°F) on the surface. This can be obtained by oven heating or by infrared heat lamps. Care must be taken not to overheat the sole to avoid distortion. Pressure during sole attachment should be adjusted carefully. Too much pressure can cause ballooning at the edges and a poor fit. Pressure should be evenly applied over the whole unit.

Bond strength of unit soles to uppers should be regularly checked in order to detect at early stages any deficiencies in the various processes and materials used.

Unit sole construction permits a wide variety of shoe styles. It is easy to produce soles of varying hardness with a wide color range from opaque to very translucent. The recipe shown in Table 19.34 has been designed to provide the feel and appearance of crepe natural rubber.

A typical formulation of an injection molding sole compound is shown in Table 19.35. The sole can be molded directly to the shoe upper or molded seperately and applied to the upper with proper adhesives. A typical extruded slab soling compound formulation is shown in Table 19.36. These compounds have good abrasion resistance and very good resistance to cut growth.

Table 19.34 Unit Soling

TPR Ba 100 phr

Alpha-methylstyrene resin (mp290°F) 18.5

Polystyrene 11.5

Precipitated silica 17

Naphthenic oil 35

Zinc stearate 0.3

Stabilizers 0.5

Pigments —

Shore a hardness 57

Melt Index, ASTM condition E 10

Specific gravity 1

aTPR B is a naphthenic oil extended (60 parts) radial TPR with 48% styrene (e.g., Solprene or Finaprene 481).



Table 19.35 Injection-Molded Soling

TPR Aa 150 phr

Clay 40

Whiting 40

Oil 60

Polystyrene 60

Polyindene resin 20

Shore A hardness 40–60

Tear strength (eb/in) 130

aTPR A is a naphthonic oil extended (50 parts) radial TPR with 40% styrene (e.g., Solprene and Finaprene 475).

I. Summary

Use of styrene-butadiene thermoplastic elastomers in footwear soling has the following characteristics:

Requires no vulcanization.

Scrap produced can be recycled.

Can be compounded in varieties of texture, color, flexibility, and firmness.

Can be made in ordinary plastics processing equipment.

Can be compounded to obtain high gloss of vinyl solings, yet provide better traction.

Can be compounded for the more rigid unit sole applications.

Satisfactory abrasion resistance.

Overall low cost.

Table 19.36 Extruded Slab Soling

TPR Ca 100 phr

Hard clay 40

Rosin 20

Polystyrene 5

Melt flow (180°C/5 kg) 5

Shore A hardness 85

Abrasion resistance Good



Cut growth resistance Very good

aTPR C is a radial TPR with 40% styrene (e.g., Solprene or Filprene 414).



References

1. H. L. Hsieh and R. H. Burr, Plastics Modifications by Blending with Diene-Styrene Block Polymers, presented at ANTEC Meeting, Soc. of Plastics Engrs., Boston, MA, May 1981. Modern Plastics, 59, No. 4, 84 (1982).

2. Shell Chemical Company, Technical Bulletin SC 165–88, Shell Kraton Polymers For Modification of Thermoplastics.

3. A. L. Bull and G. Holden, J. Elastomers and Plastics, 9, 281 (1977).

4. Phillips Chemical Company, Technical Bulletin S-P-4.

5. R. H. Burr, Modification of ABS, Phillips Petroleum Company Bulletin 825-P.

6. A. R. Schultz and B. M. Beach, J. Appl. Polym. Sci., 21, 2305 (1977).

7. G. D. Cooper and A. Katchman, US patent 4,139,574.

8. D. W. Gilmore and M. J. Modic, Plastics Engineering, 45, No. 4 (1989).

9. A. South, Jr., US patent 4,020,036 and paper presented at Soc. of Plastics Engrs. Meeting, Detroit, MI, Nov. 6–8, 1979.

10. Phillips Chemical Company, Technical Bulletin on Solprene 312 Rubber—A Low Profile Additive For Molded Polyester Applications.

11. Shell Chemical Company, Technical Bulletin SC 789–88.

12. G. Kraus and D. S. Hall, in Block Polymers, Science and Technology, D. J. Meier, Ed., MMI Symposium Series, Vol. 3, 1983, p. 167.

13. G. Kraus, Proc. 2nd International Symposium on Roofs and Roofing, Soc. of Chem. Ind., London, September 1981, p. 19.

14. G. Kraus and K. W. Rollmann, paper presented at International Rubber Conference, Nunberg, Germany, September, 1980.

15. M. G. Bouldin, J. H. Collins, and A. Berker, Rubber Chem. & Tech., 64, No. 4, 577 (1991).

16. Shell Chemical Company, Technical Bulletin SC 974–88, Kraton Polymers: Advancing the Performance of Hot Mix Asphalt Concrete Pavement.

17. E. Diana, L. Gargani and L. Vitalini Sacconi, Rubber World, 206, No. 2, 44 (1992).

18. Phillips Chemical Company, Technical Bulletins 306 and 309.

19. D. R. Paul and S. Newman, Eds., Polymer Blends, Vol. 1 and 2, Academic Press, New York, 1978.

20. O. Olabisi, L. Robeson, and M. T. Shaw, Eds., Polymer—Polymer Miscibility, Academic Press, New York, 1979.

21. D. R. Paul and J. W. Barlow, J. Macromol Sci. Rev. Macromol. Chem., C18, 109 (1980).

22. J. W. Barlow and D. R. Paul, Polym. Eng. Sci., 24, 525 (1984).



32. G. Riess, J. Periard, and A. Banderet, in Colloidal and Morphological Behavior of Block and Graft Copolymers, G. E. Molau, Ed., Plenum, New York, 1972, p. 173.

33. P. Gailiard, M. Ossenbach-Santer, and G. Riess, in Polymer Compatibility and Incompatibility—Principles and Practices, K. Solc, Ed., MMI Press Symp. Series, Vol. 2, Harwood Academic, New York, 1982, p. 289.

34. G. Hurtrez, D. J. Wilson, and G. Riess, in Polymer Blends and Mixtures, D. J. Walsh, J. S. Higgins, and A. Maconnachie, Eds., NATO ASI Series E, Vol. 89, Martinus Nijhoff, Dordrecht, 1985, p. 149.

35. T. Inoue, T. Soen, T. Hashimoto, and H. Kawai, Macromolecules, 3, 87 (1970).

36. M. Moritani, T. Inoue, M. Motegi, and H. Kawai, Macromolecules, 3, 433 (1970).

37. T. Inoue, T. Soen, T. Hashimoto, and H. Kawai, in Block Copolymers, S. L. Aggarwal, Ed., Plenum, New York, 1970, p. 53.

38. B. Ptaszynski, J. Terrisse, and A. Skoulios, Makromol. Chem., 176, 3483 (1975).

39. D. R. Paul, in Thermoplastic Elastomers—A Comprehensive Review, N. R. Legge, G. Holden, and H. E. Schroeder, Eds., Hanser Publishers, New York, 1987, Ch. 12.

40. R. Fayt, R. Jerome, and Ph. Teyssie, J. Polym. Sci. Polym. Lett. Ed., 19, 79 (1981).

41. R. Fayta. R. Jerome, and Ph. Teyssie, J. Polym. Sci. Polym. Phys. Ed., 19, 1269 (1981).

42. R. Fayt. R. Jerome, and Ph. Teyssie, J. Polym. Sci. Polym. Phys. Ed., 20, 2209 (1982).

43. R. Fayt, P. Hadjiandreou, and Ph. Teyssie, J. Polym. Sci. Polym. Chem. Ed., 23, 337 (1985).

44. D. Heikens, N. Hoen, W. Barentsen, P. Piet, and H. Ladan, J. Polym. Sci. Polym. Symp., 62 309 (1978).

45. W. J. Coumans, D. Heikens, and S. D. Sjoerdsma, Polymer, 21, 103 (1980).

46. W. J. Coumans and D. Heikens, Polymer, 21, 957 (1980).

47. S. D. Sjoerdsma, J. Dalmolen, A.C.A.M. Bleijenberg, and D. Heikens, Polymer, 21, 1469 (1980).

48. S. D. Sjoerdsma, A.C.A.M. Bleijenberg, and D. Heikens, Polymer, 22, 619 (1981).

49. C. R. Lindsey, D. R. Paul, and J. W. Barlow, J. Appl. Polym. Sci., 26, 1 (1981).

50. T. D. Traugott, J. W. Barlow, and D. R. Paul, J. Appl. Polym. Sci., 28, 2947 (1983).

51. Phillips Chemical Company, Bulletins S-A-4, S-A-8, S-A-9, S-A-12 and S-A-13.

52. O. L. Marrs, R. P. Zelinski, and R. C. Doss, J. Elastomers Plastics, 6, 246 (1974).

53. G. Kraus and J. T. Gruver, J. Polym. Sci., Part A, 3, 105 (1965).

54. G. Kraus, F. E. Naylor, and K. W. Rollmann, J. Polym. Sci., Part A-2, 9, 1839 (1971).

55. G. Kraus, F. B. Jones, O. L. Marrs, and K. W. Rollmann, J. Adhesion, 8, 235 (1977).



63. J. R. Erickson, Rubber Plast. News, p. 16 (Sept. 9, 1985).

64. B. D. Simpson and P. R. Fowler, Adhesion Age, 17(9), 32 (1974).

65. D. M. Mitchell and R. Sabia, Proceeding of the 29th International Wire and Cable Symposium, Nov. 1980, p. 15.

66. G. Holden and S. S. Chin, paper presented at the Adhesive and Sealants Conference, Washington, DC, March 1986.

67. G. Holden, paper presented at the Adhesives and Sealants Seminar, Chicago, IL, Oct. 1982.

68. H. E. Railsback, C. C. Baird, J. R. Haws, and R. C. Wheat, Rubber Age, 94, 583 (1964).

69. Phillips Petroleum Company, Technical Bulletin 513, Solprene Rubbers in Vulcanized Footwear Applications.

70. R. H. Burr, Compounding and Processing Solprene Radial Block Polymers, Phillips Petroleum Company Bulletin 804-P-TR.

71. Phillips Petroleum Company, Technical Bulletin 236, Solprene Elastomers in Integral Molded Soling; Technical Bulletin 301, Solprene Radial Teleblock Copolymers in Injection Molded Soling; Technical Bulletin 505-A, Solprene 475 and 481 Plastomers— Versatility and Economy in Footwear Soling; Technical Bulletin 307, Injection Molding Solprene Plastomers.



20 Clear Impact-Resistant Polystyrene

I. Introduction

In 1972 Phillips Petroleum Company started up a new plant to produce a new family of styrene plastics known as K-Resin BDS polymer. The resins are clear and tough and their moderate pricing places them between the low-cost resins such as polystyrene, polyethylene, and polypropylene, which are either clear or tough, but not both, and high-priced resins such as cellulosics, clear ABS, and polycarbonate, which are both clear and tough. The exceptional clarity and impact resistance make them desirable replacements for conventional polystyrene and other clear resins in many applications. Applications thus far realized are in toys, housewares storage units, lids, medical packaging and devices, films, and a wide variety of packaging uses including blister packs, injection-molded tubs, bottles, and boxes with integral hinges [1].

These resins are copolymers of styrene and butadiene prepared in hydrocarbon solution by “living” anionic polymerization process. K-Resin polymers are available in two different types. KR01 is used almost exclusively for injection molding applications. Although it is not as tough as KR03, KR01 exhibits significantly higher impact than crystal polystyrene. It provides the advantages of warpage resistance, stiffness, and surface hardness compared to KR03. In the KR03, KR04, KR05, and KR10 series of resins, all the polymers are chemically



equivalent. The different grades reflect the decreasing gel/fisheye content of the resin.

KR03 is used in injection molding, where gels are not visible in the finished parts.

KR04 contains a lower level of gel and is used most in the blended sheet extrusion market. Blending with other clear resins will dilute the gel in the sheet.

KR05 is used for blow molding, neat sheet extrusion, profile extrusion, and injection blow molding.

KR10 is selected for film extrusion and is segregated as the lowest gel level.

The KR03, KR04, KR05, and KR10 contain a microcrystalline wax that acts as an antiblock. While the wax provides processing benefits, it does make K-Resin polymers difficult to decorate. Some grades are available without wax. The nominal physical properties of K-resin polymers are shown in Table 20.1.

Table 20.1 Nominal Physical Properties of K-Resin Polymers

Property Test method KR01 KR03

Density (g/ml) D1505 1.01 1.01

Floor rate (g/10 min) D1238, G 8.0 8.0

Tensile strength (2 inch/min,psi [MPa])

D638 4400 (30.3) 3700 (25.5)

Elongation (2 inch/min, %) D638 20 160

Flexural modulus (psi [MPa]) D790 215,600 (1482) 205,000 (1413)

Flexural yield strength (psi [MPa]) D790 6400 (44.1) 4900 (33.8)

Heat deflection temp. (°F 264 psi) D648 170 163

Izod impact strength ( '' specimenfeet-pound/inch notched)

D252 0.4 0.75

Shore D hardness D2740 75 65

Vicat softening point (°F) D1525 200 188

Light transmission (%) D1746 90 90

Dielectric constant D150 2.5 2.5

Power factor D150 0.0004 0.0010

Dielectric strength (V/) D149 300 300

Haze, 100 mil (%) D1746 1–3 1–3

Moisture absorption, 24 h (%) D520 0.08 0.09

Expansion coefficient



II. Appearance

Clarity and impact strength are the two properties that characterize these resins. The clarity, as measured by haze, is excellent. These resins are colorless, while some competitive resins, such as cellulosic and SAN, frequently have a bluish or yellowish cast. Impact polystyrenes, on the other hand, are opaque. Table 20.2 summarizes data on appearance.

The visual perception of clarity is dependent upon two factors: the intensity and distortion of light as it passes through a part. Since K-Resin polymers are amorphous, they disrupt light less than many dense, crystalline polymers and thus exhibit very low haze values (1–3%) and excellent light transmission (89–91%). K-Resin DBS polymers, which will yellow when subjected to long-term direct exposure to sunlight can also yellow in less severe ultraviolet (UV) exposure. The yellowing can be significant stabilizers such as Tinuvin P, a substituted benzotriazole produced by Ciba-Gigy Company. K-Resin parts stored in dark storage showed no significant yellowing over at least a two-year period.

III. Impact Strength

The polymers are notch-sensitive and, therefore, the conventional notch Izod test cannot be used to determine their ultimate impact strength [3]. In fact, by that test the impact strength is only comparable to general purpose polystyrene, but when measured by unnotched Izod or the falling dart test, the resins are substantially above general purpose polystyrene [1].

In order to obtain the ultimate impact strengths of these resins, they require orientation [1,3]. Biaxial orientation is beneficial for part toughness. Monoaxial orientation from extrusion, however, is detrimental to part strength. As the draw-

Table 20.2 Comparison of Appearances of Various Resins

Resin Haze, % (ASTM D1746) Color

K-Resin

KR01 1–5 Water white

KR03 1–5 Water white

Polystyrene

General-purpose 2–3 Water white

Medium-impact Opaque White

High-impact Opaque White

SAN 1.5 Yellowish

Cellulose acetate 7.5 Bluish

Source: Ref. 1.



Table 20.3 Comparison of the Mechanical Properties of Various Resins

Density

(ASTM D792–66)

Tensile (ASTM D638–68)

Elongation

(ASTM D638–68)

Flexural modulus (ASTM D790–66) (

Resin (g/cc) psi MPa (%) psi MPa p

K-resin

KR01 1.01 4400 (3.03) 20 215,000 (1482) 6

KR03 1.01 3700 (25.5) 160 205,000 (1413) 4

Polystyrene

General-purpose 1.05 7600 (52.4) 4 480,000 (3310) 1

Medium-impact 1.04 4400 (30.3) 32 390,000 (2690) 5

High-impact 1.04 2600 (17.9) 43 220,000 (1517) 6

Cellulose acetate 1.20 7500 (49.0) 47 241,000 (1662) 7

SAN 1.08 6900 (47.6) 6 480,000 (3310) 15,



down ratio from the die increased, sheet and part strength plummeted. Orientation of the sheet in the direction of extrusion causes it to be increasingly brittle across its width. This tendency for K-Resin sheet to be tougher in the direction of extruion suggests that thermoformed hinges should always be aligned in that direction. Drawing K-Resin sheet from a large die opening will diminish impact strength in the sheet and thermoformed parts. Biaxial orientation during the thermoforming operation, however, promotes sheet and part toughness.

IV. Mechanical Properties

The most notable differences from polystyrene in mechanical properties are the lower tensile, flexural modulus, and hardness values and the higher elongation and flexural strength for the copolymers. All of these differences are due to the presence of butadiene in the resin (Table 20.3). In fact, the mechanical properties are similar to those for the conventional impact polystyrenes that also contain butadiene, but are opaque.

The density of the BDS resin, which is about 1.01 g/cc, coupled with its moderate price, yields a favorable cost-volume ratio in comparison to most of the other clear plastics.

V. Hinge Flex Life

An unusual and very interesting property of these BDS resins is the flexural hinge life (Table 20.4). The KR01 resin is superior to the KR03 resin in hinge flex life, but both contrast strongly with polystyrene, which cannot be flexed. Although the hinge life of these resins does not equal that of polypropylene, it is more than sufficient to make practical certain types of packaging units with integral hinges that receive limited opening and closing[1].

The hinge life is dependent upon the thickness of the hinge. Electron micrographs of KR01-type indicate the morphology to be a random array of polybutadiene domains in polystyrene matrix in the unflexed state and after flexing, the polybutadiene domains show some orientation with the appearance of fine light lines between the arrays of polybutadiene. These are thought to be microcrazes, which would account for the good lifetime in flexure of the resins[1,4].

VI. Chemical Resistance And Stress Crack Resistance

The chemical resistance of these BDS resins is similar to the polystyrene (Table 20.5). The resins are soluble in or highly swollen by most organic solvents; however, the resins are not affected by methanol, ethanol, or aqueous solutions containing these alcohols.

Containers made of BDS resin can stress crack, especially in polystyrene



Table 20.4 Hinge Flex Lifea of K-Resin BDS Polymer

Hinge thickness

Resin mm In Number of flexes

GP styrene 0.13 0.005 1

0.26 0.010 1

KR01 0.13 0.005 235

0.26 0.010 564

0.39 0.015 605

0.52 0.020 535

0.69 0.025 307

KR03 0.13 0.005 145

0.26 0.010 200

0.39 0.015 188

0.52 0.020 110

0.69 0.025 59

a Flexed 180 degrees with 1.5 kg load on a Tinius Olsen Folding Endurance Tester.

Source: Ref. 4.

blends, when the container is subjected to a stress crack medium such as oil or fats. In general, dry foods, meat products, and gelatin products may be packaged with no problems, but oily products such as butter and margarine may not. Unsaturated oils cause more rapid failure than saturated oils. Other factors that contribute to the speed at which failure may occur include molded-in stresses (thermoformed containers crack much more quickly than injection molded), part design, part loading, and storage conditions.

More extensive testing results with a broad spectrum of household products, food, and common chemicals are available[5].

VII. Polymer Blends

K-Resin polymers can be blended with many other polymers. The most widespread blend applications is with crystal polystyrene in sheet extrusion and thermoforming applications. Blending of K-Resin with polystyrene will allow extrusion of film, thin sheet and injection molding of items with a high degree of clarity (Fig. 20.1). A marginal degree of toughness and durability is gained over polystyrene alone. Generally, for extruded film the impact properties such as dart drop, elongation and Spencer Impact drop off rapidly as the polystyrene content



Table 20.5 Chemical Resistance of K-Resin BDS Polymer

Chemical Dissolve Soften Permeatesa

Organic solvents

Hydrocarbons X

Ketones X

Esters X

Ethers X

Light alcohols X X

Heavy alcohols X

Glycols X X

Oils

Mineral X X

Vegetable X X

Water-based

Water X

Dairy products X

Syrups X

Detergents X

Salt solutions X

aPermeation of 3% per year or more.

Source: Ref. 4.



Figure 20.1 Clarity of K-Resin blends. (From Ref. 6.)



Figure 20.2 Effect of polystyrene ratio in

K-Resin polymer blends on dart drop impact strength. (From Ref. 6.)

increases (Fig. 20.2). On the other hand, strength properties (i.e., tensile strength) actually increase with increasing polystyrene. The resistance to tear propagation shows little change until the polystyrene content is quite high. Other materials have successfully blended with K-Resin polymers in injection molding; K-Resin polymers impart gloss to high-impact polystyrene (HIPS) [7], economics to acrylnitrile-butadiene-styrene resin (ABS), pearlescence with polypropylene (PP), and improved impact to styrene-acryinitrile resin (SAN); K-Resin/polycarbonate blends possess excellent impact strength. Compared to polycarbonate, the blends also provide good processibility, gloss, stiffness, and hardness at reduced cost [8].

VIII. Processibility

One of the most attractive facts of K-Resin polymers is the ease of processing [1,9]. They can be formed on a wide variety of conventional equipment with a relatively broad range or “window” of process parameters. Their rheological properties are shown in Figure 20.3. The viscosities of the KR01 and KR03-type resins bracket those of general-purpose polystyrene and cellulose propionate. In addition, the shear responses of all four resins are quite similar. Therefore, processing techniques suitable for these other resins are generally suitable for the



Figure 20.3 Rheological properties measured

at 200°C (CIL data according to ASTM D1703–62). (From Ref. 1.)

BDS resins. In general, their processing characteristics most closely resemble medium-impact polystyrene. The BDS resins exhibit very low die swell, low warpage, and shrink [1,2].

These resins are stable during processing and do not present any major difficulties. Since they contain both polybutadiene and polystyrene, degradation can occur by two mechanisms: visbreaking of polystyrene and crosslinking of polybutadiene. Since both processes are promoted by heat and shear, the shear-heat should be minimized. Experience has shown that the crosslinking of the polybutadiene segments predominates as the major form of degradation.

The thermal stability of the polymer is measured by a heat soak in a melt index that compares the melt flow before and after a 30 min exposure at 250°C. This will predict the tendency of the polymer to degrade. The results of this test indicate that as the polymer proceeds through several processing steps, it tends to diminish its resistance to degradation (Table 20.6).

Spiral flow analysis of the KR01 and KR03 correlated well with the standard melt flow analysis and showed both resins to have relatively high levels of regrind stability (Table 20.7).

The flow and optical properties were virtually unaffected by the reprocessing, as highlighted in Figures 20.4–20.6.

The toughness of these resins makes some processing techniques, such as intricate injection molds and deep draws in thermoforming, possible that cannot



Table 20.6 Thermal Stability Test by Heat Soaking

Regrind pass no.

Polymer type Virgin 1 3 5 7

KR01 75–8–0780%

MF change (30 min @ 250°C) -16.4 -68.8 -96.0 — —

MW/Mn (thous) 150/97 156/98 149/91 168/96 180/95

H.I. 1.55 1.60 1.64 1.76 1.89

KR03

75–8–0821 % MF change (30 min @ 250°C)

+3.1 +13.2 -37.8 -59.1 -82.8

MW/Mn (thous) 175/109 178/107 180/106 193/103 187/105

H.I. 1.60 1.66 1.71 1.87 1.78

be used with some competitive resins such as SAN and general-purpose polystyrene.

IX. Injection Molding

BDS-type polymers may be molded on either plunger or reciprocating-screw type machines [10,11]. Screw type machines are preferred for consistent melt temperature and homogeneity. Plunger type machines are generally reserved for specific effects such as tortoise shell patterns requiring nonhomogeneous melt. As melt temperature increases up to 260°C, the flow of the material increases (Fig. 20.7). Above 260°C, material flow becomes erratic. For optimum injection molding, the

Table 20.7 Spiral Flow Analysis for Regrind Stability Test

Spiral flow

Flow length (in)

Specimen Wt (g)

Regrind pass no. KR01 KR03 KR01 KR03

Virgin 25.73 27.15 8.58 9.10

1 25.76 — 8.59 —

3 26.13 28.00 8.72 9.41

5 26.23 — 8.77 —

7 26.46 29.17 8.87 9.81



Figure 20.4 KR01 flow properties after multiple regrind passes. (From

Ref. 6.)

melt temperature should be the minimum that will permit filling the mold, which is usually between 193 and 233°C.

Mold temperatures ranging from 10 to 60°C may be used when molding K-Resin polymers, although optimum clarity for most parts occurs at mold temperatures between 27 and 49°C. Whereas surface gloss and reproduction of mold detail are typically maximized at the higher mold temperatures, higher impact resistance and reduced cycle time are achieved at the lower temperature. K-Resin response to injection pressure is much like other plastics. The flow length of the resin increases as injection pressure increases (Fig. 20.8). Typically the minimum injection pressure and injection rate that will fill out the part should be

Figure 20.5 KR03 flow properties after multiple regrind passes. (From

Ref. 6.)



Figure 20.6 KR01 and KR03 optical properties after multiple regrind

passes. (From Ref. 6.)

Figure 20.7 Trend of spiral flow as function of melt temperature.



Figure 20.8 Trend of spiral flow as function of injection pressure.

used. Since BDS polymers are amorphous, their shrinkage rate is relatively low. The cooling system should be designed to balance shrinkage throughout the part and thus minimize warpage.

X. Blow Molding

Injection blow molding or extrusion blow molding equipment may be used [12]. Generally, blow molding equipment used for high-density polyethylene (HDPE) may be used with the resins with minor modifications to the dies and molds. Actually, polyvinyl chloride (PVC) molds with low shrinkage allowances are more suitable than those of HDPE, because the shrinkage characteristics of these resins are similar to those for PVC. Optimum clarity and gloss are typically obtained at a melt temperature of 190 to 195°C, a low blow pressure in the range of 1.4–4.2 kg/cm2, and a mold temperature of about 240°C. For best wall thickness uniformity, a maximum blow-up ratio of 3:1 is recommended.



XI. Blown Film

Clear, glossy BDS polymer film is well suited for vegetable wrap, skin packaging, and shrink wrap applications requiring good rigidity and impact strength. In most ways, blown film extrusion of BDS polymers is very similar to that of HDPE [13]. Within limits, the impact strength of the film increases as both blow-up and drawn down ratios increase, so tooling must have a generous die gap, preferably 35–50 mils.

K-Resin film has excellent optical clarity, 25 µm film having a light transmission of about 95% and a haze level of only about 2%. In addition to clarity, it also has good impact strength, 25 µm film exhibiting a dart impact of about 400 g and an elongation of about 150%.

XII. Sheet Extrusion

Sheet extrusion can be accomplished using single stage and two-stage screws on equipment normally used for HDPE, PS and the cellulosics [14]. The important consideration is that the screw must be proportioned to minimize the shear heat input to the melt. The melt temperature should be limited to about 220°C. BDS polymer sheet can be successfully produced using either a standard or flex lip die, although the latter has some advantages, particularly in gauge control for thinner sheet. The polish rolls are typically set at about 82°C for optimum sheet gloss.

XIII. Thermoforming

Sheet produced from BDS polymer or polymer blends is thermoformed into a variety of thick or thin wall and deep or shallow draw parts [14]. Although sheet can actually be formed over a wide range of temperatures, control of the sheet temperature in the approximate range of 120–150°C is typically required to obtain a combination of optimum clarity and appearance, and uniform wall distribution. Although colder temperatures will generally give better clarity, the chosen thermoforming temperature is typically determined by a compromise between clarity and the required detail of the formed part. Forming techniques applicable to other thermoplastic sheet can be used including simple drape, male and female (with or without plug assist), all in combination with vacuum and/or pressure.

Molds should be temperature controlled, with a temperature of approximately 38°C recommended. Polished mold surfaces are a must for obtaining good clarity.

XIV. Medical Applications

Mathis [16] reported that BDS polymers are a clear choice for medical packaging and devices [15]. In addition to exceptional clarity and shatter resistance, K-Resin



polymers meet the food requirements of the Food and Drug Administration and the United States Department of Agriculture; meet the requirements of U.S. Pharmacopoeia Class VI-50; are compatible with blood; demonstrate no cytotoxic, mutagenic, or irritant potential; and are sterilizable by both gamma irradiation and ethylene oxide.

XV. Block Architecture

The clear impact-resistant styrenic resins (BDS) are basically a mixture of block copolymers of styrene and butadiene formed by incremental additions of alkyllithium initiator and monomers followed by some form of chain coupling. The original versions were developed by Kitchen and Szalla [17]. Here is a fine example of how a useful product is tailor-made utilizing the living nature of the anionic polymerization. The macrostructure of these resins is quite complex. The molecular weight distribution of this type of resins is very broad and some show bimodal or multimodal gel permeation chromatographic (GPC) curves. As illustrations, several of the key examples based on the information shown in patents are listed in Table 20.8. The GPC curves at each individual synthesis steps of the first two examples in Table 20.8 are shown in Figures 20.9 and 20.10.

Figure 20.9a shows the GPC curve of the polymer sample obtained following the polymerization of styrene (S-). Figure 20.9b shows the GPC curve of the polymer sample obtained after the butadiene has been polymerized onto the polystyrene segment (SB-). Figure 20.9c shows the GPC curve of the final coupled product. Figure 20.10a shows the GPC curve of the polymer sample obtained following the polymerization of the first styrene addition (S1-). Figure 20.10b shows the GPC curve of the polymer sample obtained after the polymerization of the second portion of the styrene, in which additional butyllithium was employed (S1S2- and S2-). Figure 20.10c shows the GPC curve of the polymer sample obtained after the butadiene has been polymerized onto the polystyrene segments (S1S2B1- and S2B1-). Figure 20.10d shows the GPC curve for the final product after coupling.

One should be aware that the sequences of the initiator and monomer additions are only part of controlling factors. Relative amounts of the initiator in each addition and relative amounts of the two monomers in each addition are also important. Generally, the total styrene contents are in the range of 70–85%. Degree of branching is yet another factor that needs to be specifically controlled.

In addition to the products produced directly from polymerization, several patents [27] disclosed the production of improved clear resins by blending two block copolymers.

In the manufacturing process, steps must be taken to minimize haze, color, odor, and fisheye (microgel). In one patent [28], it is claimed that clear, haze-free, colorless, impact-resistant resinous copolymers are produced by terminating the coupled polymers with water and linear saturated aliphatic dicarboxylic acids. In



Table 20.8 Examples of Clear Impact-Resistant Styrenic Resins

Examplea

Order of additionb

Linear polymer before couplingc

References

A L1;S1;B1 (S-B-)- Typical radial or star-shaped thermoplastic elastomer (see Chap. 18)

B L1;S1;L2;S2;B1 (S1-S2B1-)- and (S2B1-)- 17, 18

B L1;S1;B1;L2;S2;B2;L3;S3;B3 (S1-B1-S2-B2-S3B3-)- and (S2-B2-S3-B3-)- and (S3-B3)-

19

B L1;S1;B1;S2;B2;L2;S3;B3 (S1-B1-S2-B2-S3-B3-)- and (S3-B3)- 19

B L1;S1;L2;S2;B1;L3;S3;B2 (S1-S2B1-S3B2)- and (S2-B1-S3-B2-)- and (S3-B2

19

C L1;S1;L2;(S2 + B1 (S1-B1 S2-)- and (B1 S2-)- 20–23

C L1;S1;L2;S2;(S3 + B1) (S1-S2-B1 S3-)- and (S2-B1 S3- 20–23

D L1;S1;L2;S2;L3;S3;L4 (S4 + B1);B2

(S1-S2-S3-S4/B1-B2-)- and S2-S3-S4/B1-B2-)- and (S3-S4/B1-B2-)- and (S4/B1 - B2-)-

24–26

D L1;S1;L2;S2;L3;(S3 + B1);B2 (S1-S2-S3/B1-B2-)- and (S2-S3/B1-B2-)- and (S3/B1-B2-)-

24–26

aB, patents assigned to Phillips; C, patents assigned to ARCO; D, patents assigned to BASF.

bL, alkyllithium initiator; S, styrene; B, butadiene; L1, first time the initiator is used; S2, second time styrene is used; B1, first time butadiene is used, etc.

cB S, tapered block copolymer formed from monomer mixture; S/B, random copolymer formed from continuous monomer addition.



Figure 20.10 GPC curves of resinous branched block copolymers (KR03

type). (From Ref. 17.)

another patent [29] claim is made that by using the combination of a lactone and a halo-silane as coupling agents, it reduces haze, yet maintains clarity and impact resistance of the resulting coupled polymers.

The suppliers of clear impact-resistant styrenic resins in 1991 are listed in Table 20.9.

XVI. Conclusion

In conclusion, BDS polymers produced by anionic polymerization such as Phillips K-Resin polymers are versatile thermoplastic resins having a wide range of applications. They can be processed using existing processing techniques and



Table 20.9 Worldwide Suppliers of Clear Impact-Resistant Styrenic Resinsa

Region Supplier Capacity (MM lbs)

United States Phillips K-Resin 270

Firestone Stereon 20

Europe BASF Styrolux 81

Fina Finaclear 17

Japan Denka-Kaguku Clearance 12

Asahi Chemicals Asaflex 10

a1991 data.

equipment and produce parts having a unique combination of properties, such as toughness, clarity, surface gloss, and hinge effect.

References

1. L. M. Fodor, A. G. Kitchen, and C. C. Baird, K-Resin BDS Polymer: A New Clear Impact-Resistant Polystyrene, in New Industrial Polymers, ACS Symposium Series 4, R. D. Deanin, Ed., Am. Chem. Soc., Washington, DC, 1972.

2. K-Resin Properties and Processing, TIB200, Phillips 66 Plastics Technical Center, Bartlesville, OK.

3. Tester For Measuring Impact of KR03 and Related Polymers, PTC-313, Phillips 66 Plastics Technical Center, Bartlesville, OK.

4. Plastics Focus, 5, (31), (1973).

5. Chemical Resistance of K-Resin Polymers, PTC-352, Phillips 66 Plastics Technical Center, Bartlesville, OK.

6. K-Resin Polymer Blends with General Purpose Polystyrene, PTC-411, Phillips 66 Plastics Technical Center, Bartlesville, OK.

7. K-Resin/Crystal Polystyrene Sheet Property Modification with High Impact Polystyrene, PTC-409, Phillips 66 Plastics Technical Center, Bartlesville, OK.

8. K-Resin/Polycarbonate Injection Molded Blends, PTC-379, Phillips 66 Plastics Technical Center, Bartlesville, OK.

9. K-Resin Processing Study, PTC-408, Phillips 66 Plastics Technical Center, Bartlesville, OK.

10. Plastics Technology, 57, April (1973).

11. K-Resin: Injection Molding, TIB-202, Phillips 66 Plastics Technical Center, Bartlesville, OK.

12. K-Resin: Blow Molding, TIB-203, Phillips 66 Plastics Technical Center, Bartlesville, OK.



15. K-Resin: Medical Applications of K-Resin Polymers, Technical Service Memorandum 292, Phillips 66 Plastics Technical Center, Bartlesville, OK.

16. R. D. Mathis, K-Resin: A Clear Choice for Medical Packaging and Devices, Presented at the SPE First Annual Medical Plastics Conference, October 18–20, 1983, New Brunswick, NJ and SPE/SPI West Coast Medical Plastics Conference, March 21–22, 1984, Anaheim, CA.

17. A. G. Kitchen and F. J. Szalla, US patent 3,639,517.

18. A. G. Kitchen, US patent 4,091,053.

19. A. G. Kitchen, US patent 4,584,346.

20. L. K. Bi and R. Milkovich, US patents 4,180,530; 4,221,884; 4,248,980; and 4,248,982.

21. R. Milkovich, K. Doak, and L. K. Bi, US patent 4,248,981.

22. L. K. Bi, R. Milkovich, and K. Doak, US patents 4,248,983; 4,248,984.

23. K. Doak, L. K. Bi, and R. Milkovich, US patent 4,346,198.

24. G. Fahrback, K. Gerberding, E. Seller, and D. Stein, US patents 4,086,298 and 4,167,545.

25. K. Gerberding, US patent 4,335,221.

26. G. Heiz, et al., US patent 4,418,180.

27. L. M. Fodor, US patents 4,051,197; 4,080,407; and 4,051,197.

28. G. A. Moczygemba, US patent 4,403,074.

29. G. A. Moczygemba and Kishore Udipi, US patent 4,405,754.



21 Nonfunctional Liquid Polybutadienes

I. Introduction

Liquid polybutadienes (liquid BR) are not a new class of materials. The earliest commercial production of liquid BR is believed to have started in the mid-1920s in Germany as Plastikator 32. Between about 1950 and the mid-1960s, many companies developed production techniques and investigated the application of liquid BRs. However, with the notable exceptions of DuPont's Budium and the Richardson Co.'s Ricon range (formerly known as Enjay Buton and later produced by Colorado Chemicals Specialties Inc.), they did not become fully commercial [1]. Phillips in the early 1950s produced Butarez liquid BR in semi-commercial quantity using sodium metal as initiator [2].

Liquid BRs can be prepared by a variety of polymerization mechanisms involving anionic, cationic, coordination, or free radical initiator systems and also by depolymerization of a high-molecular-weight (MW) polymer. The preparation, modification, and applications of nonfunctional liquid polybutadiene were reviewed by Furukawa and Yamashita [3] and Luxton [1].



II. Preparation Of Nonfunctional Liquid Polybutadienes By Anionic Polymerization

A. Overview

Organosodium and organolithium compounds or the metals themselves are the initiators most frequently used. Just as in the case of preparing high-MW polybutadienes with these anionic initiators, a wide range of microstructure varying from 10% vinyl to 90% vinyl configurations can be produced. The metal, solvent, the polymerization temperature and the use and the nature of polar promoter are the main factors influencing the amounts of 1,4-addition vs. 1,2-addition. These factors are not greatly different from the preparations of high-molecular-weight polymers.

For example, liquid BRs made with alkyllithium initiators in hydrocarbon solvent have 10–20 vinyl configuration. In general, all other conditions are the same. The low-molecular-weight polymer would have slightly higher vinyl content than the corresponding high-molecular-weight product due to the perhaps 100-fold, higher level of alkyllithium used in the former case. It is very common in the production of liquid BRs with alkyllithium initiator that a Lewis base additive (“modifier,” “promoter”) such as an ether or tertiary amine is used to enhance the transmetallation reaction and thus chain-transfer reaction. This leads to medium-vinyl polybutadienes. By combining the Lewis base additive and low polymerization temperature, even higher vinyl contents can be obtained [4].

Sodium-based initiators in hydrocarbon solvent produce polybutadienes with vinyl configurations in the range of 30–70%. Combining sodium metal, a promoter such as tetrahydrofuran (THF), and at a low polymerization temperature, very high vinyl (-90%) liquid polybutadienes can be produced [5,6]. For both lithium-based and sodium-based systems in the presence of a Lewis base, low temperature favors the 1,2-addition to give vinyl configuration [7,8].

There is one major difference between the production of high- and low-molecular-weight BRs by the anionic mechanism. For high-molecular-weight polymers, the cost attributable to the initiator relative to the total raw materials is usually very small. In a “living polymer” type of mechanism, it is a termination-free and transfer-free system. It takes 1 mole of the initiator such as alkyllithium to produce 1 mole of the polymer. Using an electron-transfer type of initiator, it takes 2 moles of the initiator to produce 1 mole of polymer. The use of alkyllithium or other lithium-based initiators directly to produce low-molecular-weight polymers is therefore not cost-effective. The relative cost of the initiator is prohibitively high in a commercial process for the production of liquid polymers. Furthermore, lithium metal and organolithium compounds are substantially more expensive than the sodium metal and corresponding organosodium compounds.

The drawback of high initiator costs can be overcome by the use of a telomerization process. The basis of telomerization is the introduction in the



polymerization recipe of a chain transfer agent or “telogen,” which possesses an acidic hydrogen atom to react with a polymeric anion. The transmetallation step involves the abstraction of a proton from the telogen. If the new anion can initiate polymerization, the mechanism provides for a recycle of the initiator. One single initiator originally introduced can recycle itself indefinitely by theory and many times in practice. Thus, low-molecular-weight polymers can be made with relatively small amounts of initiator. Toluene is the most commonly used telogen in commercial operations. It can serve both the roles of telogen and solvent or partial solvent. Sodium and potassium compounds transmetallate the acidic hydrogen with ease, particularly at elevated temperatures. With lithium-based initiators, a polar promoter is necessary to bring about the transmetallation reaction.

In contrast to living polymer, the molecular weight of a telomer is governed by the ratio of propagation to the rate of transmetallation, and the breadth of the molecular weight distribution is minimized if this ratio can be kept constant throughout the process [1]. The rate of propagation is usually kept relatively constant by adding the butadiene monomer in a continuous manner and thereby controlling monomer concentration. Since the activation energy for the metallation is greater than for propagation, an increase in temperature for a given addition rate of butadiene results in a product of lower molecular weight. The rate of transmetallation can also be varied relative to propagation by adjusting the concentration of telogen or of polar promoter [9].

A polar promoter is necessary when an organolithium initiator is used in telo-merization process with toluene as telogen. Typical promoters are N,N,N',N'-tetramethylethylene diamine (TMEDA) [10–12], and N,N-dimethylaminoethoxy-ethane [11].

The use of alkoxides of heavier alkali metals in conjunction with organo-lithium initiators and telogen is another viable procedure to produce liquid BRs [9,13,14]. Hsieh and Wofford [15] proposed that in a polymerization initiated with combinations of alkyllithium and heavier alkali metal alkoxides, dynamic tauto-meric equilibria between carbon-metal bonds and oxygen-metal bonds exist and lead to quite different propagating centers than with the alkyllithium alone. In fact, it has been reported that organosodium and organopotassium compounds have been isolated from the reactions of organolithium with sodium and potassium alkoxides, respectively [13,14,16,17]. The combinations of an organolithium compound and a sodium or potassium alkoxide is often referred to as “Superbase.” These agents are highly reactive and often applied in organic synthesis and in polymer chemistry [18,19]. The “Superbase” is an active transmetallation agent and it is applied not only in the production of low-molecular-weight polymers but also in the preparation of graft polymers via metallation of the polymer chains. One can view the activities of the “Superbases” as the “in-situ” formations of the highly active organosodium or organopotassium compound [1,15]. Lochmann and Petraneck [20] found in a very recent study that even higher yields



and higher reaction rates, as well as more selective metallation in the side chain of alkylbenzenes, can be achieved by “Superbases” from alkyllithium compounds and potassium alkoxides that contain an enhanced amount of an alkoxide with a structure more bulky than that of the often used tert-butoxide.

With the use of “Superbases” for the preparation of liquid BRs, the vinyl configurations of the products approach polymers prepared with corresponding heavier alkali metals, generally in the range of 40–70% [15].

The use of a promoter is generally unnecessary when sodium metal or organosodium compound is used as initiator in telomerization reactions [21,22]; presumably, this is because of the greater reactivity of a polybutadienyl-sodium chain end to transmetallation.

When organosodium initiators were used in mixtures of toluene and THF, only low-molecular-weight telomers of butadiene were produced, even at relatively low temperatures [23,24]. The use of sodium-based initiators, with or without a promoter, yields telomers with 60–70% vinyl configuration. The vinyl contents of the telomers can be substantially higher when they are prepared in a solvent mixture containing a Lewis base such as THF at subzero polymerization temperature.

The low-molecular-weight polymers prepared by the telomerization process (telomers) have broader molecular weight distribution than the “living polymers” produced without termination and transfer reactions. The molecular weight distribution of a telomer is around 2.0, in accord with the statistical nature of the chain termination reaction, while the typical molecular weight distribution of a “living polymer” is generally in the range of 1.0–1.2.

III. Polymerization And Telomerization Of Dienes Involving Amine-chelated Lithium Catalysts

A. Polymerizations

Langer and Eberhardt showed independently that monomeric one-to-one complexes of alkyllithiums, such as n-butyllithium, with certain diamines, such as TMEDA and sparteine, a diamine containing tertiary bridgehead nitrogen atoms, drastically increases the polymerization rates of various monomers such as ethylene and butadiene. A chelated structure has been proposed to account for the observed rate increase in which a monomeric or unassociated, more highly charge-separated, complex functions as the initiating agent.

Langer [12,25–27] was the first investigator to synthesize polyethylene resins using an N-chelated alkyllithium compound. Langer observed that by using the unaged chelated initiator, polyethylene products with much lower molecular weights were produced. The molecular weights in this low range could be controlled by varying the temperature because of the occurrence of chain-transfer reactions back to the diamine during polymerization (Table 21.1).



Table 21.1 Ethylene Polymerization: Effect of Polymerization Temperaturea

Temperature (° C) Molecular weight

85 11,700

70 23,000

50 32,500

30 42,500

a2 Mmoles BuLi . TMEDA, 150 ml heptane, 1000 psig, 4 h.

Source: Ref. 25.

Eberhardt [28,29] was able to produce polyethylene waxes with molecular weight around 1300–2000 by polymerizing ethylene at 100–120°C under pressure in n-octane using nBuLi-TMEDA (or sparteine) complexes. Eberhardt postulated chain transfer to ethylene during polymerization to account for the formation of more than 1 mole of polymer per mole of initiator and for the presence of terminal unsaturation in the polymer.

Alkyllithium-chelating diamine complexes, such as n-BuLi-TMEDA, are very active diene polymerization initiators. Antkowiak and co-workers [8] showed that bi- and polydentate polar modifiers such as TMEDA and diglyme have a much greater influence on microstructure than do monodentate Lewis bases such as THF or triethylamine. At the same time, temperature exerts a great effect on microstructure in the presence of these more powerful modifiers (Figs. 21.1 and 21.2). Those figures show that the effect of diglyme and TMEDA are almost identical, suggesting that their mode of action is the same.

Langer [27] showed that a sharp decrease in both polymerization rate and vinyl content of the polybutadiene is obtained as the number of carbon atoms separating the nitrogen atoms in the chelating diamine is varied away from two or three (Table 21.2). He also showed that increasing the substituent size on the nitrogen atoms of the diamine decreases the polymerization rates, but does not affect microstructure significantly at low temperatures. In addition, increases in the number of nitrogens in the chelating amine from two to four cause no change in vinyl microstructure.

Hay and co-workers [30] showed that polymerization of butadiene at low temperatures by the n-BuLi-TMEDA initiator at ratios less than 0.5 was “living” type, giving polymers with predictable molecular weights and molecular weight distribution.

It is generally agreed [31] that both chelate type and temperature effects (Figs. 21.1 and 21.2) indicate that specific lithium solvation rather than a general



Figure 21.1 Influence of temperatures on diglyme-modified

butadiene polymerizations. (From Ref. 8, used with permission from John

Wiley & Sons.)

solvent effect is operable here. Also, this high degree of lithium solvation results in propagation from a true anionic reacting site located mainly at the secondary carbon of the terminal, active delocalized allyl carbanion in the growing polymer.

B. Telomerizations

The key element of telomerization is the transmetallation reaction. Equal molar amounts of n-BuLi and an amine, such as TMEDA, transmetalate toluene readily at ordinary temperatures to form a benzyllithium amine complex. Toluene is the telogen in this transmetallation process.

Butte and Eberhardt [32] studied the effectiveness of various amines as



Figure 21.2 Influence of temperatures on TMDA-modified

butadiene polymerizations. (From Ref. 8, used with permission from

John Wiley & Sons.)

Table 21.2 Polymerization of Butadienea

No. of C atoms separating Ns in diamine

Polymer produced g/g BuLi in 2 h at 25°C

% Vinyl microstructure

1 104 25

2 7750 80

3 544 80

4 88 25

5 106 25

aConditions: 2 Mmoles BuLi-TMEDA (1:1), 2 moles C4H6 in n-C5.

Source: Ref. 26.



Table 21.3 Extent of Metallation of Toluenea

Amine/BuLi Extent of metallation

Amine (molar ratio) (%)b

Et3N 3 2

MeN(CH2CH2)2NMe 1 5

Me2NCH2NMe21 6

Me2NCH2CH2NMe21 77

Me2NCH2CH2CH(Me)NMe21 60

Sparteine 1 90

N(CH2CH2)3N 1 43

N(CH2CH2)3N 2 66

a1 h at 60°C, 1 M n-BuLi.

bPercentage of theory based on n-BuLi.

Source: Ref. 32.

promoters for transmetallations by comparing the amounts of benzyllithium formed under identical conditions (Table 21.3).

The marked influence of TMEDA on the reactivity of n-BuLi can be ascribed to the excellent donor characteristics of the bridgehead nitrogen atom [33] that disrupts the n-BuLi aggregates [34]. Furthermore the coordination of lithium by the amine polarizes the carbon-lithium bond (solvation), thereby easing lithium-hydrogen interchange.

Using ethylene as monomer, the influence of the telogen upon molecular weight of the telomer was examined [32]. Since, under equivalent conditions, the molecular weight depends only on the transmetallation rate, the hydrocarbons can be ranked in order of their decreasing kinetic acidity: toluene > xylene > benzene > ethylbenzene > isobutylene.

For the production of liquid polybutadienes, the same influences of the types of amines and telogens apply as in the preparation of low-molecular-weight waxy polyethylenes. The use of highly efficient promoters such as TMEDA produces polybutadienes with increased vinyl content. As shown in Figure 21.2, vinyl configuration in polybutadiene can be as high as 80%. Since telomerizations are usually carried out under conditions of continuous butadiene addition, the system can suffer from monomer starvation. In this circumstance, promoters such as TMEDA can encourage a living chain end to undergo a cyclization reaction [10,11,35]. Quack and Fetters have shown that telomers produced with TMEDA as a promoter have appreciable levels of a 4-vinylcyclopentane unit in addition to the normal microstructures resulted from 1,2- and 1,4-additions [36]. They postulated an intramolecular back biting reaction of the chain end with a preceding vinyl unit as the mechanism of cyclization.



IV. Telomerization Of Dienes Involving Amine-chelated Sodium Catalysts

Bunting and Langer [37] have shown that telomerization of butadiene and isoprene with aromatics and olefins (telogens) proceeds rapidly at 0–100°C using organosodium catalysts in combination with aliphatic tertiary chelating polyamines containing two to six nitrogens. Chain transfer increased with increasing complexing ability of the chelating agent, increasing chelating agent concentration, increasing acidity of the telogen, increasing temperature, and decreasing monomer concentration.

V. Commercial Nonfunctional Liquid Polybutadienes

Nonfunctional liquid polybutadienes have been produced with anionic, cationic, and coordination-catalyzed polymerizations and are listed in Table 21.4.

VI. Chemical Modifications Of Nonfunctional Liquid Polybutadienes

Nonfunctional liquid polybutadienes contain high levels of unsaturation. The iodine number of these polymers is usually in the range of 400–450. For this reason they can be chemically modified in a variety of ways. In fact, the low-molecular-weight polybutadienes are easier to modify chemically than high-molecular-weight polymers; higher concentrations of reagents can be used with minimum levels of solvent. Among various chemical modifications, maleinization [1,38–40], epoxidation [1,41–47], and chlorination [1,48–50] are probably the most commercially important reactions. Other chemical modifications include hydrogenation [51], grafting [52,53], silylation [54,55], hydroformylation [56], addition of phenolic structure via Friedel-Craft catalysts [57,58], hydroboration [54], chlorohydrin formation [59], and simple blowing with air or oxygen to introduce a plurality of functional groups.

VII. Applications Of Nonfunctional Polybutadienes

Luxton in his review [1] stated that three main features of liquid BRs have an important bearing on their application. First, the bulk and solution viscosity are important in relation to designing formulations with the minimum levels of solvent or reactive diluent. Typical viscosities of liquid polybutadienes are shown in Table 21.5. Second, the high level of unsaturation, in addition to facilitating chemical modifications, enables the liquid BRs to be readily cured. Third, the hydrocarbon backbone results in a polymer, which, after cure, is highly resistant to hydrolysis and other chemical attacks. This feature has proven important in surface coating applications.



Table 21.4 Commercial Nonfunctional Liquid Polybutadienes

Trade name

Manufacturing company or

supplier

Technique of preparation

Microstructure

Approx. ranMWs avai

(no, avg., g

Polyol Chemische WeikeHuls A.G.

Coordination catalysis High 1,4; cis-1,4 ca.75%

1500–30

Lithenea Revertex, Limited Anionic polymerization(organo-Li), telomers and living polymers

Medium to high 1,4;mixed cis and trans; also cyclic grade

1000–80

Riconb Colorado ChemicalSpecialties, Inc.

Anionic polymerization(Na)

High vinyl-1,2 ca.70%

1000–60

Nisso-PBa Nippon Soda Co.,Ltd.

Anionic polymerization(Na or organo-Na)

Very high vinyl-1,2 ca.90%

1000–40

Hystla Hystl DevelopmentCorp.

As Nisso-PB

Nisseki LPB Nippon Oil Co. Anionic polymerization(organo-Na)

High vinyl-1,2 ca. 60–70%

1000–40

Budium E. I. Du Pont de Nemours and Co.

Cationic polymerization High 1,4; trans-1,4 ca.80%

Not know

Plastikator 32 Chemische WerkeBuna

Anionic polymerization(K)

High vinyl-1,2 ca. 32,0

aGrades with functional end groups available; lithene grade is monofunctional.

bLow-molecular-weight copolymers of butadiene and styrene are also available.

cBR of molecular weight 4500–9500 has a typical viscosity of 17 poise (25°C) as a 60 wt.% solution in heptane.

dNormally supplied as a solution in mineral spirits.

Source: Ref. 1.



Table 21.5 Typical Viscosities of Liquid Polybutadienes

Approx. number Polybutadiene microstructure (%) Brookfield LVT

average MW trans viscosity

(g mole-1) Vinyl cis Cyclica (Pa.s, 25°C)

BR Telomers (molecular weight distribution = 1.6–2.2)

900 40–50 30–40 15–25 — 0.3

1300 40–50 30–40 15–25 — 0.7

2600 40–50 30–40 15–25 — 8.5

1500 20–30 40–50 20–30 — 0.7

3400 20–30 40–50 20–30 — 8.5

1000 40–50 15–25 10–20 15–20 4.0

1800 40–50 15–25 10–20 15–20 45.0 (35°C)

PBD living polymers (molecular weight distribution = 1.0–1.2)

5000 15–20 45–55 25–35 — 40

5000 40–50 30–40 15–25 — 80

5000 60–70 20–30 10–20 — 350

5000 30–35 25–30 25–30 40–45 >1000

a4-Vinylcyclopentane chain units.

Source: Ref. 1.

Major applications for nonfunctional liquid BRs are aqueous surface coatings, nonaqueous surface coatings, thermoset resins, additives to other rubber or resin systems, and other miscellaneous applications. Detailed discussions of these applications have been reviewed by Furukawa [3] and Luxton [1].

References

1. Alan R. Luxton, Rubber Chem. & Tech., 54, 596 (1981).

2. W. W. Crouch, private communication.

3. J. Furukawa and S. Yamashita, High Polymers, 231, Pt. 2, Ch. 9 (1968).

4. Firestone Tire and Rubber Co., Br. patent 1,343,940.

5. Nippon Soda Co., Br. patents 1,255,764 and 1,253,757.

6. Esso Res. Eng. Co., Br. patent 802,308.

7. C. A. Uraneck, J. Polym. Sci., A-1, 9, 2273 (1971).




16. L. Lochmann, J. Pospisil, and D. Lim, Tetrahedron Letters, 2, 257 (1966).

17. L. Lochmann and D. Lim, J. Organometal Chem., 28, 153 (1971).

18. L. Lochmann and J. Trekoval, Collect. Czech Chem. Commun., 53, 76 (1988).

19. M. Schlosser, Pure Appl. Chem., 60, 1627 (1988).

20. L. Lochmann and J. Petraneck, Tetrahedron Letters, 32, No. 11, 1483 (1991).

21. Y. Atsuki, H. Hara, and N. Imai, US patent 3,789,090.

22. H. Hara, A. Kaiya, and Y. Atsuki, US patent 4,155,942.

23. A. DeChirico, A. Proni, A. Roggero, and M. Bruzzone, Angew. Makromol. Chem., 79, 185 (1979).

24. S. Kume, A. Takahashi, G. Nishikawa, M. Hatano, and S. Kambara, Makromol. Chem., 84, 137 and 147 (1965).

25. A. W. Langer, Jr., Trans N.Y. Acad. Sci., 27, 741 (1965).

26. A. W. Langer, Jr., Polymer Div. Preprints, American Chemical Society, 132 (1967).

27. A. W. Langer, Jr., First Akron Summit Polymer Conference, Preprint (1970).

28. G. G. Eberhardt and W. R. Davis, J. Polym. Sci., Part A, 3, 3753 (1965).

29. G. G. Eberhardt, US patent 3,567,703.

30. J. N. Hay, et al., Faraday Trans., 1, 1 (1972).

31. C. W. Kamienski, Polymerizations Using N-Chelated Alkali Metal Catalysts, in Polyamine Chelated Alkali Metal Compounds, A. W. Langer, Jr., Ed., Advances In Chemistry Series 130, American Chemical Society, Washington, DC, 1974, Chap. 7.

32. W. A. Butte and G. G. Eberhardt, Telomerization Reactions Involving Amine Chelated Lithium Catalysts, in Polyamine Chelated Alkali Metal Compounds, A. W. Langer, Jr., Ed., Advances In Chemistry Series 130, American Chemical Society, Washington, DC, 1974, Chap. 9.

33. H. C. Brown and S. J. Sujishi, J. Am. Chem. Soc., 70, 2878 (1948).

34. D. Margerison and J. P. Newport, Trans. Farady Soc., 59, 2058 (1963).

35. A. F. Halasa, US patent 3,966,691.

36. G. Quack and L. J. Fetters, Macromolecules, 11, 369 (1978).

37. W. Bunting and A. W. Langer, Jr., Telomerization of conjugated diolefins with aromatics and olefins using chelated organosodium catalysts, in Polyamine Chelated Alkali Metal Compounds, A. W. Langer, Jr., Ed., Advances In Chemistry Series 130, American Chemical Society, Washington, DC, 1974, Chap. 10.

38. C. Pinazzi, J. C. Danjard, and R. Pautaat, Rubber Chem. & Tech., 36, 282 (1963).



49. P. J. Canterino, Ind. Eng. Chem., 49, 712 (1957).

50. J. Royo, L. Gonzalez, L. Ibarra, and M. Barbero, Makromol. Chem., 168, 41 (1973).

51. Y. Camberlin, J. Gole, J.P. Pascault, J. P. Durand, and F. Dawano, Makromol. Chem., 180, 2309 (1979).

52. Nippon Oil Co. Ltd., Br. patents 1,438,370; and 1,438,718.

53. A. F. Halsa, Metallation and grafting by anionic techniques, in Polyamine Chelated Metal Compounds, A. W. Langer, Ed., Advances in Chemistry Series 130, Am. Chem. Soc., Washington, D.C., 1974, Chap. 8.

54. C. Pinazzi, J. C. Brosse, A. Pleurdeau, and D. Reyx, Appl. Polym. Symp., 26, 73 (1975).

55. J. Miron, P. Bhatt, and I. Skeist, J. Adhesion, 4, 275 (1972).

56. J. K. Mertzweiller and Tenney, US patent 3,311,596.

57. Chemische Werke Huls A. G., Br. patent 1,428,999.

58. F. E. Kempter and E. Schupp, US patent 4,189,450.

59. A. Norshay and A. H. Gleason, US patent 3,317,479.



22 Telechelic Elastomers and Prepolymers

I. Telechelic Elastomers

A. Introduction

The vulcanization of rubber is a complex process that produces a three-dimensional network structure. A network structure produced by any of the conventional vulcanization reactions may possess several types of flaws or imperfections. Among these are free chain ends, two of which exist for every primary polymer molecule. The two free chain ends per polymer molecule in the vulcanizates contribute little to the physical properties as predicted by Flory [1]. A schematic representation of a vulcanizate with the free ends emphasized is given in Figure 22.1a. An estimate of the fraction of rubber existing as free chain ends is given by the simple relationship, 2Mc/M, where Mc is the average molecular weight between crosslinkages and M is the average molecular weight of the primary molecules. In a normal styrene/butadiene rubber (SBR) tread vulcanizate this fraction probably amounts to over one-tenth of the polymer molecules. The incorporation of this fraction of the molecules into the network structure should contribute significantly to the properties of the vulcanizates [2,3]. This contribution is difficult to estimate because a terminal group chain linkage is trifunctional whereas an internal crosslink is tetrafunctional. The two types of linkages are illustrated in Figure 22.1b.

In order to study the end group problem a simple means is needed for



Figure 22.1 Schematic representation

of a conventional network structure of

(a) a vulcanizate emphasizing free chain ends and

(b) a cured telechelic polymer with chain ends incorporated

into the structure.

preparing polymers possessing different kinds of terminal reactive groups. Since polymers possessing terminal reactive groups should be of considerable interest and value, a word has been coined to avoid a cumbersome description of these types of polymers [4]. The term “telechelic” (Greek telos, meaning far, plus chele, meaning claw) was originally proposed by Uraneck [4,5] for polymer molecules possessing two functional terminal groups. The functional group is identified as a radical according to standard nomenclature practice and is used as a prefix. For example, carboxy-telechelic polybutadiene (α,ω-dicarboxy polybutadiene) is a butadiene polymer with possesses two carboxyl terminal groups per molecule.

Preparation of telechelic polymers can be done by an emulsion (free-radical) and a solution (anionic or cationic) procedure. Examples of characterization of the polymers and preparation of a mercaptotelechelic copolymer in an emulsion system and of mercapto-, hydroxy-, and aziridinyl-telechelic elastomers in solution system by the use of a bis(alkali metalorgano) compound as the initiator were reported. The elastomers were cured in peroxide or sulfur-accelerator formula-



tions. The telechelic elastomers exhibited enhanced stress-strain and dynamic properties in comparison to those of the controls. In tread formulations, outstanding properties were obtained for the mercapto- and aziridinyl-telechelic butadiene-styrene copolymers [4,5].

B. Synthesis

In 1956, Szwarc [6] published his concept of “living polymerization” with an ion-radical initiator by which polymer chains are growing from two anionic ends after the coupling of the radical ends of the two corresponding living oligomers shortly after initiation. This concept lead to the development of block copolymers and polymers with functional end groups [7,8]. In practice, the use of sodium-based initiator, polar solvent such as tetrahydrofuran (THF) and low reaction temperature made the process unattractive in large-scale operation. Furthermore, using Szwarc's description some years later, the use of sodium and THF resulted in the tendency of the chain ends to commit “suicide” (unintentional termination reaction or transfer reaction), unless the temperature is kept very low.

The two key factors for a viable industrial process for the production of telechelic polymers are the availability of a hydrocarbon-soluble, lithium-based initiator with difunctionality and a process of introducing the functional end groups with minimum side reactions. The initiators developed in the late 1950s and early 1960s were made in the presence of ethers. These dilithium compounds such as lithium-stilbene adduct (1,2-dilithio-1,2-diphenylethane) or “living” oligomers of dienes (Li-(diene-)n-Li) gave a telechelic polybutadiene with 25–60% vinyl configuration. Later, initiators prepared in 100% hydrocarbon solvent were example, Tung and his co-workers [9] reported the synthesis of a thermally stable difunctional initiator that is soluble in hydrocarbon solvents such as cyclohexane by reacting 1,3-bis(phenyl ethenyl)benzene (DDPE) with two moles of sec-BuLi in hydrocarbon solvents. This reaction (1) is also free of oligomeric products.



developed by which polybutadiene with 10–12% vinyl can be produced. For

The other key element of the preparations of telechelic polymers is the functionalization reaction. The polymer molecules with two “living” chain ends (e.g. Li+-M ˜˜˜ M-Li+) are allowed to react with a reagent to introduce the desired functional end groups. In a typical case, carbon dioxide is used to give the carboxylate group; ethylene oxide or formaldehyde is used to give primary alcoholate group, propylene oxide is used to give secondary alcoholate group, ethylene sulfide is used to give thiolate group, and so on.

The sequence of reactions (2–4) representing of carboxy-telechelic polymers is essentially as follows:

The chemistry of functionalization is extensively discussed in Chapter 11 of this book. A comprehensive list of various functional groups is available in Goethals' book on telechelic polymers [10].

Phillips Petroleum Company pioneered the industrial production, curing, and applications of terminally active polymers by means of dilithium initiation followed by functionalization. The research and development work was started in 1957 and resulted in a series of patents [11–20] and publications [4,5].

Schultz and Halasa [21–23] developed an alternative approach to preparing telechelic polymers by using initiators with blocked functional groups. After the completion of polymerization, the polymer molecules contain one living end (e.g., ˜˜˜C-Li) and one blocked functional end (e.g., ˜˜˜C-O-X or ˜˜˜C-N-X2, where X is the blocking moiety). These molecules are then coupled to form the telechelic



polymers after the subsequent removal of the blocking moiety. For example, p-lithio-N,N-bis(trimethylsilyl)aniline (structure 1) in which the primary amine

group has been silylated (blocked) was used as initiator for butadiene polymerization [22]. Polymerization was carried out homogeneously in hexane/ether mixtures to make monofunctional living polymers of low polydispersity (1.1–1.3), but of 40–50% vinyl content, or heterogeneously in hexane/toluene mixtures to make higher polydispersity (1.5–2.0), but low vinyl content (10%). Difunctionality was achieved by coupling with dimethyl dichlorosilane, but the efficiency averaged only 70–80%. Finally, the primary amine end-groups were regenerated by acid hydrolysis. By applying similar principles, primary hydroxy telechelic polymers were prepared by using lithium aliphatic and aromatic hydrocarbon acetals and ketals as initiator [21]. This novel approach, which suffers from the low solubility of the initiator in hydrocarbon solvent, the low efficiency of coupling reaction, and overall complexity of the process, is not used commercially. The use of initiators such as Li-O-C6H4-Li [24] for the preparation of polymers with multifunctional hydroxyl end groups have likewise been studied but not used in production.

II. Telechelic Prepolymers

A. Introduction

The commercial production of a liquid carboxy-telechelic (CTL) polybutadiene, Butarez CTL, and a hydroxy-telechelic (HT) polybutadiene, Butarez HT, was started in 1962 in Borger, Texas, by Phillips Petroleum Company. These liquid polymers are still in production today. Some of the physical properties [25–27], production details [28], and uses, as in rocket binders [29], of these low-molecular-weight telechelic prepolymers have been described. Butarez CTL and HT liquid polybutadienes were soon joined by other nearly similar products such as Telegen CT and HT of General Tire (production was stopped after a few years).

These telechelic polybutadienes generally have molecular weights in the range of 2000–8000 and are viscous liquid at room temperature. Because of the intermolecular associations of the chain ends via hydrogen bonding, these telechelic polybutadienes have higher bulk viscosity than the nonfunctional analogues. At higher temperatures, the differences become very small.

Unlike most of the polymers prepared by hydrocarbon soluble organolithium initiators, these low-molecular-weight telechelic polybutadienes have broader



Figure 22.2 GPC curves of Butarez CTL.

molecular weight distribution. The gel permeation chromatographic (GPC) curve of Butarez CTL (Fig. 22.2) indicates a bimodal distribution of molecular weights. The cause of the bimodal distribution was coupling of the polymer chains, as will be discussed later.

Most of the products used dilithium initiators prepared in the presence of ethers. Thus, the final polymers contain vinyl unsaturation in the range of 25–45%. One exception is Nisso, a Nippon Soda product, which has ˜90% vinyl configuration [5]. This product is made in tetrahydrofuran (THF) at very low temperature employing a sodium-based initiator. This product, in its applications, served as a prepolymer to produce a high-molecular-weight product via multilinking reactions with a di- or tricoupling agent. At the same time, the active vinyl side chains serve as the sites for further chemical reactions, such as cyclization with the neighboring vinyl group.

In recent years, there has been a push from the aerospace industry to produce liquid telechelic polybutadienes with lower vinyl content than the polymers such as Butarez CTL or HT that have been used as solid rocket binder since the early 1960s. The main reason for this push is the need to lower the Tg of the polymers to meet the requirements of storing, carrying, or firing of the rockets at ultralow temperatures. Several very good dilithium initiators prepared in pure hydrocarbons have been developed in recent years. Polymers of 12–15% vinyl can be produced and are being used in selected applications by the aerospace industry.

Functionality or “telechelicity” is another critical factor for a successful product of this type. As a prepolymer for further chain-linkings, functionality of 2.0 would be ideal. Polymers made with dilithium initiators would give the theoretical functionality of 2.0, assuming no premature termination and 100% functionalization. In reality, this ideal condition can rarely be achieved in a commercial process. The commercial liquid carboxy-telechelic and hydroxy-



telechelic polybutadienes typically have an average functionality of 1.8–1.9. In some applications, functionality slightly higher than 2.0 is desired. Morrison and Kamienski[30] developed a catalyst system by reacting 1,3-divinylbenzene (DVB) with sec-butyllithium. DVB can easily oligomerize and, by adjusting the ratios of the two reagents and varying reaction conditions, functionalities higher than 2.0 (average) can be achieved. These catalysts can be used to produce telechelic polymers with mixed functionalities.

III. Carboxy Telechelic Polybutadienes

The chemistry involved in the production of liquid telechelic polybutadienes is fairly straightforward and have been described elsewhere in this book. The technology involved, however, is somewhat unique. In fact, because of certain peculiarities of these polymers and their intermediate products, special precautions and procedures were needed.

A. Initiation

The dilithium initiators must be hydrocarbon soluble. In the 1950s–1970s, when the products were developed, a truly hydrocarbon-soluble dilithium compound was not available. The oligomers of dienes, such as 2,3-dimethyl-butadiene, with lithium ends were prepared in ether with the aid of a transfer agent such as methylnaphthalene[16]. To make it more soluble, often more butadiene or isoprene was added to form a longer chain. Other initiators[31–34] developed in the late 1970s were based on divinyl compounds. Examples of these divinyl compounds other than divinyl benzene[30] are:



In these preparations, sec-butyllithium was generally used. A tertiary amine was often added to promote the reaction and improve the solubility. In some cases, small aliquot of diene monomer was added to form a seeded catalyst to improve the solubility. In one case[34], excess butyllithium was used to increase the yield and improve hydrocarbon solubility. Unfortunately, the unreacted butyllithium would lead to monofunctional polymer. As far as we know, none of these initiators are being used to produce telechelic polymers commercially. One important difference in using a dilithium initiator for making liquid polymers vs. making rubbery polymers is that the level of the initiator used for the former is two orders of magnitude higher.

B. Polymerization

In the presence of a small amount of polar compound, generally present in the initiator solution, or in the absence of any polar compound, the living polymer solution is extremely viscous due to association. Since the polymer molecular is really a long-chain dianion, the association leads, in theory, an infinite molecular weight giant molecule. This is not true when the polymerization is carried out in the presence of high levels of polar compounds. Association is minimized under this condition. Generally, the polymer concentration in polymerization solution is 10–15%.

C. Functionalization

Functionalization reactions on a low-molecular-weight polymeric dianion are problematic. The difficulties arise from unusually high solution viscosity, very high anion concentration, and the formation of pseudogel.

The abnormally high solution viscosity, due to the association of chain-ends can be reduced to a manageable condition by the introduction of small amounts of polar compounds just before the functionalization.

The very high anion concentration and the formation of pseudogel not only make it physically difficult to carry out the functionalization reactions, but also lead to side reactions. The moment carboxylate or hydroxylates are formed by reacting the living polymer solution with CO2 or ethylene oxide, respectively, a pseudogel is formed instantly. It makes the whole mass immobile. It is called pseudogel because it has the appearance of a gel, but it returns to solution after acidification. The pseudogel is formed due to the extensive associations of the carboxylate or hydroxylate.

Pseudogel prevents, in extreme cases, complete functionalization due to the immobility of part of the polymer solution while other parts are still unreacted with CO2 or ethylene oxide. The unreacted living polybutadiene molecules can react with polymeric carboxylates to give unwanted side products[35,36] affecting the average molecular weight and molecular weight distribution. Of course it also affects functionality.



The best method to carry out carboxylation is using a large excess of CO2 in gas form with a Pownell mixing T-tube [37,38]. For example, an 8–12% polymer solution in which a small amount of a polar compound, such as THF, was added to fluidize the solution, was cooled. The solution entered the top arm of the T-tube under nitrogen pressure (10–20 psig), the CO2 gas entered the side arm of the T-tube under its own pressure (around 20 psig), and the carboxylated polymer was forced out in the form of snowlike small particles. The amounts of gaseous CO2 used is in large excess plus the sudden expansion of the CO2 which significantly lowers the temperature and the contact time is very short, all of which help to minimize side reactions.

Acidification, which instantly converts the pseudogel to a low-viscosity solution, can be accomplished by the use of either anhydrous HCl in benzene [37] or, sometimes, HCl/isopropyl alcohol mixture [39]. Aqueous HCl is not recommended because water could cause emulsification of the polymer solution, thus making polymer purification and recovery more difficult [12,38].

Patent literature [17,40] also claims that methanol, but not water, is the reagent of choice to remove LiCl. The highly concentrated polymer precipitate containing solvent and methanol was distilled to recycle the polymerization solvent [41,42]. Another process for the removal of the salt residue from the terminally reactive polymers is by dispersing a small quantity of water to the anhydrous HCl-acidified polymer solution, and subsequently contacting the resulting mixture with anhydrous calcium chloride [43].

D. Chain Extension and Crosslinking

The liquid carboxy telechelic polybutadienes are almost exclusively used as the binder for solid rocket fuel by the aerospace industry. Butarez CTL by Phillips was the first product produced for this application and is still in production after more than 30 years.

The terminally active polymer can be chain-extended and crosslinked by an aziridinyl compound [44]. This process can be accelerated by the inclusion of lithium oleate [45]. In contrast, tributylamine could slow down the reactions [46]. There are many kinds of aziridinyl compounds (see Table 22.1), but the common one is tri[1–(2-methyl)-aziridinyl] phosphine oxide (MAPO), which is mainly applied to the binder of the rocket propellant. However, MAPO will react violently with the oxidant-ammonia superchlorinate. So the binder, carboxy telechelic polybutadiene, and the vulcanization agent such as MAPO must be pre-



mixed according to the following formula, in which HOOC-poly-COOH denotes αϕ-carboxylpolybutadiene. A list of aziridinyl compounds is given in Table 22.1.

The curing reaction led to the chain extension and the molecular weight reached 105. The activation energy of the reaction was low (E = 13.8 kcal/mole) [48]. So the reaction took place at 50°C and good results could be obtained, but the reaction would last longer [49]. In order to obtain excellent mechanical properties, 100–200% excess MAPO was required. Usually, if the reaction temperature is 80°C the linking reaction will last for several days. When the temperature was about 100°C the reaction time could be reduced, but the side reactions also increased. By using lithium oleate as catalyst, the linking reaction can be successfully carried out at 50–70°C [45].

During the linking reaction with MAPO, it is possible to have side reactions: self-polymerization of MAPO, the reaction between MAPO and the trace water present in the polymer, and the reaction with trace hydroxyl-containing compounds. The self-polymerization of MAPO resulted in a higher functionality of the tri-aziridinyl compound. It was reported [50] that the average value is 3.3.

The tensile strength of the vulcanizate by MAPO was 50–250 pound/inch2 and the elongation was 125–575%. After adding carbon black or active silicon dioxide, the tensile strength could reach 1550 pound/inch2 and the elongation would be 400% [37]. Before vulcanization by MAPO, if the α,ϕ-carboxylpolybutadiene is pretreated by aniline derivatives the tensile strength can greatly increase, but the elongation will be decreased [35].

If the linking agent is polyepoxy compounds, such as dioxyethylene cyclohexane or liquid epoxy resin, the vulcanizate will not lose its shape even at 100°C or above [51]. Usually, the resistance to aging and the stability of the linked product by these linking agents are better than those by polyaziridinyl compounds. However, the reaction between the epoxy group and carboxyl group is very slow. The linking reaction will take a long time. Using catalyst and elevating tempera-



ture can help the process. The catalysts often used are tertial amine [39,48], chromium soap [48,50,52], and ion acetylacetonate [50,52], among others.

IV. Liquid Hydroxy-telechelic Polybutadiene

A. General Descriptions

The process steps for the production of hydroxy-telechelic polymers is similar to the steps required to produce the carboxy analogues. The key difference is the functionalization reactions. For subsequent linking reactions, the order of reac-

tivity is allyl alcohol > 1° alcohol > 2° alcohol > 3° alcohol. In some applications, a slower reaction rate is advantageous. In others, a rapid reaction rate is desirable.

In functionalization reactions as shown above, pseudogel is also instantly



formed and causes mixing difficulties. It is less critical than the carboxylation reaction, since the hydroxylation reactions are mostly free from the side reactions.

The microstructure of the polybutadiene has much effect on properties. This is independent of the end groups. Vinyl content affects the bulk viscosity of the telechelic prepolymer. Vinyl content also affects the Tg of the polymer and, thus, low-temperature properties.

Not all the commercial liquid polybutadienes with hydroxyl end groups are produced by anionic polymerization. One major product line produced and marketed by Atochem North America under the trade name of Poly BD is prepared with hydrogen peroxide as initiator [53–55]. It is a free-radical polymerization process. These products have functionalities in the range of 2.1–2.5 and microstructure of a typical free-radical polymerized polybutadiene, namely ˜20% cis, ˜60% trans, and ˜20% vinyl.

The GPC curves of Phillips Butarez HTS, a secondary alcohol telechelic polybutadiene produced by anionic mechanism, and ARCO's (now Atochem North America's) Poly bd R-45HT sample, a free-radical polymerization product, are shown in Figures 22.3 and 22.4.

Figure 22.3 GPC curves of an anionic-initiated

hydroxy-telechelic polybutadiene (Butarez HTS).



Figure 22.4 GPC curves of a free-radical-initiated

hydroxy-telechelic polybutadiene (Poly bd R-45HT).

B. Linking and Crosslinking

For hydroxy-telechelic polymers the reagents used for linking and crosslinking are isocyanates. If polyisocyanate is used, the cross-network linked polymer can

be produced. This reaction is the basis for the production of polyurethane elastomers, plastic resins, and fibers. Most common prepolymers for polyurethanes are α,ω-dihydroxy esters or ethers. In recent years, hydroxy-telechelic polybutadienes have been used in some applications to produce polyurethanes. By having a pure hydrocarbon backbone, these polyurethanes have improved water-resistance, heat-stability, and aging-resistance.

In addition to the different reactivity of different hydroxyl groups toward isocyanate, the main chain structure of the isocyanate also has an effect on the reaction rate. Several common isocyanates used for the linking reactions are shown in Table 22.2.

The chain extension reaction or the chain crosslinking reaction of hydroxy-telechelic polybutadiene is similar to the synthesis of polyurethane. It can be a one-step or two-step process. For the simpler one-step process, nearly equal molar amounts of alcohol and isocyanate are allowed to react to give the final product with less tensile strength and, generally, poorer overall physical properties.



For the two-step process, the polymer is allowed to react with excess isocyanate to produce a prepolymer with isocyanate end groups. This prepolymer is then used to further react with a polyalcohol or polyamine to complete the linking and/or crosslinking reactions. When polyamine is applied, the urea group, is formed (polyurea). The existence of urea group can produce hydrogen bonding, so that there is a substantial increase in the strength of the vulcanizate. If the vulcanizate contains reinforcement agent such as carbon black, the effect of the hydrogen bonding self-reinforcement will be reduced or lost.

Usually the same two isocyanate groups on a single aromatic ring, such as in the case of toluene diifocyanate (TDI), have different reactivity and they also have effects on each other. The second isocyanate group in TDI has lower activity than the first one. So the excess isocyanate used in the first step of the two-step process can produce the prepolymer-bearing isocyanate group at both ends.

Stang [56] developed a unique curing process for hydroxy-telechelic polybutadiene that involves the use of alkali metals, alkali metal hydrides, and organoalkali metal compounds. This curing process is particularly useful in the field of rocket propellants. It virtually eliminates the formation of gas pockets and other voids within the cured polymer structure caused by reactions of curing agents such as isocyanate with high-energy additives in the propellant formulation. Thus, it provides even burning rate and maximum strength.

C. Applications

Hydroxy-telechelic polybutadiene in the 2000–5000 molecular weight range is also used as binder for solid propellants [57]. In this application, secondary alcohol is often preferred because its slower reaction rate with polyisocyanate increases the “pot life.” “Pot life” is a term used in the rocket solid fuel industry to describe the time between mixing together all the ingredients and the formed propellant inside the rocket wall. Lower vinyl content is necessary in many rocket applications, such as in air-to-air missiles, where very low temperatures would be encountered.

Another major application is in re-enterable splicing compounds for wire and cable used in telecommunications and encapsulants for electronic components. Sealants and asphalt modifications are among other commercial applications.

V. Conclusions

Telechelic polymers were initially developed to improve the performance of rubber vulcanizates by eliminating free chain ends. The living polymerization and the readiness of functionalization of the living ends made it ideal to synthesize polymers with reactive end groups. While the improvements of the vulcanizate by this process was clearly demonstrated [5], for various commercial reasons, mostly cost and convenience, it was not adopted by the industry.



On the other hand, the prepolymers or liquid polymers found their way to large-quantity applications. This is particularly true in the aerospace industry, where the liquid telechelic polybutadienes have been extensively used as binder for solid propellant since 1960.

References

1. P. J. Flory, Ind. Eng. Chem., 38, 417 (1946).

2. R. M. Pierson, A. J. Costanza, and A. H. Weinstein, J. Polym. Sci., 77, 221 (1955).

3. A. J. Costanza, R. J. Coleman, R. M. Pierson, C. S. Marvel, and C. King, J. Polym. Sci., 17, 319 (1955).

4. C. A. Uraneck, H. L. Hsieh, and O. G. Buck, J. Polym. Sci., 46, 535 (1960).

5. C. A. Uraneck, H. L. Hsieh, and R. J. Sonnenfeld, J. Appl. Polym. Sci., 13, 149 (1969).

6. M. Szwarc, Nature (London), 176, 1168 (1956).

7. M. Szwarc, Carbanions, Living Polymers and Electron Transfer Processes, Interscience, New York, 1968.

8. H. Brady, D. H. Richards, and M. Szwarc, Chem. Ind., 1473 (1958).

9. L. H. Tung, G. Y-S Lo, J. W. Rakskyo, and E. Beyer, US patents 4,200, 718; 4,201, 729; and 4,205,016.

10. E. J. Goethals, Review table, in Telechelic Polymers: Synthesis and Applications, E. J. Goethals, Ed., CRC Press Inc., Boca Raton, 1989; also Chapter 3 by D. R. Richards, G. C. Eastmond, and M. J. Stewart.

11. H. L. Hsieh, US patents 3,147,313; 3,175,997; 3,177,190; and 3,251,812.

12. R. P. Zelinski and C. W. Strobel, US patent 3,108,994.

13. R. P. Zelinski, H. L. Hsieh, and C. W. Strobel, US patent 3,109,871.

14. C. A. Uraneck, J. N. Short, R. P. Zelinski, and H. L. Hsieh, US patent 3,322,738.

15. C. A. Uraneck, J. N. Short, and R. P. Zelinski, US patent 3,135,716.

16. C. W. Strobel, US patent 3,157,604.


18. J. N. Short and R. P. Zelinski, US patent 3,269,978.

19. J. T. Gruver and G. Kraus, US patent 3,354,116.

20. J. W. deary, US patent 3,048,568.

21. A. F. Halasa and D. N. Schulz, US patents 3,862,100; and 4,052,370.

22. D. N. Schulz and A. F. Halasa, J. Polym. Sci., A1 (15), 2401 (1977).

23. D. N. Schulz, A. F. Halasa, and A. F. Oberster, J. Polym. Sci., A1 (12), 153 (1974).



32. E. Leitz and H. Hocker, Makromol. Chem., 184, 1893 (1983).

33. P. Lutz, E. Frantce, and P. Rempp, Polymer, 23, 1953 (1982).

34. P. Guyot, J. C. Favier, H. Vytterhoven, M. Fantanilk, and P. Sigwalt, Polymer, 22, 1724 (1981).

35. D. P. Wyman, V. R. Allen, and T. Altares, J. Polym. Sci., A-2, 4545 (1964).

36. P. Mansson, J. Polym. Sci., Chem. Ed., 18, 1945 (1980).

37. Phillips Petroleum Company, Br. patent 921,803; and J.N. Short US patent 3,210,333.

38. C. A. Wentz and E. E. Hopper, Ind. Eng. Chem. Prod. Res. Dev., 6, 209 (1967).

39. G. Kraus and J. T. Gruver, US patent 3,362,931.


41. C. A. Wentz, Jr. and E. E. Hopper, US patent 3,281,335.

42. C. A. Wentz, Jr., US patent 3,309,348.


44. J. T. Gruver and G. Kraus, US patent 3,354,116.

45. A. Adicoff and A. Y. Yukelson, US patent 3,434,990.

46. R. S. Drake and W. J. Mecarhy, Rubber World, 159(1), 51 (1968).

47. D. M. French, Rubber Chem. & Tech., 42, 71 (1969).

48. A. Adicoff and A. Y. Yukelson, J. Appl. Polym. Sci., 10, 159 (1966).

49. R. Yamazaki and H. Tokui, Bull. Chem. Soc., Japan, 38, 2174 (1965); 39, 2290 (1966).

50. D. M. French and R. A. H. Strecker, J. Appl. Polym. Sci., 12, 1697 (1968).

51. D. M. French and L. Rosborough, J. Appl. Polym. Sci., 10, 272 (1966).

52. R. A. H. Strecker, J. Appl. Polym. Sci., 13, 2439 (1969).

53. K. L. Seligmon, US patent 2,877,212.

54. P. W. Ryan and J. Q. Verdol, US patent 3,796,762.

55. O. W. Burke, A. A. Kiger, and P. Davis, US patents 3,673,168; 3,674,743; and 3,808,281.

56. P. L. Stang, US patent 3,341,509.

57. A. G. Ajaz, Rubber Chem. & Tech., 68(3), 481 (1995).



VI POLAR MONOMERS



23 Anionic Polymerization of Methyl Methacrylate and Related Polar Monomers

I. Anionic Polymerization Of Alkyl Methacrylates

A. Introduction

The anionic polymerization of polar vinyl monomers is often complicated by side reactions of the monomer with both anionic initiators and growing anionic chain ends, as well as chain termination and chain transfer reactions 1–10. However, synthesis of polymers with well-defined structures can be effected under carefully controlled conditions, often requiring the use of low polymerization temperatures to minimize or eliminate chain termination and transfer reactions. The challenge of extending the synthetic potential of living anionic polymerization to polar vinyl monomers and to effect polymerizations at higher temperatures has led to investigation and optimization of the effects of counterion, solvent, temperature, Lewis base additives, and inorganic salts. This chapter will consider in detail the anionic polymerization of alkyl methacrylates, alkyl acrylates, and related polar vinyl monomers.

B. Polymerizability of Alkyl Methacrylate Monomers

The polymerizability of methacrylates and acrylates can be considered in terms of the general structures, 1–3, for methacrylates, acrylates, and 2-substituted acrylates, respectively. The basic question is what types of alkyl and functional groups



can be tolerated in an acrylate or methacrylate monomers to maintain or promote anionic polymerizability and the synthesis of polymers with well-defined structures. In the methacrylate series, Structure 1, it is possible to incorporate a wide

variety of functional groups as part of the substituents on the ester group, R1, and maintain controlled anionic polymerizations of these monomers. A list illustrating the range of methacrylate monomers that generally can be polymerized to give polymers with well-defined structures is shown in Table 23.1. It is interesting to note that the normal anionic polymerization of alkyl methacrylates (-78°C,



tetrahydrofuran [THF]) can tolerate a wide range of functional groups in the ester alkyl group.

Thus, a wide range of glass transition temperatures, physical properties, solubility, and chemical reactivity can be produced by anionic polymerization of these substituted methacrylates. In addition, the methacryloyl group is often the choice for polymerizable end groups in macromonomers (see Chap. 11 and Chap. 14); therefore, methacryloyl-functionalized macromonomers prepared from a wide variety of backbone-forming monomers have been prepared. Not all of the monomers represented in Table 23.1 undergo controlled, living anionic polymerization to yield polymers with well-defined structures and low degrees of compositional heterogeneity as outlined in Chapter 4. However, they can often be polymerized to high conversion and are generally suitable as the second or last block-forming monomer since often the lack of a well-defined structure for this block is less important than the ability to incorporate this block into the backbone structure [30].

With respect to the controlled anionic polymerization of acrylates, Structure 2, t-butyl acrylate, Structure 4, is the monomer that has been investigated most

thoroughly. The t-butyl group minimizes side reactions involving either the initiating anionic species or the growing ester enolate anions with the ester groups in both the acrylate and the methacrylate monomers. In general, it has proven difficult to find anionic polymerization conditions for controlled polymerization of acrylate monomers because they are even more reactive than methacrylate monomers, and also because the resulting poly(alkyl acrylates) have enolizable hydrogens along the polymer backbone. These enolizable hydrogens can react with the propagating chain-end ester enolate anions to form the corresponding in-chain ester enolate anions, as shown in Equation 23.1; this will correspond to

(23.1)

either termination or chain transfer reactions depending on whether or not the resulting ester enolate anion reacts with monomer to initiate polymerization and branching.

A few examples of polymerizable monomers corresponding to the general



structure, 3, have been reported. For the esters of α-substituted acrylates (3, R3 = H), polymerizations are limited by the ceiling temperatures for these monomers [31]. For example, the yield of polymer decreases with increasing temperature for the anionic polymerization of methyl α-ethylacrylate (5); no polymer was obtained at temperatures of 30°C or higher [32]. Similar behavior is reported for

the anionic polymerization of methyl α-propylacrylate [33]. The ceiling temperature for methyl α-phenylacrylate (6) is -8°C for neat monomer [34]; thus, it is reported that no polymer is formed above 0°C [35]. The anionic polymerization of tert-alkyl crotonates (e.g., t-butyl crotonate, Structure 7) has been reported [36];

however, the molecular weight distributions tended to be broad and chain transfer to monomer is a likely side reaction because of the acidic, allylic and enolic δ-hydrogens [37].

C. Initiators

The proper choice of initiator for anionic polymerization of polar vinyl monomers is of critical importance to obtain polymers with predictable, well-defined structures. In addition to the desired Michael-type addition reaction of the initiator with a monomer such as methyl methacrylate (Eq. 23.2), the initiator can also

(23.2)

react with the ester carbonyl group to form the corresponding ketone derivative as shown in Equation 23.3 (M corresponds to the counterion). The competition

(23.3)



between these two modes of addition depends on the general reactivity of the initiator and also the steric requirements of the initiator. As a general rule, an initiator should have approximately the same reactivity/stability as the propagating chain end carbanionic species. One measure of the stability of a carbanion is the pKa of the corresponding conjugate acid (see Chap. 2). The pKa of ethyl acetate (estimated: 30–31 in dimethylsulfoxide [DMSO]; 27–28 in H2O) [38] can be used as a model for the pKa of the conjugate acid of the ester enolate anion in anionic polymerization of alkyl methacrylates. The pKa values of carbon acids of interest with respect to initiators are listed in Table 23.2.

The results of studies of alkyllithium initiators are instructive as an example of inappropriate initiators for methyl methacrylate (MMA) polymerization because they are too reactive [40]. The reaction of MMA with n-butyllithium in toluene at -78°C produces approximately 51% of lithium methoxide by attack at the carbonyl carbon (see Eq. 23.3). Furthermore, both a polymer and an oligomer fraction are formed, both of which have butyl group initiator fragments at their chain ends; however, both fractions also contain approximately 0.7–1.0 additional butyl group fragments corresponding to C4H9CO— from the incorporation of butyl isopropenyl ketone also produced by attack at the ester carbonyl group (see Eq. 23.3). Independent studies indicated that although the butyl isopropenyl ketone is more reactive than MMA and is incorporated early in the polymerization, the resulting chain end is more stable and less reactive toward addition of MMA. It is clear that the consumption of 50% of the initiator by reaction at the ester carbonyl group is not acceptable.

The most generally useful initiator for anionic polymerization of MMA and related compounds is 1,1-diphenylhexyllithium (8), which is formed by the quantitative and facile addition of butyllithium with 1,1-diphenylethylene (DPE) as shown in Equation 23.4 [41,42]. The corresponding 1,1-diphenylalkylpotassium

(23.4)

chain end formed by addition of poly(styryl)potassium to DPE was shown to be an efficient initiator for MMA polymerization at -78°C in THF by Rempp and co-workers [43]; the resulting block copolymers exhibited narrow molecular weight distributions (Mw/Mn = 1.05–1.1). Lithium methoxide formation is reduced to 17%, corresponding to ester carbonyl addition, when 1,1-diphenylhexyllithium (Structure 8) is used in place of butyllithium (51% LiOCH3 formation) [40]. It should be noted that polymerizations of MMA in toluene are more complex than in THF; broad and often multimodal molecular weight distributions indicative of a multiplicity of active propagating species are typically obtained in toluene



[2,4,44]. The usefulness of 1,1-diphenylalkyllithium initiators to polymerize MMA efficiently with minimal attack at the ester carbonyl groups is in accord with the pKa of diphenylmethane (32), the conjugate acid of this carbanion, which is approximately the same as that of the propagating ester enolate anion (30,31), as shown in Table 23.2. Of course, the increased steric requirements of the diphenyl-alkyllithium initiator also reduce the rate of addition to the ester carbonyl group.

An interesting application of 1,1-diphenylalkyllithium initiators is the use of the corresponding bis(trimethylsilyl) derivative of 1,1-diphenylethylene [45]; the 18 trimethylsilyl protons were used to estimate the molecular weight of the resulting PMMA polymers. Functionalized diphenylalkyllithium initiators are useful for preparation of α-functionalized PMMA as well as α,ω-telechelic polymers by coupling reactions. Thus, the reaction of ethyl 3-lithiopropyl acetaldehyde ethyl acetal with 1,1-diphenylethylene provides an initiator with a protected hydroxyl functionality (9) (see Chap. 11) as shown in Equation 23.5 [7,14]. A

(23.5)

general procedure for the preparation of functionalized PMMA is based on the use of substituted 1,1-diphenylethylenes as described in Chapter 11. For example, the reaction of sec-butyllithium with 1-(4-dimethylaminophenyl)-1-phenylethylene forms the corresponding amine-functionalized initiator (10), (as shown in Eq. 23.6) [46].

The α-methylstyryl anion is a surprisingly effective initiating species for MMA polymerization. On the basis of pKa (43 for toluene; see Table 23.2), it



(23.6)

might be expected that this species would be too reactive and consequently unselective with respect to Michael addition, compared with ester carbonyl addition, as is observed for the styryl anion [43]. However, oligomeric α-methylstyryllithium functions as a reasonably efficient initiator for MMA polymerization at -78°C in THF [12].

Ester enolate anions should be the ideal initiators for alkyl methacrylate polymerizations. Since rates of initiation and propagation should be essentially the same for ester enolate anion initiators, polymers with narrow molecular weight distributions should be formed (see Chap. 5). Lochmann and co-workers [47] have developed procedures for the preparation of lithium ester enolates, as shown in Equation 23.7. Except for the methyl and t-butyl lithioisobutyrates, the lithium

(23.7)

ester enolates were fairly soluble in aromatic hydrocarbons. In hydrocarbon media, these lithium ester enolates are highly associated; for example, the degree of aggregation of ethyl α-lithioisobutyrate was 6 in benzene [48]. Although the ester enolates were generally very soluble in THF, they exhibited limited stability at room temperature because of self-condensation reactions. Even in THF, the lithium ester enolates were highly associated; for example the degrees of aggregation of both methyl and ethyl α-lithioisobutyrate are 3.5 in THF [48]. When these ester enolate anions were used to initiate polymerization of MMA at room temperature in toluene, only approximately 40% monomer conversion was achieved and the molecular weight distributions tended to be broad [49]. Essentially quantitative conversion was obtained with these ester enolates in the presence of added sodium or potassium t-butoxides.

Fluorenyl anions (pKa = 22.6, Table 23.2) are somewhat more stable and less



reactive than the diphenylalkyl anions; however, they have been used to initiate the anionic polymerization of MMA [50]. Fluorenyl anions are readily prepared by reaction with organoalkali compounds as shown in Equation 23.8. It should be

(23.8)

noted that carbanion salts of 9-methylfluorene are more useful than those of fluorene, because after initiation the fluorenyl residue still retains an acidic hydrogen that can participate in chain transfer reactions [50,51]. Because of the decreased reactivity of 9-methylfluorenyl anions, their rates of initiation are much slower than the rates of MMA propagation (e.g., kp/ki 200 at -72°C) [50]; therefore, initial molecular weight distributions were reported to be broad (MwIMn 1.35), but narrow molecular weight distributions were observed at higher conversions [50].

A new and unexpected class of initiators for acrylate and methacrylate polymerizations has been reported. These are described as metal-free initiators, of which a prototypical example is the tetrabutylammonium salt of dimethyl malonate as shown in Equation 23.9 [52–54]. It is reported that this initiator can

(23.9)

polymerize butyl acrylate in high yield (> 95% conversion) and with molecular weight distributions as low as MwIMn = 1.16 [52]. In view of the relatively acidic nature of these compounds (pKa 16; see Table 23.2), the ability of these carbanions to initiate acrylate polymerization is quite surprising. The carbanion from 2-nitropropane (Eq. 23.10) was likewise reported to be optimal for the

(23.10)

polymerization of methyl methacrylate, providing an almost narrow molecular weight distribution (MwIMn = 1.17) [52]. A P4-phosphazene base (11) has been reported to react with ethyl acetate to form an ester enolate with a soft, highly delocalized counterion (12), as shown in Equation 23.11 [54]. Using this initiator, PMMA was prepared in THF at 60°C with MwIMn = 1.11 [54].

Alkoxide salts have been reported to initiate polymerization of methyl methacrylate with varying degrees of efficiency which seems to depend on the counterion and solvent. As indicated in Table 23.2, alcohols in dipolar aprotic solvents have approximately the same acidity (pKa = 29) as the enolic hydrogens of esters. For example, in toluene at 20°C, although no polymer is formed in the presence of



(23.11)

either lithium or sodium t-butoxide, a 6.5% yield of PMMA is obtained with potassium t-butoxide [49]. The potassium salt of the ω-alkoxide from poly-(ethylene oxide) was reported to polymerize methyl methacrylate with high efficiency (> 92% yield) to form the corresponding block copolymer at room temperature in THF [55]. The [222]-cryptand complexes of both potassium methoxide and potassium t-butoxide are reported to polymerize methyl methacrylate in high yield over a wide temperature range [56]. Unfortunately the molecular weight distributions were all quite broad (Mw/Mn > 1.36) in all cases. Similar results were obtained with the corresponding crown ether complexes [57].

Organomagnesium compounds as initiators for polymerization of methyl methacrylate and related compounds have been investigated in considerable detail [24,58]. Unfortunately, although it is possible to control stereochemistry and obtain either highly isotactic or highly syndiotactic PMMA using organomagnesium initiators, the mechanisms of these polymerizations are very complex. Part of the complexity arises from the Schlenk equilibrium between various species in a typical Grignard reagent, as shown in Equation 23.12. Thus, these systems

(23.12)

have multiple active species with different reactivities and stereospecificities. As a consequence, the molecular weight is often not controllable and the molecular weight distributions tend to be broad or multimodal [24]. Thus, for polymerizations in toluene at -78°C initiated by n-BuMgBr, i-BuMgBr, sec-BuMgBr, and t-BuMgBr, the polymer yields and molecular weight distributions are 14% (Mw/Mn = 11.2), 33% (Mw/Mn = 2.29), 100% (MwMn = 1.29), and 100% (Mw/Mn = 1.18), respectively [59]. Narrow-molecular-weight distribution PMMA can be obtained with t-BuMgBr prepared in diethyl ether, although the reaction times are days under these conditions. For this Grignard initiator, the presence of an excess of MgBr2 drives the Schlenk equilibrium (Eq. 23.12) to the right so that there is only one active propagating species, t-BuMgBr [16,58].



The “ate” complexes of t-butyllithium with trialkylaluminum compounds such as triethyl-, tributyl- and trioctylaluminum ([A1]/[Li] > 3) are effective initiators for methyl methacrylate in toluene at -78°C [60]. These “ate“ complexes (see Chap. 6) provide narrow molecular weight distributions and high conversions in a period of 24 h.

Other “ate” complexes (see Scheme 23.1) have been developed for MMA polymerization that are effective at ambient temperatures in toluene [3,61]. The

Scheme 23.1

structure of the initiation species (13) is only a schematic representation. End-group analysis showed that only the t-butyl groups are present at the PMMA chain ends; however, only approximately 70% of the t-butyl groups initiate polymer chains. It was proposed that a dimeric species (higher oligomer of 13) is also formed, but that only one of its two t-butyl groups can initiate polymerization. Relatively narrow molecular weight distribution (Mw/Mn > 1.09) polymers were reported for polymerizations effected at 0°C and above [61].

A variety of substituted aluminum porphyrins (14) initiate the controlled, living polymerization of methacrylates and acrylates [3,62–65]. Using the methyl-substituted aluminum porphyrin (X = -CH3), the polymerization of MMA required irradiation with a xenon arc lamp to initiate polymerization; however, PMMA could be prepared with controlled molecular weight and relatively narrow molecular weight distribution (Mw/Mn = 1.06–1.20) [62,63]. Propagation via a (porphinato)aluminum enolate was proposed based on 1H nuclear magnetic



resonance (NMR) analysis [62]. Using the propanethiolate-substituted aluminum porphyrin (Structure 14, X = -SCH2CH2CH3), the quantitative polymerization of t-butyl acrylate in methylene chloride could be effected rapidly (0.5 h) in diffuse light over a wide temperature range (-90 to 20°C) to give polymers with somewhat broad molecular weight distributions (Mw/Mn = 1.13–1.48) [64]. The propanethiolate-substituted aluminum porphyrin (14, X = -SCH2CH2CH3) initiated the polymerization of MMA in the absence of irradiation [65]. In the presence of the hindered, bulky “Lewis acid” (15), the polymerization of MMA initiated by

the thiolate-substituted aluminum porphyrin was accelerated (t < 90 s) to give controlled molecular weights and somewhat narrow molecular weight distributions (Mw/Mn = 1.1–1.2) [65].

D. Kinetics of Polymerization

The kinetics of the anionic polymerization of alkyl methacrylates, in general, and methyl methacrylate, in particular, are complicated by the presence of a multiplicity of active species, each of which can participate in propagation with monomer under certain conditions, as shown by the general Winstein spectrum of active species (see Chap. 3) in Scheme 23.2. In general, polar solvents tend to shift the equilibria in Scheme 23.2 to the right (i.e., to lower degrees of aggregation or towards formation of more dissociated species) (see Chap. 2). In nonpolar media the formation of aggregates with varying degrees of association is generally observed, as shown in Table 23.3 for lithium ester enolates and related compounds. As



Scheme 23.2

observed for alkyllithium compounds, the degree of aggregation is sensitive to the steric environment around the coordinating metal center, and lower degrees of association are observed in THF vs. benzene. It is important to note that even for the sterically hindered lithium salt of t-butyl isobutyrate (entry 3 in Table 23.3), the degree of association is 2.3 in THF and 1.7 for the dimeric analog (entry 5 in Table 23.3). In addition the lithium salt of methyl 3,3-dimethylpropionate is associated into dimers even in THF at -108°C [66]. These results attest to the strength and importance of chain end association of lithium ester enolates.

Kinetics in a Nonpolar Solvent: Toluene

The alkyllithium-initiated polymerization of methyl methacrylates in toluene is of interest because these polymerizations provide PMMA with highly isotactic struc-



tures [2,24,67]. Unfortunately, these reactions are also characterized by broad or multimodal molecular weight distributions and termination reactions even at low temperatures. Using an appropriate initiator such as the adduct of butyllithium with 1,1-diphenylethylene (i.e., 1,1-diphenylhexyllithium [DPHLi]), side reactions of the initiator with the monomer can be minimized, as discussed in Sec. C. For a mechanism in which the concentration of active centers, [PLi], is constant and for a propagation step that is first order in monomer, a linear plot of 1n [MMA]o/[MMA] vs. time should be observed in accord with Equations 23.13 and 23.14. Plots of monomer concentration as a function of time for DPHLi-initiated

–d[MMA]/dt = kp[PLi][MMA] (23.13)

(23.14)

MMA polymerization in toluene are shown in Figure 23.1 for the data in Table 23.4 [42]. In spite of the expected complexities for this system with respect to

Figure 23.1 Rates of methyl methacrylate (MMA) consumption at three

polymerization temperatures. — —, initial [MMA]; , 70% conversion to polymer; run numbers refer to reactions conditions

listed in Table 23.4. A, run 1;

run 6; B, run 5; C, run 3; run 2; D, run 4. (From Ref. 42; reprinted by permission of the Royal Society of Chemistry.)



Table 23.4 Experimental Data for the Kinetics of DPHLi-Initiated MMA Polymerizationa

Run

[DPHLi]

(mole/L) × 103

Temperature (°

C)

Kobs

(Eq. 23.14) (min-1 × 102)

1 3.2 -30 11.1

2 1.6 -30 5.6

3 1.6 -30 5.5

4 0.8 -30 2.8

5 2.4 -80 0.29

6 2.4 0 13.0

a [MMA] = 0.125 mole/L for all runs.

Source: Ref. 42.

chain-end association, the results shown in Figure 23.1 and in Table 23.4 indicate that these polymerizations exhibit linear first-order plots for monomer conversion (up to 70% conversion) as a function of time, and apparent first-order dependence on chain end concentration, even at 0°C. Independent analysis indicated that approximately 10% of the initiator was consumed by attack at the ester carbonyl group to form lithium methoxide, which is also in accord with the data of Hatada and co-workers [40] previously discussed in the section on initiators. Furthermore, a petroleum ether-soluble fraction of low-molecular-weight material is formed initially but does not increase in yield or molecular weight with conversion (Mn 800–900 g/mol). For the main product, which corresponds to the petroleum ether-insoluble fraction (85% isotactic triads), a very broad molecular weight distribution was calculated (Mw/Mn 35).

The linear dependence displayed in Figure 23.1 and in accord with Equation 23.14 is a necessary but not sufficient condition for a living polymerization, which proceeds in the absence of chain termination and chain transfer (see Chap. 4). However, the fact that the amount of lithium methoxide increased with time for polymerizations at -30°C and above indicates that this is not necessarily a sensitive probe for detecting termination [42].

The surprising aspect of these results is that in spite of the expected aggregation of the chain ends into highly associated species (see Table 23.3), no concentration dependence of kp in chain end concentration was reported [42]. If the unreactive, associated species is the predominant form in solution and only the unassociated species reacts with monomer, the rate of polymerization would be expected to exhibit a fractional order dependence on chain end concentration (1/n) for a degree of association of n (see Chap. 7). The observation of a first-order dependence on chain end concentration (over only a fourfold variation in initiator concentration, however [42]) is not consistent with an unreactive, predominantly



aggregated form of the chain ends in solution. A possible explanation for this result is the fact that lithium methoxide is formed initially and during the polymerization. The data in Table 23.3 indicate that the degree of aggregation of the chain ends is decreased by cross-association with lithium alkoxide. If this cross-associated species is the predominant form in solution and is capable of reacting directly with monomer, a first-order dependence on active chain-end concentration would be expected. Wiles and Bywater [68] examined the effect of added lithium methoxide and reported that this alkoxide accelerated the rate of polymerization.

In contrast to the complexities associated with the anionic polymerization of methyl methacrylate in nonpolar solvents, the polymerization of t-butyl methacrylate in toluene with lithium as counterion can be used to prepare polymers with well-defined structures [2,4,69]. In accord with Equation 23.14, the kinetics exhibited a linear dependence of 1n [M]o/[M] with time, which is consistent with a first-order dependence on monomer concentration; this dependence is also consistent with the absence of significant termination or transfer reactions because the linearity implies that [PLi] is constant (see Chap. 4) [4]. However, this conclusion is mitigated by the fact that analogous linear plots were obtained for methyl methacrylate, in which independent monitoring of lithium methoxide generation indicated that significant ester carbonyl reactions (inter or intramolecular) were occurring [42]. A linear plot of Xn vs. conversion was also used to indicate the absence of chain termination or chain transfer reactions [4]; however, this linear plot is not sensitive to termination reactions (see Chap. 4) [70]. It is reported that the rate of polymerization of t-butyl methacrylate in toluene is first order with respect to initiator [2]. Although bimodal molecular weight distributions were obtained at low conversion, at high conversion the peaks merged and “rather narrow” [4] molecular weight distributions were reported (Mw/Mn > 1.15). In conclusion, the kinetics of alkyl methacrylate polymerization in toluene appear to be quite simple (i.e., first order in initiator [active chain ends] and in monomer). However, the high degree of association of model ester enolate anions suggests that these systems are worthy of further investigation over a broad range of active chain end concentration.

Kinetics in Polar Solvents

Both kinetic and conductance studies established that at least two types of species, free ions and ions pairs, are involved in propagation for the anionic polymerization of MMA with alkali metal counterions in THF at T < -75°C [71–74]. The dissociation constants for formation of free ions (schematically represented in Equation 23.15) are listed in Table 23.5. The free ion propagation rate constants are approximately 102–103 times larger than the ion pair rate constants [72].

(23.15)



Therefore, subsequent studies have usually been carried out in the presence of a common ion salt (e.g., cesium triphenylcyanoborate, sodium tetraphenylborate, potassium tetraphenylborate and lithium tetraphenylborate for cesium, sodium, potassium, and lithium ions, respectively) [71,72,74–76]. Under these conditions the half-lives for polymerization are on the order of seconds. These rates for ion pair propagation are approximately 1000 times faster than the corresponding rates of polymerization in toluene [4].

For cumyl cesium as initiator, kinetic data obtained in THF using a flow tube reactor provided plots of 1n [MMA]o/[MMA] vs. time that were linear, consistent with first-order dependence on monomer concentration (see Eq. 23.2) and the absence of significant chain termination reactions [77]. Linear plots of Xn vs. conversion indicated the absence of chain transfer reactions (see Chap. 4) at temperatures up to +20°C [76]. It was implicitly assumed that the propagation reaction was also first order in active center concentration, although the data to support the assumption were limited [77]. The molecular weight distributions were reported to be in the range of 1.01 < Mw/Mn < 1.1.

For oligomeric α-methylstyrylsodium as initiator in THF, similar results were obtained [50]. It was stated that the rate constant (k±) was independent of initiator concentration based on examination of only two concentrations of initiator at -73.3°C; for initiator concentrations of 4.77 × 10-4 M and 2.42 × 10-4 M, the calculated values of k± were 184 L/mol s and 198 L/mol s, respectively. In general, this body of kinetic work focused on the temperature dependence of the rate constants rather than examining a wide range of initiator concentrations. Molecular weight distributions were reported to be quite narrow at -100°C (Mw/Mn = 1.03) and at -51°C (Mw/Mn = 1.05).

A recent study of the dependence of propagation rate constants on Xn using metalloesters (alkali ester enolates) as initiators in THF at -46°C is one of the few



Table 23.6 Effects of Initiator Concentration on kp (±) for Anionic Polymerization of MMA Initiated by Methyl α-sodioisobutyrate in THF at —46°C

103 [initiator] kp (±) 103 [initiator] kp (±)

(mol/L) (L/mol s) (mol/L) (L/mol s)

13.5 100 12.0 210

20.8 100 6.0 260

14.7a 100 1.7 400

20.0 100 0.4a,b 500

10.0a 235

a Kinetics performed in the presence of added sodium tetraphenylborate.

b Ref. 50.

Source: Ref. 75.

kinetic studies that examined a wider range of concentrations[75]. The data are shown in Table 23.6 and indicate that this dependence on initiator concentration is not influenced by the addition of common ion salt. It was assumed that this concentration dependence was probably the result of association of ion pairs into dimeric associates, as shown in Equation 23.16, although no definitive quantitative analysis with respect to the equilibrium constant or the importance of propagation by the associated species could be determined from the data. The ion pair rate constant (k±) was estimated to be 5.50 × 102 L/mol s at -46°C in THF from the extrapolation of the dependence of kp (Eqs. 23.13,23.14) on chain end concentration.

(23.16)

It has been reported that the polymerization of MMA using diphenylmethyllithium or diphenylmethylpotassium proceeds in an ideal manner within the temperature range (°C) of -101 to -45[76]. It was stated that first-order plots for the conversion of monomer were linear. Molecular weight distributions were narrower for potassium than for lithium (Mw/Mn > 1.1). A recent kinetic study of the dependence of monomer addition rate on chain length concluded that the observed propagation constant (Xn > 5) was independent of chain end concentration, which is surprising in view of the known aggregation of model lithium ester enolates shown in Table 23.3. Recent kinetic studies at 25°C were effected at higher concentrations of initiator (i.e., 0.05, 0.1, and 0.2 mol/L [78]). At these concentrations, the observed rate constant, kobs (Eq. 23.14), exhibited a linear dependence on the inverse of the square root of the concentration of active chain ends, in accord with the mechanism shown in Scheme 23.3 and the kinetic equations



Scheme 23.3

shown in Equations 23.17–23.20. The calculated rate constant for the associated species, ka, was calculated to be 120 L/mol s and the equilibrium constant of

(23.17)

(23.18)

(23.19)

(23.20)

dissociation, Kd, was calculated to be 10-3 mol/L. The ion pair rate constant, k±, was taken from earlier kinetic studies at low concentrations and had a value of 1.5 × 103 L/mol s. Even though the concentration of unassociated ion pairs corresponds to only 10% of the active chain ends, these species account for approximately 74% of the monomer consumption. The remainder of the propagation occurs through the associated (dimeric) ion pairs. Kinetic studies at -65°C indicate that there are approximately equal amounts of associated and unassociated species; however, it was stated that the kinetic contribution of the associated species was negligible[79]. The dissociation equilibrium constant was calculated to be 0.5 × 10-3 mol/l at -65°C; this corresponds to a fraction of unassociated ion pairs equal to 0.5.

Representative kinetic data for anionic MMA propagation by ion pairs are shown in Table 23.7. The general order of reactivity is Cs K Na > > Li. This order is observed in both THF and glyme and is maintained over a wide range of temperatures. It is interesting to note that this same order of reactivity as a function of counterion is observed in the anionic polymerization of styrene in hydrocarbon solvents, although the differences between counterions were slight [82]. In general, one would expect that for the equilibrium between contact and solvent-separated ion pairs (Eq. 23.21) the formation of a higher fraction of solvent-separated ion pairs would be favored for the smaller cation, lithium (see Chap. 3). Furthermore, since more dissociated species are more reactive, in general, it



(23.21)

would be expected that if solvent-separated ion pairs were involved, faster rates of polymerization would be observed for lithium vs. cesium cations. This order (Li+ > Cs+) is observed in THF for anionic polymerization of styrene [83,84]. The fact that this is not the observed order for MMA polymerization in THF is consistent with the conclusion that contact ion pairs are the predominant propagating species in the anionic polymerization of MMA using alkali metal counterions [85]. Several factors may be responsible for the difference between MMA and styrene. The styryl anion is a more delocalized anion and therefore it would interact less strongly with the counterion, which would promote dissociation and formation of solvent-separated ion pairs. The ester enolate anion has the negative charge located primarily on oxygen and this would tend to promote more electrostatic interaction with the counterion.

The data for t-butyl methacrylate (TBMA) in Table 23.7 are also consistent with the data and conclusions regarding the active propagating species in these anionic polymerizations in THF [80]. Thus, the rate constant for propagation is higher for cesium than for sodium as counterion. In addition, the rate of propagation is slower for TBMA than for MMA, by a factor of approximately 10.

The temperature dependence of the ion-pair propagation rate constants (k±) has been investigated in detail. Linear Arrhenius plots have been observed for all counterions. This result is consistent with the involvement of only one type of ion pair, since if the equilibrium between contact and solvent-separated ions pairs were involved one might expect non-Arrhenius-type temperature dependence, as was observed for the ionic polymerization of styrene in THF [83,84].

E. Effects of Lewis Base and Salt Additives

One of the most dramatic developments in the anionic polymerization of acrylate and methacrylate monomers was the discovery by Teyssie and co-workers [86] that by addition of lithium chloride it was possible to effect the controlled polymerization of t-butyl acrylate. Thus, using oligomeric (α-methylstyryl)-lithium as initiator in THF at -78°C, the molecular weight distribution (Mw/Mn) of the polymer was 3.61 in the absence of lithium chloride but 1.2 in the presence of lithium chloride ([LiCl]/[RLi] = 5), as illustrated in Figure 23.2. In the presence of 10 equivalents of LiCl, t-butyl acrylate was polymerized with 100% conversion and 95% initiator efficiency to provide a polymer with quite narrow molecular weight distribution (Mw/Mn = 1.05) [12]. The advantages of addition of lithium chloride were also indicated for analogous studies with methyl methacrylate under



Table 23.7 Ion Pair Propagation Rate Constants for Methyl Methacrylate and t-Butyl Methacrylate (**)a Polymerization

Counterion

Solvent

Temp. °C Chain end concentration (mol/L × 10-3)

Kp± (L/mol s)

References

Sodium THF -40 800 4

** -47.7 0.57 31.4a 80

-51 0.405 444 50

-99 0.44 31.8 50

** -101.3 0.46 0.203a 80

Glymeb -40 2550 4

-54.5 1.2 1930 82

Na[222] c THF -98 270 74

Cesium THF -40 860 4

** -41.8 0.73 194a 80

-67.4 0.15 374 77

** -91 0.65 7.25a 80

-97.3 .53 32.7 77

Glymeb -40 2680 4

-51 0.45 1730 81

Lithium THF 25 50–200 1500 78

-40 100 4

-46 0.78–27.2 70 75

-65 0.1–2 46 79

Glymeb -40 16 4

Potassium THF -40 750 4

at-Butyl methacrylate kinetic data are indicated with asterisks (**).

b1,2-dimethyoxyethane.

cSodium complexed with heterobicyclic cryptand [222].

the same conditions and with addition of undiluted MMA [87]. In the absence of lithium chloride the molecular weight distribution was 1.20 and the initiator efficiency was only 66%; however, in



Figure 23.2 Influence of LiCl on the MW

distribution as seen in gel permeation chromatograms: (a) no salt, Mw/Mn = 3.61;

(b) [LiCl]/[RLi] = 5.02, Mw/

Mn = 1.20. Conditions:

(a) [RLi] = 6.6 × 10-3 mol/L; (b) [RLi] = 6.6 × 10-3 mol/L and [LiCl]

= 33.46 × 10-3 mol/L in THF at -78°C. (From Ref.

86; reprinted by permission of the American Chemical Society.)

compounds it was found that alkali metal alkoxides reduce the rate of self-condensation of lithium ester enolates and the rate of intramolecular cyclization (the principal termination reaction) relative to propagation reactions [92,93]. Furthermore, the addition of lithium t-butoxide (3 equivalents) also decreased the rate of MMA polymerization initiated by methyl α-lithioisobutyrate in THF at 20°C [93].

A combination of spectroscopic and kinetic studies has helped us to understand the effects of added lithium chloride and alkoxides that promote controlled anionic polymerization of acrylates and methacrylates. Based on the known tetrameric degree of association of lithium ester enolates, as shown in Table 23.3, the observation of the temperature dependence and multiplicity of peaks in the 1H and 7Li NMR spectra of methyl α-lithioisobutyrate (MiBLi) in THF (0.05–2.0



M) was interpreted in terms of a temperature- and concentration-dependent equilibrium between tetramers (7Li NMR; δ = 0.52 ppm) and dimers (7Li NMR; δ = -0.71 ppm) as shown in Equation 23.22 [94]. It was concluded that this equilib

(23.22)

rium corresponds to a slow exchange process. This equilibrium shifts to the right, favoring dimer, when either the temperature or concentration is lowered. Upon addition of LiCl, the 7Li NMR spectra exhibit one peak at δ = -0.52 ppm, which was assigned to the tetramer, and a new peak at δ -0.43 ppm when the ratio (R) [LiCl]/[MiBLi] = 0.5, which was assigned to a mixed complex between LiCl and the lithium ester enolate [95]. The observation of two well-resolved signals was interpreted in terms of a slow exchange between these two species. The peak corresponding to the tetramer is not observed when the ratio, R, [LiCl]/[MiBLi] > 1. The disappearance of the peak corresponding to dimer when R < 1 suggested that lithium chloride complexes preferentially with the dimer. It is noteworthy that the peak corresponding to the “mixed complex” shifts downfield with increasing values of R. The “mixed complex” was assigned the structure of a mixed dimer LiCl/MiBLi. When R > 2, the temperature dependence of the spectra were interpreted in terms of formation of mixed trimers, (LiCl)2MiBLi, and tetramers, (LiCL)3/MiBLi. In the 13C NMR spectra, the α-carbon resonance changes from δ 73.5 ppm to δ = 66.5 for the mixed complex when R = 2. This upfield shift is consistent with increasing negative charge on the alpha carbon in the mixed complex with lithium chloride relative to the tetrameric aggregate. The equilibria that have been postulated for these systems are shown in Scheme 23.4.

Scheme 23.4

13C NMR studies of the mixed complexation of methyl α-lithioisobutyrate (tetrameric) with lithium tert-butoxide (tetrameric) in THF were interpreted in terms of the equilibria involving mixed tetramers as shown in Scheme 23.5 [96]. Multiple resonances were observed both for the tertiary carbon in lithium

Scheme 23.5



t-butoxide (three mixed complexes at δ = 66.3, 66.4, and 66.5 ppm) and for the “carbonyl carbon,” α-carbon and (CH3)2 carbons in methyl α-lithioisobutyrate. The α-carbon resonance shifted upfield by 1.5 ppm upon complexation with lithium tert-butoxide. Analogous to the observed effects of lithium chloride [95], this was interpreted in terms of an increase in negative charge density at the Cα in the complex. It was also concluded from the temperature dependence of the 13C NMR spectra of the complexes that the exchange between the various mixed complexes was slow on the 13C NMR time scale. This is in contrast to the situation with lithium chloride in which only one type of predominant LiCl complex was observed.

Kinetic studies of the lithium ester enolate-initiated polymerization of t-butyl acrylate in THF have provided insight into the role of added lithium chloride and lithium t-butoxide in promoting controlled polymerizations [97–99]. One problem in obtaining polyacrylates with controlled structures is the fact that these anionic polymerizations are extremely fast. In THF at 23°C using t-butyl α-lithioisobutyrate as initiator, the half-lives for polymerization of t-butyl acrylate are 0.01 s without any additives, 0.05 s for added LiCl (R = 3), and 3 s for t-BuOLi (R = 3) as additive [97]. In the absence of additives, the rate of polymerization is so fast that uniform mixing of initiator and monomer is a problem. First-order time-conversion plots for these systems are shown in Figure 23.3. A summary of the rate constants, initiator efficiencies, and molecular weight characterization

Figure 23.3 First-order time-conversion plots of the anionic polymerization

of t-butyl acrylate initiated by t-butyl α-lithioisobutyrate (BLiB). , Without additives; , with LiCl ([LiCl]/[BLiB] = 3); , with

lithium t-butoxide ([LiOBu]/[BLiB] = 3). (From Ref. 97; reprinted by permission of Hüthig and Wepf Verlag.



Table 23.8 Rate Constants, Initiator Efficiencies, and Molecular Weight Characterization Data for the t-Butyl α-Lithioisobutyrate-Initiated Polymerization of t-Butyl Acrylate in THF at 20°C

Parameter No additive LiCla t-BuOLia

kapp,o (s-1)b 48.0 14.4 0.23

10-3kp (L/mol. s) 42.5 11.4 0.17

kt (sec-1)c 0.75 0.75 0.25

103 (kt/kp) (mol/L) 0.02 0.07 0.68

kt (s-1)d 6.0 2.8 –

103 (kt/kp) (mol/L) 0.14 0.25 –

Xne 177 158 146

103 [P*] (mol/L) 1.13 1.27 1.37

ff 0.28 0.32 0.34

Mw/Mn (crude polymer) 1.35 1.08 1.77

aR = 3 ([additive]/[RLi]).

bInitial slope of the first-order time-conversion plot: kapp,o = kp [P*]o.

cDetermined from the fraction of β-oxoester end groups by means of UV spectroscopy at 256 nm (see [100]).

dDetermined from the curvature of the monomer time-conversion plots.

eAt full monomer conversion.

fInitiator efficiency.

data for these systems is shown in Table 23.8. One important point with respect to this data is the fact that there is poor agreement between the two methods used to estimate the termination rate constant. In the first method, termination is assumed to be unimolecular and associated with an intramolecular condensation reaction to form a six-membered ring β-ketoester as shown in Equation 23.23. In the second

method, a first-order rate constant for termination was estimated from the curvature of the time-



and the reason that both of these estimates of termination may be flawed is that they both neglect bimolecular termination by transfer of an enolic hydrogen to a growing anionic center as shown in Equation 23.24. The consequences of this

(23.24)

reaction would depend on whether the resulting in-chain ester enolate can add monomer in competition with the less-hindered active chain ends. Regardless of this aspect of the termination process, several very important conclusions can be drawn from the data in Table 23.8. The major kinetic effects of the addition of either lithium chloride or lithium t-butoxide are to decrease the rate of propagation as shown by the first two rows in Table 23.8. The decrease is approximately a factor of 200 for lithium t-butoxide and a factor of three for lithium chloride. Quite surprising was that lithium chloride only reduces the rate of termination by a factor of two, and lithium t-butoxide only decreases kt by a factor of three. Neither of these effects, changes in propagation and/or termination rates, provides a satisfactory explanation for the dramatic effects of lithium chloride to provide a controlled polymerization of t-butyl acrylate. The most dramatic observed effects of these additives were on the molecular weight distributions, i.e., Mw/Mn = 1.35, 1.08 and 1.77 without additive, with LiCl and in the presence of t-BuOLi, respectively, when R = 3. It should be noted that these results were obtained in a flow tube reactor. Broader molecular weight distributions were obtained in a tank reactor or using ampoules [97].

A self-consistent explanation for the effects of lithium chloride on the controlled polymerization of t-butyl acrylate and alkyl methacrylates is provided by the assumption that the reason for the broad molecular weight distributions is that there are two or more species reacting with monomer at different rates and that their interconversion is slow compared to the rate of monomer addition [98,100]. First it is important to remember that lithium ester enolates are associated in solution into dimers and tetramers as shown by the data in Table 23.3. As discussed in the previous section, 7Li and 13C NMR studies by Teyssie and co-



workers [94] for methyl α-lithioisobutyrate were interpreted in terms of an aggregation equilibrium, presumably between tetramers and dimers, in THF with slow exchange between the two species. Thus, if addition of monomer is fast for the less associated (or unassociated [99]) ester enolate chain ends and the rate of exchange between these two species is slow, a broad polydispersity will result. Analogous studies in the presence of lithium chloride established that mixed complexes are formed between lithium chloride and methyl α-lithio-isobutyrate (see Scheme 23.4) [95]. Only one peak is observed in the 7Li NMR at values of R > 1, which does shift with R, however. Thus, although there is evidence from 13C NMR for several mixed species with lithium chloride, the NMR results are consistent with relatively rapid interconversion between them. Thus, lithium chloride slows down the rate of polymerization and converts the chain end either to one type of mixed aggregate or to several mixed aggregates that interconvert competitively with chain propagation to provide a relatively narrow molecular weight distribution. Finally, the kinetic and NMR data also provide an explanation for the fact that lithium t-butoxide addition does not lead to narrow molecular weight distributions. The interaction of lithium t-butoxide leads to the formation of several mixed complexes (see Scheme 23.5) whose rate of interconversion was very slow even at 0°C [96]. However, Dvoranek and Vlcek [101] have reported that t-butyl acrylate can be polymerized in a stirred batch reactor at -60°C in a mixture of toluene/THF (9/1, vol/vol) to yield narrower molecular weight distributions (Mw/Mn = 1.21) in the presence of 10 equivalents of lithium t-butoxide relative to the initiator, t-butyl α-lithioisobutyrate. An even narrower distribution polymer (Mw/Mn = 1.12) was obtained with less THF (toluene/THF = 19/1, vol/vol).

Complexation of the alkali metal counterion by crown ethers has also been reported to be effective in promoting the controlled anionic polymerization of methyl methacrylate at higher temperatures [102]. For example, the addition of one equivalent of dicyclohexyl-18-crown-6 to the diphenylmethylsodium-initiated polymerization of MMA in THF at -20°C exhibited an initiator efficiency of 97%, 100% conversion, and a narrow molecular weight distribution (Mw/Mn = 1.04). At 0°C in THF in the presence of two equivalents of crown ether, the conversion was 100%, the initiator efficiency was 96%, and the molecular weight distribution was narrow (Mw/Mn = 1.05). An analogous polymerization in toluene using 2 equivalents of crown ether exhibited an initiator efficiency of 91%, 100% conversion, and a narrow molecular weight distribution (Mw/Mn = 1.05); even at 0°C, the same narrow distribution was obtained, but the initiator efficiency decreased slightly to 91% and the conversion was 98%. Thus, the addition of crown ether complexing agent is one of the most effective methods of effecting controlled polymerization of methyl methacrylate at temperatures in the range of 0°C. These results are consistent with the effects of initiators discussed previously in this chapter (see Sec. C), that is, controlled polymerization of methacrylates is promoted by large, bulky, noninteracting counterions.



F. Termination Reactions

The anionic polymerizations of acrylates and methacrylates is complicated by the occurrence of termination reactions and other side reactions. In the initiation step, the initiator can react with the monomer at the carbonyl group of the ester that will form the corresponding ketone and alkoxide as shown in Equation 23.3. This side

(23.3)

reaction, which limits the ability to control molecular weight precisely by consuming initiator without forming a growing chain, can be minimized by using appropriate initiators that are less reactive and more sterically hindered (see Eq. 23.4), as discussed previously in Sec. C. In addition, this side reaction is less important for monomers that have more sterically hindered ester alkyl groups such as t-butyl methacrylate and t-butyl acrylate. This same type of reaction can occur, in principle, between the growing ester enolate anion at the chain end and monomer in competition with monomer addition, as shown in Equation 23.25. Although this

(23.25)

type of reaction was postulated to be the exclusive mode of termination [103], it has been concluded recently that there is no evidence for this [2,50].

Another possible side reaction is the intermolecular reaction of the active ester enolate anions at the chain end with ester groups on other polymer chains as shown in Equation 23.26. It was proposed that this intermolecular reaction would

(23.26)



lead to branched polymer of relatively high molecular weight [50]. Because of the absence of high-molecular-weight tailing in the size exclusion chromatographic (SEC) curves, even at 0°C [104], it was concluded that this reaction was not important [2,50]. The crowded steric environment around in-chain ester groups would also lead one to expect that this would not be an important side reaction.

The kinetics of the termination reaction for MMA polymerization initiated by α-methylstyrylsodium have been studied in the presence of a common ion salt in THF; the kinetics were determined via labeling the active ester enolate chain ends by quenching with tritiated acetic acid (CH3CO2T) [50]. The decrease in active chain end concentration with time exhibited a linear first-order dependence on active chain end concentration, and the kinetic data are shown in Table 23.9. These results are consistent with the conclusion that the primary mode of chain termination is associated with a unimolecular back-biting reaction of the ester enolate anion with the penultimate ester group to form a six-membered ring, β-keto ester group at the chain end as shown in Equation 23.27. Although the rate

(23.27)

of unimolecular termination (kt[P-]) is not zero even at low temperatures, it is smaller than the rate of propagation (kp[MMA][P-]) by a factor of approximately 104. The relative importance of this termination reaction relative to propagation increases, however, as the temperature is raised.

The kinetics of termination have been reported to depend on the initial monomer concentration [85,104]. It was observed that only a fraction of the chains become terminated and this fraction increases with increasing initial monomer concentration. When monomer was slowly distilled into the reactor, the formation of “deactivating agent” was suppressed. This suggests that exothermicity may contribute to formation of side products, such as those resulting from ester carbonyl addition (see Eq. 23.3) rather than Michael addition to the double bond.

Several general aspects of the termination reaction have been enumerated [104]. Controlled polymerizations can be effected at low temperatures (T < -75°C) in polar solvents. At elevated temperatures or in nonpolar solvents such as toluene, termination reactions become important and limit the ability to prepare polymers with well-defined structures. The least termination is observed in the



Table 23.9 Termination and Propagation Rate Constants for α-Methylstyryl-Initiated Polymerization of MMA

Temp (°C)

[MMA] 0

(mol/L)[PMMA- Na+ ]×104 (mol/L)

Kt

(sec-1)akp (±) b

(L/mol s)

-42.6 0.241 2.17 0.106 —

-51.0 0.243 4.05 0.037 444

-60.6 0.247 4.22 0.013 317

-73.3 0.248 4.77 0.006 184

aFirst-order rate constant for termination as measured by quenching active chain ends with CH3CO2T.

bSecond order propagation rate constants assuming first-order dependence on chain end concentration.

Source: Ref. 50.

most solvating solvents; the amount of termination increases in the order dimethoxyethane < tetrahydrofuran < tetrahydropyran. Using increasing polydispersity as an indicator of the importance of termination reactions [85,103], polydispersity (and presumably termination) increases with increasing size of the counterion [2]. In this context it is important to also note that the additions of lithium chloride [87], lithium t-butoxide [101], and crown ethers [102] promote more controlled polymerization (e.g., narrow molecular weight distributions), although it has been shown that not all of these effects are due to decreased rates of termination (see Table 23.8). For example, lithium t-butoxide does decrease the rate of termination, while lithium chloride functions primarily by decreasing chain end association and promoting interconversion between active chain end sites [99].

G. Stereochemistry of Polymerization

Like the anionic polymerization of dienes (see Chap. 9), the anionic polymerization of alkyl methacrylates, especially methyl methacrylate, is dependent on the counterion, solvent and, to a certain extent, temperature [67,105,106]. Representative data for the stereochemistry of polymerization of alkyl methacrylates obtained in toluene and in polar solvents are provided in Tables 23.10 and 23.11, respectively. In general, the anionic polymerization of alkyl methacrylates in toluene solution with lithium as the counterion is highly isotactic (70–90%) and the isotacticity increases with the steric requirements of the alkyl ester group. Thus, the highest isotactic triad content (99% mm) is obtained with the diphenylmethyl ester; the isotactic triad content for the trityl ester (tiphenylmethyl) is only slightly lower (96% mm). Grignard reagents, in particular t-butylmagnesium bromide or isobutylmagnesium bromide, provide highly isotactic polymers also



Table 23.10 Stereochemistry of Anionic Polymerization of Alkyl Methacrylates in Toluene Solution

Microstructure (triads)

Alkyl ester group

Initiator

Temp. (°C)

mm

mr

rr

Reference

CH3 DPHLia -78 87 10 3 42

86 10 4 40

BuLi 68 19 13 40

t-BuLi/Et3Al 0 10 90 60

C2H5 DPHLia 89 10 1 107

iso-C3H7BuLi 88 6 6 108

tert-C4H9BuLi -70 90 5 5 109

(C6H5)2CH BuLi -78 99 1 0 110

(C6H5)3C BuLi 96 2 2 110

(CH3)3Si BuLi -70 89 8 3 111

Allyl DPHLI -80 90 10 0 112

2,3-epoxypropyl

BuLi -78 33 38 29 113

CH3 t-C4H9MgBr 96.7 3 0.3 114

n-C4H9MgBr 11 15.3 73.7 114

i-C4H9MgBr 92.5 5.4 2.1 114

LiEtBb 20 71 21 8 89

LiEtB/3BuOLic 74 19 7 89

NaEtB/BuONad 33 38 29 89

KEtB/BuOKe 30 49 21 89

t-BuLi/Al(BHT)-(iB) 2 (Structure 23.13)

0 2 26 72 61

a1,1-Diphenylhexyllithium.

bα-Lithio ethyl isobutyrate.

cα-Lithio ethyl isobutyrate plus lithium t-butoxide.



Table 23.11 Stereochemistry of Anionic Polymerization of Alkyl Methacrylates in Polar Solvents

Microstructure (triads, %)

Alkyl ester group

Counterion

Solvent

Temp.(°C)

Mm

Mr

rr

Reference

CH3Li THF -85 1 15 84 76

-45 1 22 77

THP -35 6 32 62 115

DME -57 1 16 83 4

Dioxane 13 10 35 55 115

Na THF -51 4 38 58 116

THP -47 22 52 26 4

DME -55 2 21 77 81

[222], DME

-98 1 20 79 4

K THF -60 9 52 39 76

Dioxane 13 14 56 30 115

Cs THF -53 5 52 42 117

DME -66 3 37 60 81

Free ion DME -98 1 20 79 4

Mga THF -78 0.2 9.6 90.2 106

(CH6H5)2Li THF -78 2 11 87 110

(C6H5)3C 94 4 2 110

(CH3)3C Li -40 12 49 39 4

Na -48 6 65 29 4

Cs -42 4 51 45 4

Radical -55 4 17.5 78.5 106

aPolymerization initiated by n-BuMgBr in the presence of 2 equivalents of TMEDA

radical polymerization of MMA is also highly syndiotactic (78.5% at -55°C). It is important to put the effects of experimental variables on stereochemistry into an energy framework before embarking on a detailed explanation of these effects. As discussed by Pino and Sutter [118],



differences in the stereochemistry of propagation. Thus, a change in free energy of activation difference of only 1.3 kcal/mole can change the stereochemistry from 50/50 = iso/syndio to 90/10 = iso/syndio. Since we have limited tools with which to predict or understand the physical and chemical basis of such factors as solvation, particularly those associated with small energy differences of this order of magnitude, it is prudent to limit phenomenological interpretations of these effects. The fact that free radical polymerization produces approximately 78% syndiotactic diads indicates that the factors responsible for this stereospecificity are not associated with specific solvation of a counterion or interaction of penultimate in-chain monomer units with the counterion.

In THF at -78°C, the presence of up to 10 equivalents of lithium chloride relative to the lithium ester enolate chain end concentration has no significant effect on PMMA tacticity [120]. In contrast, in a 9:1 mixture of toluene and THF at -78°C, added lithium chloride increases the syndiotactic placements (rr) at the expense of isotactic placements (mm) as shown in Table 23.13 [120]. The effects of added lithium chloride in promoting controlled polymerization of MMA are more dramatic than the effects on the stereochemistry of polymerization, as shown by the molecular weight distribution data also listed in Table 23.13.

The importance of control of PMMA stereochemistry can be judged by noting that the glass transition temperature of PMMA depends strongly on the



Table 23.13 Effect of Lithium Chloride on the PMMA Stereochemistry and Molecular Weight Distribution for 1,1-Diphenylhexyllithium (DPHLi)-Initiated Polymerization of MMA in 9:1 Toluene/THF Mixtures at -78°C

PMMA microstructure (triads, %)

[LiCl]/[DPHLi] mm mr rr Mw/Mn

0 17 25 58 2.5a

1.0 6 25 69 1.7

2.0 4 24 73 —

3.0 3 25 72 —

4.0 1.15

aBimodal molecular weight distribution; α-methylstyryllithium initiator; see [87].

Source: Ref. 120.

microstructure [106]. Thus, the measured Tg for 99% mm PMMA is reported to be 50°C, the Tg for PMMA with 96–98% r dyads is 135°C [106]. To obtain a PMMA with higher upper use temperature, compared with radical-initiated polymers, polymers with the highest syndiotactic microstructure are desired.

H. Synthesis of Block Copolymers

As described in Chapter 12, one of the most important applications of living polymerizations is the synthesis of block copolymers by sequential monomer addition. One critical factor to consider for successful synthesis of block copolymers of alkyl methacrylates and related compounds is the relative stability of the propagating ester enolate anion. It is estimated that the pKa value of ethyl acetate is in the range of 30–31 in the dipolar aprotic solvent dimethylsulfoxide (DMSO) (see Table 2.3). Thus, although living polymers of styrene and diene monomers (pKa 43–44; see Table 5.3) will effectively initiate block copolymerization of alkyl methacrylates under appropriate conditions, the resulting ester enolate anionic chain ends are too stable to reinitiate block copolymerization of styrenes and dienes. However, in principle, these ester enolate anions will initiate block copolymerization to form more stable propagating anions such as those resulting from epoxides, thiiranes, siloxanes, and lactones (see Table 5.3). A list of the types of block copolymers that can be prepared with alkyl methacrylates and acrylates is shown in Table 12.4.

One of the most interesting recent developments in block copolymer synthe-



sis is the ability to prepare block copolymers of acrylates by using “stabilizing” additives such as lithium chloride, as illustrated in Scheme 23.6 [121]. The ability

Scheme 23.6

to prepare well-defined, block copolymers of acrylates is a truly remarkable achievement. Analogous procedures have been utilized to prepare AB-type block copolymers of t-butyl acrylate with A blocks formed from styrene, methyl methacrylate, and vinylpyridine [12,122].

Although, in general, ester enolate anions will not initiate styrene polymerization, an exception is t-butyl 4-vinylbenzoate (20), whose block copolymerization can be initiated by a living poly(t-butyl methacrylate) anion [123].

The delocalization of charge into the ester group obviously provides enough stability to the “benzyl”carbanion that this monomer can be initiated with the methacrylate anion. The polymerization was effected in THF at -78°C to give block copolymers with Mw/Mn = 1.1–1.3.

As described in Chapter 12, an alternative method of synthesizing triblock copolymers is by two-step sequential monomer addition followed by coupling. For example, it has been reported that living lithium poly(methyl methacrylate) ester enolate can be coupled with para-xylene dibromide in THF at -78°C to



Scheme 23.7

produce the corresponding dimer in 90–95% yield as shown in Scheme 23.7 [7,14,124]. By analogy with this result, living diblock copolymers such as poly(styrene-block-methacrylate) ester enolates could also be coupled to provide the corresponding PS-b-PMMA-b-PS triblock copolymers. Other difunctional coupling agents should also provide efficient coupling (see Table 12.2).

I. Star Polymer Syntheses

The methodology of living anionic polymerization of alkyl methacrylates can also be used to prepare regular star-branched polymers and copolymers with well-defined structures (see Chap. 13). In principle, the living ester enolate anions can be reacted with multifunctional, electrophilic linking agents to generate the corresponding star-branched polymers with uniform arm lengths. However, the living lithium ester enolate of poly(methyl methacrylate) reacted very slowly and inefficiently with 1,3,5-tri(bromomethyl)benzene to form the corresponding three-armed, star-branched polymers [7]. In order to improve the efficiency of this linking reaction, the PMMA chain end was first converted to the corresponding t-butyl methacrylate ester enolate chain end by reaction with a 10–15 units of t-butyl methacrylate as shown in Scheme 23.8 [125]. The conversion to the t-butyl methacrylate chain end allowed the temperature to be raised to 15°C for the linking reaction. The linking efficiency was 70% for this PMMA-b-t-BuMA polymer and > 85% for the homopolymer of t-butyl methacrylate.

J. Chain-End Functionalized Polymers

As described in detail in Chapter 11, the methodology of alkyllithium-initiated living anionic polymerization of styrenes and dienes is a versatile methodology for the synthesis of the corresponding chain-end functionalized polymers. In analogous fashion, it should be possible to prepare the corresponding chain-end functionalized poly(alkyl methacrylates) by using functionalized initiators or



Scheme 23.8

using postpolymerization reactions with electrophilic functionalizing agents. Both of these approaches have been described in a recent report on the synthesis of telechelic PMMA with both an α-(initiating chain end) hydroxyl group and an ω-(terminal chain end) carboxyl group as shown in Scheme 23.9 [126]. The synthesis of the functionalized initiator was based on the low ceiling temperature of α-methylstyrene-type monomers. However, as discussed in Chapter 4 regarding the synthesis of dilithium initiators from m-diisopropenylbenzene, oligomerization is generally observed; therefore, one would expect a range of initiating end functionality (i.e., n > 1). When two equivalents of the silyl-protected, α-methyl-styrene derivative were used to prepare the functionalized initiator, 1H NMR analysis of the final polymer indicated that the OH-functionality was 1.1 after removing the silyl-protecting group. In addition, the carboxyl-group functionality was reported to be 1.0 by potentiometric titration. The preparation of α,ω-dicarboxy-poly(t-butyl acrylate) by termination of the corresponding difunctional living polymer with carbon dioxide has been reported [127]. The functionality was 1.9.

Hydroxyl-functionalized PMMA has also been prepared by using a protected acetal initiator, ethyl 3-lithiopropyl acetaldehyde ethyl acetal, to form a functionalized 1,1-diphenylalkyllithium initiator (22), as shown in Scheme 23.10 [7,14]. A quantitative yield of hydroxyl-functionalized polymer was reported with



Scheme 23.9

Mw/Mn = 1.2. Coupling the functionalized polymer with para-xylene dibromide produced the corresponding telechelic α,ω-dihydroxyl-functionalized PMMA in 90% yield [14]. A similar approach was used to prepare the α-allyl-functionalized PMMA by first reacting allyllithium with 1,1-diphenylethylene to prepare the corresponding 1,1-diphenyl-4-pentenyllithium initiator [7,14].

A variety of α-functionalized poly(methyl methacrylates) can also be prepared by using substituted 1,1-diphenylethylenes as described in Chapter 11. For example, an α-3° amine-functionalized PMMA was prepared by using the amine-functionalized initiator (23) formed in situ from reaction of sec-butyllithium with 1-(4-dimethylaminophenyl)-1-phenylethylene as shown in Scheme 23.11 [46]. The resulting ω-3° amine-functionalized PMMA (24) exhibited a narrow molecular weight distribution (Mw/Mn = 1.06) and an amine functionality (titration) equal to 1.0.

K. Macromonomers

Analogous to the corresponding living polymerizations of styrene and dienes, macromonomers of PMMA can be readily prepared by a variety of initiation



Scheme 23.10

(functionalized initiators) and termination reactions. For example, an ω-styryl-functionalized PMMA has been synthesized by termination of living PMMA with vinylbenzyl iodide as shown in Equation 23.28 [7]. Ultraviolet-visible and 1H

(23.17)

NMR spectral analyses indicated that the macromonomer was obtained in 92% yield using a twofold excess of iodide and a reaction time of several hours. The corresponding vinylbenzyl bromide reacted similarly, but the chloride failed to react [7]. However, a styrene functionality of 1.12 has been reported for the tritylsodium-initiated polymerization of t-butyl methacrylate after termination with vinylbenzyl chloride at -78°C in THF [128]; analogous functionalization of trityllithium-initiated PMMA provided a styrene functionality of 0.86–0.92 after 2 h at -78°C [129]. Using the addition of lithium chloride to effect controlled polymerization of t-butyl acrylate, several groups have reported the efficient synthesis of the corresponding macromonomers by termination with vinylbenzyl bromide, for example [12,130].



Scheme 23.11

II. Summary

The controlled anionic polymerization of a variety of alkyl methacrylates can be effected using sterically hindered initiators such as 1,1-diphenylalkyllithiums in polar solvents such as THF at low temperatures (-78°C). Poly(alkyl methacrylates) can be prepared with predictable molecular weights and narrow molecular weight distributions; block copolymers, chain-end functionalized polymers, and star-branched polymers can also be prepared. The ability to perform these polymerizations at higher temperatures is enhanced by using bulky alkyl ester groups, such as t-butyl methacrylate, which can be polymerized at 25°C; by using larger counterions, which decrease intramolecular termination rates; and by addition of complexing additives such as crown ethers, lithium chloride, and lithium alkoxides. It is difficult to obtain controlled polymerization in nonpolar solvents such as toluene, even at low temperatures. The stereochemistry of methyl methacrylate can be changed from predominantly isotactic in nonpolar solvents such as toluene to predominantly syndiotactic in polar media such as THF. The controlled polymerization of t-butyl acrylate can be effected by addition of analogous complex-



ing additives such as lithium chloride and lithium alkoxide while using low temperatures and sterically hindered initiators.

References

1. S. Bywater, in Encyclopedia of Polymer Science and Engineering, 2nd ed., J. I. Kroschwitz, Ed., Wiley-Interscience, New York, 1985, Vol. 2, p. 17.

2. A. H. E. Müller, in Comprehensive Polymer Science, Vol. 3, Chain Polymerization I, G. C. Eastmond, A. Ledwith, S. Russo, and P. Sigwalt, Eds., Pergamon Press, Elmsford, New York, 1989, p. 387.

3. T. P. Davis, D. M. Haddleton, and S. N. Richards, J. Macromol. Sci., Rev. Macromol. Chem. Phys., C34, 243 (1994).

4. A. H. E. Müller, in Recent Advances in Anionic Polymerization, T. E. Hogen-Esch and J. Smid, Eds., Elsevier, New York, 1987, p. 205.

5. B. L. Erusalimskii, Mechanisms of Ionic Polymerization. Current Problems, Consultants Bureau, New York, 1986.

6. M. van Beylen, S. Bywater, G. Smets, M. Szwarc, and D. J. Worsfold, Adv. Polym. Sci., 86, 87 (1988).

7. G. D. Andrews and L. R. Melby, in New Monomers and Polymers, B. M. Culbertson and C. U. Pittman, Jr., Eds., Plenum Press, 1984, p. 357.

8. T. E. Long, R. D. Allen, and J. E. McGrath, in Recent Advances in Mechanistic and Synthetic Aspects of Polymerization, M. Fontanille and A. Guyot, Eds., D. Reidel Publishing, Dordrecht, Holland, 1987, p. 79.

9. R. Jerome, R. Forte, S. K. Varshney, R. Fayt, and Ph. Teyssie, in Recent Advances in Mechanistic and Synthetic Aspects of Polymerization, M. Fontanille and A. Guyot, Eds., D. Reidel Publishing, Dordrecht, Holland, 1987, p. 101.

10. Ph. Teyssie, R. Fayt, J. P. Hautekeer, C. Jacobs, R. Jerome, L. Leemans, and S. K. Varshney, Makromol. Chem., Macromol. Symp., 32, 61 (1990).

11. D. Kunkel, A. H. E. Müller, M. Janata, and L. Lochmann, Makromol. Chem., Macromol. Symp., 60, 315 (1992).

12. S. Varshney, C. Jacobs, J.-P. Hautekeer, P. Bayard, R. Jerome, R. Fayt, and Ph. Teyssie, Macromolecules, 24, 4997 (1991).

13. R. D. Allen, T. E. Long, and J. E. McGrath, Polym. Bull., 15, 127 (1986).

14. B. C. Anderson, G. D. Andrews, P. Arthur, Jr., H. W. Jacobson, L. R. Melby, A. J. Playtis, and W. H. Sharkey, Macromolecules, 14, 1599 (1981).

15. K. Ute, T. Asada, N. Miyatake, and K. Hatada, Makromol, Chem., Macromol. Symp., 67, 147 (1993).

16. T. Kitayama, N. Fujimoto, and K. Hatada, Makromol. Chem., Macromol. Symp., 67, 137 (1993).



22. A. Hirao, H. Kato, K. Yamaguchi, and S. Nakahama, Macromolecules, 19, 1294 (1986).

23. K. Hatada, H. Sigino, H. Ise, T. Kitayama, Y. Okamoto, and H. Yuki, Polym. J., 12, 55 (1980).


25. H. Ozaki, A. Hirao, and S. Nakahama, Macromolecules, 25, 1391 (1992).

26. H. Mori, A. Hirao, and S. Nakahama, Macromolecules, 27, 35 (1994).

27. A. Hirao and S. Nakahama, Prog. Polym. Sci., 17, 283 (1992).

28. T. Nakano, T. Hasegawa, and Y. Okamoto, Macromolecules, 26, 5494 (1993).

29. R. Bohnert and H. Finkelmann, Makromol. Chem., Rapid. Commun., 14, 139 (1993).

30. G. Riess and G. Hurtrez, in Encyclopedia of Polymer Science and Engineering, 2nd ed., J. I. Kroschwitz, Ed., Wiley, New York, 1985, Vol. 2, p. 324.

31. K. J. Ivin and W. K. Busfield, in Encyclopedia of Polymer Science and Engineering, 2nd ed., J. I. Kroschwitz, Ed., Wiley, New York, 1985, Vol. 2, p. 555.

32. K. Hatada, S. Kokan, T. Niinomi, K. Miyaji, and H. Yuki, J. Polym. Sci., Polym. Chem. Ed., 13, 2117 (1975).

33. H. Yuki, K. Hatada, T. Niinomi, and K. Miyaji, Polym. J., 1, 130 (1970).

34. H. Hopff, H. Lussi, and L. Borla, Makromol. Chem., 81, 268 (1965).

35. H. Yuki, K. Hatada, T. Niinomi, M. Hashimoto, and J. Ohshima, Polym. J., 2, 629 (1971).

36. A. Matsumoto, A. Horie, and T. Otsu, Polym. J., 23, 211 (1991).

37. R. P. Quirk and J. Ren, Makromol. Chem., Macromol. Symp., 88, 17 (1994).



40. K. Hatada, T. Kitayama, K. Fumikawa, K. Ohta, and H. Yuki, in Anionic Polymerization. Kinetics, Mechanisms, and Synthesis, J. E. McGrath, Ed., ACS Symposium Series 166, American Chemical Society, Washington, D.C., 1981, p. 327.

41. D. M. Wiles and S. Bywater, J. Polym. Sci., Part B: Polym. Lett., 2, 1175 (1964).


43. D. Freyss, P. Rempp, and H. Benoit, J. Polym. Sci., Part B: Polym. Lett., 2, 217 (1964).

44. K.-E. Piejko and H. Höcker, Makromol. Chem., Rapid Commun., 3, 243 (1982).


46. R. P. Quirk and L.-F. Zhu, Br. Polym. J., 23, 47 (1990).

47. L. Lochmann and D. Lim, J. Organometal. Chem., 50, 9 (1973).



55. D. Garg, S. Höring, and J. Ulbricht, Makromol. Chem., Rapid Commun., 5, 615 (1984).

56. M. Viguier, A. Collet, and F. Schue, Polym. J., 14, 137 (1982).

57. M. Viguier, A. Collet, F. Schue, and B. Mula, in Recent Advances in Anionic Polymerization, T. E. Hogen-Esch and J. Smid, Eds., Elsevier, New York, 1987, p. 249.

58. P. E. M. Allen and D. R. G. Williams, Ind. Eng. Chem. Prod. Res. Dev., 24, 334 (1985).

59. K. Hatada, K. Ute, K. Tanaka, Y. Okamoto, and T. Kitayama, Polym. J., 18, 1037 (1986).

60. T. Kitayama, T. Shinozaki, T. Sakamoto, M. Yamamoto, and K. Hatada, Makromol. Chem., Supplement 15, 167 (1989).

61. D. G. Ballard, R. J. Bowles, D. M. Haddleton, S. N. Richards, R. Sellens, and D. L. Twose, Macromolecules, 25, 5907 (1992).

62. M. Kuroki, T. Aida, and S. Inoue, J. Am. Chem. Soc., 109, 4737 (1987).

63. M. Kuroki, S. Nashimoto, T. Aida, and S. Inoue, Macromolecules, 21, 3114 (1988).

64. Y. Hosokawa, M. Kuroki, T. Aida, and S. Inoue, Macromolecules, 24, 824 (1991).

65. T. Adachi, H. Sugimoto, T. Aida, and S. Inoue, Macromolecules, 26, 1238 (1993).

66. W. Bauer and D. Seebach, Helv. Chim. Acta, 67, 1972 (1984).

67. S. Bywater, in Comprehensive Polymer Science, Vol. 3, Chain Polymerization I, G. C. Eastmond, A. Ledwith, S. Russo, and P. Sigwalt, Eds., Pergamon Press, Elmsford, New York, 1989, p. 433.

68. D. M. Wiles and S. Bywater, J. Phys. Chem., 68, 1983 (1964).

69. A. H. E. Müller, H. Jeuck, C. Johann, and P. Kolz, Polym. Prepr., Am. Chem. Soc., Div. Polym. Chem., 27(1), 153 (1986).

70. R. P. Quirk and B. Lee, Polym. Internat., 27, 359 (1992).

71. G. Löhr and G. V. Schulz, Makromol. Chem., 172, 137 (1973).

72. G. Löhr and G. V. Schulz, Eur Polym. J., 10, 121 (1974).

73. J. E. Figueruelo, Makromol. Chem., 131, 63 (1970).

74. C. Johann and A. H. E. Müller, Makromol. Chem., Rapid Commun., 2, 687 (1981).

75. Ch. B. Tsvetanov, A. H. E. Müller, and G. V. Schulz, Macromolecules, 18, 863 (1985).

76. H. Jeuck and A. H. E. Müller, Makromol. Chem., Rapid Commun., 3, 121 (1982).

77. R. Kraft, A. H. E. Müller, V. Warzelhan, H. Höcker, and G. V. Schulz, Macromolecules, 11, 1093 (1978).

78. A. H. E. Müller, L. Lochmann, and J. Trekoval, Makromol. Chem., 187, 1473 (1986).

79. D. Kunkel, A. H. E. Müller, M. Janata, and L. Lochmann, Makromol. Chem., Macromol.



84. L. L. Böhn, M. Chmelir, G. Löhr, B. J. Schmitt, and G. V. Schulz, Adv. Polym. Sci., 9, 1 (1972).

85. A. H. E. Müller, in Anionic Polymerization. Kinetics, Mechanisms, and Synthesis, J. E. McGrath, Ed., ACS Symposium Series 166, American Chemical Society, Washington, D.C., 1981, p. 441.


87. S. K. Varshney, J. P. Hautekeer, R. Fayt, R. Jerome, and Ph. Teyssie, Macromolecules, 23, 2618 (1990).

88. L. Lochmann, M. Rodova, and J. Trekoval, J. Polym. Sci., Polym. Chem. Ed., 12, 2091 (1974).

89. L. Lochmann, D. Doskocilova, and J. Trekoval, Coll. Czech, Chem. Commun., 42, 1355 (1977).

90. L. Lochmann, J. Kolarik, D. Doskocilova, S. Vozka, and J. Trekoval, J. Polym. Sci., Polym. Chem. Ed., 17, 1727 (1979).

91. L. Lochmann, S. Pokomy, J. Trekoval, H.-J. Adler, and W. Berger, Makromol. Chem., 184, 2021 (1983).

92. L. Lochmann, M. Rodova, J. Petranek, and D. Lim, J. Polym. Sci., Polym. Chem. Ed., 12, 2295 (1974).

93. L. Lochmann, M. Janata, L. Machova, P. Vicek, J. Mitera, and A. H. E. Müller, Polym. Prepr., Am. Chem. Soc., Div. Polym. Chem., 29(2), 29 (1988).

94. J. S. Wang, R. Jerome, R. Warine, and Ph. Teyssie, Macromolecules, 26, 1402 (1993).

95. J. S. Wang, R. Warine, R. Jerome, and Ph. Teyssie, Macromolecules, 26, 6776 (1993).

96. J. S. Wang, R. Jerome, R. Warine, and Ph. Teyssie, Macromolecules, 27, 1691 (1994).

97. M. Janata, L. Lochmann, P. Vlcek, J. Dybal, and A. H. E. Müller, Makromol. Chem., 193, 101 (1992).


99. K. Kunkel, A. H. E. Müller, M. Janata, and L. Lochmann, Makromol. Chem., Makromol. Symp., 60, 315 (1992).


101. L. Dvoranek and P. Vlcek, Polym. Bull., 31, 393 (1993).

102. S. K. Varshney, R. Jerome, P. Bayard, C. Jacobs, R. Fayt, and Ph. Teyssie, Macromolecules, 25, 4457 (1992).

103. G. Löhr, A. H. E. Müller, V. Warzelhan, and G. V. Schulz, Makromol. Chem., 175, 497 (1974).

104. F. J. Gerner, H. Höcker, A. H. E. Müller, and G. V. Schulz, Eur. Polym. J., 20, 349 (1984).

105. H. Yuki and K. Hatada, Adv. Polym. Sci., 31, 1 (1979).



110. H. Yuki, K. Hatada, T. Niinomi, and Y. Kikuchi, Polym. J., 1, 36 (1970).

111. N. N. Aylward, J. Polym. Sci., A-1, 8, 319 (1970).

112. D. M. Wiles and S. Brownstein, J. Polym. Sci. B, 3, 951 (1965).

113. Y. Iwakura, F. Toda, T. Ito, and K. Aoshima, J. Polym. Sci. B, 5, 29 (1967).

114. K. Hatada, K. Ute, K. Tanaka, Y. Okamoto, and T. Kitayama, Polym. J., 18, 1037 (1986).

115. J. P. Pascault, J. Kawak, J. Gole, and Q. T. Pham, Eur. Polym. J., 10, 1107 (1974).

116. V. Warzelhan, H. Höcker, and G. V. Schulz, Makromol. Chem., 181, 149 (1980).

117. A. H. E. Müller, H. Höcker, and G. V. Schulz, Macromolecules, 10, 1086 (1977).

118. P. Pino and U. W. Suter, Polymer, 17, 977 (1976).

119. E. Eliel, Stereochemistry of Carbon Compounds, McGraw Hill, New York, 1962.

120. J. Wang, R. Jerome, R. Warine, and Ph. Teyssie, Macromolecules, 26, 5984 (1993).

121. J.-P. Hautekeer, S. K. Varshney, R. Fayt, C. Jacobs, R. Jerome, and Ph. Teyssie, Macromolecules, 23, 3893 (1990).


123. J. Wang, Ph. Bayard, R. Jerome, Ph. Teyssie, and S. K. Varshney, Macromolecules, 26, 2386 (1993).

124. L.-F. Zhu, Ph.D. dissertation, University of Akron, 1991.

125. T. Q. Nguyen and H.-H. Kausch, Makromol. Chem., Rapid Commun., 6, 391 (1985).

126. M. Ohata, S. Ikeda, S. Akatani, and Y. Isono, Macromolecules, 25, 5131 (1992).

127. F. Fontaine, J. Ledent, R. Sobry, E. Francois, R. Jerome, and Ph. Teyssie, Macromolecules, 26, 1480 (1993).

128. K. Ishizu, K. Mitsutani, and T. Fukutomi, J. Polym. Sci., Part C., Polym. Lett., 25, 287 (1987).

129. K. Ishizu and T. Fukutomi, J. Polym. Sci.: Part A: Polym. Chem., 27, 1259 (1989).

130. K. Antolin, J.-P. Lamps, P. Rempp, and Y. Gnanou, Polymer, 31, 967 (1990).



24 Block Polymers Prepared Via Anionic Ring-Opening Polymerization

I. Introduction

A. Ring-Opening Polymerization: General

Along with condensation (a step-reaction), and addition (a chain-reaction), ring-opening reaction constitutes one of the most important methodologies for the formation of macromolecules. Ring-opening polymerizations are unique in that these reactions do not produce small molecules as by-products as in the condensation polymerizations; nor do they involve the exothermic driving force of converting the multiple bonds to single bonds as in the typical chain-type olefin polymerizations. Ring-opening polymerizations can be classified as either step-growth or chain reaction[1]; the distinction is often a function of how molecular weight varies with conversion (Scheme 24.1). A comprehensive treatise of all aspects of

Scheme 24.1

ring-opening polymerizations in a three-volume series was published in 1984 edited by Ivin and Saegusa[2]. McGrath[1] edited an 1984 ACS Symposium Series publication in which the entire field of ring-opening polymerization based



on the symposium lectures of 26 experts was presented. A recent publication [3] edited by Brunelle serves as an excellent updated supplement to the definitive Ivin/Saegusa series[2].

The polymerizability of cyclic monomers depends on thermodynamic and mechanistic or kinetic factors[4–6]. The release of ring strain by polymerization provides the principal driving force for the polymerization of cyclic monomers. Fulfillment of the thermodynamic requirements is necessary, but a reaction path on which monomer molecules could be converted into the linear one must also exist.

Many cyclic compounds have been successfully polymerized. The ring size of the monomer, the substituents on the ring and the selection of the catalyst are the key factors of polymerizability. Thus, cyclic ethers, amines, amides, esters, carbonates, anhydrides, thioethers, sulfides, siloxanes, phosphonites, olefins and many others have been converted to olegimers and high-molecular-weight linear polymers by ring-opening reactions. Many important commercial materials are products of ring-opening polymerization. These products have found many utilities and applications as textile fibers, thermoplastic resins, high-performance elastomers, water-soluble polymers, oil additives, intermediates for segmented urethane and urea foams, intermediates for thermoplastic elastomers, additives in cosmetics, biodegradable materials, biomaterials, and others. Certain products are produced in large volume such as polycaprolactam (Nylon 6), polyacetal resin (polyoxymethylene), polyethylene oxide, polypropylene oxide, polytetramethylene oxide (polytetrahydrofuran), and others. Some specialty products such as polysiloxanes, polyamines, polyphosphazenes, polyglycolides, polylatides, and others possess unique properties and find utilities in diverse markets.

Ring-opening polymerization can be initiated by a wide variety of catalysts. In some cases, spontaneous polymerization and copolymerization are known. Generally, radical, cationic, anionic, and covalent nucleophilic catalysts are effective. Other catalysts such as “pseudoanionic” types are known to initiate polymerization of oxiranes, lactones, lactides, and others. Catalysts based on W, Mo Ru, Rc, Ti, and Ta are specifically effective for the polymerization of cyclic olefins. Aluminum porphrins lead to either living or “immortal” ring-opening polymerization.

B. Anionic Ring-Opening Polymerization

Anionic ring-opening polymerization involve end activation either before or after ring opening being possible[5], as illustrated in Schemes 24.2 and 24.3. Each

Scheme 24.2



Scheme 23.3

propagation step involves a nucleophilic attack to the anionic active center, located at the end of the growing macromolecule, on the heterocyclic monomer [7]. By selecting conditions in which side reactions are nearly nonexistent, or at least their rates are much slower than the propagation rate, anionic ring-opening polymerizations can have the characters of living polymerizations. In fact, Flory [8] reported the anionic polymerization of ethylene oxide in 1940, describing the characteristics of what is now referred to as living polymerization.

Typical anionic initiators used in ring-opening polymerization are radical anions such as sodium naphthalide; carbanions such as butyllithium, fluorenylsodium, tetrabutylammonium compounds; alcoholates, silonates, thiolates, carboxylates, lactamates of alkali metals; amines and phosphines. The detailed chemistry of these initiators such as initiation mechanisms, kinetics, and their applications can be found in several reviews and overviews [7,9–12]. The selection of the right catalyst for the polymerization of a cyclic monomer is critical. If one assumes that the monomer is thermodynamically favorable for the ring-opening reaction, the end growth generated from the reaction must be favorable to the formation of polymer molecules. Many of the anionic initiators such as alcoholates, carboxylates, thiolates, silanolates, and others are in fact the models of the growing species.

The structure of the active center (growth site) is dependent on the pathway of the ring scission of the cyclic compound. As an illustration, lactones can undergo α or β ring scission (Scheme 24.4). Lactones such as ζ-valerolactone and

Scheme 24.4

ε-caprolactone are known to undergo a ring scission to form alcoholate end groups. On the other hand, β-propiolactone and its substituted derivative undergo β-ring scission to form carboxylate end groups. One thus uses alcoholates as initiator for the polymerization of valerolactone or caprolactone, but carboxylates for the polymerization of propiolactone.



Pseudoanionic catalysts are based on multivalent metals, most commonly Al and Zn. The initiation reaction takes place not via charged species but via polarized covalent bond. It involves a concerted insertion with concurrent cleavage of a covalent polymer-catalyst bond [13]. For this reason these catalysts are also referred to as “coordinated anionic.” Examples of this class of initiators are R3Al, Zn[OAl(OR)2]2, Ti(OR)4, Sn(OR)4, R2AlOR', among others. The pioneering work of Vandenberg, Tsuruta, and Teyssie is cited by Slomokowski and Dudas in their review [7]. The active species generated by these multivalent metal catalysis are less reactive than corresponding alkalimetal species and undesirable side reactions are often avoided. Both Teyssie [14,15] and Penczek [16] have demonstrated that the polymerization of ε-caprolactone with these initiators is living, at least, within the time of complete conversion of monomers to polymers. However, with a further increase in temperature and/or reaction time, both intermolecular transesterfication and intramolecular tranesterfication occur resulting in broader molecular weight distribution of the product and formation of cyclic olegomers, respectively. Teyssie and his co-workers have reported that the addition of a Lewis base, such as pyridine, effectively delays the occurence of the secondary transesterfication reactions in ε-caprolactone and lactide polymerizations [17,18]. Penczek recently discussed the factors affecting “livingness” in polymerizations intiated with aluminum alkoxides [19].

In this chapter the formations of many block copolymers utilizing anionic ring-opening polymerization are illustrated. This is an area of considerable interest among polymer chemists and technologists because of the unusual and unique properties exhibited by these copolymers. These copolymers contain distinctly different blocks that are linked together in one single macromolecule. For example, a crystalline block is linked with an amporphous block or a hydrophilic block is linked with a hydrophobic block.

II. Synthesis Of Block Polymers Via Anionic Ring-opening Polymerization

A. Polymers with Polyether Block

Ethylene oxide polymerizes readily to form (-O-CH2-CH2)n in the presence of anionic initiators such as carbazylpotassium [20], flurenylsodium [21], ethyldiethyleneoxy cesium [22], CH3COOK-DBC (dibenzo-18-crown-6 ether) [23], and fluoradenyl alkali metals [24]. Lithium alkyls (RLi) and lithium alkoxides (ROLi) generally do not initiate the polymerization of ethylene oxide [25–29], and sodium metal is only active under high pressure and longer polymerization time [30].

Propylene oxide can also be readily polymerized with some anionic intitiators. However, the anionic polymerization of propylene oxide is complicated by a



transfer reaction involving hydrogen abstraction from the methyl group of the propylene oxide to form allyl alcohol and, thereby, initiation of additional chains.

Quirk [31] successfully prepared a well-defined AB diblock copolymer in which A is polystyrene and B is polyethylene oxide (poly (styrene-b-ethylene oxide)). Living polystrene chains (PSLi) initiated with sec-BuLi were terminated with ethylene oxide in benzene to form alkoxide chain ends (PSCH2 CH2 OLi). This polymer was dissolved in a 2:1 (v:v) mixture of benzene and dimethylsulf-oxide. Using high vacuum techniques, ethylene oxide was added and polymerization was carried out at 40°C in 4 days and followed by 60°C for an additional 4 days. The isolated product yielded 83% overall indicating 70% conversion of ethylene oxide. Both the narrow MWD of the diblock and the absence of any observable peak (by GPC) corresponding to the original PS block indicates that (a) the hydroxyethylation reaction occurs essentially quantitatively in benzene solution; and (b) no evidence of chain-temination or chain-transfer is apparent in the polymerization of ethylene oxide with Li as counterion in a mixture of benzene and dimethylsulfoxide (DMSO). Quirk concluded that a dipolar aprotic solvent such as DMSO provides the necessary solvation and polarity to render lithium alkoxide an effective initiation for ethylene oxide polymerization.

By using potassium-naphthalene adduct as initiator and by incremental additions of monomers, block copolymers of A-B-C-B-A (where A = polyethylene oxide, B = polystyrene or polyisoprene, and C = polyisoprene or polystyrene) were prepared in tetrahydrofuran (THF) [2]. In these examples, C, the center block, was formed first; followed by B and then A. Vacuum techniques were used. Styrene was polymerized at -70°C in 2 h and ethylene oxide, the last increment, was polymerized in 6.5 h at room temperature. Compositions, molecular weights and molecular weight distributions of the final and intermediate polymers were examined by gel permeation chromatography (GPC), infrared (IR), and PMR, and osmometry and concluded the formations of the designed block copolymers.

Polystyrene-polyethylene oxide diblock copolymers were prepared by using cumylpotassium or diphenylmethylpotassium as the initiator and by sequential monomer additions [28,30,33,34]. The use of cumylpotassium leads to copolymers containing high molecular impurities (PEO-PS-PEO triblock copolymers) and fluorescent species. The preparation of diphenylmethylpotassium from diphenylmethane and naphthylpotassium leads to the production of difunctional byproducts. The presence of these byproducts results in copolymers with higher polydispersity. Using an improved method to prepare cumylpotassium [35], diblock copolymers with low polydispersity were prepared free of triblock products. The improved method also avoided the production of fluorescent byproducts that give copolymers with unwanted luminescent properties. The improved catalyst was prepared from liquid K-Na alloys containing ˜29–83% K by reacting the alloy with cumylmethyl ether. When the K content of the alloy dropped to <



29% as a result of reaction, the alloy became solid and was filtered off together with KO Me.

Ethylene oxide-2-vinylpyridine diblock copolymers were prepared in THF with diphenylmethylpotassium as initiator [36]. The polymers were fractionated from benzene solution by progressive precipitation with 2,2,4-trimethylpentane at 35°C. These hydrophilic-hydrophobic block copolymers had a well-defined structure and low polydispersity in both molecular weight and chemical composition.

Diblock and triblock copolymers of AB, ABA, and BAB types were synthesized where A = tert-butylmethacrylate and B = ethylene oxide by sequential monomer additions and with diphenylmethylpotassium and potassium-naphthalene adduct as initiator [37]. Wang et al. reported that the order of the monomer addition is not important. The products had predictable molecular weight, narrow MWD, and no homopolymer formation indicating no transesterfication reactions. Similar series of diblock, triblock as well as A-B-A-B-A multiblock copolymers of ethylene oxide and tert-butylmethacrylate were prepared in THF by sequential monomer addition with cumylpotassium and diphenylmethylpotassium as initiator [38]. Reuter et al. also observed that the order of monomer addition is not a factor. Ethylene oxide was polymerized at 25–40°C and tert-butylmethacrylate was polymerized at 25°C. Block copolymers containing both hydrophilic semi-crystalline and hydrophobe amorphous segments have interesting properties and are suitable for applications such as surface active agents, compatibilizers, solubilizers, and membrane precursors. An attempt to make ethylene oxide and methylmethacrylate block copolymers was less successful [38]. With methyl-methacrylate, termination and transfer reactions occurred at temperatures above ˜20°C. Polyethylene oxide was not soluble in THF below 20°C. Transesterfication reactions were also present. The block copolymer was inhomogeneous.

Numerous publications reported the synthesis of ethylene oxide-propylene oxide block copolymers. One interesting report [39] is the synthesis of A-B block copolymers with A = EO/PO random copolymer with low ratio and B = EO/PO random copolymer with high ratio. The balance of hydrophilicity and hydro-phobicity can be achieved by adjusting the monomer ratios in each block. By using potassium alkoxide plus 18-crown-6-ether as initiator, polymerization was carried out in THF at 40°C. Under these conditions, the undesirable transfer reaction, hydrogen abstraction from the methyl group of the propylene oxide, formed allylalcohol and thereby initiations of additional chains were suppressed. These di- and triblock copolymers possessed interesting properties in aqueous solutions. In a dilute solution, polymers associate into micelles. In a concentrated solution, the association becomes stiff gels. The association behaviors are determined by their composition and chain length.

Star block polymers containing polystyrene block and polyethylene oxide block are of interest. These polymers behave typically as amphiphilic macro-



molecules. Several studies on the emulsifying and crystalline properties of amphiphilic block copolymers have been reported [40–43]. The phase transfer catalytic effect [44] and the mesomorphic behavior in toluene solution [45] have also been studied. One could prepare the star block polymers by employing a multifunctional initiator such as the one prepared from divinylbenzene and sec-BuLi and by incremental addition of monomers in high dilution (styrene first). This approach would have polystyrene as the core blocks. Star block polymers of (PS)2 (PEO)2 and (PS)2 (PEO) were also prepared by sequence of reactions shown in Scheme 24.5 [45].

Scheme 24.5

These copolymers were purified by extraction and chracterized by GPC, IR, PMR, and torsional braid analysis. Their crystalline, emulsifying and complexing properties, and phase transfer catalytic effect were discussed. Gia, Jerome, and Teyssie [45] tried several approaches to prepare (PS) (PEO)2 star block polymers by first preparing a polystyrene having a naphthalene end group. For example, they tried the use of naphthyllithium or naphthylmethyllithium as initiator to introduce the end group or terminate the living polystyrene with naphthylbromide or naphthylemthylbromide. In both cases, secondary reactions made these approaches unsuitable. However, by reacting the living polystyrene first with MgBr and then coupling with naphthylmethylbromide (Scheme 24.6), naphthalene functionality of > 90% was achieved.

Scheme 24.6

The metallation of naphthalene-terminated polyisoprene by potassium in THF at room temperature proceeded through the transient formation of the anion radical and was followed by the quantitative occurrence of two anions per naphthalene molecule.



By addition of ethylene oxide to the polyisoprene dianion, (PI) (PEO)2 star block polymer was synthesized. The similar dianion formed with polystyrene or polytertbutylstyrene was unstable. The naphthalene end group was released and the polymeric vinyl anion was liberated and finally isomerized as illustrated in Scheme 24.7. By addition of few units of isoprene at the end of polystyrene chain

Scheme 24.7

before the introduction of naphthalene end group, the conjunction between the aromatic molecules was removed. Thus, (PS) (PEO)2 and (PTBS) (PEO)2 can also be prepared by the same basic procedure.

B. Polymers with Polyester Block

Homopolymers of β-propiolactone and β-butyrolactone and their substituted compounds can be readily polymerized via ring-opening reaction with many anionic initiators. ε-Caprolactone can be polymerized with equal ease. Pseudo-anionic initiators are very effective in lactone polymerization. The key to the successful preparation of well-defined block copolymers with low polydispersity involving lactones lies in the ability to have the propagation reaction much faster that the side reactions. These will be discussed in more detail later.

Pivalolactone

Pivalolactone, α, α-dimethyl-β-propiolactone, (PVL) can be polymerized by initiators such as tertiary amines, phosphines, butaine ((CH3)3N+CH2CO-S2), cyclic tertiary amines, and others.

Polyprivolactone, with its easy formation, rapid crystallizability, and good physical properties, created considerable attention in the polymer industry. Sharkey [47], in his review article on polymerization of PVL, discussed the use of tetraalkylamonium carboxylates as initiator in aproticorganic solvent, of which THF is preferred. The polymer is linear and has very low polydispersity; the living propagating terminus has an exceptionally long lifetime. Block/graft copolymers containing elastic polyisoprene in the center and crystallizable polypropiolactone end blocks were prepared [47–50]. The PVL blocks crystallize into discontineous domains serving as the physical crosslinks in a characteristic thermoplastic elastomer style. These polymers were easily melt-processable, and could be readily



melt spun into elastic fibers of good strength, high elongation, and high resilience. The synthesis routes are illustrated in Schemes 24.8 and 24.9. The diinitiator was

Scheme 24.8

Scheme 24.9

synthesized by addition of sec-BuLi to 1,3-diisopropenylbenzene followed by reaction with isoprene and modified with triethylamine. Isoprene was polymerized in hydrocarbon solvent to give high-cis configuration. By using essentially these same techniques, Foss and colleagues prepared poly(isobutyleneg-PVL) samples [51,52]. Isobutylene was copolymerized with 1–2% methylstyrene by a cationic intiator. Metallation of the methyl group followed by CO2, H+, Bu4NOH reactions provided the sites for PVL grafting. These graft polymers were examined as films and fibers.

Lenz and co-workers [53] synthesized ABA triblock copolymers in which B = poly(α-methyl-α-butyl-β-propiolactone) and A = polypivalolactone with a difunctional initiator, tetrabutylamonium salt of sabacic acid, Bu4NO2C(CH2)8CO2NBu4. Sundet [54] modified vinyl/acrylic copolymers containing methacrylic acid by reacting with tetrabutylamonium hydroxide and used the carboxylic salts so developed to initiate polymerization of PVL. Graft polymers were thus prepared. Caywood [55] prepared graft polymers by first saponifying the ester groups of polyacrylates to create the sites for further reaction with tetravutylamonium hydroxide, followed by the polymerization of PVL. Caywood [55] reported that the physical properties of the graft copolymers were similar to those of the parent elastomeric polyacrylates that had been compounded with carbon black and chemically crosslinked.



EPDM rubbers (poly(ethylene/propylene/1,4-hexa-diene)) were first modified by reacting with thioglycolic acid or maleic cenhydride, followed by conversion of the acid groups or anhydride groups in the product to tetrabutylamonium salts and use of the salts to initiate the polymerization of PVL [56,57]. The graft polymers show the reinforcement plus crosslinking effects of the PVL block.

Yamashita [58–60] studied the polymerization of PVL with carboxylates and alkoxides of alkali metals extensively. These workers found that addition of PVL to polystyrisodium or polystyrylethoxysodium in THF resulted in homopolymer mixtures. With addition of PVL to polystyrene dicarboxylate or polytetrahydrofuran dicarboxylate in THF containing DMSO, block copolymers were formed, but some unreacted homopolymers remained. Yamashita considered the polymerization of PVL with carboxylate a living system as shown in Scheme 24.10.

Scheme 24.10

Others [62] also considered the polymerization of PVL as living for its lack of termination reaction. If ki > kp, block polymers are formed. If ki < kp, block polymers plus some polystyrenes are mixed in the product. Propagating carboxylate anion is more basic than the initiating carboxylate anion as shown in Scheme 24.11. This is consistent with the results reported by Hall [61] that the rates of

Scheme 24.11

reaction of α,α-disubstituted β-lactone are pivalate > acetate > benzoate. The kp/ki decreases with the increase of the nucleophilicity of the carboxylates. Polarity of the solvent or nature of the counter ion did not change the situation of slow initiation. Thus the block efficiency is determined by the slow initiation, the molar feed ratios of pivalolactone, and the conversion.

While the mechanisms of the intiation and propagation of β-lactones with carboxylate are fairly well established, the attack of a strong nuclophile (R- or RO-) on the carbonyl group of the lactone causing acyl-oxygen fission is not



clearly defined. Although polymerization is known to occur with these initiators, the exact nature of the propagating species has not been identified. Yamashita [60] suggested that some form of monomer transfer reaction occurred to form a new initiating species favorable for the polymerization of PVL. Alkoxide groups from ROM and RM are not favorable for propagation and, not surprisingly, black polymers cannot be made this way; only a mixture of homopolymers is formed.

Alkoxide and naphthalenides of alkalimetals complexed with crown ethers, cryptands, and similar compounds are powerful initiators for the ring-opening polymerization of many cyclic monomers including β-lactones. It is generally understood that the active species of these complexed initiators are mostly free ions and loose ion pairs. There is clear evidence that the polymerization of β-lactones by these compounds does not contain any metal or other initiator moieties in the polymer chains. Instead unsaturated and hydroxyl end groups were identified [63–66]. The proposed mechanism involving the incorporation of the initiator into the growing polymer chains [67] needs to be readdressed.

Regardless of the identity of the actual intiating species, alkoxides end groups are not favorable for the synthesizing block polymers by sequential addition of β-lactones, whether the metal is complexed or not.

ε-Caprolactone

A wide variety of compounds are known to initiate the polymerization of caprolactone (CL). The chemistry of CL polymerization has been extensively studied and reviewed [59,68,69]. In their earlier attempts to prepare block polymers of diene, styrene, and capilactone, Hsieh and his colleagues [70] learned that polyvinyllithium chains converted to have ethylenyloxide end groups with ethyleneoxide, polymerized CL very rapidly; small amounts of CL couple the living polyvinyl chains; and the coupled species polymerized CL also rapidly, as illustrated in Schemes 24.12 and 24.13.

Scheme 24.12

Scheme 24.13



The polymerizations were carried out in cyclohexane at 30–70°C. The lactone was polymerized very rapidly and completed in a very few minutes. Unless the product was quickly terminated and recovered, redistribution of the polycaprolactone chain (intermolecular transesterification), and depolymerization (intramolecular transesterification or back-biting) occurred. This phenomenon was manifested by the increase in acetone solubles of the product (cyclic oligomers) and the dramatic changes in the shape of the GPC curves. When the product was left in the solution at 70°C long enough, nearly all the polymerized CL units were gone from the product. To make the process viable in a commercial operation it was discovered that the introduction of a trace amount of isocyanate, either in the CL or added shortly after the addition of CL to the living polymer solution, prevented the undesirable side reactions and made the desired product possible [71–73]. The extremely rapid polymerization of CL, followed by almost instantaneous termination with isocyanate eliminated the transesterification reactions. The star block polymers of (S-B)m -(CL)n type are thermoplastic elastomers (TPR) with similar mechanical properties of comparable pure S-B-S or (S-B)n-X types. There is one difference in these two TPRs, however; the caprolactone-containing elastomer had good ozone reistance. Blends of polycaprolactone and S-B-S polymers did not give an ozone-resistant TPR. The exact reason for this is not known. Clark and Childers [74] hot mixed sytrene/acrylonitrile copolymer (SAN resin) and rubbery (S-B)m -(CL)n star block copolymer were made as described above, in the presence of an organic peroxide that produced impact-resistance resins resembling ABS polymer. The polymer alloy had high tensile strength, fair melt flow, and excellent Izod impact. By adjusting the S-B or S/B composition so that the refractive index of this block is equal to the value of the SAN material, blends were produced having good clarity along with good properties. Block polymers of S-CL and B-CL are often used as “oil-in-oil” type of emulsifiers in polymer blends, whose mechanical properties can be improved if better homogenicity is achieved by the emulsifier additives [75,76].

Both Young [68] and Yamashita [69] discussed the two major side reactions that occurred in anionic polymerization of CL in detail. The back-biting (intra-molecular ester interchange) and the scrambling (intermolecular esterification) are show in Schemes 24.14 and 24.15. Yamashita concluded, however, that tailored block copolymers can be synthesized under kinetically controlled conditions.

Scheme 24.14



Scheme 24.15

Hsieh tried to improve their earlier process of synthesizing CL-containing block polymers by taking the approach of reducing the nucleophilicity of the alkoxide chain end to cope with the ester exchange interferences. Hsieh and Wang [77] succeeded in providing a less nucleophilic blocking sites for the process. Stepwise addition of an alkylaluminum halide, which follows that an ethylene oxide to the alkyllithium-initiated living prepolymer provides the desired oxy-aluminum chain end as a modified ring-opening site for cyclic esters. The improved process is illustrated in Scheme 24.16.

Conversion, IR analysis, acetone extraction, GPC, and Rheovibron measurements confirmed the integrity of the block structure. By using diepoxides, lactones

Scheme 24.16



(½ molar ratio to RLi initiator), or CO2 as coupling agent for RLi-initiated living prepolymer prior to the addition of a alkyaluminum halide and CL, multiarmed block (star block) polymers could also be prepared.

Lactone-Containing Block Polymers Made with Pseudoanionic Initiators

A general discussion of block copolymerization with psuedoanionic initiators was made by Teyssie [78]. To illustrate the versitility of these initiators, several examples are given below. ε-Caprolactone and DL-lactide diblock polymer was prepared using [(BuO)2AlO]2 Zn catalyst [79]. In the presence of an Al/Zn bimetallic alkoxide complex, three types of triblock copolymers, ABA, ACC, and ACB, where A, B, and C are polycaprolactone, poly (DL-lactide), and poly-glycolide, respectively, were correspondingly synthesized in toluene at 90°C by sequential monomer addition [80]. Teyssie and co-workers also reported the synthesis of triblock polymer of CL and B(butadiene) in CL-B-CL form using a bimetallic oxoalkoxide catalyst, (RO)2Al-O-Zn-O-Al(OR)2 in an involved stepwise process [81], ε-caprolactone and β-propiolactone block copolymer with a bimetallic oxoalkoxide catalyst [82], and CL and lactide block copolymer with (iso-Pro)3 Al [83].

Other Related Block Polymers

The block polymers of 4-methyl-2-oxetanone and 2-oxetanone were prepared in the presence of K solutions in THF containing 18-crown-6-ether [84]. Diblock of ζ-valero-lactone and L-lactide were prepared polymers initiated by MeOK in THF at 20–25°C [85,86]. It was reported [87] that the polymerization of mixtures of 5,5-dimethyl-1,3-dioxane-2-one and pivalolactone in toluene with K dihydronaphthalene as initiator resulted in formation of block copolymers with yields of ˜90%. The substituted 1,3-dioxane-2-one was polymerized first and subsequently the lactone. By means of [13C] nuclear magnetic resonance (NMR) spectroscopy the block structure was confirmed. Reaction of Na polyethylene-glycolate with L-lactide resulted in ABA triblock copolymers [88]. Tapered block copolymer of 2,2-dimethyltrimethylene carbonate with caprolactone was prepared in toluene using sec-BuLi or K dehydronaphtlylide as catalyst. The tapered block system was due to the higher reactivity of the carbonate than the lactone [89].

C. Polymers with Polythioether Block

The chemistry of anionic polymerization of propylene sulfide was extensively studied by Boileau and Sigwalt [90–99]. Propylene sulfide gives polymers with perfectly stable thiolate living ends in ethereal solvents. Edmonds patents [100] disclosed the preparations of homopolymer of trimethylene sulfide (thiacyclobutane) and copolymers of styrene and trimethylene sulfide with alkali metals or organoalkali metal compounds such as BuLi. The copolymerization process involving the direct polymerization of the two monomer (one is a vinyl and the other



is a hetrocyclic compound) is intriguing. The copolymers' sequence distribution was not disclosed; nor was there any discussion of the nature of the propagating species.

Morton and his co-workers [101–106] studied the initiation and propagation reactions of ethylene sulfide and thiocyclobutane initiated with BuLi. These authors identified the active living ends as shown in Scheme 24.17.

Scheme 24.17

The two cyclic sulfides have different reaction pathways with alkyllithium compounds. Thiirane (three-membered ring), such as ethylene sulfide, gives thiolate species and thietanes (four-membered ring) give carbanions. The thiolate end groups cannot initiate the polymerization of a vinyl monomer. The carbanions from the reactions of thietanes and alkyllithium compounds are active toward both vinyl and cyclic sulfide monomers. This is an important consideration when one is designing a scheme to prepare block polymers of certain block sequences by sequential monmer addition. The formation of carbanion from trimethylene sulfide and RLi also explains Edmonds earlier copolymerization results [100].

Morton et al. prepared a series of diblock and triblock copolymers using RLi, ROLi, RSLi, and R(Li)2 as initiator and by sequential monomer addition. Of particular interest are ABA triblock polymers having rubbery polydience center block and crystalline end blocks of polythiabutane or polythiapropane. The Tm for the former polysulfide is 55°C and the Tm for the latter polysulfide is ˜200°C. The triblock polymers with crystalline end blocks were compared with triblock polymers having glassy end blocks. In general, the ethylene sulfide block polymers were much weaker and less extensible thant the corresponding SBS or SIS polymers. The thiacyclobutane block polymers showed no measurable strength at all.

D. Polymers with Polysiloxane Block

Cyclic siloxanes (commonly designated as Dn, when D denotes the [-SiMe2O-] unit) undergo anionic polymerization in the presence of strong nucleophiles to



give a mixture of linear polymer with a broad spectrum of cyclic species. The polymerization is complicated by side reactions such as back-biting (intramolecular chain transfer) and intermolecular chain transfer. An overview of the polymerization of cyclosiloxanes [106] and a comprehensive review on the same subject [107] are excellent sources for detailed information on the chemistry of initiation, propagation, inter- and intramolecular chain transfers, cross-over on copolymerization, and other subjects.

Alkyllithium compounds react with hexa-methyllcyclotrisiloxane (D3) to form the linear adduct, R-(Si(CH3)2-O)3Li, but no polymerization occurred even in the presence of large excess of D3 in a hydrocarbon solvent. When a donor solvent like THF, diglyme, triglyme, DME, HMPA, or DMSO is added, a reasonably rapid polymerization starts to give a near monodispersed polymer. Cryptated lithium silanolate polymerizes D3 with a greatly enhance rate compared to that of other systems. The amount of cyclic byproducts is very low when the polymer yield reaches the maximum value [91,108].

The synthesis of well-defined sytrene-siloxane [109–113] and a-methylstyrene-siloxane [114] block polymers has been reported. Block polymers of methyl methacrylate and siloxane were prepared by polymerizing the acylate monomer first [109,115]. This procedure produced low siloxane incorporation in the block polymer and the product containing polysiloxane homopolymer, which

Scheme 24.18



was initiated by MeOLi as formed from the initiation and propagation reactions of methyl methacrylate. An improved procedure was developed by which siloxane was polymerized first [116]. The block polymer produced by this procedure showed a much greater siloxane incorporation and purer product. The sequence of reactions is illustrated in Scheme 24.18.

The steric hindered 1,1-diphenylmethyl carbanion is not very active toward cyclic siloxane monomer relative to the unhindered alkoxy anion in the first stage and the silanolate is inactive to initiate the polymerization of MMA in the second stage.

Hagen-Esch and Mason [117] used an acetal-protected initiator to form polymethyl methacrylate first and then generated lithium alkoxide sites from the initiator moiety for D3 polymerization as shown in Scheme 24.19.

Scheme 24.19

By using the same initiator and applying the same strategy, AB and ABA block polymers of 2-vinylpyridine and poly-dimethylsiloxane were prepared [118].

Yin and Hogen-Esch [119] reported the first synthesis of a well-defined monodisperse macrocyclic PS-PDMS block polymer. The linear PDMS-b-PS-b-PDMS precursor was prepared through initiation of styrene by lithium naphthalenide in THF at -78°C, followed by addition of hexamethylayclotrisiloxane (D3) to the living Li(PS)Li. The cyclization was carried out by reaction with Me2SiCl2 in high dilution. With higher MW polystyrenes, side reactions were observed. These side reactions could be prevented by capping the Li(PS)Li with 2,2,5,5-tetramethyl-2,5-disila-1-oxacyclopentane (EDS) before the addition of D3. The capping reaction removed the slow initiation of D3 by PSLi. With the



more active initiating end groups, D3 polymerized rapidly with essentially no side reactions. The linear triblock polymer (the precursor of the macrocyclic block polymer) was terminated with Me2SiCl. The macrocyclic block polymer based on the linear PDMS-b-PEDS-b-PS-b-PEDS-b-PDMS precursor was also synthesized. After coupling, this is a macrocyclic tetrablock. It was shown by proton [13C] and [29Si] NMR that there are distinct differences between a macrocyclic and its leaner block copolymers.

Scheme 24.20

E. Other Block Polymers

Polysilylene Block

Polystyrene-b-poly(methylphenylsilylene), PS-b-PMPS, polymer was prepared by the reaction of living polystyryl anions with chloro endcapped PMPS [120]. The same block polymer was synthesized by sequential monomer addition, styrene first, followed by Me4Ph4Si4 in benzene with sec-BuLi as initiator. 12-Crown-4 ether was introduced to the solution prior to the addition of the MPS monomer [121]. PI-b-PMPS polymer was also made in the similar manner. Sakamoto et al. made block copolymers of PMMA and [SiMe(n-Bu)SiMe2]n by subsequent addition of MMA to the active end groups of the polysilylene chains [122].

Polyether-b-Nylon 6

The anionic polymerization of caprolactam was first reported by Joyce and Ritter in 1941 [123]. The mechanism was first proposed by Hall [124], Motters [125], and Sebenda [126] independently. The polymerization catalyzed by metalcaprolactamate is characterized by a high rate of propagation but with a long induction period associated with the formation of the acyllatam growth center. The polymerization can be activated by the addition of suitable cocatalysts that essentially eliminate the induction period. The block copolymerization requires the attachment of cocatalyst moiety to polyether chain ends so that caprolactam chain-initiation can take place. Some mechanistic aspects and examples of the anionic block copolymerizations of caprolactam and polyether diols are discussed by Akkapeddi and colleagues [127].



Polystyrene (or Polydiene)-b-Polytetrahydrofuran

Direct reaction of a living anionic polymer with a living cationic polymer could lead to diblock, triblock, and multiblock copolymers (see Scheme 24.21) [128,129].

Scheme 24.21

Another methodology involves the coupling of two polymer chains via anionic linking groups. For example, the tertiary amine end group is introduced to an anionically polymerized polystyrene. The amine-terminated polymer can be isolated and redissolved in THF. Living cationically polymerized poly THF is allowed to react with the amine-terminated polystyrene to form the block copolymer. The linking reaction was reported to be very fast at room temperature and virtually quantitative [130]. The two examples in Scheme 24.22 do not

Scheme 24.22

involve anionic ring-opening polymerization. The cyclic monomer is polymerized cationically and the product is linked to the anionically polymerized vinyl polymer. They are included here as an illustration of an indirect method of preparing block polymers.

III. Block Polymers From Aluminum Porphyrin Initiators

Inoue and Ada developed a novel class of catalysts based on aluminum (also zinc) porphyrins for the polymerization of cyclic monomers [131]. For example, polymerization of oxirane initiated with a porphyrin aluminum compound proceeds via a porphinato aluminum alcoholate as the growing species, affording polyethers with narrow MWD (Scheme 24.23), in which the number of polymer



Scheme 24.23

molecules is in agreement with that of the initiator molecules because of the absence of a chain transfer reaction [132,133]. When a protic compound (HY) is added to the system, the number of the polymer molecules is increased proportionally to the amount of the added HY, retaining the narrow MWD [134]. In this case, a rapid reversible chain transfer occurred (Scheme 24.24).

Scheme 24.24

The transfer reaction is fully reversible and very rapid (severalfold faster than chain growth reaction). Furthermore, the reaction product has the ability to reinitiate a new chain (Scheme 24.25).

Scheme 24.25

Thus, this system still provides polymers with narrow MWD, but with the number of polymer molecules higher than the number of initiator molecules. Inoue and Aida considered the first system with no transfer reaction as living and the later system with rapid reversible transfer reaction as “immortal.” Several aluminum block polymers with and without a HY were prepared with these porphyrins [131,135]. Representative examples are shown in Table 24.1.



References

1. J. E. McGrath, Ed., Ring-Opening Polymerization: Kinetics, Mechanisms, and Synthesis, Am. Chem. Soc. Symp. Series, 286, Am. Chem. Soc., Washington, D.C., 1985.

2. K. J. Ivin and T. Saegusa, Eds., Ring-Opening Polymerization, Vol. 1–3, Elsevier, London, 1984.

3. D. J. Brunell, Ed., Ring-Opening Polymerization: Mechanisms, Catalysis, Structure, Utility, Henser, Munich, 1993.

4. J. E. McGrath, in Ring-Opening Polymerization: Kinetics, Mechanisms, and Synthesis, J. E. McGrath, Ed., Am. Chem. Soc. Symp. Series, 286, Am. Chem. Soc., Washington, D.C., 1985.

5. D. J. Brunell, in Ring-Opening Polymerization: Mechanisms, Catalysis, Structure, Utility, Henser, Munich, 1993.

6. M. J. Bowden, in Introduction to Polymer Science, F. A. Bovey and F. H. Winslow, Eds. Academic Press, New York, 1979, Ch. 2, Sec. 3.

7. S. Slomkowski and A. Duda, in Ring-Opening Polymerization: Mechanisms, Catalysis, Structure, Utility, D. J. Brunnell, Ed., Henser, Munich, 1993.

8. P. J. Flory, Nature (London), 62, 1561 (1940).

9. S. Inoue and T. Aida, in Ring-Opening Polymerization, K. J. Ivin and T. Sawgusa, Eds., Vol. 1, Elsevier, London, 1984, p. 185.

10. P. Sigwalt and N. Sparsky, in Ring-Opening Polymerization, K. J. Ivin and T. Saegusa, Eds., Vol. 2, Elsevier, London, 1984, p. 603.

11. S. Boileau, in Comprehensive Polymer Science, G. Allen, J. C. Bevington, and G. C. Eastmond, Eds., Pergamon Press, Oxford, 1986, Vol. 3, part 1, p. 467.

12. S. Penczek and S. Slomskowski, in Recent Advances in Anionic Polymerization, T. E. Hogen-Esch and J. Smid, Eds., Elsevier, New York/Amsterdam/London, 1987, p. 275.

13. G. L. Brode and J. V. Kolesks, Macromol. Sci., A6(6), 1109 (1972).

14. T. Oukadi, Ch. Stevens, and Ph. Teyssie, Makromol. Chem. Suppl., 1, 191 (1975).

15. J. M. Vion, R. Jerome, Ph. Teyssie, H. Aubin, and R. Prudhomme, Macromolecules, 19, 1828 (1986).

16. A. Duda, Z. Florjanczyk, A. Hofman, S. Slomkowski, and S. Penzek, Macromolecules, 23, 1640 (1990).

17. Ph. Doubois, L. Barakat, R. Jerome, and Ph. Teyssie, Macromolecules, 26, 4407 (1993).

18. Ph. Doubins, R. Jerome, and Ph. Teyssie, Polym. Chem. Washington, D.C., 35, No. 2, 536 (1994).

19. S. Penczek, A. Duda, and T. Brela, Poly. Preprints, Am. Chem. Soc., Washington, D.C., 35, No. 2, 508 (1994).

20. Y. Kawakami, S. Salbara, F. Suzuki, T. Abe, and Y. Yamashita, Polym. Bull., 25, 521 (1991).



24. C. J. Chang, R. F. Kiesel, and T. E. Hogan-Esch, J. Am. Chem. Soc., 95, 8446 (1973).

25. L. E. St. Pierre, and C. C. Price, J. Am. Chem. Soc., 78, 3432 (1956).

26. F. Cubasso and A. Zilkha, J. Macromol. Sci.-Chem. A8(8), 1313 (1974).

27. Y. Guilbert and J. Brossas, Poly. Bull., 1, 293 (1979).

28. D. H. Richards and M. Sczwarc, Trans Faraday Soc., 55, 1644 (1959).

29. K. S. Kazanskii, A. A. Solovyanov, and S. G. Eukelis, Eur. Poly. J., 1, 142 (1971).

30. M. C. Barker and B. Vincent, Colloid Surf., 8, 289 (1984).

31. R. P. Quirk, and N. S. Seung, in Ring-Opening Polymerization: Kinetics, Mechanisms, and Synthesis, J. E. McGrath, Ed., Am. Chem. Soc. Symp. Series, 286, Am Chem. Soc., Washington, D.C., 1985.

32. D. W. Koetsier, A. Bantjes, J. Feijen, and D. J. Lyman, J. Polym. Sci.-Chem., 16, 511 (1978).

33. G. Finciz, P. Rempp, and J. Parrod, Bull. Soc. Chem. Fr., 262 (1962).

34. J. J. O'Malley and R. H. Marchesault. Macromol. Synth., 4, 35 (1972).

35. Z. Hruska, G. Hurtrey, S. Water, and G. Riess, Polymer, 33(11), 2447 (1992).

36. P. Marie, J. P. Lingelser, and Y. Gallot, Makromol. Chem. 183(12), 2961 (1982).

37. J. Wang, S. K. Varshney, R. Jerome, and Ph. Teyssie, J. Polym. Sci., A, Polym. Chem., 30, 2251 (1992).

38. H. Reuter, I. V. Berlinova, S. Horing, and J. Ulbricht, Eur. Polym. J., 27(7), 673 (1991).

39. Y. Deng, C. Price, and C. Booth, Eur. Polym. J., 30(1), 103 (1994).

40. G. Riess, J. Nervo, and D. Rogez, Polym. Eng. Sci., 17, 634 (1977).

41. B. Lotz and A. J. Kavacs, Z. Z. Kolloid, Polym., 209, 115 (1966).

42. H. Q. Xie, and P. G. Zhou, in Multicomponent Polymer Materials, Adv. Chem. Ser. 211, D. R. Paul and C. H. Sparling, Eds., Am. Chem. Soc., Washington, D.C. 1986, p. 139.

43. J. Kelly, W. M. Mackenize, D. C. Sherington, and G. Riess, Polymer, 20, 1048 (1979).

44. H. Q. Xie and P. G. Zhou, Polym. Eng. Sci., 25, 32 (1985).

45. H. Q. Xie and J. Xia, Makromol. Chem., 188, 2543 (1987).

46. H. B. Gia, R. Jerome, and Ph. Teyssie, J. Polym. Sci. Chem. Ed., 18, 3483 (1980).

47. W. H. Sharkey, in Ring-Opening Polymerization: Kinetics, Mechanisms, and Synthesis, J. E. McGrath, Ed., Am. Chem. Soc. Symp. Series, 286, Am. Chem. Soc., Washington, D.C., 1985.

48. W. H. Sharkey, US patent 3,557,255.

49. R. P. Foss, US patent 3,821,331.



57. R. C. Tham, W. H. Buck, S. W. Caywood, J. M. Meyer, and B. C. Anderson, Macromol. Chem., 58/59, 345 (1977).

58. Y. Yamashita and H. Toshioki, J. Polym. Sci., Polym. Chem. Ed., 11(2), 425 (1973).

59. Y. Yamashita, Polym. Preprints, Am. Chem. Soc., Div. of Polym. Chem., Washington, D.C., 21(1), 51 (1980).

60. Y. Yamashita, T. Tsuda, H. Ishida, A. Uchikawa, and Y. Kuriyama, Makromol. Chem., 113, 139 (1968).

61. H. K. Hall, Jr., Macromolecules, 2, 488 (1969).

62. R. Thiebaut, N. Fisher, Y. Etienne, and J. Coste, Ind. Plast. Mod. (Paris), 14(2), 13 (1960).

63. P. Kurcok, Z. Zedlinski, and M. Kowalczuk. J. Org. Chem., 58, 4219 (1993).

64. P. Kurcok, M. Kowalczuk, H. Hennek, and Z. Jedlinski, Macromolecules, 25, 2017 (1992).

65. J. Dale and J. Schwartz, J. Acta. Chem. Scand, B40, 559 (1986).

66. H. R. Kricheldorf and N. Schwamagl, Macromol. Sci., Chem., A26, 951 (1988).

67. A. Hofman, S. Slomkowski, and S. Penczek, Macromol. Chem., 185, 91 (1984).

68. R. H. Young, M. Matzner, and L. A. Pilato, in Ring-Opening Polymerization, T. Saegusa and E. Goethals, Eds., Am. Chem. Soc., ACS Symp. Series, 59, Washington, D.C., 1977, Ch. 11.

69. Y. Yamashita, in Anionic Polymerization: Kinetics, Mechanisms, and Synthesis, J. E. McGrath, Ed., ACS Symp. Series 166, Am. Chem. Soc., Washington, D.C., 1981, Ch. 14.

70. H. L. Hsieh, J. Appl. Polym. Sci., 22, 1119 (1978).

71. F. X. Mueller, Jr., and H. L. Hsieh, US patents 3,585,257 and 3,639,519.

72. H. L. Hsieh, F. X. Mueller, Jr., and W. H. Busler, US patent 3,764,639.

73. F. E. Naylor, US patent 3,598,799.

74. E. Clark and C. W. Childers, J. Appl. Polym. Sci., 22, 108 (1978).

75. D. R. Paul, in Polymer Blends, D. R. Paul and S. Newman, Eds., Academic Press, New York, 1978, Vol. 2, Ch. 12.

76. D. R. Paul and J. W. Barlow, J. Macromol. Sci., Rev. Macromol, Chem., C18(1), 109 (1980).

77. H. L. Hsieh and I. W. Wang, in Ring-Opening Polymerization: Kinetics, Mechanisms, and Synthesis, J. E. McGrath, Ed., Am. Chem. Soc. Symp. Series, 286, Am. Chem. Soc., Washington, D.C., 1985.

78. Ph. Teyssie, J. P. Bioul, A. Hamitou, J. Heuschen, L. Hocks, R. J. Jerome, and T. Oukadi, in Riing-Opening Polymerization, T. Saegula and E. Goethals, Eds., ACS Symp. Series 59, Am. Chem. Soc., Washington, D.C., 1977. Ch. 12.

79. X. D. Feng, C. X. Song, and W. Y. Chen, J. Polym. Sci., Polym. Lett. Ed., 21(8), 593 (1983).



83. C. Jacobs, P. Dubois, R. Jerome, and Ph. Teyssie, Macromolecules, 24(1), 3027 (1991).

84. J. Zbigniew, M. Kowalczuk, P. Macek, L. Brzoskowska, and J. Franek, Makromol. Chem., 188(7), 1575 (1987).

85. Z. Jedlinski, P. Kurcok, M. Kowalczuk, and J. Frank, Spec. Publ.-R. Soc. Chem., 109 (Biodegradable Polymers and Plastics), 259 (1992).

86. P. Kurcok, J. Penczek, J. Franek, and J. Zbigniew, Macromolecules, 25(9), 2285 (1992).

87. H. Keul, H. Hoecker, E. Leitz, K. H. Ott, and L. Morbitzer, Makromol. Chem., 191(9), 1975 (1990).

88. Z. Jedlinski, P. Kurcok, W. Walach, H. Janeczek, and I. Radecka, Makromol. Chem., 194(6), 1681 (1993).

89. H. Keul, H. Hoecker, E. Leitz, E. Ott, K. Heinz, and L. Morbitzer, Makromol. Chem., 189(10), 2303 (1988).

90. S. Boileau, G. Champetier, and P. Sigwalt, Makromolek. Chem., 69, 180 (1963).

91. S. Boileau, in Anionic Polymerization: Kinetics, Mechanisms, and Synthesis, J. E. McGrath, Ed., ACS Symp. Series 166, Am. Chem. Soc., Washington, D.C., 1981, Ch. 19.

92. P. Hemery, S. Boileau, P. Sigwalt, and B. Kaemp, J. Polym. Sci. (B), 13, 49 (1975).

93. P. Hemery, S. Boileau, and P. Sigwalt, J. Polym. Sci., Polym. Symp., 52, 189 (1975).

94. P. Sigwalt and S. Boileau, J. Polym. Sci., Polym. Symp., 62, 51 (1978).

95. J. C. Favier, S. Boileau, and P. Sigwalt, Eur. Polym. J., 4, 3 (1968).

96. P. Guerin, P. Hemery, S. Boileau, and P. Sigwalt, Eur. Polym. J., 7, 953 (1971).

97. P. Guerin, S. Boileau, and P. Sigwalt, Eur. Polym. J., 7, 1119 (1971).

98. P. Hemery, S. Boileau, and P. Sigwalt, Eur. Polym. J., 7, 1581 (1971).

99. G. Tersac, S. Boileau, and P. Sigwalt, Makromol. Chem., 149, 153 (1971).

100. J. T. Edmonds, Jr., US patents 3,320,217 and 3,328,361.

101. R. Kammereck, L. Fetters, and M. Morton, Polym. Preprints, Am. Chem. Soc., Div. of Polym. Chem., Washington, D.C., 11(1), 72 (1970).

102. M. Morton and R. F. Kammereck, J. Am. Chem. Soc., 92, 3217 (1970).

103. M. Morton, R. F. Kammereck, and L. J. Fetters, Macromolecules, 4, 11 (1970).

104. M. Morton, R. F. Kammereck, and L. J. Fetters, Brit. Polym. J., 3(3), 120 (1971).

105. M. Morton and S. L. Mikesell, J. Macromol. Sci. Chem., 7(7), 1391 (1973).

106. J. E. McGrath, J. R. Riffle, A. K. Banthia, I. Yilgor, and G. L. Wilkes, in Initiation of Polymerization, F. E. Bailey, Jr., Ed., ACS Symp. Series 212, Am. Chem. Soc., Washington, D.C. 1983, Ch. 13.



INDEX

A

Abrasion resistance, 456

Acidity constants, 33

carbon acids,

aqueous solution, 37, 38

cyclohexylamine, 41, 42

dimethyl sulfoxide solution, 38-41

electrochemical determination, 42

gas phase, 33-37

tetrahydrofuran, 41

solvent effects, 34-37, 41

monomers,

conjugate acids, 100

Acrylate, alkyl, 98

t-butyl acrylate, 101, 327, 643, 659, 663, 674

Acrylonitrile-butadiene-styrene (ABS) resins, 542

Adhesives, applications, 565-571

compounding additives, 563-564

[Adhesives, applications]

morphology and viscoelastic behavior of, 565-566

solubility and solution properties of, 555-563

stabilizers, 564-565

Aggregation of alkyllithiums, 13-17

allyllithium, 21

benzyllithium, 15



[Aggregation of alkyllithiums]

polar solution, 15-17

Table 1.4, 16

polymeric organolithiums

Table 1.5, 20

disagreement, 160-162, 249

stereochemistry effects, 207

poly(styryl)lithium, 15

pyramidal inversion in, 57, 61, 62

variables, 15

Aldehyde functionalization, 277

Alkali metals, 5, 7

alkoxide derivatives

additives, 116

copolymerization initiators, 110

metalation grafting, 374

initiators, 103

organoalkali compounds

allyl derivatives, 22, 23

x-ray structures, 8

Alkoxides

copolymerization additives, 110, 318

initiators, methacrylate, 648

kinetic effects propagation, 167-169

methacrylate polymerization, 654, 660

stereochemistry effects, 226

Alkylene oxides (epoxides, oxiranes)

as functionalization agent, 267, 632

as monomer, 688-692, 704-705



Alkylene sulfides

as monomer, 696-699

Alkyllithium compounds

aggregation

hydrocarbon solutions, 13-15

polar solutions, 15-17

[Alkyllithium compounds]

C-Li bonding, 5

bond lengths, 10-13

chiral, 59-61

configurational inversion, 56, 57

1H NMR studies, 61

13C NMR studies, 62

initiators, 108-110

relative reactivity, 109

stability, 174

structural models, 9

x-ray crystal structures, 8-13

Alkyllithium/alkali metal alkoxide complex, 409, 468, 609-610

(see also Super base)

Allyl carbanion, 19

charge distribution, 212

rotation barrier, 22

Allyllithiums, 19-24

association, 160


initiators, 677

1H and 13C NMR, 21, 23

rotation barrier, 21, 22



Association (see Aggregation)

Ate complexes, 143

initiators, 143-146, 650

B

Barium alkoxides diene microstructure, 228

Barium-based catalysts, 469

Benzyl carbanion, 17

Benzyllithium association, 15

13C NMR, 18

J13C-H solvent dependence, 63

1H NMR, 19

rotation barrier, 63

UV-visible spectra, 19

x-ray structures, 17, 18

BDS resins (clear, medium-impact butadiene/styrene resins)

appearance of, 589

block architecture of, 601-603

blow molding, 599

blown film, 600

chemical resistance of, 591-592

hinge flex life of, 591

impact strength of, 589-591

injection molding, 596-599

introduction, 413, 587-589

mechanical properties of, 591

medical applications of, 600-601

polymer blends, 592-594

processibility of, 594-596



stress crack resistance of, 591-592

thermoforming, 600

Bidentate ligands

effects on diene stereochemistry, 217, 218

Bifunctional lithium initiators, 412, 484-486, 623-624, 627-628, 693 (see also Difunctional and dilithium initiators)

Bismorpholinoethane (BME)

diene stereochemistry, 217, 218

1,3-Bis(1-phenylethenyl)benzene

difunctional initiator, 113,317

difunctional linking agent, 354-360

1,2-Bispiperidinoethane (DIPIP)



2,2' -Bis(4,4,6-trimethyl-1,3-dioxane) (DIDIOX)


Blends

medium-cis polybutadiene and natural rubber, 432

medium-cis polybutadiene and SBR 1712, 437

medium-vinyl polybutadiene and SBR 1712, 432

tapered S/B block copolymer and natural rubber, 452

tapered S/B block copolymer and emulsion SBR, 452

Block copolymers, 307

t-butyl acrylate, 674

control, living polymerization, 74

criterion, living polymerization, 84

diblock, 447-453

from aluminum porphyrin initiators, 703-705

Lewis base addition, 312

made by ring-opening polymerization, 688-703



[Block copolymers]

nomenclature, 307, 308, 476-477

polar monomers, 321, 324

polyether-b-Nylon 6, 702

SIBR, 467-469

star (radial), 463, 477, 484-490, 691, 696, 698

styrene/diene, 321, 322

tapered, 242, 246, 251

triblock (A-B-A type), 696, 698, 410-411, 475-491

triblock synthesis

difunctional initiators, 316-319

three-step, 309-313

two-step, 313-316, 319

with polyester block, 692-698

with polyether block, 688-692, 702

with polysiloxane block, 699-702

with polytetrahydrofuran block, 703

with polythioether block, 698-699

Branching, 402-404, 436-438, 463, 484-491

poly(butadienyl)lithium, 180

Branch polymer definition, 333

1,3-Butadiene, 428-429, 461

macromonomer copolymerization, 385

t-Butyl acrylate, 643

block copolymers, 327, 674

lithium chloride effects

kinetics, 663

polymerization, 659

t-Butyl methacrylate



kinetics, polymerization, 660

tetrahydrofuran, 659

[t-Butyl methacryalate]

toluene, 655

sec-Butyllithium

dibutylmagnesium complex, 145

retention of configuration, 60

thermal decomposition, 174, 175

tert-Butylstyrene, 491

t-Butyllithium/diethylzinc, 144

t-Butyllithium/trialkylaluminum methacrylate initiators, 650

n-Butylsodium/dibutylmagnesium, 145

C

ε-Caprolactone as monomer, 696-698

Carbanion

definition, 3, 33

enantiomeric forms, 56

stability

aqueous solution, 37, 38

cyclohexylamine, 41, 42

dimethylsulfoxide, 38-41

gas phase, 33-37

pyramidal inversion, 56

stereochemistry, 56

tetrahydrofuran, 41

Carbon acid, 33

Carbon acidity, 33

aqueous solution, 33-37



[Carbon-lithium bond]

bond lengths, 10-13

Ceiling temperature, 83, 107, 492, 676

Chain-end functionalization, 261


control, living polymerization, 75

functionalized initiators, 75

Chain-extension, 629-632, 634-636

Chain reaction

definition, 72

polymerization, 72

Chain transfer, 182

agents, 189

allenes, alkynes, 190

constants, 185

definition, 73

Lewis base effects, 188

reversible, 89

to monomer, 192

1,3-cyclohexadiene, 192

p-methylstyrene, 192

to solvent

alkenes, 189

ammonia, 183

ethylbenzene, 189

toluene, 184,189

Charge distribution, 59-61

benzyllithium, 19

methyllithium, 5



poly(styryl)lithium, 19

Chelating Lewis bases

diene microstructure, 218

N-chelated alkyllithium, 610-615

Chemical modification, 615 (see also Hydrogenation)

Chiral alkyllithiums

sec-butyllithium, 60

2-methylcyclopropyllithium, 60

2-octyllithium, 59

[Chiral alkyllithiums]

2,2-diphenyl-1-methylpropyllithium, 60

2-benzyl-1-(1,3-dioxobutyl)-propyllithium, 61

Chlorosilane(s) as linking agent, 463, 487-488, 491

Common ion effects

free ion formation, 54

ion pair equilibria, 39, 50

kinetics, 54

Cold flow, 400, 402, 404

Compositional heterogeneity

copolymers, 237, 240, 246

graft copolymers, 370, 376

Configurational inversion

neohexyllithium, 61

2-methylbutyllithium, 61

Contact ion pair, 48 (see also Tight ion pair)

fluorenyl anions, 51, 52

Continuous process, 465

Controlled polymer synthesis, 73

Coordinated anionic initiators, 688, 698



[Copolymerization]

monomer reactivity ratios, 247

styrene/butadiene, 240-246

styrene/isoprene, 240

Copolymerization equation, 237, 138

macromonomers, 382, 385

Copolymers

compositional heterogeneity, 237, 240, 246

structure analysis, 255

tapered block, 242, 246, 251

Coulomb's law, 47

Counterion effects

diene stereochemistry, 199


thermodynamic parameters, 54

UV-visible spectra, ion pairs, 52

Coupling agents, 315

Coupling reactions

graft copolymer synthesis, 379

triblock copolymer synthesis, 313

Cross-association, RLi, 135

kinetic effects, 135, 138, 139, 165-168

Crown ethers, 688, 702

ion pairing effects, 39

methacrylate polymerization, 659

Cryptands

effect on diene stereochemistry, 218

effect on ion pairing, 39, 52

external solvation, 53




Cumyl cesium, 656

Cumyl potassium, 114

Cyanoacrylates, alkyl, 99

Cyclic siloxane(s) as monomer, 699-702

1,3-Cyclohexadiene

chain transfer, 192

Cyclohexane, 429, 462

Cyclohexylamine, 40

dielectric constant, 41

ion pair acidity, 41, 42

D

Decomposition

alkyllithiums, 174

Delocalization, effects on

acidity, 36


solvent interactions, 41

Dibutylmagnesium, 144, 229

Dielectric constant

cyclohexylamine, 41

dimethoxyethane, 51

dimethylsulfoxide, 38

effect on ion pairing, 48, 52

temperature dependence, 54

tetrahydrofuran, 41

tetrahydropyran, 51

1,3-Dienes, 97



m-Diisopropenylbenzene, 111, 112, 320

copolymerization, 377

graft copolymers, 377

Dilithium initiators, 113 (see also Difunctional initiators)

Dimethylsulfoxide

acidity scale, 38-41


pKa, 39

1,1-Diphenylethylene

4,4-bis(trimethylsilyl), 656

carbonation, 264

copolymerization, 246, 289

divinylbenzene copolymers, 378

functionalization general method, 279

functionalized initiators, 646, 676, 677

grafting, 380

initiators, 645

macromonomers, 360-365, 387

RLi addition, 133-135

sulfonation, 276

1,1-Diphenylhexyllithium, 101

methacrylate initiator, 545, 653

1,1-diphenylmethyl carbanions initiators, 115,657,666

1,2-Dipiperidinoethane (see 1,2-Bispiperidinoethane)

Divinylbenzene(s) (DVB), 409, 463, 488, 491, 627

copolymerizations, 343-344, 378

difunctional initiators, 112

1,1-diphenylethylene copolymers, 378

linking reactions, 335-343



heteroarm star polymers, 346-347,

multifunctional initiators, 344

Domain(s), 478-479, 494, 499

[Domain(s)]

size, 449

Donor solvent(s), 700-701

E

Electron deficient compounds, 8

bonding, 10

Enolate anions, 24-25

stability, 25

stereochemistry, 58

Equilibrium polymerizations, 87

Ester enolate anions, 25-26

initiators, 115, 647, 660

stability, 25

Ether solvents

side reactions, 102, 103

Ethyl anion

stability, 4

Ethylenation, 131-133

Ethylene, 94, 133

Ethyl vinyl acetate, 514

Experimental procedures, 113

F

Fluorenyl anions

initiators, methacrylate, 647



[Free ions]

methacrylate polymerization, 655,656

Footwear, 452, 467, 571-582

Functionalization, 261, 624, 628-629,632-633

alkyl methacrylate, 675-677

amination, 271, 283

carbonation, 263

carboxyl-group, 286

functionalized initiators, 262, 266, 268, 272, 280, 676

general reactions, 278

1,1-diphenylethylenes, 279

silyl halides, 279

hydroxylation, 267

oxidation, 277

phenol-group, 281, 287

sulfonation, 275

Functionalized initiators gelation, 262, 266, 268, 272, 280

α,ω-dilithium

functionalization, 263

G

Glass transition temperature (Tg)

polybutadienes, 222, 223, 433-435, 459-460

poly(methyl methacrylate), 672

SIBRs, 467-469

solution SBRs, 457-459

triblocks, 477, 491-493

GPC (SEC), 401, 402, 403, 410, 488, 489, 496, 626, 633, 634



Graded block copolymers (see Tapered block copolymers)

Graft copolymers, 413-416

definition, 369

model systems, 386-388

synthesis methods

by addition, 377-379

by coupling, 379-381

[Graft copolymers]

macromonomer procedure, 381-388

metalation grafting, 372-377

Grafting

efficiency, 374

from, 371-377

onto, 371, 379-381

through, 371, 381-386

Green strength (raw strength), 441, 465, 479

Grignard reagents

methacrylate initiators, 649

H

Hammett rho

1,1-diphenylethylene, RLi addition, 135

solvent effects, 36

styrene, RLi initiation, 152

Heteroarm star polymers

1,1-diphenylethylene linking, 353-365

divinylbenzene linking, 346-347

macromonomer synthesis, 387

silyl halide coupling, 350-353

Heterocyclic monomers, 94



[Hydrocarbon acidity]

gas phase, 33-37

tetrahydrofuran, 41

Hydrogenation, 470, 491-494

Hydroxylation, 267

I

Immortal ring-opening polymerization, 693-705

Impact polystyrene, 428, 431

Indicator method

carbon acidity, 38

Initiation

definition, 72

induction periods, 137

kinetic orders, 138

kinetics, styrene, 136, 156


mechanism

model reactions, 131-135

solvent effects, 139

Initiators

alkali metals, 103

alkyllithium, 108-110

aromatic radical anions, 104

complete consumption, 155


difunctional, 111

functionalized, 75, 113, 262, 266, 268, 272, 280

methacrylate polymerization, 644-651



reactivity, 100, 146-148, 155

Intrinsic electronic effects, 42

Intrinsic structural effects, 35-37

Inversion barrier

methyl anion, 4

2-methylbutyllithium, 61

neohexyllithium, 61

7-phenylnorbornyllithium, 62

Ionic species

Winstein spectrum, 47

Ion pairs and ion pairing, 38

[Ion pairs and ion pairing]

contact, 48

definition, 48

detection, 39, 50, 51

dissociation constants, 54

effect on acidity, 39

equilibria, 49

fluorenyl anions, 50-54

loose, 48

solvent-separated, 48

tight, 48

variables

charge localization, 42

counterion, 39, 51, 52

crown ether, 39, 52, 53

cryptand, 39


solvent, 51, 52



[Lewis bases]

decomposition, 180, 181


ethylenation, 131, 132

initiation rates, 140

enthalpies of interaction

alkyllithiums, 16

polymeric organolithiums, 17

kinetic effects, 132, 140

Linking (coupling), 402-404, 436-438, 484-491

agents

star polymers, 335

Liquid polybutadienes

monofunctional, 607-619

telechelic, 625-637

Lithium alkyl isobutyrate initiators, 647

association, 662

kinetics, methacrylate, 664

Lithium alkoxide

aggregation, 143

copolymerization, effects on, 252


dilithium initiators, 318

isotactic polystyrene, 231

kinetic effects

acrylate polymerization, 664

initiation, 140-143


propagation, 165-168



thermal decomposition, RLi, 174

Lithium chloride, 101

acrylate polymerization, 327, 659, 664

methacrylate polymerization, 654, 660, 672, 673

Lithium hydride, 174

Lithium 2-(2-methoxyethoxy)-ethoxide, 101

Living linking reactions

star polymers, 346-347, 353-365

Living polymerization

definition of, 72

equilibrium polymerizations and, 87

experimental criteria, 76

block copolymer formation, 82

chain-end functionalization, 85

kinetics, 85-87

Mn control, 80

Mn versus conversion plot, 77

molecular weight distribution, 81

monomer conversion, 77

number of polymer molecules, 79

laboratory time scale, 88, 173

Loose ion pair, 48 (see also Solvent-separated ion pair)

Macromonomers (Macromers®) 413-416

alkyl methacrylate, 677-679

anionic copolymerization, 384

butadiene copolymerization, 385

copolymerization, 381-388

1,1-diphenylethylene, 360-365, 387

homopolymerization, 383



Menthyllithium, 137

Metalation grafting, 372

Metal-free initiators, 648

Metaloester initiators, 647, 656

Methacrylate, alkyl, 98

block copolymers, 673-675

t-butyl methacrylate, 101

functionalized polymers, 675-677

initiators, 644-651

polymerizable monomers, 642

polymerization, 101

star polymers, 675

stereochemistry, 670, 671

termination reactions, 667-669

Methyl anion

inversion barrier, 4

structure, 4

stability, 4

2-Methylbutyllithium

configurational inversion, 61

Methyllithium


structure, 5

Methyl methacrylate

butyllithium reactions, 645

functionalized initiators, 646

polymerization

kinetics, 652

rate constants, 660



stereochemistry, 101

termination reactions, 667-669

α-Methylstyrene, 491


equilibrium polymerization, 87

molecular weight distribution, 83

oligomers, initiators, 104, 107

solvent

triblock copolymer synthesis, 320

[α-Methylstyrene]

p-Methylstyrene

chain transfer, 192

Microstructure

BRs, 397-399, 422, 424-425, 438-439, 469

IRs, 398-399

liquid polybutadienes, 608, 612-613, 616-617

polydienes

grafting efficiency, 375

SBRs, 409,469

SIBRs, 467-468

telechelic prepolymers, 626

Modifications

of ABS resin, 542

of asphalt, 545-554

of polycarbonate, 543

of polyethylene, 537-540

of polyphenylene ether, 542-543

of polypropylene, 540-542

of polystyrene, 526, 534-537



Monomer conversion


Monomer reactivity ratios, 238, 239, 244, 247

Monomers

conjugate acid pKa, 100

functional groups, 95

heterocyclic, 99

polymerizability, 93

reactivity, 95, 100

substituents, 94, 96-99

Monomer sequence distributions, S/B, 449, 463-464

Multifunctional initiator, 409

divinylbenzene, 349

N

Neohexyllithium

configurational inversion, 61

3-Neopentylallyllithium, 23

association, 23, 161

chain-end configuration, 214

C-Li bonding, 63

1H NMR, 23

stereoisomers, 63

Neopentylmethylallyllithium

13C NMR, 220

O

2-Octyllithium



chirality, 59

retention of configuration, 59

Organometallic initiators

mixed, 143-146

Oxidation, 277

Oxidative degradation

polyisoprene graft copolymers, 374

tapered block copolymers, 242, 255

P

Pentamethyldiethylenetriamine


Phase separation, 478-479, 499

Phenol functionalization, 282, 287

Phosphazene base (P4)

initiator, 648

Physical crosslinking, 450, 479, 494

Piperylene, 442-443

Pivalolactone, 692-695

Polar additives, 407-409, 464-465, 608, 610-614, 688 (see Lewis bases)

Polarization

effect on acidity, 36

Poly(alkyl methacrylates)


Polybutadiene

glass transition temperature, 222, 223

microstructure

counterion effects, 199, 200


spectroscopic determination, 198



[Poly(butadienyl)lithium]

thermal stability, 177-180

UV spectra, 221

Poly(t-butyl acrylate)

block copolymers, 327

Poly(t-butyl methacrylate), 655

macromonomers, 678

star polymer, 675


Polycarbonate, 543

Polydiene

in styrenic TPE, 514

stereochemistry, 210,

Poly(dienyl)lithium


Polyethylene, 514, 537-540

Polyisoprene

microstructure

counterion effects, 200

grafting efficiency, 375

polar solvent effects, 212

spectroscopic determination, 198

Polyisoprene rubbers (IR)

by anionic process, 438-442

high-cis, 400

Poly(isoprenyl)lithium

association, 20

concentration effects, 162

controversy, 160-162




thermal stability, 177-180

Polymerization process

anionic, 426-430

cis-polybutadiene, 422-426

Poly(methyl methacrylate)

block copolymers, 327, 673

chain-end functionalization, 675

macromonomers, 675

star polymer, 675

[Poly(methyl methacrylate)]


lithium chloride effects, 670, 671, 673

Poly(α-methylstyryl)lithium

methacrylate initiator, 646, 669

thermal stability, 176

Poly(α-methylstyryl)sodium, 656

Polyphenylene ether, 542

Polypropylene, 514, 540-542

Polystyrene

blends with styrenic TPE, 534-537

blends with BDS resins, 592-594

high impact resins, 428, 431-432

impact, 428, 431


Table 9.14, 230

isotactic, 231

Poly(styryl)lithium

association, 15, 19, 20



[Propagation]

kinetic orders, 158, 159

Lewis base, 164

reactivity

styrenes vs. dienes, 163

salt effects, 165

Pyramidal inversion

alkyllithium, 57, 61, 62

allyl carbanion, 4-6

Pyridine, 101

R

Racemization

2-methyl-1-phenyl-1-butanone, 58

Radical anion initiators, 104-106, 110

Reaction intermediates

definition, 47, 72

Reaction orders, kinetics

initiation, 137


propagation, 157-163

Regiospecificity

isoprene addition, 204, 221

Reversible chain transfer, 89

Reversible termination, 88

Ring-opening polymerization

anionic, 686-688

block copolymers, 688-705

general, 685-686



immortal, 693-705

Rolling loss, 457

S

Salt effects

diene microstructure, 223

S-B-S triblock copolymers (see styrenic TPEs)

Schlenk equilibrium, 649

S-EB-S triblock copolymers, 491-493

Sheet molding compounds, 543-545

SIBRs, 467-469

Solid rocket binder, 412, 636

Solubility

alkyllithium compounds

hydrocarbon solution, 13

polar solutions, 15

Solution SBRs, 403-411, 453-467

Solvent effects

acidity constants, 35-37

delocalization, 36

ion pair equilibria, 52

linear free energy relationships, 36

methacrylate polymerization stereochemistry, 669-672

polarization, 36

reaction orders, kinetics

initiation, 137

propagation, 158

Solvents, 102, 103

Solvent-separated ion pair, 48, 52, 55 (see also Loose ion pair)



[Star-branched polymers (radial polymers)]

living linking reactions, 346-347

multifunctional initiators, 344-346

silyl halide linking, 347-350

Stereochemistry

polydiene

concentration effects, 201

counterion effects, 199

1,1-diphenylethylene linking, 353-360


macromonomers, 360-365

mechanism, 203


temperature effects, 202

poly(alkyl methacrylate), 669-673

lithium chloride effects, 673

polystyrene, 230-231

Stereoisomerism

alkyllithium, 57

Steric effects

cation solvation, 52

ion pair equilibria, 52

Styrene, 462

polymerization, 100

substituted, 96, 97

triblock copolymers, 319

Styrene-acrylonitrile resins (SAN), 542

Styrene-butadiene diblock copolymers

applications, 450-453



introduction, 447-448

properties, 448-450

Styrene-isoprene-butadiene rubbers (SIBRs), 466-469

Styrene/butadiene random copolymers (SBRs)

applications, 466-467

[Styrene/butadiene random copolymers (SBRs)]

commercial rubbers, 455-459

compounding, 465-466

introduction, 453-455

manufacturing process, 459-464

processing alternatives, 464-465

solution versus emulsion, 453-455

trans rubbers, 469

Styrenic thermoplastic elastomers (styrenic TPEs)

in ABS modifications, 542

in adhesives, 555-571

in asphalt modifications, 545-554

commercial products of, 503-512

compounding of, 513-519

domain size and domain structure of, 478-479, 494

environmental resistance and chemical resistance of, 508-510

features of, 525

in footwear applications, 571-582

general applications of, 524-527

history of, 476-483

hydrogenated, 491-493

introduction, 475

permeability of, 510-511

in polycarbonate modifications, 543



[Styrenic thermoplastic elastomers (styrenic TPEs)]

in polystyrene modifications, 526, 534-537

processing of, 519-524

producers of, 483

in SAN resins modifications, 542

in sheet molding compound modifications, 543-545

solution properties of, 501-502

structure/morphology of, 499-501

synthesis of, 484-491

tensile properties of, 494-499

thermal characteristics of, 504, 509

typical electrical properties of, 508, 511

typical mechanical properties of, 504, 507-508

viscoelastic properties of, 502

viscosity and rheological properties of, 510, 512

Sulfonation, 275

Superbases, 409, 468, 609-610

(see also Alkyllithium/alkali metal alkoxide complex)

Synthetic rubbers (elastomers)

classification of, 395-396

definitions of, 395

T

Tapered block copolymers, 242, 246, 251

Telechelic polymers

applications, 636

carboxy, 627-632

chain extension and crosslinking, 629-632, 634-636

hydroxy, 632-637



[Telechelic polymers]

high-vinyl, 626

introduction, 412, 621-623

prepolymers, 625-637

synthesis, 623-625

Telogen (chain-transfer agent), 608-615

Telomerization, 608-615

Termination

alkyllithium, 174

polar solvents, 180

t-butyl acrylate polymerization, 664

definition of, 72

methacrylate polymerization, 661, 667-669

kinetics, 664, 669

reversible, 88

Tetrabutylammonium dimethyl malonate

methacrylate initiator, 648

Tetrahydrofuran


ion pair acidity, 41

RLi decomposition, 102, 103, 181

N,N,N',N'-Tetramethyl-1,2-diaminocyclopentane (TMDC)


N,N,N',N'-Tetramethylethylene-diamine (TMEDA), 468, 609-615



Thermal decomposition

alkyllithiums, 174

poly(butadienyl)lithium, 177



Tight ion pair (see also contact ion pair), 48

Time living polymerization, 88, 173

Tire applications, 400, 626, 432-438, 455-459

Transmetalation, 609-610

Triblock copolymer synthesis

dilithium initiators, 316

three-step, 309

two-step


coupling, 313

V

Vinylcyclopentane units polybutadiene, 222

Vinyl microstructure

chain end concentration effects, 208

DIPIP effects, 221

glass transition temperature, 223

TMEDA effects, 221

Vinylpyridine, 97

polymerization, 101

Viscosity

effect of long-chain branching, 402, 404, 411, 555-563

Newtonian, 402, 404

non-Newtonian, 402, 404

Viscosity index-improving polymers, 412-413, 469-472

W

Wet traction, 456



nlreader.dll?bookid=12873&filena...

Documents