ANIONIC POLYMERIZATION AND CHAIN-END FUNCTIONALIZATION CHEMISTRY
ABSTRACT
Anionic polymerization, especially alkyllithium-initiated polymerization of styrenes and dienes, is a truly living polymerization that proceeds in the absence of the kinetic steps of chain termination and chain transfer. The major discoveries in the science and technology of anionic polymerization are chronicled herein. My introduction to this fascinating science is also described. This includes research and training at Phillips Petroleum Company (1974) with Dr. Henry Hsieh and a sabbatical leave at The University of Akron (1976–1977) with Professors Maurice Morton and Lewis J. Fetters. Also detailed is the initiation of my formal anionic polymerization research career at Midland Macromolecular Institute (MMI) in 1979 with outstanding colleagues Drs. Dale Meier, Karl Solc, and Hans Georg Elias. At MMI, I started research on the use of living, alkyllithium-initiated polymerization to prepare chain-end functionalized polymers. This developed into one of my most important research areas. Based on my research experience and publications at MMI, I was appointed Professor of Polymer Science at The University of Akron in 1983. This was the most significant development in my professional career. The University of Akron offered a graduate polymer research program with outstanding, world-class colleagues and facilities; excellent graduate students; a moderate, primarily graduate teaching responsibility; a supportive university administration; and an international reputation in polymer research and education. As described herein, at The University of Akron I was able to develop a comprehensive research program on anionic polymerization, especially in the area of functionalized elastomers, which was the basis for the American Chemical Society Rubber Division Award of the Goodyear Medal to me. A summary of the many research contributions of my outstanding graduate students, visiting scientists, and postdoctoral associates is included.
I. UNIQUENESS OF ANIONIC POLYMERIZATION
The science of anionic polymerization, especially alkyllithium-initiated polymerization, for vinyl and diene monomers is a fascinating subject and has been my main research interest. In 1956, Szwarc and coworkers1,2 elucidated the concept of living anionic polymerization. Living polymerizations are chain reaction polymerizations that proceed in the absence of chain termination and chain transfer reactions to provide polymer chains that quantitatively retain their reactive chain ends when all of the monomer has been consumed. This was preceded by the remarkable discovery by Stavely and coworkers3,4 at Firestone Tire and Rubber Company (Stavely, Goodyear Medal, 1972) that the use of lithium metal to initiate the polymerization of neat isoprene resulted in an elastomer that possessed many of the attributes of NR, that is, high cis-1,4-microstructure and stress-induced crystallization. The contrasting behavior of other alkali metal counterions (e.g., Na+, K+, Ru+, and Cs+) compared with the lithium counterion is that in hydrocarbon solution lithium is unique in producing polydienes, especially polybutadiene and polyisoprene, with high 1,4-microstructure, whereas the other counterions produce polybutadienes and polyisoprenes with high 1,2- or 4,3-microstructure, respectively.5–7 As a consequence, lithium-based polydienes have low glass transition temperatures (Tg; e.g., from −60 to −90 °C), whereas other alkali-based dienes have much higher Tg values.8,9 The Tg has a dramatic effect on physical properties of elastomers and compounded rubbers. These developments were followed by the groundbreaking investigation of the use of hydrocarbon-soluble alkyllithium compounds in hydrocarbon solution to prepare high 1,4-polyisoprene by Hsieh and Tobolsky.10 During this period, the obvious commercial potential of this technology was being explored by almost all of the synthetic rubber companies (e.g., Goodyear Tire & Rubber Company, Firestone Tire & Rubber Company, B.F. Goodrich Company, Shell, General Tire, and Phillips Petroleum). Many of these companies were involved in the U.S. Synthetic Rubber Program during World War II.11
One of the attributes of living anionic polymerization that was illustrated by Szwarc and coworkers12 was the fact that after all of the monomer has been consumed, the reactive anionic chain end can initiate the polymerization of another monomer to generate a block copolymer, for example, polystyrene-block-polyisoprene or polystyrene-block-polybutadiene. These reactive chain ends can also be reacted with electrophilic reagents to generate chain-end functionalized polymers.12 Chain-end functionalization chemistry developed into a major focus of our research throughout my career.13 After extensive investigations, companies such as Shell and Phillips discovered that triblock copolymers such as polystyrene-block-polyisoprene-block-polystyrene (SIS) and polystyrene-block-polybutadiene-block-polystyrene (SBS) exhibited the properties of vulcanized elastomers (long-range, reversible extensibility) with the ability to be processed in the melt like thermoplastics; that is, they discovered thermoplastic elastomers (TPEs).14 Thus, the 1950s and 1960s were exciting times for both academic and industrial scientists working in the area of alkyllithium-initiated polymerizations.
II. POLYMER EDUCATION AT THE UNIVERSITY OF ARKANSAS AND PHILLIPS PETROLEUM COMPANY
My exposure to this exciting field began with my first academic appointment as an Assistant Professor of Chemistry at the University of Arkansas in 1969. One of my first research projects with my first graduate student, Dennis Kester, was to investigate how Lewis bases such as amines and ethers interact with organolithium compounds in hydrocarbon solution by using high-dilution solution calorimetry.15–17 We used techniques developed as a postdoctoral research associate with Professor Edward M. Arnett at Mellon Institute and the University of Pittsburgh. One of the tenets governing successful academic research is that it is necessary to secure funding. Upon investigating the literature on applications of organolithium compounds, I discovered that alkyllithium compounds were used as polymerization initiators for both commercial applications and fundamental academic research. Unfortunately, my academic training at the University of Illinois had not exposed me to polymerization chemistry. To learn about this subject, I convinced the department chair to allow me to teach an undergraduate/graduate course on polymer chemistry. Since I was learning about polymer science at the same time, I was generally only one lecture ahead of the students! However, because of the excitement and challenges of this new subject, I never looked back in my quest for further knowledge on polymerization chemistry.
To obtain practical information about alkyllithium-initiated polymerizations, I contacted Dr. Henry Hsieh at Phillips Petroleum Company to work in their research laboratory in Bartlesville, Oklahoma, during summer 1974. This was a pivotal decision for my career. On a daily basis, I was exposed to state-of-the-art laboratory research on alkyllithium-initiated styrene–diene polymerizations. I had the opportunity to have all of my questions on these subjects answered by practicing polymer scientists. I also had access to reports on a variety of topics relating to alkyllithium-initiated polymerizations. Most importantly, I found a friend and colleague in Dr. Henry Hsieh, one of the research giants in this field. Approximately 20 yr later, Dr. Hsieh and I wrote a tome on this subject, Anionic Polymerization. Principles and Practical Applications.18 I wrote my portion of the book in Strasbourg, France, in 1991 at the Centre National de la Recherche Scientifique research laboratories of another pioneering research scientist in anionic polymerization, Professor Paul Rempp, and his colleagues Yves Gnanou and Pierre Lutz.
III. POLYMER RESEARCH AND EDUCATION AT THE UNIVERSITY OF AKRON WITH PROFESSORS MAURICE MORTON AND LEWIS J. FETTERS
After my experience at Phillips Petroleum Company, it was obvious to me that I needed to learn more about anionic polymerization, especially laboratory techniques. I contacted Professor Maurice Morton at The University of Akron (Goodyear Medal, 1985) to explore the possibility of coming to Akron during my sabbatical leave (1976–1977) to work with him and Professor Lewis J. Fetters. This was one of the most exciting experiences of my career. I worked closely with Professor Fetters and his graduate students and postdoctoral associates to learn the techniques of glassblowing and high-vacuum manipulations for effecting controlled, alkyllithium-initiated polymerization in an all-glass, sealed apparatus. I was able to learn about the fundamentals of the chemistry and physics of polymers by auditing most of the graduate courses in the Department of Polymer Science during my stay in Akron. I was also able to collaborate with Dr. Patricia Dreyfuss and learn the fundamentals of cationic polymerization of tetrahydrofuran (THF). It was interesting for me to recall that while I was in graduate school at the University of Illinois under the guidance of Professor David Y. Curtin, I was encouraged to spend summer 1964 working at 3M Company with another outstanding scientist, Sam Smith, on amine termination of living cationic polymerization of THF. Sam Smith was the co-inventor of Scotchguard®.19
One of the great lessons that I learned at Akron was the importance of working on alkyllithium-initiated polymerizations in hydrocarbon media. The Morton–Fetters team emphasized this because this science and technology was used in industry to prepare commercial elastomers such as polybutadiene, polyisoprene, solution SBR (sSBR), and thermoplastic elastomers (SBS and SIS) and liquid rubbers.18 It was obvious to me that academic research results obtained with organolithium compounds in hydrocarbon media would be of interest to industry and could perhaps lead to industrial support of my research.
IV. ANIONIC POLYMERIZATION RESEARCH AT MIDLAND MACROMOLECULAR INSTITUTE
Upon returning to the University of Arkansas, it was apparent that I would not be able to reach my potential in anionic polymerization research in the absence of a strong supportive environment such as that at The University of Akron. I was fortunate to obtain a position as a Research Scientist at Midland Macromolecular Institute (MMI) in Midland, Michigan, in 1979. My anionic polymerization research flourished in this excellent polymer environment in the presence of such outstanding scientists as Dale Meier (TPE expert; American Chemical Society [ACS] Rubber Division, Thermoplastic Elastomer Award, 2007), Karl Solc (TPE theory), and the prolific researcher and writer Hans Georg Elias. With the help of M.S. graduate students from Central Michigan University, we began what was to become one of my main research interests: anionic chain-end functionalization chemistry. For example, we developed an anionic polymerization method for quantitative primary amine, ω-chain-end functionalization.20 One of the most simple and useful functionalization reactions is the treatment of carbanionic species such as Grignard reagents with carbon dioxide to form the corresponding carboxylic acid derivatives. However, when polymeric organolithium compounds were reacted with carbon dioxide in hydrocarbon solution, the desired ω-carboxyl–functionalized polymers were contaminated with the corresponding dimers and trimers as shown in Eq. 1.21
For example, the yields of carboxylated polymer, dimer, and trimer are 47, 27, and 26%, respectfully, for poly(styryl)lithium (PSLi), whereas the corresponding yields are 27, 23, and 50% for the poly(butadienyl)lithium (PBDLi) chain end. This was one of the first examples we encountered illustrating the sensitivity of chain-end functionalization reactions to the nature of the anionic chain end. The formation of dimeric and trimeric side products was attributed to the known strong chain-end association of polymeric organolithium compounds. For example, PSLi is primarily associated into dimers in benzene solution, whereas recent studies are consistent with predominant tetrameric aggregation of poly(dienyl)lithiums in hydrocarbon solution.22 This was our first foray into investigating a variety of reaction variables to optimize various chain-end functionalization reactions. We have found that variables such as chain-end concentration, mode of reagent additions (normal and inverse), stirring rate, temperature, addition of Lewis bases, addition of lithium salts, and anionic polymer chain-end structure can have dramatic effects on the efficiencies of these electrophilic functionalization reactions. For example, addition of sufficient quantities of Lewis bases such as THF and N,N,N′,N′-tetramethylethylenediamine (TMEDA) can reduce or even eliminate the association of polymeric organolithiums.23,24 Using this information, Quirk and Yin found that post-polymerization addition of THF or TMEDA was effective in eliminating dimer and trimer formation for carbonation such that the carboxylated polymers were obtained in yields of >99% for PSLi, PBDLi, and poly(isoprenyl)lithium (PILi).25 These bases were added after complete monomer consumption because polymeric organolithium compounds exhibit decreased chain-end stability in the presence of Lewis bases.18 Furthermore, added Lewis bases dramatically affect the microstructure of polydienes, generating increased levels of 1,2-enchainment (for butadiene) and 4,3-enchainment (for isoprene).18 At MMI, we also explored peroxide functionalizations with molecular oxygen26 and halogen chain-end functionalization. Continuing work begun in Akron, we explored the anionic polymerization of myrcene, a monomer that was derived from a renewable natural resource, turpentine, but that is now available from microbial fermentation (Amryis).27 Thermoplastic elastomers based on myrcene, that is, polystyrene-block-polymyrcene-block-polystyrene, were prepared.28,29 In collaboration with Dr. Dale Meier, a process was developed for phase-specific curing of thermoplastic elastomers.30 With the help of an outstanding technical associate, Dennis McFay, we investigated the interaction of Lewis bases with polymeric organolithium compounds by using high-dilution solution calorimetry.31,32 The interaction of Lewis bases with polymeric organolithium compounds effects dramatic changes in chain-end association, kinetics, diene microstructure, and copolymerization parameters.18
V. ACADEMIC RESEARCH CAREER AT THE UNIVERSITY OF AKRON
It was at this point in my career that I realized that I was the recipient of remarkably good fortune. My friend and colleague Professor Lewis J. Fetters decided to leave The University of Akron and pursue research opportunities at a new polymer research facility in Annandale, New Jersey, with Exxon. This provided an opportunity for me to apply for this vacant faculty position. Based on my research experience and publications at MMI, and with the help and encouragement of Professor H. James Harwood, I was appointed a Professor of Polymer Science at The University of Akron in 1983 to continue the strong anionic polymerization research tradition of Professors Maurice Morton and Lewis J. Fetters. I cannot emphasize how important this opportunity was for the success of my research career. It was invaluable to me to be able to flourish in a graduate polymer research program with outstanding, world-class colleagues and facilities; excellent graduate students; a moderate, primarily graduate teaching responsibility; a supportive university administration; and an international reputation in polymer research and education.
A. anionic functionalization reactions with ethylene oxide
With one of my first graduate students, Jing-Jing Ma, we investigated the unique, specific functionalization reaction (one type of chemistry that is useful for only one functional group) of polymeric organolithium compounds with ethylene oxide to form the corresponding terminal chain-end (ω), hydroxyl-functionalized polymers without oligomerization of ethylene oxide after 12 h at 25 °C.33,34 This was surprising because ethylene oxide is, in principle, an anionically polymerizable monomer. For example, ethylene oxide readily undergoes anionic polymerization with other alkali metal counterions.35 This is another example (see diene microstructure, “Uniqueness of Anionic Polymerization”) of the uniqueness of lithium as a counterion in anionic polymerization. It is interesting to note that later work by matrix-assisted laser desorption/ionization (MALDI)-time of flight mass spectrometry found that some oligomerization occurred in cyclohexane, but only with PBDLi, not with PSLi.36 These results are in agreement with the report by Mays and coworkers.37 No oligomerization was observed when the functionalization was terminated after several minutes versus the normal 12 h reaction time. Figure 1 shows the structures of other epoxides and carbonyl compounds that have been used as terminating agents to quantitatively introduce hydroxyl and other functional groups at polymer chain ends via anionic polymerization.
Citation: Rubber Chemistry and Technology 93, 1; 10.5254/rct.20.79987
B. anionic polymerization chemistry of 1,1-diphenylethylene: difunctional initiators and linking agents
During my tenure at MMI, I had the opportunity to interact with Drs. Lu Ho Tung and Arnold Gatsky at Dow Chemical Company. Dow had developed a process for commercial anionic synthesis of thermoplastic elastomers based on the difunctional initiator formed from the reaction of 1,3-bis(1-phenylethenyl)benzene (DDPE) with 2 moles of butyllithium as shown in Scheme 1.38 This process was one of the factors leading to the joint venture between Dow and Exxon (DEXCO). The surprising ability of relatively stable, sterically hindered 1,1-diphenylalkylithium compounds to initiate styrene and diene polymerization, forming much less stable anions, can be traced back to the pioneering work by Fetters and Morton.39,40
Citation: Rubber Chemistry and Technology 93, 1; 10.5254/rct.20.79987
Quirk and Ma41 found that the difunctional Dow initiator formed by reaction of DDPE with 2 moles of sec-butyllithium did not cleanly initiate the polymerization of butadiene in contrast to the results reported in the Dow patents and publications.38 In fact, bimodal molecular weight distributions were obtained. Upon careful reexamination of the Dow patents, it was realized that in contrast to our procedures using high purity reagents and high vacuum techniques, the Dow experiments were carried out under nitrogen and impurities were deactivated by using additional sec-butyllithium (more than the stoichiometric amount).38 It was deduced that these procedures to remove impurities would presumably form the corresponding lithium alkoxides. In fact, when lithium alkoxide was added to our DDPE/2 sec-BuLi initiator using high purity reagents and high vacuum techniques, monomodal, well-controlled anionic polymerizations of butadiene resulted. This technique, that is, addition of lithium alkoxide, was also adapted for the use of DDPE as an initiator for isoprene polymerization.42
One of the advantages of the use of hydrocarbon-soluble, dilithium initiators is that they provide a two-step method for the synthesis of thermoplastic elastomers such as poly(styrene-block-diene-block-styrene) triblock copolymers, as shown previously in Scheme 1.43 An interesting extension of this DDPE chemistry was to investigate the use of this chemistry to prepare a trifunctional initiator, based on tris-1,3,5-(1-phenylethenyl)benzene (TDPE) as shown in Scheme 2. Quite surprisingly the reaction of 3 moles of sec-butyllithium with TDPE generated a reactive, hydrocarbon-soluble, trifunctional organolithium initiator for styrene and diene polymerizations and for the two-step synthesis of a star-branched, thermoplastic elastomer as shown in Scheme 2.44,45
Citation: Rubber Chemistry and Technology 93, 1; 10.5254/rct.20.79987
With a very capable and experienced graduate student, Frederick Ignatz-Hoover46 (2009 Melvin Mooney Distinguished Technology Award, Rubber Division, ACS), we investigated the reaction of stoichiometric amounts (2 moles) of polymeric organolithium compounds with DDPE. As shown in Scheme 3, this process forms a polymeric dilithium species (B) that can be used to initiate anionic polymerization of another monomer to form star-branched polymers (C). Thus, the reaction of polymeric organolithium reagents with DDPE is a living linking reaction in contrast to most other linking chemistries, such as the use of silicon halides that terminate the chains. This subject has been reviewed by Hadjichristidis and coworkers.47
Citation: Rubber Chemistry and Technology 93, 1; 10.5254/rct.20.79987
C. general functionalization reactions based on functionalized 1,1-diphenylethylenes
The unique anionic polymerization chemistry of 1,1-diphenylethylene (DPE) was an important component of many of our research investigations at The University of Akron.40 As described previously, one of our continuing research interests was in the development of efficient anionic polymerization methods for the introduction of chain-end functional groups. The goal was to develop general functionalization reactions, that is, efficient functionalization reactions independent of the type of functional group that could be introduced. Because DPE is a non-homopolymerizable monomer, only monoaddition occurs even when excess DPE is added to a polymeric organolithium compound. Thus, it was of interest to investigate the use of substituted DPEs for terminal (ω) chain-end functionalization of polymeric organolithium compounds. The first example of this type of functionalization chemistry involved the termination of PSLi with 1-(4-dimethylaminophenyl)-1-diphenylethylene (see Eq. 2).48 After termination with methanol, the resulting ω-amine–functionalized polymer was obtained in quantitative yield.
Figure 2 contains a list of all of the functional groups that have been added quantitatively to polymeric organolithium compounds by using substituted 1,1-diphenylethylenes to obtain the corresponding ω-chain-end functionalized polymers in quantitative yield.40 It is noteworthy that we were excited to apply this functionalization chemistry for fluorescent labeling of polymers after a colleague at The University of Akron, Professor Wayne Mattice, mentioned that it would be possible to investigate polymer dynamics in micelles by using fluorescence spectroscopy if we could synthesize polymers with different fluorescent groups at the chain ends.49 As shown in Figure 2, we developed very useful methods using 1,1-diphenyethyene–type chemistry to prepare polymers with naphthalene50 and pyrene51,52 groups at the chain ends. These methods for fluorescent labeling using DPE chemistry have also been used by other researchers.40
Citation: Rubber Chemistry and Technology 93, 1; 10.5254/rct.20.79987
A landmark breakthrough in elastomer synthesis and tire technology occurred when researchers at Nippon Zeon reported that the termination of high molecular weight polymeric dienyllithium and styrene-random-dienylithium (sSBR) compounds with 4,4′-bisdiethylaminobenzophenone resulted in formation of chain-end functionalized polymers that dramatically improved the properties of the corresponding compounded elastomers in tire applications.53 These effects were attributed to more efficient dispersion of the carbon black fillers, although these functional groups were present in relatively small concentrations (only at the chain ends of polymers with molecular weights in the range of 200 000 to 400 000 g/mol). This report led to an increase in interest for elastomer companies to investigate the use of other chain-end functional elastomers to improve tire properties. Because we had been investigating methods to incorporate a wide variety of chain-end functional groups using alkyllithium-initiated polymerization in hydrocarbon solution, we benefited from this interest in terms of support for our research and consulting interactions with companies involved in elastomer synthesis.
D. specific anionic functionalization reactions
Figure 3 illustrates the wide variety of polymer chain-end functional groups that can be incorporated efficiently using specific functionalization procedures for each different functional group. A careful reexamination of the reaction of polymeric organolithium compounds (specifically PSLi) with 4,4′-bis(diethylamino)benzophenone showed that this reaction formed the corresponding amine-functionalized polymer in 94% yield in addition to 6% of functionalized polymer and 1% of the dimer.54 Similar results were found for the corresponding reaction with benzophenone (82–94% functionalized polymer, 4–10% dimer, and 2–8% nonfunctional polymer). The formation of dimer and nonfunctionalized polymer was ascribed to the intervention of an electron transfer pathway for this functionalization reaction. Figure 1 includes a compilation of the carbonyl compounds that have been investigated for introduction of hydroxyl and other functional groups at polymer chain ends.
Citation: Rubber Chemistry and Technology 93, 1; 10.5254/rct.20.79987
Based on the exciting results obtained by Nippon Zeon53 regarding the benefits of amine chain-end functionalization of elastomers for tire applications, a variety of specific functionalization reactions were investigated to produce amine chain-end functionalized polymers. The reactions of polymeric organolithium compounds with imine compounds proved to be one of the most useful methods for preparing a variety of ω-amine–functionalized polymers as shown in Figure 4 and illustrated in Eq. 3. Quantitative amine functionalization was observed with PBDLi; that is, functionality was 0.98 by amine end-group titration and no unfunctionalized polybutadiene was observed by thin-layer chromatography.55,56 In contrast to results obtained for N-benzylidenetrimethylsilylamine,57 no evidence of Cannizzaro-type side reactions was observed; that is, no amine-functionalized dimer and acetophenone-type functionalized polymer were observed.
Citation: Rubber Chemistry and Technology 93, 1; 10.5254/rct.20.79987
It was always of interest to find new general functionalization reactions. The reaction of polymeric organolithium compounds with functionalized alkyl chlorides was one reaction that was investigated in detail as a general functionalization reaction.58,59 When a systematic study of the reaction of polymeric organolithium compounds with 3-dimethylaminopropyl chloride was carried out in hydrocarbon solution, the yield of amine-functionalized polymers was 67, 85, and 90% for PSLi, PILi, and PBDLi, respectively.58 These results illustrate again how the structure of the chain end can have a dramatic effect on the efficiency of chain-end functionalization reactions. The analytical results for PSLi functionalization are shown in Eq. 4.
The reaction was complicated by (a) formation of dimer (13%) from lithium–chloride exchange followed by Wurtz coupling and (b) formation of nonfunctional polymer presumably resulting from E-2 elimination of HCl by the polymer anion. Fortunately, many variables were available to optimize this reaction, as mentioned previously. One variable that had not been used in hydrocarbon solution was the addition of lithium chloride. Lithium chloride previously had been shown to have a dramatic effect on the chain-end stability of acrylates and methacrylates such that controlled anionic polymerization could be achieved.60 A breakthrough in developing the reaction with substituted alkyl chlorides as a general functionalization reaction was realized when one equivalent of lithium chloride was added to the PSLi reaction mixture and quantitative functionalization was obtained.59 For PILi it was necessary to add 10 equivalents of lithium chloride to obtain quantitative functionalization. Figure 5 contains a list of the chain-end functional groups and protected functional groups that have been incorporated efficiently into polymers using this general functionalization chemistry, that is, using the corresponding α-halo-ω-functional alkanes as terminating agents for polymeric organolithium compounds. An advantage of this chain-end functionalization chemistry is that a variety of substituted alkyl chlorides are available commercially.
Citation: Rubber Chemistry and Technology 93, 1; 10.5254/rct.20.79987
E. functionalization reactions using functionalized initiators based on substituted 1,1-diphenylethylenes
Another useful application of DPE chemistry, a non-homopolymerizable monomer, was to use the reactions of substituted 1,1-diphenylethylenes with sec-butyllithium to form substituted 1,1-diphenylalkyllithium compounds for use as initiators for anionic polymerization of styrenes and dienes as shown in Eq. 5.40
For example, Quirk and Zhu48 showed that when initiator formed by reaction of sec-butylithium with 1-(4-dimethylaminophenyl)-1-phenylethylene was reacted with styrene (Scheme 4), α-dimethylaminopolystyrene was obtained after methanol termination. This polymer [Eq. 5, M = styrene, X = N(CH3)2, Y = H] exhibited a narrow molecular weight distribution (MWD; Mw/Mn = 1.04) and amine end group functionality of 1.2 as determined by titration. The narrow MWD was somewhat surprising given the stability of the 1,1-diphenylalkyllithium initiator [pKa of (C6H5)2CH2 = 32] relative to the much less stable styryllithium chain end formed [pKa of C6H5CH3 = 43].40 An extensive investigation of 1,1-diphenylalkyllithium initiators with styrene by Quirk and Lee61 showed that narrow MWDs can be obtained for Mn (out) > 7000 g/mol. It has also been demonstrated that 1,1-diphenylalkylithium initiators are much more reactive and efficient for initiating polymerization of butadiene and isoprene and narrow MWD polydienes can be formed for Mn (out) > 2900 g/mol. When the α-dimethylaminopoly(styryl)lithium shown in Scheme 4 was reacted with a second mole of 1-(4-dimethylaminophenyl)-1-phenylethylene instead of termination with methanol, a telechelic α,ω-bisdimethylaminopolystyrene was formed after methanol termination.48 The amine functionality of this telechelic α,ω-diamine was determined to be 2.1 by end-group titration.
Citation: Rubber Chemistry and Technology 93, 1; 10.5254/rct.20.79987
F. functionalization with substituted 1,1-diphenylethylenes as comonomers
It is also worth noting that substituted 1,1-diphenylethylenes can also be used as comonomers with styrenes and dienes to form in-chain functional polymers.40 Because 1,1-diphenylethylenes cannot homopolymerize, their monomer reactivity ratios with styrenes and dienes are zero. Thus, when 1,1-diphenylethylenes copolymerize with styrenes or dienes, an alternating-type copolymer is obtained (r1r2 = 0). The copolymerization of DPE with styrene forms an almost perfectly alternating copolymer (r1 = 0.44).62 The importance of these copolymerizations is that copolymerization of DPE with styrene can form TPE end blocks that have Tg values in the range of 169–175 °C with [styrene]/[DPE] = 1.1–1.3.63 This has been shown to increase tensile properties and increase the upper use temperatures for ABA triblock copolymers with diene center blocks and DPE/styrene end blocks.63,64 Styrene can be readily copolymerized with substituted DPEs also to form in-chain functionalized polymers.40 Quirk and Zhu65 have shown that the copolymerization of styrene with a 0.3 molar excess of 1-(4-dimethylaminophenyl)-1-phenylethylene produced a copolymer (Mn =1.6 × 104 g/mol) with 24 amine groups per chain rstyrene = 5.6).
The monomer reactivity ratios for copolymerizations of dienes with DPE with lithium as counterion are quite different from those of styrene copolymerizations: r1(butadiene) = 54; r1(isoprene) = 37.66–68 Thus, dienes exhibit dramatic unreactivity toward addition to DPE. However, in the presence of Lewis bases, it is possible to copolymerize dienes with substituted DPEs to form in-chain functionalized polymers.40 Quirk and Kuang69 investigated the copolymerization of butadiene (M1) with excess 1-(4-dimethylaminophenyl)-1-phenylethylene using sec-butyllithium as initiator in benzene. Anisole, triethylamine, and tert-butyl methyl ether were used in ratios of [base]/[RLi] = 60, 20, and 30, respectively. Copolymers were obtained with narrow molecular weight distributions (Mw/Mn = 1.02–1.03) and 8, 12, and 30 amine groups per chain for anisole, triethylamine, and tert-butyl methyl ether, respectively. The corresponding butadiene monomer reactivity ratios (r1) and microstructures [% 1,2 (vinyl) enchainment] were 42 (14%), 33 (29%), and 14 (58%) for anisole, triethylamine, and tert-butyl methyl ether, respectively. It is important to note that analogous copolymerizations in the presence of potassium tert-amyloxide ([KOR]/[Li] = 0.025), a randomizer for SBR copolymerizations, readily incorporate amine groups into the polybutadiene copolymer (r1 = 1.35) and the amount of 1,2-microstructure was only 28%.69
Thus, the addition reactions of simple and polymeric organolithium compounds with substituted 1,1-diphenylethylenes provide general methods for the synthesis of a wide variety of functionalized polymers. Polymers can be prepared with functional groups at the initiating chain end, at the terminating chain end, within the polymer chain by copolymerization, and at the junction between blocks.40
The unique chemistry of substituted 1,1-diphenylethylenes [i.e., 1-(4-dimethylaminophenyl)-1-phenylethylene] has been used commercially by Sumitomo70 to prepare amine-substituted SBRs with 1 (initiating end), 2 (terminating end), or 3 (in-chain) amine functional groups as shown in Scheme 5. Hankook Tire71 has used similar procedures to prepare α-tertiary amine-functionalized SBRs.
Citation: Rubber Chemistry and Technology 93, 1; 10.5254/rct.20.79987
G. collaborative research with fmc, lithium division, on new functionalized alkyllithium initiators
As a result of our work on functionalized initiators based on 1,1-diphenylethylenes as well as our work on the general syntheses and characterization of functionalized polymers, we began a research collaboration with FMC, Lithium Division, to investigate the polymer chemistry of several new functionalized alkyllithium initiators. Over a 6 yr period of research support from FMC, 22 patents were issued based on the research collaboration between The University of Akron and FMC, Lithium Division. The accomplishments of this collaborative research were recognized in 2000 by the awarding of the Cooperative Research Award of the Polymeric Materials Science and Engineering Division of the ACS jointly to R. P. Quirk and Dr. James Schwindeman of FMC, Lithium Division.
Figure 6 shows the structures of the hydrocarbon-soluble, protected, hydroxyl-functional alkyllithium initiators prepared by FMC and investigated by Quirk and coworkers at The University of Akron.72 All of the alkyllithium initiators shown in Figure 6 were useful for the controlled anionic polymerization of butadiene, isoprene, and styrene monomers. As part of these investigations, the thermal stabilities were investigated, and methods for the efficient removal of the hydroxyl protecting groups were developed.
Citation: Rubber Chemistry and Technology 93, 1; 10.5254/rct.20.79987
H. development of a new, general functionalization methodology based on silyl hydride chemistry
The reactions of polymeric organolithium compounds with silyl halides are very efficient reactions that are not complicated by competing side reactions.47,73,74 DeSimone and coworkers75,76 demonstrated that these reactions could provide a general functionalization methodology to prepare a variety of end-functionalized polymers by reactions with silyl halides containing either functional groups or protected functional groups. For example, this methodology was applied successfully for the synthesis of primary amine functionalized polystyrenes, polyisoprenes, and polydimethylsiloxanes as shown in Eq. 6 for polystyrene (Mn = 2200–21 300 D) by using a silyl chloride with a bis-trimethylsilyl–protected amine group.76
The amine functionalities determined by titration after deprotection varied from 0.87 to 1.04. The structures were also confirmed by MALDI-secondary ion mass spectrometry analysis, and it was stated that significant amounts of unfunctionalized material were absent in the spectra. This termination method using substituted silyl chlorides provides a useful general methodology for the synthesis of a wide variety of functional end groups.75,76 However, this methodology requires the use of protection and deprotection protocols for most functional groups of interest because of their reactivity with organolithium compounds.77 In addition, the long-term stability of these protected, functionalized silyl chlorides was found to be a problem.78
We were intrigued by the synthetic methods that were used to prepare the substituted silyl chlorides. The key step that was of interest to our research was the addition reaction of chlorodimethylsilane to a functionalized alkene in the presence of a platinum catalyst as shown in Eq. 7.
This reaction suggested to us that an alternative functionalization procedure would be to first prepare a polymeric silyl hydride by reaction of a polymeric organolithium compound with chlorodimethylsilane and then react the polymeric silyl hydride with a substituted alkene as shown in Scheme 6.79 The first reaction is one of the most efficient reactions in organolithium chemistry, that is, reaction of a polymeric organolithium compound with a silyl chloride.47,73,74 The second reaction, hydrosilation, is also a very efficient reaction that does not require a protection/deprotection protocol for most functional groups.80 This is unique among the functionalization reactions that we have investigated previously, because most functional groups are reactive toward organolithium chain ends and require protection.77 These advantages opened up a much wider variety of functional groups that could be incorporated in polymer chains as shown in Figure 7; groups such as perfluoro, epoxide, phenol, alcohol, amine, and ester are notable functional groups in the category. The versatility of the hydrosilation functionalization procedure is that not only can a wide variety of new functional groups be incorporated at the chain end but also functional groups can be incorporated within the polymer chain,78 at both the α-81 and ω-chain ends,79 in-chain,82 and at the interface between blocks.83 The syntheses of the silyl hydride precursors for the chain ends and in-chain (or between blocks) are shown in Scheme 7. One of the most powerful aspects of this hydrosilation methodology is that it elevates the ability to prepare a variety of well-defined chain-end functional polymers because one polymeric silyl hydride precursor (well-defined, controlled Mn and narrow Mw/Mn) can be used to prepare a panel of functional polymers with the only difference being the chain-end functional group. This methodology was used in a collaborative study with the Colmenero group in Spain to isolate and investigate the effect of functional groups on the dielectric relaxation behavior of polystyrene.84–86 The hydrosilation functionalization procedure also suggested another general methodology for synthesis of in-chain functional polymers, the use of a silyl hydride–functionalized monomer, that is, p-dimethylsilylstyrene as shown in Scheme 8. In general, the use of substituted styrenes requires the use of protection/deprotection protocols for most functional groups. However, the use of a silyl hydride–functionalized styrene comonomer means that the functional group is not incorporated until after the alkyllithium-initiated polymerization, thus no protection is required for the functional groups. A detailed investigation of the applications of silyl hydride–functionalized polystyrenes for alkyllithium-initiated styrene and diene homopolymerizations and copolymerizations was initiated in cooperation with Dynasol Elastomeros. A joint patent was recently issued based on this research.87
Citation: Rubber Chemistry and Technology 93, 1; 10.5254/rct.20.79987
Citation: Rubber Chemistry and Technology 93, 1; 10.5254/rct.20.79987
Citation: Rubber Chemistry and Technology 93, 1; 10.5254/rct.20.79987
Citation: Rubber Chemistry and Technology 93, 1; 10.5254/rct.20.79987
VI. CONCLUSIONS
Beginning at Michigan Molecular Institute in 1979 and continuing from 1983 to 2010 at The University of Akron, our research group has investigated new methods for the introduction of chain-end and in-chain functional groups into polymer chains by using alkyllithium-initiated anionic polymerization of styrenes and dienes in hydrocarbon media. Alkyllithium-initiated anionic polymerization in hydrocarbon media corresponds to the procedures used commercially to produce polydienes; SBRs; styrene–diene–styrene triblock copolymer TPEs; diblock copolymers; star-branched, hydrogenated polydienes (viscosity improvers); and liquid rubbers. Because of this correspondence between our procedures and commercial processes, we benefited from continuing interest and research support from many companies, especially Exxon, Mobil, Exxon/Mobil, Sartomer, Dow Chemical Company, Dynasol Elastomers, Negromex, Polysar, B.F. Goodrich, FMC Lithium Division, BASF Corporation (Wyandotte), BASF Aktiengesellschaft (Ludwigshafen), Bridgestone-Firestone, and Goodyear. With this support and the excellent research environment at The University of Akron, Department of Polymer Science, we have developed quantitative specific procedures for synthesis of a wide variety of chain-end functional groups of interest. In addition, we have developed general functionalization procedures (independent of the functional group) by using reactions with substituted alkyl chlorides and substituted 1,1-diphenylethylenes. Using silyl chloride functionalization and hydrosilation chemistry, a general functionalization methodology has been developed that does not require the use of protecting groups for most functional groups of interest. The awardee acknowledges the sine qua non contributions of the graduate students, postdoctoral associates, and visiting scientists who truly performed the requisite research.
Structures of epoxides and carbonyl compounds that have been used as terminating agents to quantitatively introduce hydroxyl and other functional groups at the chain end.
Two-step synthesis of polystyrene-b-polybutadiene-b-polystyrene triblock copolymer from the initiator formed from DDPE plus 2 moles of sec-butyllithium.
Synthesis of star-branched, polystyrene-b-polybutadiene block copolymer using the trifunctional initiator formed by reaction of 2 moles of sec-butyllithium with tris-1,3,5-(1-phenylethenyl)benzene.
Living linking reactions of bis-1,3-(1-phenylethenyl)benzene with PSLi followed by addition of butadiene to form a heteroarm, star-branched copolymer.
Functional groups that can be added quantitatively to polymeric organolithium compounds by using substituted 1,1-diphenylethylenes to obtain ω-chain-end functionalized polymers.
Reagents and resulting functional groups used for specific chain-end functionalization of polymeric organolithium compounds.
Reagents that have been used for specific amine functionalization reactions by termination of polymeric organolithium compounds.
Substituted alkyl chlorides and resulting functional groups incorporated at polymer chain ends by using the general functionalization reaction with substituted alkyl chlorides.
Synthesis of α-dimethylaminopolystyrene by reaction of the initiator formed from butyllithium plus 1-(4-dimethylaminophenyl)-1-phenylethylene and formation of α,ω-dimethylaminopolystyrene by reaction with 1-(4-dimethylaminophenyl)-1-phenylethylene before methanol termination.
Sumitomo patented process for preparing SBR with one, two, or three dimethylamino groups by using DPE functionalization chemistry.
Hydrocarbon-soluble, protected, hydroxyl-functionalized alkyllithium initiators prepared by FMC and investigated at The University of Akron.
General functionalization methodology using termination of polymeric organolithium compounds with chlorodimethylsilane followed by reaction of the polymeric silyl hydride with a substituted alkene in the presence ofKarstedt's catalyst.
Synthesis methods for silyl-hydride-functionalized polymers (in-chain, chain-ends).
Copolymerizations of styrene with p-dimethylsilylstyrene for in-chain functionalization using hydrosilation chemistry.
Functional end groups added to polymer chain ends using the reaction of polymeric organolithium compounds with chlorodimethysilane followed by reaction of the resulting polymeric silyl hydride with a functionalized alkene.
Contributor Notes