Polyisoprene lithium

The use of alkaU metals for anionic polymerization of diene monomers is primarily of historical interest. A patent disclosure issued in 1911 (16) detailed the use of metallic sodium to polymerize isoprene and other dienes. Independentiy and simultaneously, the use of sodium metal to polymerize butadiene, isoprene, and 2,3-dimethyl-l,3-butadiene was described (17). Interest in alkaU metal-initiated polymerization of 1,3-dienes culminated in the discovery (18) at Firestone Tire and Rubber Co. that polymerization of neat isoprene with lithium dispersion produced high i7j -l,4-polyisoprene, similar in stmcture and properties to Hevea natural mbber (see ELASTOLffiRS,SYNTHETic-POLYisoPRENE Rubber, natural). 

From the time that isoprene was isolated from the pyrolysis products of natural mbber (1), scientific researchers have been attempting to reverse the process. In 1879, Bouchardat prepared a synthetic mbbery product by treating isoprene with hydrochloric acid (2). It was not until 1954—1955 that methods were found to prepare a high i i -polyisoprene which dupHcates the stmcture of natural mbber. In one method (3,4) a Ziegler-type catalyst of tri alkyl aluminum and titanium tetrachloride was used to polymerize isoprene in an air-free, moisture-free hydrocarbon solvent to an all i7j -l,4-polyisoprene. A polyisoprene with 90% 1,4-units was synthesized with lithium catalysts as early as 1949 (5). 

Al—Ti Catalyst for cis-l,4-PoIyisoprene. Of the many catalysts that polymerize isoprene, four have attained commercial importance. One is a coordination catalyst based on an aluminum alkyl and a vanadium salt which produces /n j -l,4-polyisoprene. A second is a lithium alkyl which produces 90% i7j -l,4-polyisoprene. Very high (99%) i7j -l,4-polyisoprene is produced with coordination catalysts consisting of a combination of titanium tetrachloride, TiCl, plus a trialkyl aluminum, R Al, or a combination of TiCl with an alane (aluminum hydride derivative) (86—88). 

The first successhil use of lithium metal for the preparation of a i7j -l,4-polyisoprene was aimounced in 1955 (50) however, lithium metal catalysis was quickly phased out in favor of hydrocarbon soluble organ olithium compounds. These initiators provide a homogeneous system with predictable results. Organ olithium initiators are used commercially in the production of i7j -l,4-polyisoprene, isoprene block polymers, and several other polymers. 

Alkali Metal Catalysts. The polymerization of isoprene with sodium metal was reported in 1911 (49,50). In hydrocarbon solvent or bulk, the polymerization of isoprene with alkaU metals occurs heterogeneously, whereas in highly polar solvents the polymerization is homogeneous (51—53). Of the alkah metals, only lithium in bulk or hydrocarbon solvent gives over 90% cis-1,4 microstmcture. Sodium or potassium metals in / -heptane give no cis-1,4 microstmcture, and 48—58 mol % /ram-1,4, 35—42% 3,4, and 7—10% 1,2 microstmcture (46). Alkali metals in benzene or tetrahydrofuran with crown ethers form solutions that readily polymerize isoprene however, the 1,4 content of the polyisoprene is low (54). For example, the polyisoprene formed with sodium metal and dicyclohexyl-18-crown-6 (crown ether) in benzene at 10°C contains 32% 1,4-, 44% 3,4-, and 24% 1,2-isoprene units (54). 

To confirm these phenomena the UV absorption spectra of lithium polyisoprene of high and of low molecular weights dissolved in cyclohexane were re-examined over... 

The observations discussed above suggest that the kinetic order of lithium poly-isoprene propagation should vary with the living polymer concentration. The effect is imperceptible in aliphatic hydrocarbons, but is observed in benzene solutions. The apparent propagation constants of lithium polyisoprene (MW 2 2 10 ) were determined in benzene and the results are displayed in Fig. 16 in the form of a plot of log kapp vs log c, c denoting the total living polymer concentration. 

The heat of dissociation in hexane solution of lithium polyisoprene, erroneously assumed to be dimeric, was reported in a 1984 review 71) to be 154.7 KJ/mole. This value, taken from the paperl05> published in 1964 by one of its authors, was based on a viscometric study. The reported viscometric data were shown i06) to yield greatly divergent AH values, depending on what value of a, the exponent relating the viscosity p of a concentrated polymer solution to DPW of the polymer (q DP ), is used in calculation. As shown by a recent compilation 1071 the experimental a values vary from 3.3 to 3.5, and another recent paper 108) reports its variation from 3.14 to 4. Even a minute variation of oe results in an enormous change of the computed AH, namely from 104.5 KJ/mole for oe = 3.38 to 209 KJ/mole for oe = 3.42. Hence, the AH = 154.7 KJ/mole, computed for a = 3.40, is meaningless. For the same reasons the value of 99.5 KJ/mole for the dissociation of the dimeric lithium polystyrene reported in the same review and obtained by the viscometric procedure is without foundation. 

The same approach supposedly demonstrated the dimeric nature of lithium polyisoprene and polybutadiene. A tenfold decrease of viscosity was claimed 97), contrary to the findings of Worsfold and By water 115) who reported a 15 fold decrease of viscosity for lithium polyisoprene on protonation of their hydrocarbon solutions. 

Closer examination of the data pertaining to polymerization of lithium polyisoprene in aliphatic hydrocarbons suggests that kpcl and kptt are negligible in comparison with k, and k, = k IS, i.e. the addition of isoprene whether to cis or trans active... 

The distinction between the rates of homo- and copolymerization apparently is misapprehended by some workers. For example, a recent review 141) discusses the results of McGrath 142) who reported butadiene to be more reactive in polymerization in hexane than isoprene, whether with respect to lithium polybutadiene or polyisoprene, although the homopropagation of lithium polyisoprene in hexane was found to be faster than of polybutadiene. The miscomprehension led to the erroneous statement14l) McGrath 142) results regarding the rate constants for butadiene and isoprene place in clear perspective the bizarre assertion 140) that butadiene will be twice as reactive as isoprene (in anionic co-polymerization). 

An interesting approach to studies of the effects of coordination on the reactivity of lithium polydienes in hydrocarbon solvents was developed by Erussalimski and his colleagues 151 154 The polymerization of lithium polyisoprene in hexane is accelerated by the addition of TMEDA152), the rate levels off at a value of R = [TMEDA]/[li-thium polyisoprene] of 8, its final value giving the absolute rate constant of propagation of the polyisoprene coordinated with TMEDA, namely 0.17 M7l s at 20 °C. 

Investigation of the polymerization at a low value of R = 0.01 allows the observation of two simultaneously proceeding propagations, one involving the unassociated lithium polyisoprenes and the other due to those coordinated with TMEDA, provided that both contribute comparably to the growth. This is claimed by the authors. Indeed, examination of their data indicates that 10% of the polyisoprene formed... 

In the low molecular weight fraction a relatively high content of the 3,4-linkages is formed, characteristic of the polymers produced in the presence of TMEDA. On this basis it was concluded that the unassociated polymers coordinated with TMEDA propagate more slowly than those unassociated and non-coordinated with the diamine, a conclusion concordant with the previously discussed findings of Fontanille84) and confirmed by Schue, 55> for lithium polyisoprene. 

It is apparent from these data that all of the polymers, including butadiene, exhibit an association as dimers, and that there is no reason to expect any higher states of association for polyisoprene or polybutadiene. This is confirmed not only by the viscosity data on the active vs. terminated "capped" polymers, but also by the fact that there was no significant increase in viscosity when the polystyryl lithium was "capped" by butadiene or isoprene, i.e., all three types of chain ends are associated in the same way, as dimers. 

The microstructure of polyisoprene prepared by lithium initiation in hydrocarbons is 95% 1,4 under all conditions. The trans 1,4 content however falls from about 20% to zero as the monomer/initiator ratio increases leading finally to a 95% cis 1,4 polymer. This variation can be explained with the following scheme. 

These efforts coupled with the much earlier work on sodium and lithium initiated polymerizations led to an appreciation of the stereospecificity of the alkyllithium initiators for diene polymerization both industrially and academically. Polymerization of isoprene to a high cis polyisoprene with butyllithium is well known and the details have been well documented 2 Control over polybutadiene structure has also been demonstrated. This report attempts to survey the unique features of anionic polymerization with an emphasis on the chemistry and its commercial applications and is not intended as a comprehensive review. 

The first report on anionic polymerization appeared in the patent literature in 1910-1911. Matthews and Strange (l) in 1910 and later Harries (2) in 1911, described the preparation of polyisoprene using sodium and potassium as initiators. They mentioned the use of lithium as a possible initiator for this polymerization, but there seems to be no description of the polymer... 

Meanwhile, development of coordination catalyst was proceeding full scale. The polyisoprene prepared using this coordination catalyst (TiClj, AIR ) proved to be more suitable in physical properties than the one made by lithium metal or organolithium compounds in hydrocarbon media. The Ziegler polyisoprene, as it was called, has greater stereoregularity and stress-induced crystallization properties than polyisoprene made by the alkyl lithium catalyst. How-... 

As of this date, there is no lithium or alkyl-lithium catalyzed polyisoprene manufactured by the leading synthetic rubber producers- in the industrial nations. However, there are several rubber producers who manufacture alkyl-lithium catalyzed synthetic polybutadiene and commercialize it under trade names like "Diene Rubber"(Firestone) "Soleprene"(Phillips Petroleum), "Tufdene"(Ashai KASA Japan). In the early stage of development of alkyl-lithium catalyzed poly-butadiene it was felt that a narrow molecular distribution was needed to give it the excellent wear properties of polybutadiene. However, it was found later that its narrow molecular distribution, coupled with the purity of the rubber, made it the choice rubber to be used in the reinforcement of plastics, such as high impact polystyrene. Till the present time, polybutadiene made by alkyl-lithium catalyst is, for many chemical and technological reasons, still the undisputed rubber in the reinforced plastics applications industries. 

The commerical polybutadiene (a highly 1,4 polymer with about equal amounts of cis and trans content) produced by anionic polymerization of 1,3-butadiene (lithium or organolithium initiation in a hydrocarbon solvent) offers some advantages compared to those manufactured by other polymerization methods (e.g., it is free from metal impurities). In addition, molecular weight distributions and microstructure can easily be modifed by applying appropriate experimental conditions. In contrast with polyisoprene, where high cis content is necessary for suitable mechanical properties, these nonstereoselective but dominantly 1,4-polybutadienes are suitable for practical applications.184,482... 

The discovery that lithium and its alkyls produce a highly cis-1,4 polyisoprene in hydrocarbon solvents (103) has led to a renewed interest in metal and metal alkyl initiated polymerization. About the same time Szwarc (109) postulated an electron transfer mechanism for the initiation of polymerization by sodium naphthalene in ether solvents. This was extended to lithium metal catalysis by Tobolsky (80) and Overberger (83) and subsequently generalized to cover all alkali metal initiation, e" + M M (1) ... 

---