Nuclear magnetic resonance studies of the propagating chain end in the organolithium polymerization of dienes. II. Isoprene and 2,3-dimethylbutadiene in hydrocarbon solvents
High-resolution studies have been carried out on the organolithium polymerization of isoprene and 2,3-dimethylbutadiene in hydrocarbon solvents. As in the case of the previously reported butadiene polymerization, no evidence could be found of any measurable amount of complexation of the isoprene with the organolithium chain end. The proton resonance spectrum of the chain-end unit in the case of isoprene indicates an exclusively 4,1 structure, rather than a 1,4 (no β hydrogen), showing no 4,3 (or 1,2) structures, despite the fact that the in-chain units contain approximately 10% 3,4 structures. Similarly, the spectrum of the poly(2,3-dimethylbutadiene) chain end confirms the peak assignments based on the butadiene and isoprene studies, and shows an exclusively 1,4 structure, despite the presence of 15-20% 1,2 in-chain units. Hence, as in the case of the butadiene, it is suggested that these 1,2 or 3,4 units result from the presence of an undetectably small proportion of π-allyl chain ends in equilibrium with the main component of σ-bonded (localized) chain ends. The isoprene chain-end spectrum permits the resolution of cis- and trans-4,1 chain-end units, which are present in approximately the same ratio as their in-chain counterparts (cis/trans = 2). The observed differential rates of decomposition of these two isomeric forms of the chain end is taken as evidence that the 2,3 bond is not labile, i.e., no 4,1-4,3 tautomerism. This further strengthens the hypothesis of a localized-delocalized equilibrium (σ-allyl-π-allyl) to account for the 3,4 unit content of the polymer, since such an equilibrium would not affect the cis-trans isomerism.