Isobutylene was initiated using 5-tert-butyl-1,3-di(2-chloro-2-propyl)benzene/TiCl4 in 60/ 40 hexane/methyl chloride: [IB]0 ) 1.0 M, [TiCl4] ) 0.12 or 0.24 M, [t-Bu-m-DCC] ) 0.0119 M, T ) -(40-80) °C. Most polymerizations contained a Lewis base or other additive, i.e., 2,4-dimethylpyridine, 2,6-di-tert-butylpyridine, tetra-n-butylammonium chloride, and/or pyridine hydrochloride. Polymerizations containing an additive yielded theoretical molecular weights, narrow polydispersity index, and apparent absence of irreversible chain termination (linear kinetic plots, ATR-FTIR spectroscopic data) and chain transfer, with two exceptions: coupled product was obtained at -40 °C, and protic initiation occurred with n-Bu4NCl alone. Polymerizations without an additive produced bimodal molecular weight distributions; however, essentially all chains were initiated from t-Bu-m-DCC. With an additive, Eact for propagation was -(5.3-5.5) kcal/mol. Removal of additives increased polymerization rate moderately at -80 °C but dramatically at -60 and -40 °C; this yielded higher Eact compared to that of systems containing additives. These results indicated that both paired and unpaired (free) ions are propagating species in absence of additives, with free ions less important at lower temperatures; free ion concentration and lifetime suggested the presence of adventitious common ions and chain transfer between free ions and tert-chloride-terminated PIB chains. The primary role of additives is suppression of free ions through in situ production, via the scavenging of protic impurities, of common ions.

Isobutylene polymerization was initiated using the cumyl chloride/TiCl4/2,4-dimethylpyridine system ([IB] ) 1.0 M, [CumCl] ) 2.3 × 10 -2 M, [TiCl4] ) 0.24 M, [DMP] ) 2.4 × 10 -3 M). Polymerization kinetics in 60/40 hexane/MeCl at -80 °C showed that an IB conversion of ∼98% was reached in 70 s. The active centers were monitored over time at high conversion, by 1 H NMR after quenching with MeOH or allyltrimethylsilane (ATMS), and by GPC after a second addition of IB. Quenching with MeOH or ATMS at times up to 70 s yielded quantitative tert-chloride or allyl chain ends, respectively. For longer periods up to 12 h, MeOH-quenched PIB showed increasing depletion of tert-chloride chain ends of the normal structure (but no appearance of olefin) and ATMS quenching yielded a gradual reduction in allylation to 78% at 12 h. Real-time 1 H NMR of an active polymerization mixture also indicated depletion of tert-chloride end groups. GPC results showed that coupling of PIB chains was not occurring and that a second charge of IB was slowly initiated. The latter process caused a fraction of the chain ends to be restored with normal tert-chloride groups. It was hypothesized that normal tert-chloride chain ends slowly undergo carbenium ion rearrangement to form a mixture of isomerized chain end structures. This process represents the dominant chain interrupting event in TiCl 4-co-initiated IB polymerization under these conditions. The rearranged structures were hypothesized to result from combinations of 1,2-hydride and 1,2-methide shifts; 1 H NMR analysis of a dehydrochlorinated sample was consistent with this interpretation. The rate of depletion of tert-chloride end groups was shown to follow first-order kinetics with an apparent rate constant of 8 × 10 -5 s -1 . The ratio of rate constants for propagation and rearrangement, kp/kr, was calculated to be 3 × 10 4 M -1 .

No abstract