Chromatographic Separation of Living Chains in Reversible-Deactivation Radical Polymerization: From Polystyrene Homopolymers to Acrylate Block Copolymers

Reversible-deactivation radical polymerization (RDRP) methods such as atom transfer radical polymerization (ATRP), reversible addition-fragmentation chain-transfer (RAFT) polymerization, and nitroxide-mediated polymerization (NMP) produce polymers with narrow molecular weight distributions (MWD) and controlled architectures. Dead chains still form through radical termination and chain-end degradation, however, and neither size-exclusion chromatography (SEC) nor NMR can physically separate them from the living chains. We have developed liquid chromatography methods that separate polymer chains by their end groups, so that the living chains in an RDRP product can be isolated and quantified at the molecular level. In our first study, the bromine ω-end group of ATRP-grown polystyrene was converted to a polar end group by azidation and a CuAAC click reaction. HPLC on a bare silica column then fully resolved the living chains from the dead chains, and the isolated living fraction showed a Poisson-type MWD, the theoretical limit of an ideal living polymerization. The same approach was later applied to polystyrenes prepared by RAFT  and NMP, and to living polyacrylates through dihydroxyl chain-end functionalization.

This talk covers our recent extension of the method from homopolymers to acrylate block copolymers. In poly(benzyl acrylate)-b-poly(tert-butyl acrylate) synthesized by ARGET ATRP, a dihydroxyl tag was attached to the ω-chain end of the living chains by a mild, catalyst-free thio-bromo click reaction with thioglycerol, and the tagged chains were isolated on bare silica. In visible-light photoiniferter RAFT polymerization, the α,ω-dihydroxyl trithiocarbonate that acts as the photoiniferter also provides the chromatographic handle. Solvent-gradient HPLC separated the RAFT agent-derived living chains, which made up about 96% of all chains even after one-pot block copolymerization, and the isolation raised the chain-end fidelity from 95% to above 99%. In both systems, MALDI-TOF MS of the isolated living fractions gave Poisson-type distributions (Đ = 1.02-1.03, close to the theoretical 1 + 1/DP). This is direct molecular-level evidence that block copolymers grown by these methods contain a nearly ideal living population. Chain-end-selective HPLC therefore works both as a quantitative analytical tool for living and dead chains and as a preparative route to block copolymers free of dead-chain impurities.