Analysis of Copolymerization with Simultaneous Reversibility and Transesterification by Stochastic Model Regression

Determining kinetic parameters for copolymerizations including depropagation and transesterification reactions is complicated by the impracticality of a complete deterministic model. Stochastic simulations are uniquely appropriate for reversible copolymerization with simultaneous transesterification and have been used to describe theoretical reversible copolymerizations, but parameter estimation on experimental data using stochastic models has not been attempted in a copolymerization context. Here, we advance the method of stochastic model regression (SMR) for fitting a stochastic simulation algorithm to experimental data. Conversion versus time data were fit to a stochastic model for five initial compositions of glycolide (fG) in rac-lactide/glycolide copolymerization resulting in poly(lactide-co-glycolide) (PLGA) and five initial compositions of rac-lactide (fL) in rac-lactide/ε-caprolactone copolymerization resulting in poly(lactide-co-ε-caprolactone) (PLCL). We determined reactivity ratios of rG = 4.4, rL = 0.34 for PLGA and rL = 12.3 and rC = 0.07 for PLCL, which establish an initial compositional gradient that is counteracted by random transesterification. Complete kinetic parameters including depropagation and transesterification rate constants resulted from our analysis by SMR and are needed for a complete description of repeat unit sequencing and the time evolution of composition (fG or fL).