Monte Carlo Simulation Studies of Polymerization of Polyurethane Block Copolymer. IV. Modeling of Experimental Data

摘要

Three different models which use the Monte Carlo method have been developed to predict the molecular weight, composition, and hard segment length distribution of polyurethane block copolymers polymerized under varying conditions. The simplest model tries to describe natural compositional heterogeneity that arises in polyurethanes due to the nature of the polymerization even under ideal conditions. The subsequent models introduce nonidealities into the simulation that actually occur in real polymerizations, including premature phase separation of the reactants and unequal reactivities of the monomer species. In this investigation, these models are used to simulate experimental data reported by MacKnight and coworkers on polybutadiene polyurethanes. The results of the modelling strongly support the hypothesis of MacKnight and coworkers that phase separation occurred near the beginning of the polymerization. The primary effect of the phase separation is to limit the degree of polymerization (molecular weight) in each phase presumably due to stoichiometric imbalances of reactants resulting from an unequal partitioning of the reactants in the two phases. The low molecular weights, in turn, lead to a broad composition distribution. The broad molecular weight distribution results from having a different average degree of polymerization in each phase, although allophonate cross-linking may also be a contributing factor. Differences between the three models and their utility for simulating experimental data are also discussed.