Multi-scale numerical modeling and simulation of microphase-separated block copolymers

Overview & Synopsis

Block copolymer (BCP) systems have unique properties and serve as a useful design platform for polymer scientists and engineers. Depending on the length of the polymer chain, the volume ratio between the blocks, and their chemical nature, these systems can undergo microphase separation, forming periodic structures down to the nanometric scale. To better understand and model BCPs, this thesis aims at developing multiscale simulation methodologies, bridging the gap between meso and nanoscale simulations. This is accomplished with a top-down multiscale approach, coupling self-consistent field theory (SCFT) and coarse-grained molecular dynamics (CGMD). Instead of starting the nanoscale simulation of BCPs from a random configuration, an algorithm named Propagator-Biased Chain Generation (PBCG) is developed, allowing for the generation of SCFT-biased initial CGMD configurations that already present phase separation, saving simulation time. To conduct the CGMD simulations, a methodology based on Bayesian optimization is developed to parameterize the interactions between monomers of different types, responsible for maintaining the SCFT-predicted structures during the CGMD simulations.