Optimization of the processing and materials composition of epoxy/block copolymer blends

Block copolymers (BCPs) remain highly interesting materials due to their ability to self-assembly into a variety of self-assembled structures either in the bulk phase or as a minority component in a different matrix material. Over the past few years, we have studied the self-assembly of poly(ethylene oxide-propylene oxide-b-ethylene oxide) BCPs in epoxy consisting of a bisphenol A diglycidyl ether (BADGE) monomer and an ionic liquid latent curing agent for use as a 3D-printable thermoset or as a high-toughness adhesive. Our previous research has focused on determining the structure-property relationships of these materials, particularly in relation to their rheological and mechanical properties. However, while our observations had shown us that the processing history of the materials was critically important as well, we had not been able to explicitly identify the relationships. Our previous approach using traditional experimental methods to determine the structure-property relationships of the materials precluded us from performing an in-depth study of the processing relationships due to the large number of potential variables and the length and cost of the experiments. In this study, we have changed our approach to leverage active learning to accelerate our ability to navigate through the complex variable space and enable the optimization of the processing methods as well as the composition to develop a deeper understanding of the structure-property-processing relationships.