Accelerated Design of Block Copolymers: An Unbiased Exploration Strategy via Fusion of Molecular Dynamics Simulations and Machine Learning

Abstract

Star block copolymers (s-BCPs) have potential applications as novel surfactants or amphiphiles for emulsification, compatbilization, chemical transformations and separations. s-BCPs are star-shaped macromolecules comprised of linear chains of different chemical blocks (e.g., solvophilic and solvophobic blocks) that are covalently joined at one junction point. Various parameters of these macromolecules can be tuned to obtain desired surface properties, including the number of arms, composition of the arms, and the degree-of-polymerization of the blocks (or the length of the arm). This makes identification of the optimal s-BCP design highly non-trivial as the total number of plausible s-BCPs architectures is experimentally or computationally intractable. In this work, we use molecular dynamics (MD) simulations coupled with reinforcement learning based Monte Carlo tree search (MCTS) to identify s-BCPs designs that minimize the interfacial tension between polar and non-polar solvents. We first validate the MCTS approach for design of small- and medium-sized s-BCPs, and then use it to efficiently identify sequences of copolymer blocks for large-sized s-BCPs. The structural origins of interfacial tension in these systems are also identified using the configurations obtained from MD simulations. Chemical insights on the arrangement of copolymer blocks that promote lower interfacial tension were mined using machine learning (ML) techniques. Overall, this work provides an efficient approach to solve design problems via fusion of simulations and ML and provide important groundwork for future experimental investigation of s-BCPs sequences for various applications.