Bayesian optimization of radical polymerization reactions in a flow synthesis system

Proportions of monomers in a copolymer will greatly affect the properties of materials. However, due to a phenomenon known as composition drift, the proportions of monomers in a copolymer can deviate from the value expected from the raw monomer ratio because of differences in monomer reactivity. It is therefore necessary to optimize the polymerization process to account for such composition drift. In the present study, styrene-methyl methacrylate copolymers were generated using a flow synthesis system and the processing variables were tuned employing Bayesian optimization (BO) to obtain a target composition. First trials of BO with generation of four candidate points per cycle, completed the optimization within five cycles. Subsequent Bayesian Optimization (BO) trial, using 40 points per cycle, identified several sets of processing conditions that could achieve the desired copolymer composition, accompanied by variations in other physical properties. To optimize the monomer composition ratio in the polymer, it was discovered from a data science perspective that the solvent-to-monomer ratio was as crucial as the styrene proportions. The role of each variable in the radical polymerization reaction was elucidated by assessing the extensive array of processing conditions while evaluating several broad trends. The proposed model confirms that specific monomer proportions can be produced in a copolymer using machine learning while investigating the reaction mechanism. In the future, the use of multi-objective BO to fine-tune the processing conditions is expected to allow optimization of the copolymer composition together with adjustment of physical properties.