Extrapolation performance improvement by quantum chemical calculations for machine-learning-based predictions of flow-synthesized binary copolymers

Abstract

The properties of polymers are highly dependent on the combination and composition ratio of the monomers used to prepare them; however, the large number of available monomers makes an exhaustive investigation of all the possible combinations difficult. In the present study, five binary copolymers were prepared by radical polymerization using a flow reactor and the prediction performance of a machine learning model constructed using the obtained data was evaluated for the interpolation and extrapolation regions. Copolymer analysis was performed using ultra-high-performance liquid chromatography, and the measurement results were analysed to calculate the monomer conversion and monomer composition ratio in the polymer, which were used as objective variables. A prediction model was constructed using the process variables during polymerization and additional molecular descriptors (i.e., molecular flags (one-hot encoding), fingerprints or quantum chemical calculation values) related to the monomer type as explanatory variables. In the interpolated regions where all monomer types used were included in the training data, the prediction accuracy was high irrespective of the molecular descriptors added to the process variables. In the extrapolation region, the model that included explanatory variables corresponding to quantum chemical calculation values representing the energy generated when radical reactions occur, showed a high prediction accuracy for each objective variable. We found that quantum chemical calculation values (especially the molecular orbital energy of monomers in the extrapolation region) are important factors in the search for new binary copolymers prepared by radical polymerization. The proposed model is expected to accelerate the development of polymers using new monomers.