Fast Predictions of Liquid-Phase Acid-Catalyzed Reaction Rates Using Molecular Dynamics Simulations and Convolutional Neural Networks

The rates of liquid-phase, acid-catalyzed reactions relevant to the upgrading of biomass into high-value chemicals are highly sensitive to solvent composition and identifying suitable solvent mixtures is theoretically and experimentally challenging. We show that the atomistic configurations of reactant-solvent environments generated by classical molecular dynamics simulations can be exploited by 3D convolutional neural networks to enable fast predictions of Brønsted acid-catalyzed reaction rates for model biomass compounds. We develop a computational implementation, which we call SolventNet, and train it using experimental reaction data for seven biomass-derived oxygenates in water-cosolvent mixtures. We show that SolventNet can predict reaction rates for additional reactants and solvent systems an order of magnitude faster than prior simulation methods. This combination of machine learning with molecular dynamics enables the rapid screening of solvent systems and identification of improved biomass conversion conditions.