(7d) Achieving Quantum Mechanical Accuracy for Modeling Reaction-diffusion Process in Liquid-phase Heterogeneous Catalysis

Due to the different time- and length scales involved, diffusion and chemical transformations are traditionally handled through different approximations. For example, density functional theory is used to model chemical transformation (with a highly idealized catalyst model), while pair-wise additive force fields are used to model the diffusion process. However, given the importance of solvation and confinement, it is essential to model the entire catalytic process at or near quantum mechanical accuracy. Recent advances in machine learning (ML) approaches have the potential to make this goal a reality. In this presentation, we will share our recent work in modeling liquid-phase reactions catalyzed by zeolites and highlight successes and limitations.