Animating Lewis Dots: the development of transferable sub-atomistic force fields for efficient, intuitive, turnkey simulations of chemical reactions

Brandeis University, Waltham MA

Investigators

Abstract

Judith Herzfeld of Brandeis University is supported by an award from the Chemical Theory, Models and Computational Methods program to develop reactive force fields for molecular simulation. Her goal is to make theory a nimbler partner with experiment in exploring and understanding the pathways of complex chemical reactions. Ideally, computational chemistry should be accurate, versatile, efficient and user-friendly. This is a tall order so computational chemistry usually prioritizes only a couple of these goals. Herzfeld and coworkers are developing a new approach that seeks to combine accurate versatility with efficiency and user-friendliness. This work is expected to provide researchers with useful computational tools for simulating chemical reactions in a broad range of realistic solvents. This research may have impacts in catalysis and advanced manufacturing. The project also provides educators with visualization tools that are useful for instructing students in the mechanisms of chemical reactions. The handiest tool for thinking and communicating about reactions has been the semi-classical "Lewis dot" picture. This century old rubric is "classical" in the sense of thinking of valence electrons and kernels as independently mobile particles, and "semi" in the sense that these particles follow non-classical rules. Herzfeld's work seeks to provide a quantitative version of this heretofore qualitative picture by developing force fields for Lewis' sub-atomistic particles that effectively and efficiently incorporate quantum effects. This project builds on their earlier work on the acid-base chemistry of non-metals (with paired valence electrons) and the ionization and spin states of nonmetals and metals (with valence electrons modeled singly). In particular, the researchers are using an improved understanding of the exchange contributions to the energy to extend the range of applicability. The work involves training, validating and demonstrating the potentials that describe the interactions between the particles. The goal is for these simulations be "turn-key" in the sense of requiring no input as to reaction intermediates or products. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.