Material Chemistry with Dynamic Charge Reactive Potentials

Computational methods are useful tools in the investigation of atomic and molecular dynamics and associated mechanisms at surfaces and interfaces. Physics-based classical potentials are a class of computational method that is useful for use in classical atomistic simulations of systems made up of thousands to many billions of atoms. These potentials consist of parameterized functions that capture aspects of atomic and molecular interactions within these material systems.

The focus of this presentation is on the third-generation charge-optimized many-body (COMB3) potential. COMB3 was developed to enable an atomic-scale description of systems that include combinations of metallic, ionic, and covalent bonding under the same framework. The framework enables the system to determine the charge state of an atom or ion and manifest the physically appropriate type(s) and strength of local bonding as a function of environment correctly and autonomously. The framework further includes a combination of atomic-specific, bond-specific, bond-angle-specific parameters; the former is the same regardless of material, and only new bond-specific and bond-angle-specific parameters are required to extend existing elements to new compounds.

This presentation will provide an overview of the COMB3 potential and illustrate its utility in the study of water-metal surfaces for Cu and Pt, as well as for Pt nanoparticle (NP) interactions. In particular, classical molecular dynamics simulations of oxidized platinum NPs in an explicit water environment were used to investigate their stability and dissolution. The results indicate that the Pt-O layer reduces the kinetic activity for Pt atom dissolution, and is projected to make dissolution more favorable for lower adsorbed oxygen coverages. These findings quantify the effect of oxygen and temperature on the dissolution of oxidized platinum NPs in an explicit water environment similar to the conditions in fuel cells and electrocatalysis. This work was supported by the U.S. Department of Energy, Basic Energy Sciences, CPIMs Program, under Award No. DE-SC0018646.