Charge Transfer Potentials

Abstract

The last two decades have witnessed a dramatic rise in computational resources that has facilitated tremendous progress in computational science. In particular, this progress has enabled the application of quantum-based methods such as Hartree-Fock (HF) theory and density functional theory (DFT) to compute the potential energy surfaces of numerous complex reactions that are critical to understanding catalytic reactions. These approaches provide high fidelity because of their explicit treatment of electronic structure; however, their computational cost increases rapidly with system size. Therefore, they are limited to a relatively small number of atoms (o500). To overcome this limitation, classical empirical methods (also known as interatomic potentials) that model molecules and materials at the atomic scale without explicitly treating electrons have been developed and have been employed in molecular dynamics (MD) and Monte Carlo (MC) simulations. Such simulations have been employed to examine catalysis at length and time scales beyond the reach of quantum-based approaches.The main strength of classical empirical potentials is their low computational cost relative to electronic-structure calculations. Recently, systems with billions of atoms have been modeled in MD simulations.