We formulated a potential energy surface for the water-hydrogen dimer using Buckingham-like potentials for repulsions and dispersions, and point charges, including the water m-site. Using 1140 water-hydrogen configurations created from numerous scans, we calculated the F12-MP2 ab initio binding energy with a Dunning quadruple zeta basis set. The resulting coordinates and energies are used as the training set for a Radial Basis Function Neural Network optimized to fit the energy differences between the Buckingham-Coulomb surface and ab initio. The average deviation between the ab initio and the resulting model energies for the training set is on the order of one ten-thousandth of a wavenumber.

Fitting a hydrogen-water dimer potential energy surface

2019 Spring National Meeting

The official archive and repository for all ACS Meetings & Expositions abstracts and select posters or presentations.