Microscopic modeling of gas-surface scattering. II. Application to argon atom adsorption on a platinum (111) surface

TL;DR

This paper combines molecular dynamics simulations with a rate equation model to analyze argon atom scattering on platinum surfaces, providing detailed insights into energy transfer, trapping, and sticking probabilities relevant for plasma-surface interactions.

Contribution

It introduces a novel integrated approach using MD simulations and rate equations to study gas-surface scattering dynamics over extended time scales.

Findings

Identification of stationary transition rates after tens of picoseconds

Detailed energy loss and distribution functions for adsorbate atoms

Quantitative sticking probabilities depending on incident energy, angle, and temperature

Abstract

A new combination of first principle molecular dynamics (MD) simulations with a rate equation model presented in the preceding paper (paper I) is applied to analyze in detail the scattering of argon atoms from a platinum (111) surface. The combined model is based on a classification of all atom trajectories according to their energies into trapped, quasi-trapped and scattering states. The number of particles in each of the three classes obeys coupled rate equations. The coefficients in the rate equations are the transition probabilities between these states which are obtained from MD simulations. While these rates are generally time-dependent, after a characteristic time scale $t_E$ of several tens of picoseconds they become stationary allowing for a rather simple analysis. Here, we investigate this time scale by analyzing in detail the temporal evolution of the energy distribution functions of the adsorbate atoms. We separately study the energy loss distribution function of the atoms and the distribution function of in-plane and perpendicular energy components. Further, we compute the sticking probability of argon atoms as a function of incident energy, angle and lattice temperature. Our model is important for plasma-surface modeling as it allows to extend accurate simulations to longer time scales.