Fingerprints of energy dissipation for exothermic surface chemical reactions: O$_2$ on Pd(100)

TL;DR

This study uses first-principles simulations to show that energy dissipation significantly influences the sticking coefficient of O₂ on Pd(100), revealing that similar experimental data can arise from different underlying mechanisms.

Contribution

It introduces a comprehensive simulation approach incorporating energy dissipation effects, demonstrating their importance in accurately modeling surface reaction kinetics.

Findings

Energy dissipation qualitatively affects sticking curves.

Good agreement with experiments achieved using GLO model.

Different mechanisms can produce similar sticking data.

Abstract

We present first-principles calculations of the sticking coefficient of O$_2$ at Pd(100) to assess the effect of phononic energy dissipation on this kinetic parameter. For this we augment dynamical simulations on six-dimensional potential energy surfaces (PESs) representing the molecular degrees of freedom with various effective accounts of surface mobility. In comparison to the prevalent frozen-surface approach energy dissipation is found to qualitatively affect the calculated sticking curves. At the level of a generalized Langevin oscillator (GLO) model we achieve good agreement with experimental data. The agreement is similarly reached for PESs based on two different semi-local density-functional theory functionals. This robustness of the simulated sticking curve does not extend to the underlying adsorption mechanism, which is predominantly directly dissociative for one functional or molecularly trapped for the other. Completely different adsorption mechanisms therewith lead to rather similar sticking curves that agree equally well with the experimental data. This highlights the danger of the prevalent practice to extract corresponding mechanistic details from simple fingerprints of measured sticking data for such exothermic surface reactions.