(179c) A Parameterization Strategy for Coverage-Dependent DFT-Based Microkinetic Modeling of Surface Catalytic Reactions

The impact of coverage dependence, arising from lateral adsorbate interactions, on microkinetic modeling predictions is well known. The influence of coverage dependence on binding/activation energies of surface reactions can be elucidated using Density Functional Theory (DFT) calculations, which treat the surface as a periodically repeated unit cell. However, the translation of these results to appropriately parameterize microkinetic models that invoke commonly-employed mean-field assumptions remains an intellectual challenge. In this work, we develop a strategy to parameterize a mean-field approach for a prototypical two-step catalytic reaction network with a generic adsorbate interaction model, using results from previously employed and accurate lattice-based kinetic Monte Carlo (kMC) simulations as the benchmark. We compare statistically averaged activation barriers from kMC simulations across a broad range of reaction conditions, against activation barriers predicted from configurationally-distinct coverage representations on a 4x4 periodic unit cell and deduce the appropriate active site and spectator arrangements from these comparisons. We posit a single descriptor, capturing the neighboring information around the active site and show that it correlates one-to-one with the steady state coverage from the kMC simulations. These results point to a lucid protocol that utilizes a handful of carefully chosen DFT calculations to construct and evaluate coverage-cognizant mean-field microkinetic models.