Machine-learning prediction of facet-dependent CO coverage on Cu electrocatalysts

Copper-based electrocatalysts, which hold great promise in selectively reducing CO2 into multicarbon products, have attracted a lot of recent interest, both experimentally and theoretically. While many studies have suggested a strong dependence of catalytic selectivity on the concentration of the *CO reaction intermediate on Cu surface, it remains challenging for a direct experimental probe of the CO coverage. This necessitates a reliable computational method that can accurately establish the theoretical coverage-dependent phase diagram of CO adsorbates on the catalyst. Here we propose a scheme composed of density functional theory (DFT) calculations, machine-learning force fields (MLFF) and graph neural networks (GNN) as a solution. This method enables a fast screening of 7 million adsorption configurations based on a small set of DFT data, with a balance between accuracy and efficiency tuned by the combinatorial use of MLFF and GNN models. We have investigated 8 different Cu facets, and discovered that the high-index facets such as (310), (210) and (322) exhibit a much higher CO coverage than the low-index counterparts such as (111), leading to an increased opportunity for C-C coupling for the former. Our results can provide a new perspective for the understanding of the fundamental role of CO coverage on Cu surface for electrochemical CO2 reduction.