Selectivity- and Activity-Aware Catalyst Descriptors for CO$_2$ Hydrogenation on Alloy Nanocatalysts using Machine-Learned Force Fields

Adsorption energy distributions (AEDs) have emerged as a powerful and increasingly adopted descriptor for catalytic performance in high-entropy alloys and, more recently, in conventional metallic alloy nanocrystal catalysts. By accounting for diverse adsorption sites and crystallographic facets, AEDs more fully represent nanoparticle-based catalytic surfaces and show strong promise for accelerating rational design and discovery of heterogeneous catalysts, especially for CO$_2$ hydrogenation. However, previous approaches have not sufficiently resolved facet-specific contributions, despite the catalytic significance and prevalence of certain Miller planes in nanoscale catalysts, limiting their applicability in predicting activity and selectivity. Here, we introduce an updated facet-resolved framework for predicting catalytic activity, which also enables insight into selectivity toward C1 products. Universal machine-learned force fields trained on Open Catalyst Project data were employed to compute adsorption energetics across 226 experimentally observed metals, binary alloys, and ternary alloys, encompassing 1.4 million adsorption sites on 2,626 crystallographically distinct surfaces. Using statistical and unsupervised learning techniques, we analyzed facet-specific AEDs to identify highly active and methanol-selective facets. Our approach provides insight into the relationship between structure and catalytic performance metrics like activity and selectivity, and presents a set of alloy compositions and their respective surface orientations for experimental validation toward highly selective CO$_2$ hydrogenation.

Source: arXiv:2605.07714v1 - http://arxiv.org/abs/2605.07714v1 PDF: https://arxiv.org/pdf/2605.07714v1 Original Link: http://arxiv.org/abs/2605.07714v1