Insights into Supported Subnanometer Catalysts Exposed to CO <i>via</i> Machine-Learning-Enabled Multiscale Modeling

Subnanometer catalysts offer high noble metal utilization and superior performance for several reactions. However, understanding their structures properties on an atomic scale under working conditions is challenging due to the large configurational space. Here, we introduce efficient multiscale framework predict stability exposed adsorbate. The integrates a comprehensive toolset including density functional theory (DFT) calculations, cluster expansion, machine learning, structure optimization. end-to-end machine-learning workflow guides DFT data generation enables significant computational acceleration. We demonstrate approach CO-adsorbed Pdn (n = 1–55) clusters CeO2(111). Simulation results reveal that CO can facilitate restructuring by stabilizing smaller planar bilayer of specific intermediate sizes, consistent with experimental reports. Metal–support interactions, preferential adsorption, nuclearity control catalyst stability. allows automatic discovery stable systematic strategy exploit in subnanometer scale.