Hybrid sampling approach to machine-learning potentials for gas adsorption: Hydrogen adsorption in MOF-303

Porous materials such as metal–organic frameworks (MOFs) are widely studied for applications in catalysis and gas adsorption, including carbon dioxide capture and hydrogen storage for energy and environmental challenges. Accurate simulation of gas adsorption in these materials remains challenging because most approaches rely on classical force fields, which lack first-principles accuracy. Here, we present an approach to generating machine-learning potentials (MLPs) tailored for grand canonical Monte Carlo (GCMC) simulations of gas adsorption. The training dataset combines snapshots from GCMC sampling, evaluated by single-point density functional theory (DFT), with trajectories from DFT-based molecular dynamics. We demonstrate this approach for hydrogen adsorption in MOF-303, achieving accurate energies and forces and robust GCMC performance. The simulation also revealed four primary and four secondary adsorption sites formed near N, O, C, and –OH motifs, highlighting the material’s structural heterogeneity and adsorption behavior. The resulting MLP enables calculation of adsorption isotherms, isosteric heat values, and diffusion coefficients across temperatures. This hybrid-sampling framework bridges DFT-level accuracy and large-scale simulation capability, providing a transferable tool for screening and designing porous materials for gas storage and separation.