Mapping adsorption on ionic surfaces via a pairwise potential-based high-throughput approach.

Understanding molecular adsorption on ionic surfaces is crucial for a variety of chemical applications, from heterogeneous catalysis to prebiotic chemistry. Traditional approaches for identifying adsorption sites often rely on computationally expensive methods such as density functional theory (DFT), which limits their applicability to chemically complex surfaces. In this work, we propose an automated high-throughput approach to obtain a complete picture of the adsorbate-surface interaction by means of pairwise Coulomb and Lennard-Jones potentials. Using a grid-based surface scan to calculate per-site potential energies of adsorption, this method efficiently predicts global adsorption minima and all potential binding modes of a surface-adsorbate system, with the only user input being the surface CIF. Our approach is validated by studying formaldehyde (HCO) adsorption on forsterite (MgSiO), a common silicate, and l-cysteine adsorption on cadmium sulfide (CdS). The predicted adsorption configurations and energies are compared with DFT values in the literature, showing good agreement and confirming the accuracy of our method. Our workflow provides a rapid means of exploring large configurational spaces and identifying stable adsorption structures, making it particularly useful for complex surfaces with multiple interaction sites. The simplicity of the model, combined with its accuracy, suggest it could be employed to discover new catalytic pathways on chemically complex ionic surfaces.