The functionalization of perovskite interfaces with small molecules is a widely employed strategy for passivating surface defects and regulating charge transfer at grain boundaries in polycrystalline metal halide perovskite thin films. Traditionally, the selection of additives for experimentation and device integration relies on chemical intuition and heuristics. Atomistic modeling, particularly density functional theory (DFT), enables the direct investigation of molecular interactions at perovskite interfaces, offering valuable insight into experimental observations and facilitating the prediction of physical property modifications induced by surface functionalization. However, modeling molecular interactions at interfaces presents several challenges, including sensitivity to molecular placement, dependence on the level of theoretical approximation, and limited transferability across different perovskite compositions. This work introduces an advanced perovskite interfacial modeling framework that harnesses the predictive capabilities of DFT through process automation and systematic database generation, providing a robust and scalable approach for understanding and optimizing perovskite surface functionalization. Using this high-throughput framework, we systematically characterize the chemical and electronic effects of surface functionalization with a diverse set of two hundred molecules, selected to maximize chemical diversity within the ZINC20 dataset. Our study focuses on the PbI2-rich (100) surface of formamidinium lead iodide (FAPbI3), generating a comprehensive dataset that includes molecular features, DFT-calculated adsorption energies, surface charge transfer, and quantitative descriptors of surface distortion. Further, we conduct detailed correlation analyses to understand how the molecular features of passivating ligands impact the FAPI surface. Notably, we find that electrotopological descriptors exhibit a strong correlation with surface charge transfer, suggesting the potential of these descriptors to predict the doping ability of perovskite additives.
