Modular Global Optimization using Molecular Graphs : A Case Study of ZnO/Cu Surface for Methanol Synthesis Reaction

Heterogeneous and electrocatalysts play a crucial role in enabling various industrial chemical transformations, with quantum chemistry calculations serving as a fundamental tool for investigating their atomic-scale properties. Advances in computational power have facilitated the study of increasingly complex catalytic systems, particularly metal-metal oxide interfaces under realistic reaction conditions. However, these studies remain inherently constrained by approximations in computational models, which often fail to fully capture the intricacies of catalytic phenomena. Additionally, the configurational space associated with such systems is too large to be systematically explored using chemical intuition alone. To address these challenges, we introduce GG, a modular graph-based Python package. This approach enables the systematic exploration of configurational space on common catalytic surfaces using molecular graph representations, allowing for efficient scaling to larger systems. We demonstrate the capabilities of GG through a case study for the ZnOxHy/Cu system to gain insights into the active site for methanol synthesis under reaction conditions. The proposed strategy is broadly applicable and can be extended to a wide range of complex atomic systems.