Fine-Tuning Dual Single-Atom Metal Sites on Graphene toward Enhanced Oxygen Reduction Reaction Activity

The oxygen reduction reaction (ORR) remains at the forefront of research in diverse energy and sustainability domains. While graphene-supported single-atom catalysts (SACs) have garnered attention for optimizing ORR efficiency, tailoring the interactions between adjacent single-atom sites presents intricate challenges. In this study, we leveraged density functional theory (DFT) calculations and cutting-edge machine learning (ML) techniques to explore 144 graphene-supported SACs, featuring interacting M₁-N₄ and M₂-N₄ moieties (M₁, M₂ = Mn, Fe, Co, Ni, Cu, Ru, Rh, Pd, Ag Ir, Pt, Au), denoted as M₁-M₂. By tailoring these interactions, we discovered 13 exceptional SACs outperforming the benchmark catalyst Fe(OH)-N₄, including the best-performing Fe-Pd and several non-noble-metal SACs like Fe-Ag, Ag-Cu, and Ag-Ag. Venturing further, our ML models effectively capture the correlation between single-atom metal properties and overpotential, offering tools for rational electrocatalyst design. Our study illuminates the path to efficient SAC-catalyzed ORR, fostering a sustainable, energy-efficient future.