Exploring the Potential of Metal-Doped Graphene as Improved Electrocatalysts for CO₂ Reduction Using Embedded Mean-Field Theory

Carbon-based materials are particularly interesting potential electrocatalysts for CO₂ reduction due to their low cost and ability to form a wide range of nanostructures. Numerous studies have shown single metal atoms embedded in a conductive graphene network to exhibit unique electrocatalytic properties. These systems present a computational challenge because, in contrast to an extended metallic surface, the electronic structure of the active site in metal-doped graphene is localized and thus requires a density functional theory (DFT) treatment beyond the generalized-gradient approximation (GGA) level. However, higher-level (e.g., hybrid-DFT) methods conducted on the whole system are prohibitively expensive. In this work, we use embedded mean-field theory (EMFT) to overcome the challenges of accurately modeling the electrocatalytic activity of metal-doped graphene at reasonable computational cost. We show that EMFT enables efficient exploration of metal-doped graphene as potentially improved electrocatalysts over pure transition metals for CO₂ reduction.