▎ 摘　　要

The non-noble transition metals (Sc, Ti, V, Cr, Mn) embedded in graphene as single-atom catalysts have been comprehensively screened for CO oxidation using density functional theory calculations. Among these options, Mn-graphene is predicted to have superior activity for CO oxidation. This conclusion is based on the binding energy between metal atom and graphene substrate, diffusion barrier of metal atom on graphene, and reaction barrier based on the transition state analysis. On the other hand, Sc-, Ti-, V-, and Cr-graphene bind O-2 too strongly. This will lead to catalyst poisoning by O for these systems. We expect that Mn-graphene should be straight forward to fabricate experimentally, and predict that it will be a novel, stable, and efficient single-atom catalyst. For Mn-graphene, the Langmuir-Hinshelwood (LH) and Eley-Rideal (ER) mechanisms of CO oxidation have been investigated. However, the CO molecules cannot interact with surface activated O-2 on graphene to form carbonate-like CO3 complexes or other intermediates. This demonstrates that CO oxidation will not proceed via the ER mechanism. The reaction mechanism for catalysis of CO oxidation occurs in two steps: The LH mechanism CO + O-2 -> OCOO -> CO2 + O followed by the ER mechanism CO + O -> CO2. The energy barriers are 0.57-0.69 eV and 0.08 eV, respectively. These barriers are comparable to or smaller than those for Ni and Mo, indicating high activity. Brief molecular dynamics simulations were also performed on this system. We predict that Mn-graphene can be used as a single-atom catalyst for CO oxidation over a broad range of temperatures. The present work should inspire experimental work on synthesis of novel single-atom catalysts.