A Comparative Study of the p(2 × 2)-CO/M(111), M=Pt,Cu,Al Chemisorption Systems

We propose a model for CO chemisorption on late transition metal, noble metal, and main-group surfaces based on the results of energy partitioning studies of surface−CO bonding for the CO/M(111), M=Pt,Cu,Al chemisorption systems. Plane-wave density functional theory was used to calculate the chemisorption geometry for CO on the top, bridge, and hollow sites of the M(111) (M=Pt,Cu,Al) surfaces and to verify the experimentally determined preference for top site CO chemisorption on all three surfaces. To construct a chemically intuitive, molecular orbital based model of surface−CO bonding, an energy partitioning analysis of surface−CO bonding was carried out within a tight binding scheme based on the extended Hückel method. Although within this one-electron formalism we are no longer able to make quantitative assesments of bonding, we are able to readily extract surface−CO bonding trends. By expanding the orbital basis on CO to include energetically low-lying nonfrontier orbitals and explicitly evaluating the role of the surface s and p bands in surface−CO bonding, we note several discrepancies between our model and traditional, frontier orbital based models of surface−CO interaction. Especially important is the role of the CO(4σ) orbital. We note that for CO chemisorption on all three surfaces, the energetic preference for top site chemisorption is the result of a balance between the stabilization associated with the formation of the surface−CO bond and chemisorption-site-dependent changes in both C−O bonding and M−M (M=Pt,Cu,Al) bonding within the surface layer on chemisorbing CO. Further, by choosing to consider CO chemisorption on the Cu(111) surface as part of a continuous transition from CO chemisorption on late transition metal surfaces to CO chemisorption on sp-metal surfaces, we are able to assess the degree to which we may refer to copper as an sp-metal.