Computational screening for selective catalysts: Cleaving the C-C bond during ethanol electro-oxidation reaction

The efficiency of direct ethanol fuel cells suffers from the partial oxidation of ethanol into acetic acid as opposed to the complete oxidation of this fuel to CO2. Herein, we support the quest for a selective catalyst for ethanol electro-oxidation to CO2, building on our previous mechanistic hypothesis based on experimental insight and DFT computations. We derive a simple descriptor of the expected selectivity towards full oxidation, Omega, as a function of the adsorption energy of atomic C and O. Three different families of catalyst surfaces are screened using this descriptor: monometallics, bimetallics and conducting metal oxides, totaling to 600 surfaces. In agreement with available experimental data, no single metal surface is more selective for total oxidation than platinum and palladium. While the selected conducting oxides were not predicted to be selective towards splitting the C-C bond, structurally-controlled monometallics (such as Pd(100)) or some bimetallics (Pd3Ag) are found to be competitive with the most stable facet, (111), of Pd and Pt. Despite this very extensive screening, no very promising catalyst has been identified. This highlights the need to identify catalysts for acetate oxidation or to exploit support effects and electrolyte engineering to profit from the full power of direct ethanol fuel cells. (C) 2018 Elsevier Ltd. All rights reserved.