An in Situ Surface-Enhanced Infrared Absorption Spectroscopy Study of Electrochemical CO2 Reduction: Selectivity Dependence on Surface C-Bound and O-Bound Reaction Intermediates

The CO2 electroreduction reaction (CO2RR) is a promising avenue to convert greenhouse gases into high-value fuels and chemicals, in addition to being an attractive method for storing intermittent renewable energy. Although polycrystalline Cu surfaces have long been known to be unique in their capabilities of catalyzing the conversion of CO2 to higher-order C1 and C2 fuels, such as hydrocarbons (CH4, C2H4, etc.) and alcohols (CH3OH, C2H5OH), product selectivity remains a challenge. Rational design of more selective catalysts would greatly benefit from a mechanistic understanding of the complex, multiproton, and multielectron conversion of CO2. In this study, we select three metal catalysts (Pt, Au, Cu) and apply in situ surface enhanced infrared absorption spectroscopy (SEIRAS) and ambient-pressure X-ray photoelectron spectroscopy (APXPS), coupled to density-functional theory (DFT) calculations, to get insight into the reaction pathway for the CO2RR. We present a comprehensive reaction mechanism for the CO2RR and show that the preferential reaction pathway can be rationalized in terms of metal-carbon (M-C) and metal-oxygen (M-O) affinity. We show that the final products are determined by the configuration of the initial intermediates, C-bound and O-bound, which can be obtained from CO2 and (H)CO3, respectively. C1 hydrocarbons are produced via OCH3,ad intermediates obtained from O-bound CO3,ad and require a catalyst with relatively high affinity for O-bound intermediates. Additionally, C2 hydrocarbon formation is suggested to result from the C-C coupling between C-bound COad and (H)COad, which requires an optimal affinity for the C-bound species, so that (H)COad can be further reduced without poisoning the catalyst surface. It is suggested that the formation of C1 alcohols (CH3OH) is the most challenging process to optimize, as stabilization of the O-bound species would both accelerate the formation of key intermediates (OCH3,ad) but also simultaneously inhibit their