Despite extensive research on methanol synthesis over Cu/ZnO, the active sites are still highly debated. With the isotope labeling techniques, experiments showed that CO<sub>2</sub> is the major carbon source for producing methanol, however density functional theory (DFT) studies suggested that the major carbon source is CO. If DFT results are correct, then we can conclude that metallic copper is not the active site. However, it is well known that conventional exchange-correlation (XC) functionals overestimate adsorption energies of molecules on transition metals and underestimate reaction barriers, due to the self-interaction error. In this work, we systematically investigated the reaction energies of the CO and CO2 hydrogenations over stepped copper surface using the random phase approximation (RPA) correlation functional. We established the accuracy of RPA by computing the adsorption energies of selected molecules on platinum surface. RPA results agree with the experiments within 0.3 eV. Hybrid functionals, which are often believed to predict better thermochemistry, overestimate more the adsorption energies than the generalized gradient approximation (GGA) functionals. RPA gives similar overall reaction barriers for CO and CO2 hydrogenations over copper, indicating that they may proceed at similar rates. Microkinetic modeling is performed to gain a detailed insight into the kinetics. Our results point out the importance of using many-body electronic structure methods, such as RPA, to resolve many challenging problems in heterogeneous catalysis, such as pinning down the active sites.

Can copper be the active site in methanol synthesis?

254th National Meeting (2017)

The official archive and repository for all ACS Meetings & Expositions abstracts and select posters or presentations.