Activation of O₂ on Cu, Ag, and Au surfaces for the epoxidation of ethylene: Dipped adcluster model study

Aiming to clarify why only silver is an effective catalyst for the partial oxidation of ethylene, we studied theoretically the reactivity and the stability of oxygen species on Cu, Ag, and Au surfaces. We used the dipped adcluster model (DAM), since electron transfer from metal to oxygen is important, and the SAC/SAC-CI method, since several electronic states are involved. We found that if Superoxide species exists on the surface, both Cu and Au surfaces show a reactivity similar to Ag surface, leading smoothly to ethylene oxide, and the barriers leading to complete oxidations should be very high. Therefore, the point is the relative stability of the various oxygen species and, in particular, the stability of the Superoxide species on the metal surface. On Cu, Superoxide is much less stable than peroxide, which again is less stable than the dissociated species, and no barrier exists for the conversion from Superoxide to peroxide. On Au, our DAM calculations show that the electron flow from the bulk metal into the adcluster does not occur, so that the molecularly adsorbed oxygen species, as well as the dissociative ones, find it difficult to exist stably on the clean surface. On Ag, Superoxide should certainly have some life time to react with ethylene to give ethylene oxide, which is considered to be the origin of the unique catalytic activity of silver for the epoxidation of ethylene. This is related to the ability of electron transfer and to the geometry of the Ag surface. We have proposed a basic idea for a new catalytic design of the epoxidation reaction.