Catalytic NO Oxidation Pathways and Redox Cycles on Dispersed Oxides of Rhodium and Cobalt

Abstract

AbstractThe elementary steps and site requirements for the oxidation of NO on Rh and Co and the oxidation state of the catalysts were probed by isotopic tracers, chemisorption methods, and kinetic measurements of the effects of the pressures of NO, O2, and NO2 on turnover rates. On both catalysts, NO oxidation rates were first order in NO and O2 and were inversely proportional to NO2 pressure, as observed on Pt and PdO. These data implied that O2 activation on an isolated vacancy (*) on the catalyst surfaces that were saturated with oxygen (O*) was the kinetically relevant step. Quasi‐equilibrated NO–NO2 interconversion steps established the coverage of * and O* and the chemical potential of oxygen during the catalysis. These chemical potentials set the oxidation state of Rh and Co clusters and were described by an O2 virtual pressure, which was determined from the formalism of non‐equilibrium thermodynamics. RhO2 and Co3O4 were the phases that were present during NO oxidation, which had several consequences for catalysis. Turnover rates increased with increasing cluster size because the vacancies that were needed for O2 activation were more abundant on large oxide clusters, which delocalized electrons better than small clusters. NO oxidation turnover rates on RhO2 and Co3O4 were higher than expected from the oxygen‐binding energy on Rh and Co metal surfaces and from the reduction potentials of Rh3+ and Co2+. These NO oxidation rates were consistent with the rates on Pt and PdO when one‐electron‐reduction processes, which were accessible for Rh4+ and Co3+ but not for Pt2+ and Pd2+, were used to describe the reactivity of RhO2 and Co3O4. One‐electron redox cycles caused the 16O2–18O2 exchange rates to be higher than the NO oxidation rates, in contrast with their analogous values on Pt and PdO, although O2 activation on the vacancies limited NO oxidation and O2 exchange on all of the catalysts. One‐electron redox cycles allowed electron sharing between metal cations and a facile route to form vacancies on RhO2 and Co3O4. This interpretation of the data highlighted the role of vacancies in kinetically relevant O2‐activation steps to explain the higher reactivity of larger metal and oxide clusters and to provide a common framework to describe NO oxidation and the active species on catalysts of practical interest.