Theoretical insight into the strong size-dependence of dry reforming of methane over Ru/CeO2

Abstract

Transitional metals catalyzed dry reforming of methane (DRM) usually shows strongly size-dependence behavior, but the atomic level understanding is lacking. The DRM over Run/CeO2(111) (n = 1,2,4,7) were studied systematically by using density functional theory calculation combined with microkinetic modeling analysis in order to explore the nature behind the Ru size effect on the DRM catalytic activity. By density functional theory calculation, it was found that the CH4 activation and CO2 activation processes in DRM reaction are size-dependence and the Ru coverage of optimal chemical performance is around 0.1–0.2 Monolayer (i.e., Ru2 and Ru4 models). The main reason is that Ru2 and Ru4 models could activate CO2 much more easily than other models either with lower Ru coverage like Ru1 or higher Ru coverage like Ru7 model. By using the energy decomposition scheme, it was shown that the Pauli repulsion between two dissociated species and the absorption energy of CO* have crucial effects for the chemical performance of the CO2 activation step, and Ru2 and Ru4 models have weak Pauli repulsion and strong absorption ability of CO. The Ru2 and Ru4 models have optimal chemical performance than other models by DRM free energy profiles. By coke formation analysis, the Ru2 model is difficult to form coke on surface owing to the similar ability for the carbon-carbon coupling step and carbon-oxygen step. Microkinetic model simulation shown that the reaction rates of DRM reaction are also size-dependence and the Ru coverage of highest rate is around 0.1–0.2 Monolayer. By those results, the Ru2 and Ru4 models has the optimal chemical performance for DRM reaction. The high catalytic activity of Ru2(Ru4) model can be attributed to the properties of balances the relationships between cleavage steps (including CH4 and CO2 cleavage) and oxidation processes (mainly CH* oxidation process). It was found that CO2 activation is the rate-controlling step of DRM in terms of the microkinetic modeling results, and the 2D cluster model (such as Ru1, Ru2) could form the interfacial oxygen vacancy more easily than 3D cluster model (such as Ru4, Ru7) by oxygen reverse spillover mechanism. Our results provide further mechanistic understanding for the behaviors of DRM reaction over various Ru coverage on CeO2(111) in order to design more efficient catalysts for DRM by adjusting the loading of transition metals.