• Publisher-DOI

Because ceria (CeO2) is a key ingredient in the formulation of many catalysts, its catalytic roles have received a great amount of attention from experiment and theory. Its primary function is to enhance the oxidation activity of catalysts, which is largely governed by the low activation barrier for creating lattice O vacancies. Such an important characteristic of ceria has been exploited in CO oxidation, methane partial oxidation, volatile organic compound oxidation, and the water–gas shift (WGS) reaction and in the context of automotive applications. A great challenge of such heterogeneously catalyzed processes remains the unambiguous identification of active sites.

In oxidation reactions, closing the catalytic cycle requires ceria reoxidation by gas-phase oxygen, which includes oxygen adsorption and activation. While the general mechanistic framework of such processes is accepted, only very recently has an atomic-level understanding of oxygen activation on ceria powders been achieved by combined experimental and theoretical studies using in situ multiwavelength Raman spectroscopy and DFT.

Recent studies have revealed that the adsorption and activation of gas-phase oxygen on ceria is strongly facet-dependent and involves different superoxide/peroxide species, which can now be unambiguously assigned to ceria surface sites using the combined Raman and DFT approach. Our results demonstrate that, as a result of oxygen dissociation, vacant ceria lattice sites are healed, highlighting the close relationship of surface processes with lattice oxygen dynamics, which is also of technical relevance in the context of oxygen storage-release applications.

A recent DFT interpretation of Raman spectra of polycrystalline ceria enables us to take account of all (sub)surface and bulk vibrational features observed in the experimental spectra and has revealed new findings of great relevance for a mechanistic understanding of ceria-based catalysts. These include the identification of surface oxygen (Ce–O) modes and the quantification of subsurface oxygen defects. Combining these theoretical insights with operando Raman experiments now allows the (sub)surface oxygen dynamics of ceria and noble metal/ceria catalysts to be monitored under the reaction conditions.

Applying these findings to Au/ceria catalysts provides univocal evidence for ceria support participation in heterogeneous catalysis. For room-temperature CO oxidation, operando Raman monitoring the (sub)surface defect dynamics clearly demonstrates the dependence of catalytic activity on the ceria reduction state. Extending the combined experimental/DFT approach to operando IR spectroscopy allows the elucidation of the nature of the active gold as (pseudo)single Au+ sites and enables us to develop a detailed mechanistic picture of the catalytic cycle. Temperature-dependent studies highlight the importance of facet-dependent defect formation energies and adsorbate stabilities (e.g., carbonates). While the latter aspects are also evidenced to play a role in the WGS reaction, the facet-dependent catalytic performance shows a correlation with the extent of gold agglomeration. Our findings are fully consistent with a redox mechanism, thus adding a new perspective to the ongoing discussion of the WGS reaction.

As outlined above for ceria-based catalysts, closely combining state-of-the-art in situ/operando spectroscopy and theory constitutes a powerful approach to rational catalyst design by providing essential mechanistic information based on an atomic-level understanding of reactions.