Fig. 3

Experimental and theoretical elucidation of palladium-promoted In₂O₃ catalysts. a XPS Pd3d core-level spectra, and b Pd–K-edge EXAFS spectra of the CP and DI catalysts in fresh form and after use in CO₂ hydrogenation for 1 h. EXAFS spectra simulated for the latter are added in red. The dashed lines mark expected positions of signals of the species indicated^52,53,54. c Abundance of Pd-containing fragments emitted from the surface of used CP and DI materials in TOF-SIMS analysis. d H₂-TPR profiles at 5 MPa of the fresh CP and DI catalysts with In₂O₃ as a reference. e XPS O1s core-level spectra of CP and DI catalysts after use for 1 and 16 h. The contribution of lattice oxygen, oxygen close to a defect, and oxygen in a hydroxyl group retrieved from deconvolution are marked in red, green, and blue, respectively, and labeled with their relative abundance in %. f In¹¹⁵ NMR spectra collected for pure In₂O₃ and the CP and DI catalysts in fresh form and after use for 16 h. g (left) Structure of a regularly terminated In₂O₃(111) surface featuring, owing to the strong anisotropy, a depression (pocket) where Pd atoms preferentially accommodate in the fresh DI catalyst. The coloring of In atoms at the surface represents the energy associated with their replacement by palladium (E_seg, in eV), which takes place upon CP. 1–3 mark the lattice positions most preferred by the promoter. (right) Initial palladium-promoted In₂O₃(111) surface models for the CP and DI catalysts and their evolution to account for structural modifications occurring upon their use for in the reaction. The energy difference between states for each transition (ΔE, in eV) and the Bader charges of the Pd atoms (q_B in |e^‒|) are indicated in blue. Atoms are shaded with a progressively lighter color upon increasing distance from the surface towards the bulk