Evolution of Oxygen Vacancy Sites in Ceria-Based High-Entropy Oxides and Their Role in N2 Activation

Abstract

In this work, a relatively new class of materials, rare earth (RE) based high entropy oxides (HEO) are discussed in terms of the evolution of the oxygen vacant sites (O_v) content in their structure as the composition changes from binary to HEO using both experimental and computational tools; the composition of HEO under focus is the CeLaPrSmGdO due to the importance of ceria-related (fluorite) materials to catalysis. To unveil key features of quinary HEO structure, ceria-based binary CePrO and CeLaO compositions as well as SiO₂, the latter as representative nonreducible oxide, were used and compared as supports for Ru (6 wt % loading). The role of the O_v in the HEO is highlighted for the ammonia production with particular emphasis on the N₂ dissociation step (N_2(ads) → N_ads) over a HEO; the latter step is considered the rate controlling one in the ammonia production. Density functional theory (DFT) calculations and ¹⁸O₂ transient isotopic experiments were used to probe the energy of formation, the population, and the easiness of formation for the O_v at 650 and 800 °C, whereas Synchrotron EXAFS, Raman, EPR, and XPS probed the Ce-O chemical environment at different length scales. In particular, it was found that the particular HEO composition eases the O_v formation in bulk, in medium (Raman), and in short (localized) order (EPR); more O_v population was found on the surface of the HEO compared to the binary reference oxide (CePrO). Additionally, HEO gives rise to smaller and less sharp faceted Ru particles, yet in stronger interaction with the HEO support and abundance of Ru-O-Ce entities (Raman and XPS). Ammonia production reaction at 400 °C and in the 10-50 bar range was performed over Ru/HEO, Ru/CePrO, Ru/CeLaO, and Ru/SiO₂ catalysts; the Ru/HEO had superior performance at 10 bar compared to the rest of catalysts. The best performing Ru/HEO catalyst was activated under different temperatures (650 vs 800 °C) so to adjust the O_v population with the lower temperature maintaining better performance for the catalyst. DFT calculations showed that the HEO active site for N adsorption involves the O_v site adjacent to the adsorption event.

Keywords: DFT; EPR; Synchrotron EXAFS; ammonia; high entropy oxides; isotopic exchange; oxygen vacancies.

Figure 1

Unit cells of (a) pure CeO₂, (b) binary CeGdO, (c) ternary CeLaGdO, and (d) quinary CeLaPrSmGdO.

Figure 2

(a) Unit cell of CeO₂ along (110) surface with vacuum. Bulk region, sublayer, and top-layers are shown. (b) Evolution of oxygen vacancy formation energy in pure CeO₂, binary systems (CeGdO, CePrO), ternary systems (CeLaGdO, CeLaPrO), and quinary (CeLaPrSmGdO) high entropy system.

Figure 3

(A) XRD diffractograms; (B–C) EPR spectra collected at 100 K; (D) Raman spectra collected over the Ru/CePrO catalyst; (E) Raman spectra collected over the Ru/HEO catalyst; (F) H₂-temperature-programmed desorption (H₂-TPD) profiles obtained over all the Ru catalysts of the present study.

Figure 4

(A) Ru 3p, (B) Ce 3d, and (C) O 1s XPS core-level spectra of the prepared catalysts.

Figure 5

(A, B) Linear combination fitting of the XANES spectra collected at the LIII edge of Ce in the HEO sample (left) and CePrO sample (right) to determine all the oxidation states of Ce in the samples; (C) Fourier transforms with their respective real part of the k3 weighted EXAFS spectra collected at the Ce LIII edge in CLPSG-HEO and CePr samples.

Figure 6

(A) HRTEM images along with Fast Fourier Transform (FFT) pattern; (B, C) facets analysis of the Ru/CePrO catalyst.

Figure 7

(A–E) STEM-HAADF RGB analysis and (F) selected area electron diffraction (SAED) over the Ru/CePrO catalyst.

Figure 8

(A–C) HRTEM images of the Ru/HEO catalyst with emphasis on Ru particles at different areas (B) and (C).

Figure 9

(A) HRTEM images of the Ru/HEO catalyst along with Fast Fourier Transform (FFT) pattern and (B–D) facets analysis.

Figure 10

(A–H) STEM-HAADF RGB analysis and (I) selected area electron diffraction (SAED) over the Ru/HEO catalyst.

Figure 11

Catalytic production of NH₃ (A) over Ru-based catalysts of this study in the 10–40 bar pressure range and 400 °C; (B) over the Ru/HEO catalyst at 10 bar, 400 °C following two different activation conditions (650 vs 800 °C).

Figure 12

(A) Transient rates of ¹⁶O₂ (μmol g^–1 s^–1) consumed during transient isothermal oxidation (TIO) at 650 °C, following the different pretreatment conditions, over HEO solid; and (B) amounts of O₂ consumed (μmol g^–1) during TIO, over both solids.

Figure 13

(A) Transient concentration (mol %) response curves of Kr, ¹⁶O₂, ¹⁶O¹⁸O, and ¹⁸O₂ gaseous species recorded during the ¹⁸O₂-TIIE experiment at 650 °C, over the calcined HEO; (B) descriptor as a function of time and pretreatment procedure, estimated during ¹⁸O₂-TIIE over HEO; and (C) descriptor obtained over the both reduced at 650 °C solids.

Figure 14

Top and side views of N₂ Adsorption configurations on Ru4/reduced HEO surface: (a) in absence of hydrogen and (b) in the presence of hydrogen. Gd is shown in purple, La in dark green, Pr in yellow, Sm in pink, Ce in light green, Ru in orange, N in white, H in blue, and O in red.

Figure 15

Top and side views of N Adsorption configurations on reduced quinary CeLaPrSmGdO (111) surface: (a) in absence of hydrogen and (b) in the presence of hydrogen. Gd is shown in purple, La in dark green, Pr in yellow, Ce in light green, Sm in pink, H in blue, N in white, and O in red.