Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.
Nature Communications volume 15, Article number: 5751 (2024 ) Cite this article cas 12008 21 8
Oxygen vacancy (Ov) is an anionic defect widely existed in metal oxide lattice, as exemplified by CeO2, TiO2, and ZnO. As Ov can modify the band structure of solid, it improves the physicochemical properties such as the semiconducting performance and catalytic behaviours. We report here a new type of Ov as an intrinsic part of a perfect crystalline surface. Such non-defect Ov stems from the irregular hexagonal sawtooth-shaped structure in the (111) plane of trivalent rare earth oxides (RE2O3). The materials with such intrinsic Ov structure exhibit excellent performance in ammonia decomposition reaction with surface Ru active sites. Extremely high H2 formation rate has been achieved at ~1 wt% of Ru loading over Sm2O3, Y2O3 and Gd2O3 surface, which is 1.5–20 times higher than reported values in the literature. The discovery of intrinsic Ov suggests great potentials of applying RE oxides in heterogeneous catalysis and surface chemistry.
Oxygen vacancy (Ov)1,2,3 is a ubiquitous anionic point defect in transition and f-element metal oxides. Commonly, the defect type of Ov requires cations with changeable multiple valence states such as Ce3+/4+ 4,5,6 and Ti3+/4+ 7,8. This defect is realized during the treatment in high temperature9 or reductive conditions8,10. The lattice or surface O2− are taken away together with the reduction of metal cations while keeping the crystal structure, leaving Ov within the lattice. On the one hand, the role of Ov in heterogeneous catalysis has been widely reported4,5,6,7,8. For example, the generation of Ov in CeO2 (cubic fluorite structure, \({Fm}\bar{3}m\) space group) accompanied by the reduction of Ce4+ to Ce3+ changes its surface charge distribution and creates electrophilic sites11, which plays a crucial role in improving catalytic performance in CO oxidation, water-gas shift and CO2 reduction5,6,9,10. On the other hand, such redox process is hardly possible with irreducible oxides such as ZrO2, SiO2 and Al2O3, thereby preventing the generation of defect-based Ov12,13,14,15.
In this work, we discover a new type of surface Ov that does not require a point defect formation nor the redox of metal cations. Such intrinsic Ov stems from the natural atomic arrangements in certain crystalline surface of rare earth oxide (RE2O3). We have analysed the structure and surface charge distribution of RE2O3 (such as Sm2O3, Y2O3 and Gd2O3) with body-centred cubic structure (Ia3 space group) based on density functional theory (DFT). An irregular hexagonal sawtooth-shaped structure is found in the (111) surface of those RE2O3, forming intrinsic surface Ov. Next to the Ov are penta-coordinated RE3+ with strong electrophilic nature. These RE2O3 with intrinsic surface Ov are loaded with Ru clusters as active metal (Ru/RE2O3, RE = Y, Sm and Gd) for ammonia decomposition reaction, in which an optimal N-binding strength is required. These Ru/RE2O3 catalysts exhibit excellent catalytic performance that is comparable to the most active Ru/CeO2 catalyst that is equipped with the defect Ov16, despite that the RE cations are not redox active. During the reaction, the intrinsic Ov has desired absorption strength of NH3 at the neighbouring RE3+, and causes Ru species more reducible, which facilitates the initial N–H breaking. Such intrinsic Ov shows promise of utilizing their novel properties of RE2O3 in catalysis, providing suitable adsorption of reaction molecules for oxidation and hydrogenation chemistry.
Rare earth oxides with cubic structure, such as Sm2O3, Y2O3, and Gd2O3 mainly expose (111) surface at high temperatures or under harsh reaction conditions5,17. We found irregular hexagonal sawtooth-shaped structures formed by three 5-coordinated RE atoms and three 4-coordinated O atoms in cubic-phase Y2O3(111), Gd2O3(111) and Sm2O3(111) (Fig. 1 a−c, e−g, i−k). These vacancy structures are slightly different from the surface point defect Ov in CeO2(111), which is surrounded by three 6-coordinated Ce (Supplementary Fig. 1). Three RE-O bonds are broken for each oxygen vacancy, and thus 25% outmost oxygen vacancy are missing on the (111) surface. Similar Ov is also observed in Sm2O3(110) and (100) surfaces (Supplementary Fig. 2). Due to the exposure of unsaturated 5-coordinated RE atoms, these Ov are electrophilic. Such electrophilic sites can adsorb and activate electron-rich molecules, such as NH3 and H2O. The adsorption strength needs to be high enough to form a stable RE-N/O bond and not too high to cause surface poisoning. We have calculated the adsorption energy values of NH3 molecules on a series of RE2O3 and compared with the standard γ-Al2O3(111) surface (Supplementary Figs. 3−5 and Supplementary Table 1). The adsorption of NH3 on the Sm2O3(111) surface (−0.44 eV) was stronger than the Sm2O3(110) surface (−0.36 eV), but weaker than that on the Sm2O3(100) surface (−0.98 eV). The moderate adsorption of NH3 on Sm2O3(111) surface (−0.44 eV) is more favourable to the activation of NH3 molecule than that on γ-Al2O3(111) surface, because the latter exhibits an excessively strong adsorption of NH3 (−1.74 eV). Besides, in order to illustrate the unique role of intrinsic Ov, which is the special spatial structure in catalysts, the adsorption of molecules at different Sm ions on the surface of Sm2O3 is investigated. The adsorption on the vacancy-related Sm sites is stronger than on the non-vacancy Sm sites of Sm2O3(111) surface (Supplementary Fig. 6 and Supplementary Table 1), where adsorption might be too weak for effective catalysis. Thus, intrinsic Ov on the Sm2O3(111) surface showed great superiority in the adsorption of molecules. Ru is a highly efficient active metal to catalyse the ammonia decomposition reaction18, so we have constructed a supported catalyst model of Ru9 clusters on Sm2O3(111) surface (Fig. 1d, h, l). The NH3 adsorption energy on Ru9/Sm2O3(111) is −0.79 eV, indicating that the addition of Ru promotes the adsorption of NH3 without causing strong binding in Ru9/Al2O3(111) (−1.88 eV). Therefore, RE2O3 support metal catalysts can be promising in catalysing reactions involving electron-rich molecules.
a−d Electron density isosurface mapped with electrostatic potential surface of Y2O3(111), Gd2O3(111), Sm2O3(111), and Ru9/Sm2O3(111), respectively, at 0.003 e·bohr−3; e–h top views, and i−l side views of optimized surface structure of Y2O3(111), Gd2O3(111), Sm2O3(111), and Ru9/Sm2O3(111), respectively. Blue circles indicate oxygen vacancy or intrinsic surface Ov structures; blue and orange areas on electrostatic potential surface indicate electrophilic and nucleophilic sites, respectively.
In order to explore the potential applications of the intrinsic surface Ov in RE2O3, a series of Ru/RE2O3 (Ru/Sm2O3, Ru/Y2O3, Ru/Gd2O3) and reference Ru/Al2O3 catalysts (~1 wt% Ru content, determined by ICP-MS, Supplementary Table 2) were prepared by colloidal deposition method. The transmission electron microscopy (TEM) images of the catalysts were shown in Supplementary Fig. 7. Ru/Sm2O3 and Ru/Gd2O3 had nanorods structure, while Ru/Y2O3 and Ru/Al2O3 had nanosheets and nanoparticles morphology, respectively. The high-angle annular dark-field scanning transmission electron microscopy (HAADF-STEM) images showed that the edge of Sm2O3 (Fig. 2d and Supplementary Fig. 8), Gd2O3 (Supplementary Fig. 9) and Y2O3 (Supplementary Fig. 10) supports after the reaction was mostly {111} plane. The high-resolution transmission electron microscopy (HRTEM) images (Supplementary Fig. 11−14) and the HAADF-STEM images (Fig. 2a−e, Supplementary Figs. 15 and 16) showed the clear presence of very small Ru clusters on the support. In the HAADF-STEM images of Ru/Sm2O3 (Fig. 2c) and Ru/Al2O3 (Supplementary Fig. 16c, d), we confirmed that the inter-planar spacing was consistent with the lattice fringe of Ru (100). According to the particle size distribution (Supplementary Figs. 11c, 12c, 13c, 14c, 17) of all catalysts, the size of Ru cluster in all catalysts was mainly 0.5–2 nm, and the average size was 1.2–1.9 nm. The energy dispersive X-ray spectroscopy (EDS) elemental mappings further confirmed the good dispersion of Ru species on the Sm2O3 (Fig. 2e, Supplementary Fig. 15d) and γ-Al2O3 (Supplementary Fig. 16e, f). The X-ray diffraction (XRD) patterns (Supplementary Fig. 18) of both the fresh and used catalysts only showed the diffraction peak of the corresponding support, which was consistent with the TEM results because the Ru species were highly dispersed with very low loadings.
a–d The aberration-corrected HAADF-STEM images of the used Ru/Sm2O3; e EDS elemental mapping results of the used Ru/Sm2O3; f temperature-dependent activities of the catalysts (Ru/Sm2O3, Ru/Y2O3, Ru/Gd2O3 and Ru/Al2O3, GHSV = 30,000 mL·g−1·h−1); g comparison of H2 yield with other Ru-based catalysts: 1: Ru/Sm2O3-p20, 2:Ru/Y2O3-p21, 3: K-Ru/MgO22, 4: K-Ru/CNTs23, 5: Ru/MgO-CNTs24, 6: Ru/c-MgO25, 7: Ru/CaAlOx-w26, 8: Ru-Ni/CeO227, 9: Ru/MgO28, 10: Ru/LaCeOx29, 11: Ru/CNF30, 12: Ru-Cs-Mg/MIL-10131, 13: Ru/Al2O332, 14: Ru-K/ZrO218, 15: Ru-K/CNT18; (h) the long-term stability (GHSV = 36,000 mL·g−1·h−1) and the cycle stability test (GHSV = 30,000 mL·g−1·h−1) of Ru/Sm2O3 catalyst.
The catalytic performances of the Ru-based catalysts (Ru/Sm2O3, Ru/Y2O3, Ru/Gd2O3, Ru/CeO2 and Ru/Al2O3) and corresponding pure oxides materials were evaluated in catalytic ammonia decomposition (Fig. 2f and Supplementary Fig. 19a). This reaction was crucial for online H2 production using liquid NH3 as the media for H2 storage and transportation19. First of all, pure metal oxide materials had similar and poor NH3 conversion, suggesting a non-catalytic process without the presence of Ru. RE2O3 supported Ru showed similar activities that were obviously higher than the Ru/Al2O3, Ru/TiO2, Ru/SiO2 and Ru/CNTs catalyst (Supplementary Fig. 19b). The NH3 conversion of Ru/RE2O3 catalyst at 450 °C was 60–72%, much higher than that of Ru/Al2O3 catalyst (only ~13%, GHSV = 30,000 mL·g−1·h−1). The turnover frequency (TOF) value of Ru/Sm2O3 (3.2 s−1) was 3 times higher than Ru/Al2O3 (1.1 s−1) at 400 °C. The apparent activation energy (Ea, Supplementary Fig. 20) of Ru/Sm2O3 (102.2 kJ·mol−1), Ru/Y2O3 (123.5 kJ·mol−1) and Ru/Gd2O3 (105.1 kJ·mol−1) was lower than that of Ru/Al2O3 (136.2 kJ·mol−1), exhibiting superiority in catalytic conversion of NH3. Interestingly, even though the SBET of Ru/RE2O3 (Supplementary Table 2) was only 1/3–1/2 of Ru/CeO216 (SBET = 106 m2·g−1), Ru/RE2O3 catalysts with intrinsic Ov had very similar activity to Ru/CeO2 with defect-based Ov. It followed that those intrinsic Ov could have similar effects to the defect-based Ov. In addition, the H2 yield with per Ru species of the Ru/RE2O3 catalysts was 1.5–20 times higher than that of other Ru-based catalysts reported18,20,21,22,23,24,25,26,27,28,29,30,31,32 (Supplementary Table 3), proving that the Ru-rare earth oxide catalysts achieved the highest noble-metal atom utilization efficiency for ammonia decomposition reaction at present (Fig. 2g). The durability of Ru/Sm2O3 catalyst was evaluated (Fig. 2h). The NH3 conversion of the Ru/Sm2O3 catalyst decayed by only 4% in 400 h test (GHSV = 36,000 mL·g−1·h−1). In addition, after six cycles of stability tests, the catalyst still maintained the same NH3 conversion at different temperatures (Supplementary Fig. 21), which also verified the excellent stability. We further tested the performance of the Ru/Sm2O3 catalyst at lower temperatures by the online mass spectrometer (Supplementary Fig. 22). It was found that the catalyst had started catalysing the reaction at a low temperature as 200 °C and had obvious catalytic activity at 250 °C, showing the extraordinary catalytic performance.
The chemical state of the catalyst surface was explored by X-ray photoelectron spectroscopy (XPS, Fig. 3a−c, Supplementary Figs. 23−25 and Supplementary Table 4), Near Edge X-ray absorption fine structure (NEXAFS) (Supplementary Fig. 26) and the in situ infrared (IR) spectroscopy in the transmission mode (Fig. 3d and Supplementary Fig. 27). The initial catalysts contained mainly cationic Ru for all catalysts. After catalysis, most of the cationic Ru in the Ru/Sm2O3 were reduced to metallic Ru, as shown in the XPS result of the used catalyst and the quasi in situ XPS experiment (Fig. 3a). In comparison, after the same treated process, the Ru on Al2O3 remained as oxidative state for the used sample (Fig. 3c), only a small amount of Ru was reduced under quasi in situ XPS measurement condition. Such comparison suggested that RE2O3 surface with intrinsic Ov helped the reduction of Ruδ+ to Ru0 in the presence of H2. In the Sm 3d (Fig. 3b) and Y 3d (Supplementary Fig. 25b) spectra, Sm3+ and Y3+ were the only species detected, respectively, confirming their non-reducible nature. The surface properties were studied with the Sm M4,5 edges and O K edge NEXAFS. Both the fresh and used catalysts had exactly the same Sm3+ and O features, suggesting a highly durable surface that maintained the intrinsic Ov (Supplementary Fig. 26). Comparing to the O K edge spectrum of bulk O in Sm2O3 standard33, the surface O in the Ru/Sm2O3 sample has reduced O 1 s → 5d-π transition comparing to the O 1 s → 5d-σ transition. This showed the slightly different coordination nature between surface and bulk O. To further validate the reduction of Ru, in situ IR experiments during CO adsorption were carried out. The main peak position of Ru/RE2O3 was concentrated between 2040 and 2050 cm−1, which was considered as the CO adsorption on Ru0 species34,35. While for Ru/Al2O3 catalysts that were concentrated at 2064 cm−1, which was considered as the CO adsorption on Ruδ+ species (Fig. 3d). The temperature-programmed reduction by H2 (H2-TPR, Supplementary Figs. 28 and 29) results showed that Ru on RE2O3 could be reduced at 123 °C whereas reduction of Ru on Al2O3 required 271 °C. Combining XPS, NEXAFS, in situ IR and H2-TPR, we concluded that Ru over RE2O3 with intrinsic Ov preferred Ru0 whereas Ru over Al2O3 remained at Run+.
a–c XPS results of the fresh and used catalysts: a Ru 3d of Ru/Sm2O3, b Sm 3d of Ru/Sm2O3, c Ru 3d of Ru/Al2O3; d the CO adsorption in situ infrared spectroscopy in the transmission mode of used catalysts at 130 K.
The combination of XPS, NEXAFS, IR, DRIFTS and H2-TPR study suggested that Ru over RE2O3 could be easily reduced to Ru0 compared with that of Ru/Al2O3. To validate the catalytic function of RE2O3 with intrinsic Ov in this process and to the NH3 decomposition reaction in general, we first used CO2-temperature programmed desorption (TPD) to investigate the electrophilic nature of the catalyst (Supplementary Fig. 30). Ru/Sm2O3 surface contained more medium base sites compared to Ru/Al2O3, indicating that Ru/Sm2O3 had more effective surface basicity29. This medium basicity was conducive to the electron transfer from Sm2O3 to Ru species, and further facilitated the dissociative adsorption of N species at Sm3+ site. The results of NH3-TPD (Supplementary Fig. 31) further proved the NH3 desorption was significantly less than that of Ru/Al2O3, which was consistent with the calculation result of adsorption energy values of NH3 molecules on the surface of catalysts (Supplementary Table 1).
To explore the reaction mechanism, we performed first-principles theoretical calculations (Fig. 4) to further study the N–H dissociation. A Ru (0001) slab was adopted to simulate the large-size Ru nanoparticles while Ru9 clusters supported on Sm2O3(111) slab and γ-Al2O3(111) were used to simulate the supported cluster catalyst (Ru/Sm2O3 and Ru/Al2O3). As shown in Fig. 4a, the N-H bond activation barrier on Ru (0001) surface was 1.16 eV, in good agreement with previous work36. Comparatively, the activation barrier of N-H bond on Ru9/Al2O3 and Ru9/Sm2O3 was lowered to 0.76 eV and 0.65 eV, respectively. From Fig. 4b, we could see that on both Ru9/Sm2O3 and Ru9/Al2O3, the N-H dissociation went through a synergistic process that NH3 adsorbed on the metal cations in the support surface via the formation N-Sm or N-Al interaction. Meanwhile, H atom in NH3 was captured by the Ru cluster, thus resulting in the N-H bond breaking. So, for Ru9/Sm2O3, the intrinsic Ov sites were responsible for adsorption of NH3, and the interface Ru atoms played the role of activating N-H bonds. After the first N-H bond break, compared to Ru9/Sm2O3 maintained a relatively active state (−1.85 eV), Ru9/Al2O3 was in a very stable state (−3.31 eV), making it difficult for subsequent reactions to occur. As shown in Fig. 4c, the calculated charge density difference showed that electrons transferred from Sm2O3 slab to Ru9 (Supplementary Fig. 32). This result suggested that Sm2O3 was alkalescence and increased the electron density of Ru to dissociate NH3 more favourably. We further calculated the projected density of states (PDOS) of the Ru9/Sm2O3, isolated NH3, and Ru (0001) surface (Fig. 4d). To activate the N-H bonds, the d-band of Ru and LUMO of NH3 must be at similar energy level, which meant Ru metals with higher d-band centre will interact with NH3 stronger37. The d-band centres of Ru9 on Sm2O3 and Ru (0001) slab were −1.37 eV and −1.95 eV, respectively, which accounts for the higher activity of Ru/Sm2O3. The excellent activity of Ru/Sm2O3 was partially from the sintering-resistant property of Sm2O3 substrate. The calculated bind energies of Ru9 with Al2O3 and Sm2O3 were −9.28 eV and −10.24 eV, respectively. Benefiting from the abundant intrinsic surface Ov sites in Sm2O3 surface, Ru clusters were held firmly, thus avoiding the coalescence process and exhibiting solid durability38,39.
a Energy profiles and b corresponding structures for NH3 adsorption and activation of N–H bond on Ru(0001), Ru9/Sm2O3(111), and Ru9/Al2O3(111), respectively; c charge density differences, ∆ρ = ρ(Ru9/Sm2O3) − ρ(Ru9) − ρ(Sm2O3), for Ru9 adsorption on Sm2O3(111) surface (yellow and blue isosurfaces denote where electronic density increase and decrease, respectively) and integration of ∆ρ in planes parallel to the surface and plotted as a function of the z coordination; (d) the projected density of states (PDOS) of the Ru9 on Ru9/Sm2O3, isolated NH3, and Ru(0001) surface (black dash lines label the d-band centre of Ru).
Based on the present results, the effects of the RE2O3 with intrinsic surface Ov structure for catalysis could be summarized as the following. Firstly, the RE2O3 with intrinsic surface Ov promoted the adsorption and activation of the catalyst on Lewis basic reactant molecules to improve the catalytic performance; secondly, the RE2O3 with intrinsic surface Ov enhanced the interaction between the active metals and the rare earth oxide supports, raising the d band of Ru and increase the energy of surface adsorbed NH2. To explore the generality of this effect, we prepared Cu/RE2O3 and Cu/Al2O3 catalysts and tested the activity of the water-gas shift (WGS) reaction (Supplementary Fig. 33), which was also crucial in industrial hydrogen production. The activity of Cu/RE2O3 has been found to be significantly higher than that of Cu/Al2O3, possibly due to the activation of H2O and O-H cleavage via intrinsic Ov.
In summary, we have discovered a new type of Ov on the surface of RE2O3. Those Ov stem from the surface symmetric and the atomic arranges and therefore intrinsic of a crystalline, which is completely different from the conventional defect-based Ov. Such RE2O3 with intrinsic Ov is found to play an important role in catalysis, such as ammonia decomposition and WGS reaction. The RE2O3 offers significant advantages, including: (1) moderate adsorption of reaction molecules such as NH3 and H2O; (2) maintaining active species in metallic state, (3) forming unique RE-N(O)-H-Ru configurations for the N(O)-H bond breaking. Such Ov -metal synergy is new for those redox inactive metal oxide supports and will bring RE2O3 on the screening system for heterogenous catalysis.
The typical synthetic method of Ru colloidal solution has been reported previously16. Dissolving 0.15 g RuCl3 (Sinopharm) in 50 mL ethylene glycol (C2H6O2; Sinopharm), and then added 0.16 g NaOH (Sinopharm) to the mixture with constant stirring for 30 min. Next, the solution was refluxed at 160 °C for 3 h. After cooling to room temperature, we obtained the Ru colloidal solution with dark brown.
Sm2O3 nanorod (Sm2O3) and CeO2 nanorod (CeO2) follow the same hydrothermal method. Dissolving NaOH (14.40 g; Sinopharm) in 40 mL deionized water, and then the solution of 3 mmol nitrate (Sm(NO3)3·6H2O (aladdin) and Ce(NO3)3·6H2O (kermel)) was added into the previous mixture and kept stirring for 30 min. Then we transferred the solution to the teflon bottle for hydrothermal reaction at 100 °C for 24 h. The precipitate produced was washed and dried overnight at 60 °C to obtain Sm(OH)3 and CeO2. The Sm(OH)3 was calcined in air at 450 °C for 4 h to obtain Sm2O3.
Y2O3 nanosheet was synthesized by hydrothermal method. Dissolving 1.149 g Y(NO3)3·6H2O (aladdin) in 60 mL deionized water. And then using 10% NaOH solution adjusted the pH to 12. Then we transferred the solution to the teflon bottle for hydrothermal reaction at 120 °C for 12 h. The precipitate produced was washed and dried overnight at 60 °C. Finally, it was calcined in the air at 500 °C for 6 h.
Gd2O3 nanorod was synthesized by hydrothermal method. Dissolving 0.02 mol Gd(NO3)3·6H2O (aladdin) in 50 mL deionized water. And then using 2.5 mol·L−1 NaOH solution adjusted the pH to 12.8. Then we transferred the solution to the teflon bottle for hydrothermal reaction at 180 °C for 24 h. The precipitate produced was washed and dried overnight at 80 °C. Finally, it was calcined in the air at 450 °C for 2 h.
The Ru-based catalysts were synthesized according to the previous methods16. The typical steps of colloidal deposition method were shown as followed. 1 g supports (Sm2O3, CeO2, Y2O3, Gd2O3 and γ-Al2O3 (commercial, Macklin)) was dissolved and dispersed in 25 mL deionized water, and then added a certain amount of Ru colloid to this mixture and kept stirring for 48 h. Next, ageing for 12 h and the precipitates were collected and washed by centrifugation. The obtained products were dried for 48 h at 60 °C. Finally, the catalysts were reduced in NH3 atmosphere at 550 °C before catalytic test.
The inductively coupled plasma mass spectrometer (ICP-MS, PerkinElmer, NexION 350X) analysis was used to detect the actual content of Ru. The N2 adsorption−desorption measurements were on a Builder SSA-4200 analyzer at −196 °C. All the samples were pretreated at 200 °C for 400 min under vacuum. The BET (according to the Brunauer, Emmett and Teeler method) specific surface area (SBET) can be calculated from that. The X-ray diffraction (XRD) was carried out on a PANalytical X’pert3 powder diffractometer (40 kV, 40 mA) using Cu Kα radiation (λ = 0.15406 nm). The diffraction angles (2θ) ranged from 10° to 90°. The thermogravimetric analysis (TGA) was carried out on a simultaneous thermal analyzer (METTLER, TGA/DSC3 + ) in N2. The transmission electron microscopy (TEM) images were conducted on a JEOL JEM-2100F microscope operating at 100 kV. The high-resolution TEM (HRTEM) was carried out under 200 kV on a FEI Talos F200s microscope. The high-angle annular dark-field scanning transmission electron microscopy (HAADF-STEM) images were obtained on a JEOL ARM200F microscope equipped with a probe-forming spherical-aberration corrector and Gatan image filter (Quantum 965). The hydrogen temperature-programmed reduction (H2-TPR) measurements were carried on a Builder PCSA-1000 instrument. After pretreated at 500 °C in air, 50 mg catalysts were reduced by 5%H2/Ar (30 mL·min−1) from room temperature to 700 °C. The H2 consumption was shown by the thermal conductivity detector (TCD). The X-ray photoelectron spectroscopy (XPS) measurements were carried out on an Thermo scientific ESCALAB Xi+ XPS spectrometer with Al Kα radiations, and with the C 1 s peak at 284.8 eV as an internal standard for all the spectra. The in situ infrared spectroscopy in the transmission mode measurements were conducted in a UHV apparatus combining a FTIR spectrometer (Bruker Vertex 70 v) with a multi-chamber UHV system. The sample was pretreated with H2 in a vacuum at 873 K for 30 min, and then exposed to CO (10−2 mbar) at 130 K to collect the spectrogram.
Near Edge X-ray absorption fine structure (NEXAFS) experiments were carried out at the VerSoX beamline (B07-C) of Diamond Light Source (DLS, UK). The beamline has a maximum resolving power hν/Δ(hν) > 5000 with a photon flux > 1010 photons s−1 from 170 eV to 2000 eV and can be operated (delivering lower flux) up to 2800 eV. The accuracy of the sample and analyser position is typically less than 10 µm. The gas pressure and composition are controlled via a butterfly valve and mass flow controllers. The endstation consists of a fixed interface flange which holds the entrance cone of the ambient-pressure electron energy analyser (SPECS Phoibos NAP-150). The samples (around 1 mg) were dispersed in water (around 1 mL) and dropped (around 2 droplets) on Au-coated Si (~1 cm × 1 cm), followed by heating at 70 °C to remove the solvent. NEXAFS spectra at Sm M4/M5 edge and O K-edge were measured in both total electron yield (TEY) mode and Auger electron yield (AEY) mode at room temperature. The measurements were performed under UHV condition.
The temperature programmed desorption (TPD) measurements of the catalysts were performed at the online mass spectrometer (TILON LC-D200M). For the typical NH3-TPD experiments, 300 mg catalysts (20–40 mesh) were pretreated at 550 °C for 1 h in NH3 atmosphere (20 mL·min−1). And then it was cooled to room temperature, and held for 1 h in the NH3 atmosphere. Next, switched to Ar and purged until the baseline was stable, collected the signal from room temperature to 800 °C. For the typical CO2-TPD experiments, 300 mg catalysts (20–40 mesh) were pretreated at 550 °C for 1 h in 5%H2/Ar (30 mL·min−1). And then it was cooled to room temperature, and held for 1 h in the CO2 atmosphere. Next, switched to Ar and purged until the baseline was stable, collected the signal from room temperature to 800 °C. For the reaction at lower temperatures, 50 mg catalysts (20–40 mesh) were pretreated at 550 °C for 1 h in NH3 atmosphere (20 mL·min−1). After cooling down, we collected the signal from 100 °C to 300 °C with a step of 50 °C in continuous pure NH3 flow.
All static calculations were carried out using spin-polarized density functional theory (DFT) with generalized gradient approximation (GGA) of Perdew–Burke–Ernzerhof (PBE) and PAW pseudopotentials as implemented in VASP 5.4.4 code40,41. DFT + U method with U = 4 eV was used to describe the localized rare earth 4 f states42. The valence orbitals were described by plane-wave basis sets with a cutoff energy of 400 eV. Considering the large size of p(4×4)-(111) slabs used in this work, the single gamma-point grid sampling was used for Brillouin Zone integration for geometry optimization, and 3 × 3 × 1 k-mesh was used for density of states calculations. Atomic positions were optimized by using the conjugate gradient algorithm until the forces were less than 0.03 eV/Å. Transition states were searched by climbing image nudged-elastic-band (CI-NEB) method with convergence criterion of 0.05 eV/Å43,44. The criterion for electronic self-consistent field convergence was set to 10−6 eV.
The catalytic performance of the catalysts was tested in a self-constructed fixed-bed flow reactor. The temperature controller (UDIAN, XIAMEN YUDIAN AUTOMATION TECHNOLOGY CO., LTD.) was used in the reactor temperature control system. During the test, 50 mg catalysts (20–40 mesh) mixed with 500 mg quartz sand (20–40 mesh) and then packed into the reactor with an inner diameter of 6 mm. Before the test, the catalysts were activated at 550 °C in pure NH3 atmosphere. Then the activity test was performed from 350 to 550 °C with a step of 50 °C for the reactor temperature (GHSV = 30,000 mL·g−1·h−1). The outlet gas was analyzed by an online gas chromato-graph (Ouhua GC 9160), and then the real-time N2 and NH3 contents were obtained. The NH3 conversion was calculated through Eq. (1).
The stability test of the catalysts was conducted at 450 °C (GHSV = 36,000 mL·g−1·h−1) for 400 h. The apparent activation energy for the reaction was determined with an equal conversion of 12.5% by tuning the flow rate and temperature.
The main data supporting the findings of this study are available within the article and its Supplementary Information. Additional data are available from the corresponding authors upon request. Source data are provided with this paper.
TOMPKINS, F. C. Superficial chemistry and solid imperfections. Nature 186, 3–6 (1960).
Xie, C. et al. Defect chemistry in heterogeneous catalysis: recognition, understanding, and utilization. ACS Catal. 10, 11082–11098 (2020).
Wang, G. et al. Oxygen defective metal oxides for energy conversion and storage. Nano Today 13, 23–39 (2017).
Article MathSciNet CAS Google Scholar
Campbell, C. T. & Peden, C. H. F. Oxygen vacancies and catalysis on ceria surfaces. Science 309, 713 (2005).
Article CAS PubMed Google Scholar
Zhang, Y. et al. Unraveling the physical chemistry and materials science of CeO2-based nanostructures. Chem 7, 2022–2059 (2021).
Article MathSciNet CAS Google Scholar
Kang, L. et al. Adsorption and activation of molecular oxygen over atomic copper(I/II) site on ceria. Nat. Commun. 11, 4008 (2020).
Article ADS CAS PubMed PubMed Central Google Scholar
Schaub, R. et al. Oxygen-mediated diffusion of oxygen vacancies on the TiO2(110) surface. Science 299, 377–379 (2003).
Article ADS CAS PubMed Google Scholar
Xu, M. et al. Insights into interfacial synergistic catalysis over Ni@TiO2–x catalyst toward water–gas shift reaction. J. Am. Chem. Soc. 140, 11241–11251 (2018).
Article CAS PubMed Google Scholar
Yu, W.-Z. et al. Construction of active site in a sintered copper–ceria nanorod catalyst. J. Am. Chem. Soc. 141, 17548–17557 (2019).
Article CAS PubMed Google Scholar
Yan, H. et al. Construction of stabilized bulk-nano interfaces for highly promoted inverse CeO2/Cu catalyst. Nat. Commun. 10, 3470 (2019).
Article ADS PubMed PubMed Central Google Scholar
Li, H.-Y. et al. Multiple configurations of the two excess 4f electrons on defective CeO2(111): Origin and implications. Phys. Rev. B 79, 193401 (2009).
Bahmanpour, A. M. et al. Cu−Al spinel as a highly active and stable catalyst for the reverse water gas shift reaction. ACS Catal. 9, 6243–6251 (2019).
Fan, R. et al. Synergistic catalysis of cluster and atomic copper induced by copper-silica interface in transfer-hydrogenation. Nano Res. 14, 4601–4609 (2021).
Article ADS CAS Google Scholar
Li, Y. et al. Selective light absorber-assisted single nickel atom catalysts for ambient sunlight-driven CO2 methanation. Nat. Commun. 10, 2359 (2019).
Article ADS PubMed PubMed Central Google Scholar
Feng, J. et al. Applications of rare earth oxides in catalytic ammonia synthesis and decomposition. Catal. Sci. Technol. 11, 6330–6343 (2021).
Hu, X.-C. et al. Ceria-supported ruthenium clusters transforming from isolated single atoms for hydrogen production via decomposition of ammonia. Appl. Catal. B. Environ. 268, 118424 (2020).
Nolan, M. et al. Density functional theory studies of the structure and electronic structure of pure and defective low index surfaces of ceria. Surf. Sci. 576, 217–229 (2005).
Article ADS CAS Google Scholar
Yin, S. et al. Investigation on the catalysis of COx-free hydrogen generation from ammonia. J. Catal. 224, 384–396 (2004).
Lucentini, I. et al. Review of the decomposition of ammonia to generate hydrogen. Ind. Eng. Chem. Res. 60, 18560–18611 (2021).
Zhang, X. et al. Metal–support interaction-modulated catalytic activity of Ru nanoparticles on Sm2O3 for efficient ammonia decomposition. Catal. Sci. Technol. 11, 2915–2923 (2021).
Feng, J. et al. Highly dispersed ruthenium nanoparticles on Y2O3 as superior catalyst for ammonia decomposition. ChemCatChem 13, 1552–1558 (2021).
Ju, X. et al. Mesoporous Ru/MgO prepared by a deposition-precipitation method as highly active catalyst for producing COx-free hydrogen from ammonia decomposition. Appl. Catal. B. Environ. 211, 167–175 (2017).
Yin, S. F. et al. Carbon nanotubes-supported Ru catalyst for the generation of COx-free hydrogen from ammonia. Catal. Today 93−95, 27–38 (2004).
Yin, S. F. et al. Magnesia–carbon nanotubes (MgO–CNTs) nanocomposite: novel support of Ru catalyst for the generation of COx-free hydrogen from ammonia. Catal. Lett. 96, 113–116 (2004).
Ju, X. et al. Highly efficient Ru/MgO catalyst with surface-enriched basic sites for production of hydrogen from ammonia decomposition. ChemCatChem 11, 4161–4170 (2019).
Zhao, J. et al. Metal–support interactions on Ru/CaAlOx catalysts derived from structural reconstruction of Ca–Al layered double hydroxides for ammonia decomposition. Chem. Commun. 55, 14410 (2019).
Lucentini, I. et al. Catalytic ammonia decomposition over Ni-Ru supported on CeO2 for hydrogen production: Effect of metal loading and kinetic analysis. Appl. Catal. B. Environ. 286, 119896 (2021).
Hu, X. et al. Hydrogen production via catalytic decomposition of NH3 using promoted MgO-supported ruthenium catalysts. Sci. China Chem. 62, 1625–1633 (2019).
Le, T. A. et al. Ru-supported lanthania-ceria composite as an efficient catalyst for COx-free H2 production from ammonia decomposition. Appl. Catal. B. Environ. 285, 119831 (2021).
Duan, X. et al. Carbon nanofiber-supported Ru catalysts for hydrogen evolution by ammonia decomposition. Chin. J. Catal. 31, 979–986 (2010).
Li, J. et al. Sub-nm ruthenium cluster as an efficient and robust catalyst for decomposition and synthesis of ammonia: Break the “size shackles”. Nano Res. 11, 4774–4785 (2018).
Karim, A. M. et al. Correlating particle size and shape of supported Ru/γ-Al2O3Catalysts with NH3 decomposition activity. J. Am. Chem. Soc. 131, 12230–12239 (2009).
Article CAS PubMed Google Scholar
ltman, A. B. et al. Evidence for 5d-σ and 5d-π covalency in lanthanide sesquioxides from oxygen K-edge X-ray absorption spectroscopy. Dalton Trans. 45, 9948–9961 (2016).
Hadjiivanov, K. et al. FTIR study of CO interaction with Ru/TiO2 catalysts. J. Catal. 176, 415–425 (1998).
Chin, S. Y., Williams, C. T. & Amiridis, M. D. FTIR studies of CO adsorption on Al2O3- and SiO2- supported Ru catalysts. J. Phys. Chem. B. 110, 871–882 (2006).
Article CAS PubMed Google Scholar
Honkala, K. et al. Ammonia synthesis from first-principles calculations. Science 307, 555–558 (2005).
Article ADS CAS PubMed Google Scholar
Hammer, B. & Nørskov, J. K. Why gold is the noblest of all the metals. Nature 376, 238–240 (1995).
Article ADS CAS Google Scholar
Curto, A. G. et al. Unidirectional emission of a quantum dot coupled to a nanoantenna. Science 329, 930–933 (2010).
Article ADS CAS PubMed Google Scholar
Hu, S. & Li, W. X. Sabatier principle of metal-support interaction for design of ultrastable metal nanocatalysts. Science 374, 1360–1365 (2021).
Article ADS CAS PubMed Google Scholar
Kresse, G. & Furthmüller, J. Efficiency of ab-initio total energy calculations for metals and semiconductors using a plane-wave basis set. Comput. Mater. Sci. 6, 15–50 (1996).
Kresse, G. & Furthmüller, J. Efficient iterative schemes for ab initio totalenergy calculations using a plane-wave basis set. Phys. Rev. B 54, 11169–11186 (1996).
Article ADS CAS Google Scholar
Capdevila-Cortada, M., Łodziana, Z. & Lόpez, N. Performance of DFT+U approaches in the study of catalytic materials. ACS Catal. 6, 8370–8379 (2016).
Mills, G., Jonsson, H. & Schenter, GK Schenter, Reversible work transition state theory: application to dissociative adsorption of hydrogen. surf. Sci. 324, 305–337 (1995).
Article ADS CAS Google Scholar
Henkelman, G., Uberuaga, B. P. & Jónsson, H. A climbing image nudged elastic band method for finding saddle points and minimum energy paths. J. Chem. Phys. 113, 9901–9904 (2000).
Article ADS CAS Google Scholar
This work was financially supported by National Key Research and Development Program of China (2021YFA1501103), the National Science Fund for Distinguished Young Scholars of China (22225110), the National Science Foundation of China (nos. 22075166, 22271177 and 22033005), the National Science Foundation of Shandong Province of China (ZR2023ZD21 and ZR2023QB187), the Young Scholars Program of Shandong University, EPSRC (EP/P02467X/1 and EP/S018204/2), Royal Society (RG160661, IES\R3\170097, IES\R1\191035, IEC\R3\193038), and by the Guangdong Provincial Key Laboratory of Catalysis (No. 2020B121201002). We thank the Center of Structural Characterizations and Property Measurements at Shandong University for the help on sample characterizations. The calculations were performed by using supercomputers at SUSTech, Tsinghua National Laboratory for Information Science and Technology, and the Computational Chemistry Laboratory of the Department of Chemistry under the Tsinghua Xuetang Talents Program.
These authors contributed equally: Kai Xu, Jin-Cheng Liu.
Key Laboratory for Colloid and Interface Chemistry, Key Laboratory of Special Aggregated Materials, School of Chemistry and Chemical Engineering, Shandong University, Jinan, 250100, China
Kai Xu, Wei-Wei Wang, Lu-Lu Zhou & Chun-Jiang Jia
Department of Chemistry and Engineering Research Center of Advanced Rare-Earth Materials of Ministry of Education, Tsinghua University, Beijing, 100084, China
Jin-Cheng Liu & Jun Li
Center for Rare Earth and Inorganic Functional Materials, School of Materials Science and Engineering & National Institute for Advanced Materials, Nankai University, Tianjin, 300350, China
College of Materials Science and Engineering, Hunan University, Changsha, 410082, China
Department of Chemical Engineering, University College London, Roberts Building, Torrington Place, London, WC1E 7JE, UK
Xuze Guan & Feng Ryan Wang
Fundamental Science Center of Rare Earths, Ganjiang Innovation Academy, Chinese Academy of Sciences, Ganzhou, 341000, China
Beijing National Laboratory for Molecular Sciences, State Key Lab of Rare Earth Materials Chemistry and Applications, PKU-HKU Joint Lab in Rare Earth Materials and Bioinorganic Chemistry, Peking University, Beijing, 100871, China
You can also search for this author in PubMed Google Scholar
You can also search for this author in PubMed Google Scholar
You can also search for this author in PubMed Google Scholar
You can also search for this author in PubMed Google Scholar
You can also search for this author in PubMed Google Scholar
You can also search for this author in PubMed Google Scholar
You can also search for this author in PubMed Google Scholar
You can also search for this author in PubMed Google Scholar
You can also search for this author in PubMed Google Scholar
You can also search for this author in PubMed Google Scholar
C.-J. J., F. R. W., J. L. and C.-H. Y. supervised the work; K. X., L.-L. Z. and C.-J. J. designed the experiments, analyzed the results. K. X., J.-C. L., C.-J. J., F. R. W. and J. L. co-wrote the manuscript; K. X. and W.-W. W. performed the in situ DRIFTS and quasi in situ XPS; J.-C. L. and J. L. performed the DFT calculations and analysed the theoretical data; K. X. performed the catalysts preparation, catalytic tests and the TPR tests; C. M. performed the aberration-corrected HAADF-STEM measurements and analysed the results. X. G. and F. R. W. performed the XAFS experiments and analysed the data.
Correspondence to Feng Ryan Wang, Jun Li or Chun-Jiang Jia.
The authors declare no competing interests.
Nature Communications thanks the anonymous reviewer(s) for their contribution to the peer review of this work. A peer review file is available.
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.
Xu, K., Liu, JC., Wang, WW. et al. Catalytic properties of trivalent rare-earth oxides with intrinsic surface oxygen vacancy. Nat Commun 15, 5751 (2024). https://doi.org/10.1038/s41467-024-49981-9
DOI: https://doi.org/10.1038/s41467-024-49981-9
Anyone you share the following link with will be able to read this content:
Sorry, a shareable link is not currently available for this article.
Provided by the Springer Nature SharedIt content-sharing initiative
Nature Communications (Nat Commun) ISSN 2041-1723 (online)
dyi3 Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.