Robust 2 nm-sized gold nanoclusters on Co₃O₄ for CO oxidation

Abstract

In this study, gold nanoparticles were dispersed on Co₃O₄ nanoplates, forming a specific Au–Co₃O₄ interface. Upon calcination at 300 °C in air, aberration-corrected STEM images evidenced that the gold nanoclusters (NCs) on Co₃O₄{111} were maintained at ca. 2.2 nm, which is similar to the size of the parent Au colloidal particles, demonstrating the stronger metal-support interaction (SMSI) on Co₃O₄{111}. Au/Co₃O₄{111} showed good catalytic activity (a full CO conversion achieved at 80 °C) and durability (over 10 hours) in CO oxidation, which was mainly due to the promotion by the surface oxygen vacancies and intrinsic defects of Co₃O₄{111} for activating O₂ and by Au⁰, Au^δ+, and Au⁺ species on the surface of gold NCs for CO activation, as evidenced by Raman and Fourier-transform infrared (FT-IR) spectroscopy analysis. Au/Co₃O₄ catalyzed CO oxidation obeyed the Langmuir–Hinshelwood mechanism at low temperatures.

---

Introduction

Carbon monoxide (CO), emitted from mobile and stationary combustion sources, is one of the major air pollutants, and its presence even in traceable amounts may cause serious environmental and health problems. Therefore, the elimination of CO, via oxidation by air, acts as one of the important technologies for air cleanup. Gold nanoparticles (NPs) at the size range of 2–5 nm, supported on reducible oxides, have been demonstrated to be extremely active for CO oxidation even at room temperature.^1,2 However, the chemical nature of gold particle sizes and the identification of active sites is still being debated. There are several parameters, including the chemical state of Au, the size of the Au-NPs, and the gold–support interface, that can intensely affect the activity of CO oxidation on gold catalysts.^3–7 To date, the interfacial perimeter of the Au NPs and oxide supports is popularly proposed as the active site for CO oxidation, where the reaction occurs via a redox mechanism. For example, on Au/TiO₂ (ref. 4 and 5) and Au/Fe₂O₃,⁷ CO is proposed to be chemically adsorbed on gold NPs and the O₂ gas is activated on the oxygen vacancy at the interface. Therefore, identification of the atomic structure of the active gold-oxide is of critical importance for the design of highly active gold catalysts. The geometrical structure of the gold-supported interface at the atomic scale is less studied.

Reducible oxides, such as Fe₂O₃, TiO₂, CeO₂, and Co₃O₄, often exhibit exceptionally high activity for CO oxidation, when they are used to support gold nanoparticles.^8–14 This is primarily because these oxides could create oxygen vacancies on their surfaces close to the Au particles, forming strong metal-support interactions (SMSI). In this context, Au/Co₃O₄ catalyst has attracted particular attention because of their exceptionally high activities for low-temperature CO oxidation.¹⁵ The key function of cobalt oxide is to disperse and stabilize gold NPs through its surface oxygen vacancies that strongly depend on the size and shape of cobalt oxide crystallites. The dispersion of gold NPs on high surface area support is expected to promote catalytic activity and stability, as it not only increases the gold loading but also improves its dispersion, which would be beneficial to avoid the sintering of Au NPs.^16,17

Herein, we prepared a nanoplate-shaped Co₃O₄ oxide with the preferential exposure of Co₃O₄{111} as the support to disperse gold nanoclusters (NCs) with a uniform size of ∼2 nm. The atomic arrangements of Au{111} were obtained by aberration-corrected STEM technology, which formed a strong metal-support interaction. The resulting Au/Co₃O₄ catalysts, exhibiting surface oxygen vacancies and intrinsic defects of Co₃O₄{111} and the Au⁰, Au^δ+, and Au⁺ species on the surface of gold NCs, showed significant activities in the CO oxidation at low reaction temperatures. The catalyzed CO oxidation over Au/Co₃O₄ obeyed the Langmuir–Hinshelwood mechanism at low temperatures.

Experimental

Synthesis of Co₃O₄ oxides

Co₃O₄ was synthesized following typical procedures reported in the literature.^18,19 CoCl₂·6H₂O was dissolved in deionized water, and triethylamine was added. After 10 min stirring, the solution was transferred and reacted at 180 °C for 20 h in an autoclave. The precipitates were filtered and washed with deionized water, followed by calcination at 350 °C for 3 h, which was then used to support the gold particles.

Preparation of Au/_Co3O4 catalysts

Au/Co₃O₄ catalysts were prepared by an immobilization method of Co₃O₄ oxides with Au:PVP (polyvinyl alcohol) colloids of ∼2 nm, which were prepared by a typical procedure.⁷ Typically, Co₃O₄ oxides were added into the aqueous solution containing the fresh Au:PVP colloids, which were stirred at room temperature for 3 h. The solid was collected by filtration and washed with distilled water. The solid sample was dried at 80 °C and calcined at 300 °C for 2 h in air to remove all the PVP ligands. Furthermore, the analysis of inductively coupled plasma mass spectroscopy (ICP-MS) showed that the actual loading of Au was 0.65 wt% in the prepared Au/Co₃O₄.

Characterization

X-ray powder diffraction patterns were recorded on a Rigaku D/MAX-2500PC with Cu Kα radiation (λ 1.5418 Å) at 40 kV and 200 mA. Transmission electron microscopy (TEM) images were recorded on a Hitachi 7700 microscope operated at 120 kV. Aberration-corrected scanning transmission electron microscopy (AC-STEM) images were obtained on a JEM-ARM200F at 200 kV. The specimen was prepared by ultrasonically dispersing the sample powder into ethanol, and drops of the suspension were deposited on a carbon-coated copper grid and dried at room temperature. Raman spectra were collected using a Renishaw inVia Raman microscope with a laser wavelength of 785 nm and a laser power of 3 mW and were taken under the same conditions after 60 seconds of exposure. TPO analysis was performed with an AutoChem II 2920 instrument (Micromeritics) and analyzed with a thermal conductor detector (TCD). Before analysis, 50 mg of the samples were pre-treated at 200 °C for 1 h in He gas to clean the surface of the samples. After cooling to room temperature, 5% of O₂/He (30 mL min⁻¹) was used as a feed gas to perform TPO from RT to 600 °C. The infrared Fourier transform spectra were recorded using a Vertex 70v spectrometer (Bruker). Spectra were recorded at 4 cm⁻¹ spectral resolution and 60 kHz scanning velocity. Au/Co₃O₄ samples were pretreated at 300 °C in an Ar atmosphere for one hour. After the system was cooled to room temperature, the flow was switched to a feed gas of 1% CO/Ar, and the infrared spectra were in situ recorded. Visible Raman spectra were collected on a Jobin-Yvon U1000 scanning double monochromator with a 532 nm single-frequency laser.

Catalytic test

CO oxidation was performed in a flow reactor. Preliminary reactions were carried out with ∼0.200 g Au/Co₃O₄ or pristine Co₃O₄ catalysts were taken in the form of 40–60 mesh particles. Before the reactions, the samples were activated at 300 °C for 1 h in the presence of 20% O₂/N₂ gas. When the reactor was cooled to room temperature, a feed gas (1% CO/20% O₂/N₂) was passed over the catalysts at a flow rate of 30 mL min⁻¹. The amounts of CO, CO₂, and O₂ in the inlet and outlet streams were analyzed using an online gas chromatograph. The CO conversion (X_CO, %) was calculated on the basis of CO consumption and CO₂ formation by eqn (1):

X_CO = (C₀ − C₁)/C₀ × 100% (1)

where, C₀ and C₁ represent the CO concentrations at the inlet and outlet of the reactor, respectively.

Results and discussion

Characterization of Au/Co₃O₄

The XRD pattern of the as-prepared Au/Co₃O₄ catalyst is shown in Fig. 1. The diffraction peaks of Co₃O₄ were observed at 19.0°, 31.3°, 36.8°, 44.8°, 55.7°, 59.4°, and 65.2°, which can be assigned to the (111), (220), (311), (222), (400), (422), (511), and (440) planes of the Co₃O₄ phase (JPDS No. 00-042-1467), respectively, with a space group of Fd3m.²⁰ The high peak intensities confirmed the high crystallization of the Co₃O₄ support. Another XRD peak at ∼38.5° corresponds to the Au(111) phase, clearly suggesting the existence of gold NCs in the Au/Co₃O₄ sample.²¹

Fig. 1 Powder XRD of the Au/Co₃O₄ sample.

Furthermore, the morphology of gold NCs and Co₃O₄ oxides and the size of Au particles were characterized by TEM methods. TEM images showed that the Co₃O₄ oxide had a hexagonal shape, with an average edge length of ∼160 nm and thickness of ∼25 nm. As shown in the AC-STEM image, Au NCs of the Au/Co₃O₄ showed a particle size of 2.2 ± 0.6 nm (Fig. 2a). Au particles were uniformly loaded onto the surface of the Co₃O₄{111} plane, based on the lattice fringes with an inter-planar spacing of 0.46 nm, which can be attributed to the Co₃O₄{111} of Co₃O₄ (Fig. 2b).^18,22

Fig. 2 (a) AC-STEM images of Au/Co₃O₄ sample. (b) The Au particle on Co₃O₄{111}. Statistics in (a) were performed by counting 202 particles.

To study the chemical state of the surface gold atoms on the gold nanoparticles, Fourier-transform infrared (FT-IR) spectroscopy for the CO adsorption on the Au/Co₃O₄ catalyst was employed at room temperature. As shown in Fig. 3a, the CO adsorption on Au/Co₃O₄ results in the appearance of three bands in the carbonyl stretching region, with their maxima located at 2106 cm⁻¹, 2120 cm⁻¹, and 2142 cm⁻¹. The adsorption band at 2106 cm⁻¹ is assigned to CO absorbed linearly on metallic gold particles (Au⁰–CO).^23,24 It can be predicted that this band characterizes weakly adsorbed species because it easily disappeared even after a short time of 2 min. The band observed at 2120 cm⁻¹ can be attributed to Au^δ+–CO (0 < δ < 1). The 2142 cm⁻¹ band frequency region can be assigned to CO adsorbed on gold exposed at the surface of the particles where gold sites are to some extent, made electropositive by the adsorbed oxygen (Au⁺–CO).²³ Thus, Au^δ+ and Au⁺ species coexist on the surface of the gold particles, which facilitates the nanogold-catalyzed CO oxidation for the adsorption and activation of CO molecules.

Fig. 3 (a) CO-FTIR of Au/Co₃O₄ catalysts. The adsorption bands at 2105 cm⁻¹, 2120 cm⁻¹, and 2142 cm⁻¹ are assigned to the Au⁰–CO, Au^δ+–CO (0 < δ < 1), and Au⁺-CO species, respectively. (b) Raman spectra of the bare Co₃O₄ oxides and the as-prepared Au/Co₃O₄ catalysts.

Furthermore, Raman analysis was carried out to clearly identify the surface oxygen vacancies of the bare Co₃O₄ oxides and Au/Co₃O₄ catalyst. As presented in Fig. 3b, the bare Co₃O₄ and Au/Co₃O₄ displayed similar Raman features: four vibration peaks (i.e., A_1g + E_g + 2 F_2g) in the range of 400–800 cm⁻¹. The Raman peaks at 484 cm⁻¹, 523 cm⁻¹, and 622 cm⁻¹ are assigned to the E_g, F2g⁽²⁾, and F⁽¹⁾_2g symmetry, respectively. Another peak at 691 cm⁻¹ with A_1g symmetry is attributed to the characteristics of the octahedral CoO₆ sites, corresponding to the unique characteristics of the spinel-type cubic Co₃O₄ phase, in good agreement with the reported literature^25,26 Furthermore, the ratio of the surface oxygen vacancies to surface lattice oxygen was calculated to be 0.46 on the Co₃O₄ support and 0.52 ion Au/Co₃O₄, based on the full width at half-maximum of A_1g peaks.²⁵ Thus, Au/Co₃O₄ exhibits obviously high defective structures, induced by the anchored Au nanoclusters, which can promote the oxidation, as the defective structures can largely promote the adsorption and activation of oxygen species during the catalytic CO oxidation.²⁷

Catalytic performance

Furthermore, Au/Co₃O₄ catalysts were evaluated in the CO oxidation reactions. As shown in Fig. 4a, the bare Co₃O₄ oxide was inactive at low reaction temperatures of 20–60 °C and showed a 31% CO conversion at 80 °C and a full conversion at 120 °C. While, the Au/Co₃O₄ catalyst gave a promising CO conversion (35%) at a low temperature of 20 °C, and the full CO conversion was achieved at 80 °C, Fig. 4a, green line. Therefore, the introduction of Au NCs largely promoted the catalytic activity in CO oxidation. The activation energy for CO oxidation was calculated to be 15.49 kJ mol⁻¹ for Au/Co₃O₄, in good agreement with that reported in the early literature.²⁸ Therefore, we deduce that the Au/Co₃O₄-catalyzed CO oxidation should obey the Langmuir–Hinshelwood mechanism at low temperatures (20–60 °C), as the oxygen molecules and CO should be adsorbed and activated over the surface oxygen vacancies and gold particles, respectively, rather than on the surface lattice oxygen sites.^29,30

Fig. 4 (a) Catalytic activity of Co₃O₄ oxides and Au/Co₃O₄ for CO oxidation as the function of temperature. Reaction conditions: the gas flow was composed of 1% CO/20% O₂/N₂ (vol.) with a flow rate of 30 mL min⁻¹, and 50 mg catalysts were used. (b) Arrhenius plots of the rate vs.(1/T) for CO oxidation reaction over Au/Co₃O₄ catalysts. (c) Durability tests of Au/Co₃O₄ catalysts in the CO oxidation at 25 °C (using 50 mg of catalyst) and 60 °C (40 mg of catalyst). Tests conditions: 1% CO/20% O₂/N₂ with a flow rate of 30 mL min⁻¹. The CO conversion were measured in the range of 25% to 45%.

Furthermore, the durability of the Au/Co₃O₄ catalysts was investigated at 25 and 60 °C. As depicted in Fig. 4c, the activity of the Au/Co₃O₄ catalysts was dropped from 2.92 to 1.87 mol_{reacted CO} g_Au⁻¹ s⁻¹ during the 12 h reaction in the CO oxidation (red line). While, Au/Co₃O₄ gave a stable activity performance of 5.26–5.39 mol_{reacted CO} g_Au⁻¹ s⁻¹ during the test (9 h, black line), which is higher than the reported value over the oxide-supported gold NCs catalysts, as shown in Table S1.†^8–12 Overall, the results indicated that Au/Co₃O₄ catalysts exhibited excellent durability for CO oxidation. No apparent sintering of Au NCs was observed in TEM analysis of the spent Au/Co₃O₄ catalysts (Fig. 5a), demonstrating that the slight activity loss was not caused by the aggregation of gold particles. The CO₂ species were clearly observed in the TPO tests of the spent Au/Co₃O₄ catalysts, compared with the pristine one, as shown in Fig. 5b. Hence, the slight deactivation at low reaction temperature (e.g., 25 °C) should be associated with the formation and accumulation of carbonate-like and H₂O/OH⁻ species at the support interface, resulting in the blockage of active sites during the CO oxidation reaction.^31–34

Fig. 5 (a) TEM image and (b) TPO analysis of the spent Au/Co₃O₄ after CO oxidation. The statistical analysis was performed by counting over 100 gold particles.

Conclusions

In summary, gold nanoclusters of ca. 2.2 nm immobilized on the nanoplate-shaped Co₃O₄ oxides were synthesized by a simple impregnation of gold colloids and oxides. STEM analysis identified that the Au particles of Au{111} and Au{100} exposures interacted with Co₃O₄{111}. Au/Co₃O₄{111} exhibited good catalytic activity and durability in the CO oxidation, which may be due to the strong metal-support interaction. FT-IR and Raman spectroscopy analyses confirmed that the surface oxygen vacancies and intrinsic defects on Co₃O₄{111} and Au^δ+ and Au⁺ species on the surface of gold NCs are responsible for the high catalytic behavior in the CO oxidation.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

We thank the financial support from the fund of National Natural Science Foundation of China (22065029 and 22172167), the Natural Science Foundation of Inner Mongolia Autonomous Region (2022MS02002), and central guidance for Local Scientific and Technological Development funds (2022ZY0081).

Notes and references

1. G. C Bond and D. T. Thompson, Catal. Rev.: Sci. Eng., 1999, 41, 319–388.
2. Z. Li, X. Zhang, Q. Shi, X. Gong, H. Xu and G. Li, Nanoscale Adv., 2021, 3, 7002–7006.
3. Z. Jin, Y. Y. Song, X. P. Fu, Q. S. Song and Ch.-J. Jia, Chin. J. Chem., 2018, 36, 639–643.
4. X. Wei, B. Shao, Y. Zhou, Y. Li, Ch. Jin, J. Liu and W. Shen, Angew. Chem., 2018, 57, 11289–11293.
5. X. Du, Y. Huang, X. Pan, B. Han, Y. Su, Q. Jiang, M. Li, H. Tang, G. Li and B. Qiao, Nat. Commun., 2020, 11, 5811.
6. Y. Wang, D. Widmann, F. Lehnert, D. Gu, F. Schuth and R. Jurgen Behm, Angew. Chem., 2017, 56, 1–7.
7. B. Shao, J. Zhang, J. Huang, B. Qiao, Y. Su, Sh. Miao, Y. Zhou, D. Li, W. Huang and W. Shen, Small Methods, 2018, 2, 1800273.
8. Y. Zhang, Z. Li, J. Zhang, L. Xu, Z. Han, A. Baiker and G. Li, Nano Res., 2023, 16, 8919–8928.
9. M. Chen and D. W. Goodman, Acc. Chem. Res., 2006, 39, 739–746.
10. A. A. Herzing, C. J. Kiely, A. F. Carley, P. Landon and G. J. Hutchings, Science, 2008, 321, 1331–1335.
11. W. Li, C. Liu, H. Abroshan, Q. Ge, X. Yang, H. Xu and G. Li, J. Phys. Chem. C, 2016, 120, 10261–10267.
12. Y. Denkwitz, Z. Zhao, U. Hormann, U. Kaiser, V. Plzak and R. J. Behm, J. Catal., 2007, 251, 437–442.
13. Z. Li, W. Li, H. Abroshan, Q. Ge, Y. Zhou, C. Zhang, G. Li and R. Jin, Nanoscale, 2018, 10, 6558–6565.
14. H. Atout, A. Bouguettouch, D. Chebli, J. Crespo, J. Dupin, J. M. López-de-Luzuriaga, H. Martínez, M. Monge, M. Elena Olmos and M. Rodríguez-Castillo, New J. Chem., 2021, 45, 11727–11736.
15. Z. Zhou, S. Kooi, M. Flytzani-Stephanopoulos and H. Saltsburg, Adv. Funct. Mater., 2008, 18, 2801–2807.
16. Y. Liu, H. Dai, J. Deng, S. Xie, H. Yang, W. Tan, Y. Jiang and G. Guo, J. Catal., 2014, 309, 408–418.
17. Q. Shi, X. Zhang, Z. Li, A. Raza and G. Li, ACS Appl. Mater. Interfaces, 2023, 15, 30161–30169.
18. T. Barakat, J. C. Rooke, E. Genty, R. Cousin, S. Siffert and B. I. Su, Energy Environ. Sci., 2013, 6, 371–391.
19. X. Wei, S. Barkaoui, J. Chen, Q. Fang, G. Cao, Z. Wu, F. Wang and G. Li, Nanoscale Adv., 2021, 3, 1741–1746.
20. U. Sengupta, A. Mukherjee, M. P. S. Mukhopadhyay, K. J. Jenkinson, A. E. H. Wheatley and A. Kar, New J. Chem., 2023, 47, 11425–11443.
21. Y. Guo, Q. Zhou, X. Chen, Y. Fu, S. Lan, M. Zhu and Y. Du, J. Mater. Sci. Technol., 2022, 119, 53–60.
22. H. Chen, M. Yang, S. Tao and G. Chen, Appl. Catal., B, 2017, 209, 648–656.
23. F. Boccuzzi, A. Chiorino and M. Manzoli, Surf. Sci., 2000, 454–456, 942–946.
24. X. Zhang, Z. Li, W. Pei, G. Li, W. Liu, P. Du, Z. Wang, Z. Qin, H. Qi, X. Liu, S. Zhou, J. Zhao, B. Yang and W. Shen, ACS Catal., 2022, 12, 3634–3643.
25. G. Zhai, J. Wang, Z. Chen, W. An and Y. Men, Chem. Eng. J., 2018, 337, 488–498.
26. S. Barkaoui, M. Haddaoui, H. Dhaouadi, N. Raouafi and F. Touati, J. Solid State Chem., 2015, 228, 226–231.
27. Y. Cao, L. Zhao, T. Gutmann, Y. Xu, L. Dong, G. Buntkowsky and F. Gao, J. Phys. Chem. C, 2018, 122, 20402–20409.
28. M. Haruta, S. Tsubota, T. Kobayashi, H. Kageyama, M. J. Genet and B. Delmon, J. Catal., 1993, 144, 175–192.
29. Y. Chen, Y. Li, W. Chen, W. Xu, Z. Han, A. Waheed, Z. Ye, G. Li and A. Baiker, Nano Res., 2022, 15, 1366–1374.
30. Y. Wang, Q. Jiang, L. Xu, Z. Han, S. Guo, G. Li and A. Baiker, ACS Appl. Mater. Interfaces, 2021, 13, 61078–61087.
31. H. Wang, H. Zhu, Z. Qin, F. Liang, G. Wang and J. Wang, J. Catal., 2009, 26, 154–162.
32. C. H. Kim and L. T. Thompson, J. Catal., 2005, 230, 66–74.
33. Q. Guo, S. Chen, Y. Liu and Y. Wang, Chem. Eng. J., 2010, 165, 846–850.
34. Q. Shi, Y. Zhang, Z. Li, Z. Han, L. Xu, A. Baiker and G. Li, Nano Res., 2023, 16, 6951–6959.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: https://doi.org/10.1039/d3na00561e

---