Supported gold catalysts have attracted significant attention due to their potential applications in many low-temperature reactions of industrial and environmental importance [1], such as CO oxidation [2], the water gas shift (WGS) reaction [3], selective oxidation of organic compounds [4], and VOC removal [5]. In particular, CO oxidation is the most extensively studied owing to its practical applications and its status as a model reaction for probing the oxidation activity and determining reaction mechanisms [6]. It is well known that the catalytic activity of Au catalysts is strongly dependent on the nature of its support [7-9]. Generally, Au nanoparticles (NPs) supported on reducible oxides (including TiO₂ [10] and CeO₂ [11]) exhibit excellent activity for CO oxidation. Strong metal-support interaction between the Au NPs and supports stabilizes the small Au NPs and creates active sites at the interfaces [12]. Moreover, the supports act as electron modifiers providing new active sites or function as active species during the catalytic reaction. These catalysts have been widely studied, but most are not suitable for large-scale applications for various reasons [13].

Although Al₂O₃ is considered to be an inert support for CO oxidation due to its non-reducibility and relatively weak interactions with gold, it is a preferable support from a practical point of view due to its high specific surface area, excellent thermal and mechanical stability, and relative inertness toward steam. By improving the synthesis method, morphology, and surface defects, more "active" Al₂O₃ supports can be produced for Au catalysts [14]. Previously, Al₂O₃ rods with abundant external mesopores were successfully synthesized and the corresponding 3 wt% Au/Al₂O₃ catalyst exhibited excellent catalytic activity for CO oxidation, achieving complete CO conversion at 18 ℃ [15]. Alternatively, adding transition metals or metal oxides can enhance the activity of Au catalysts by adjusting the electronic properties of Au and promoting oxygen activation during the reaction [16-18]. Nickel-based oxides have been developed as active materials for CO oxidation. For example, nanosized nickel ferrite powder (NiFe₂O₄) is highly reactive because of its strong CO adsorption [19]. NiO supported on ceria-alumina mixed oxides exhibited good catalytic performance at subzero temperatures [20]. Experimental studies and theoretical calculations have demonstrated that the introduction of Ni to Au catalysts can promote oxygen adsorption and activation on Au NPs [21, 22]. Moreover, NiO prevents Au NPs from sintering during high temperature pretreatments [23].

Herein, for preparing an efficient novel catalyst based on Au/Al₂O₃ with improved catalytic performance and low Au loading, transition metal Ni was doped into the Al₂O₃ support as an additive for obtaining Au/Ni_xAl catalysts. The catalysts were tested for the CO oxidation reaction and characterized by XRD, TEM, H₂-TPR, O₂-TPD, and DRIFTS. The Ni content and pretreatment atmosphere of the catalysts were investigated to further understand the promotion effects of Ni species.

A series of Ni-doped Al₂O₃ supports were prepared using the precipitation-hydrothermal method, denoted as Ni_xAl, where x represented the molar ratio of Ni and Al. Considering the synthesis of Ni_0.05Al as an example, 0.75 g Al(NO₃)₃·9H₂O and 0.03 g Ni(NO₃)₂·6H₂O were dissolved in 50 mL of deionized water, to which 50 mL of (NH₄)₂CO₃ solution (0.18 mol/L) was added. The obtained mixture with a pH of 9–10 was transferred into a 150 mL Teflon-lined stainless-steel autoclave and maintained at 100 ℃ for 24 h. After cooling to room temperature in air, the precipitants were centrifuged and washed with deionized water and ethanol, and subsequently dried at 80 ℃. These precipitants were further calcined in air at 500 ℃ for 2 h at a heating rate of 1 ℃/min to obtain the Ni_0.05Al support. The pure Al₂O₃ support was prepared following the same steps without adding Ni precursors and is referred to as Al.

The Au/Ni_xAl catalysts were prepared using the typical deposition-precipitation (DP) method with 1 wt% Au loading. The HAuCl₄ solution was added dropwise to an aqueous suspension of Ni_xAl, and the pH of the resulting mixture was adjusted to 8–9 using a (NH₄)₂CO₃ solution and maintained at 60 ℃ for 2 h. After washing and drying, the precursors were calcined under an air or H₂/N₂ atmosphere at 250 ℃ at a heating rate of 5 ℃ /min, to obtain the Au/Ni_xAl and Au/Ni_xAl-R catalysts, respectively.

The catalytic activity was evaluated using a fixed-bed quartz reactor using 50 mg of the catalyst. The total flow rate of the reaction gas was 67 mL/min with a composition of 1% CO, 20% O₂, and 79% N₂. The composition of the effluent gas was analyzed using an online GC-7890 gas chromatograph equipped with a thermal conductivity detector (TCD) and a 5 molecular sieve column. T_100% represents the temperature required for 100% CO conversion.

XRD analysis of the samples was performed using an X'Pert3 powder diffractometer (Cu K_α, λ = 1.54056 ). The diffraction patterns were collected using Cu K_α radiation (40 kV, 40 mA) over a 2θ range of 10°–90°. The Brunauer-Emmett-Teller (BET) surface area was measured using a Micromeritics Tristar 3000 instrument. The samples were degassed at 200 ℃ for 6 h before analysis and the nitrogen adsorption and desorption tests were subsequently performed at a liquid nitrogen temperature of –196 ℃. Transmission electron microscopy (TEM) and high-angel annular dark field (HAADF) images of the catalysts were obtained using a Tecnai F30 electron microscope equipped with a FEG gun operating at 300 kV. The X-ray photoelectron spectroscopy (XPS) measurements were performed using a Thermo Scientific ESCALAB XI+ spectrometer.

H₂-TPR and O₂-TPD tests were performed using a Micromeritics Autochem II 2920 apparatus. For H₂-TPR, the precursors without further calcination were first treated at 100 ℃ under an Ar atmosphere for 60 min. After cooling to 50 ℃, the samples were heated to 300 ℃ at a ramp rate of 10 ℃/min under an 8% H₂/Ar flow. For O₂-TPD, the reduced catalysts were pretreated under an 8% H₂/Ar flow for 2 h at 250 ℃ before cooling to 40 ℃. Next, the samples were exposed to a 5% O₂/He stream for 1 h. After blowing with He for 20 min, the temperature was then ramped to 500 ℃ at a heating rate of 10 ℃/min.

FTIR experiments were conducted using a Bruker Vertex 70v spectrometer equipped with an MCT detector and ZnSe windows. The unpretreated samples were heated in the in-situ reaction cell to 250 ℃ for 30 min under air or H₂/Ar flow. After the background spectra were obtained at 30 ℃ under a He atmosphere, a flow of 5% CO/He was introduced into the cell for 20 min, which was then purged with He and the infrared spectra were recorded.

The CO oxidation performance of the prepared Au/Ni_xAl catalysts is shown in Fig. 1a. After doping with Ni, the catalysts showed significantly higher CO conversion than that of Au/Al under the same reaction conditions, with the bare Ni_0.05Al exhibiting negligible activity. The molar ratio of Ni/Al significantly influenced the reactivity, with increasing Ni/Al from 0.01 to 0.05, T_100% decreased from 100 to 40 ℃. When Ni/Al was further increased to 0.1, T_100% increased to 120 ℃. This indicates that the addition of a suitable amount of nickel can improve the activity of Au/Al for CO oxidation. Because the catalyst pretreatment atmosphere is also known to significantly influence catalytic activity [23, 24], the reactivities of the catalysts pretreated under oxidative and reductive atmospheres were investigated (Fig. 1b). Au/Ni_0.05Al-R exhibited the best catalytic activity of the prepared samples with a T_100% of 20 ℃, lower than that of Au/Ni_0.05Al (T_100% = 40 ℃) and comparable with previously reported results (Table S1), whereas the T_100% of Au/Al-R was lower than that of Au/Al. These results indicated that the reductive catalyst pretreatment atmosphere improved the CO oxidation activity.

The textural properties of the different supports were measured by N₂ adsorption. As shown in Fig. 2a, the bare Al₂O₃ exhibited type IV nitrogen adsorption/desorption isotherms with H2-shaped hysteresis loops, suggesting the presence of typical ink-bottle mesopores. However, Ni_0.05Al showed type IV isotherms with H3-shaped hysteresis loops, indicating the presence of slit-shaped mesopores. The textural parameters of the different samples are summarized in Table S2. The specific surface areas were 404 and 297 m² g^–1, whereas the pore size distributions were centered at 4.5 and 10.1 nm for the Al₂O₃ and Ni_0.05Al, respectively. These results indicated that the co-precipitation influenced the final pore structure of the Ni_0.05Al and the crystalline structure. XRD patterns of the supports were measured to determine the influence of nickel on the crystal structure of Al₂O₃ and the results are shown in Fig. 2b. For bare Al₂O₃, two weak diffraction peaks at 45.8° and 67.3° were observed and were attributed to the γ-Al₂O₃ phase (COD 00-049-0134). After doping with Ni, the intensity of these diffraction peaks strengthened, and a broad peak at 37.2° appeared, which was attributed to the formation of Al₂O₃ with improved crystallinity [25]. Additional crystal nuclei formed during nucleation after doping with nickel. Thus, the crystal growth was facilitated in the hydrothermal reaction, resulting in larger crystalline grains and improved crystallinity [26]. After loading Au, the position of the diffraction peak remained unchanged and no peaks assigned to Ni species or Au NPs were detected, indicating that they were highly dispersed on the support.

TEM measurements were used to define the morphology of the as-synthesized catalysts and the gold particle sizes. Images of representative samples and the particle size distributions are shown in Fig. 3a and 3b as well as Fig. S1. The Al₂O₃ support showed a closely packed sheet-like structure, which changed slightly after the introduction of Ni. For both catalysts, the Au NPs were highly dispersed on the support with average sizes of 3.6 and 2.4 nm for the Au/Al-R and Au/Ni_0.05Al-R samples, respectively. The high magnification images showed that the Au NPs supported on Ni_0.05Al were relatively small. To further distinguish the differences in particle size and distribution between the Au NPs supported on the different supports, the catalyst structure were studied with 3 wt% Au loading and the results are presented in Fig. 3c and 3d as well as Fig. S2. For the 3 wt% Au/Ni_0.05Al-R, the Au NPs were distributed much more uniformly than that of 3 wt% Au/Al and Au NPs larger than 5 nm were not observed. This indicated that the incorporation of Ni stabilized the small gold particles, resulting in a reduction of average size and uniform distribution of Au NPs, which may contribute to the improved activity. For Au/Al and Au/Ni_0.05Al, no obvious aggregated metallic NPs were observed in the TEM and HAADF images (Fig. S3), indicating high dispersions of both the Ni species and Au NPs.

H₂-TPR experiments were performed to elucidate the reduction properties of the precursors. As shown in Fig. 4a, the reduction peaks were centered at 205 and 189 ℃ for Au/Al and Au/Ni_0.05Al, respectively, which was attributed to the reduction of Au³⁺ to Au⁰. The reduction peak of Au³⁺ shifted to lower temperatures when Ni was introduced, indicating that the Ni species improved the reducibility of Au in the precursors. It was previously reported that the incorporation of dopants with valence states below +4 results in higher lattice oxygen mobility which enhances the reduction processes [27]. Furthermore, O₂-TPD was performed to investigate the adsorbed oxygen species which are generally believed crucial for CO oxidation performance. According to Fig. 4b, the desorption peaks at 200–350 ℃ were mainly assigned to dissociatively adsorbed oxygen species (O^–) [28]. In this region, the desorption peak area of the Au/Ni_0.05Al-R catalyst was significantly larger than that of Au/Al-R, indicating that the incorporation of Ni remarkably enhanced the capacity for supplying oxygen species. Moreover, for Au/Ni_0.05Al-R, new desorption peaks appeared at 80 and 188 ℃, arising from surface oxygen species weakly interacting with Au NPs and Ni species, respectively [24, 29]. This phenomenon demonstrated that the Au/Ni_0.05Al-R adsorbed more active oxygen species and exhibited the strongest ability for utilizing oxygen species of all prepared catalysts, which coincided with catalytic performance and H₂-TPR results.

To understand the surface electron properties of the prepared Au catalysts, in situ FTIR analysis was performed using CO as probe molecule and the results are shown in Fig. 5a and 5b. After purging with He for 2 min, the band at approximately 2100 cm^‒1 was detected for both Au/Al-R and Au/Ni_0.05Al-R and was ascribed to the linear adsorption of CO on metallic Au [30]. For Au/Ni_0.05Al-R, a new band at approximately 2050 cm^‒1 was assigned to the adsorption of CO on Ni[31]. Upon purging with He, the bands at approximately 2100 cm^‒1 decreased gradually as a function of purging time. However, the peak intensity of Au/Ni_0.05Al-R decreased more slowly than that of Au/Al-R. The Au-CO band of Au/Ni_0.05Al-R remained after purging for 20 min, while the same band for Au/Al-R disappeared under the same conditions. These results indicated that the electronic structure of the Au species was modified by Ni addition, resulting in stronger CO adsorption. The effect of the pretreatment atmosphere on the surface electron properties was also investigated [32]. Compared to the Au/Ni_0.05Al catalyst, the Au-CO band was red-shifted for Au/Ni_0.05Al-R, indicating the presence of more negatively charged Au species on the catalyst pretreated under a reductive atmosphere [30], leading to better CO oxidation reactivity.

To investigate the chemical state of the Au and Ni species, XPS experiments were performed and the results are shown in Fig. S4. The photoelectronic splitting of gold in Au/Ni_0.05Al-R showed binding energies located at 84.2 and 88.2 eV, which were assigned to the 4f_7/2 and 4f_5/2 orbitals of metallic gold, respectively. The main peaks of Au 4f_7/2 and 4f_5/2 in metallic gold are usually centered at 84.0 and 87.9 eV, respectively, and the slight shift to higher binding energy was attributed to electron transfer from Au to Ni [33]. For the Ni species, the Ni 2p_3/2 spectra showed two peaks centered at 856.1 and 861.8 eV, a characteristic Ni 2p_1/2 orbital band with a binding energy of 873.5 eV, and another sub-band at 879.9 eV. These bands were ascribed to the main peaks of NiO and corresponding satellite peak, respectively [23].

Because of the lack of oxygen mobility in Al₂O₃, the Langmuir-Hinshelwood mechanism is often assumed for CO oxidation over Au/Al₂O₃, for which CO adsorption and O₂ activation are both critical steps [34]. When the Au NPs were loaded onto the surface of Al₂O₃, CO oxidation occurred only on the Au NP surfaces. However, when Ni species were introduced into the framework of Al₂O₃, the synergetic effect between gold and nickel may result in higher catalytic activity. First, CO molecules could be adsorbed both on the surface of the Au NPs and nickel species, and the introduction of Ni distinctly strengthened CO adsorption on the Au NPs. Second, the presence of nickel formed new oxygen vacancies, allowing O₂ molecules to be adsorbed and activated more easily [33]. Third, the interaction between Au NPs and Ni_xAl decreased the size of the gold particles, which produced more active sites.

Herein, nickel-doped alumina nanosheets were synthesized via co-precipitation and used for the deposition of Au NPs. The obtained catalysts showed excellent performance for CO oxidation with 100% conversion of CO at 20 ℃, much lower than that of the bare-Al₂O₃-supported Au catalyst. Combining TEM, H₂-TPR, O₂-TPD, and other characterization techniques, it was demonstrated that the incorporation of an optimized amount of Ni promoted the dispersion of Au NPs, strengthened CO adsorption, and provided reactive oxygen species to facilitate CO oxidation. Moreover, the catalysts pretreated under a reductive atmosphere facilitated CO oxidation. This study provides novel insights for developing other supports for heterogeneous catalysis.