Ni-Doped Amorphous Al₂O₃ for One-Pot Synthesis of Azoxybenzene via Nitrobenzene Reduction with Sodium Borohydride

Abstract

Ni-doped amorphous Al₂O₃ catalysts were successfully prepared for the one-step reduction of nitrobenzene to azoxybenzene using NaBH₄ as hydrogen donors under mild conditions. The amorphous Ni₁Al₈₇O_x catalyst achieved a highly efficient azoxybenzene production rate of 1.89 mol·g-Ni⁻¹·h⁻¹, significantly outperforming its Ni/γ-Al₂O₃ counterpart. On the basis of the MAS NMR and XPS characterization results, the enhanced catalytic performance is associated with Ni incorporation during the preparation of amorphous Ni₁Al₈₇O_x, which introduces abundant unsaturated pentacoordinate Al species with oxygen vacancies and stabilizes Ni^δ+ sites against over-reduction. Notably, Ni₁Al₈₇O_x loaded on commercial ZSM-5 supports maintained an azoxybenzene yield of 9.02 mol·g-Ni⁻¹·h⁻¹, highlighting the strong potential for further scalable applications.

1. Introduction

Azoxybenzene serves as an important building block in the fields of dyestuffs, pharmaceutical agents, and materials science [1,2]. Conventional synthetic routes generally follow a two-step protocol: complete hydrogenation of nitrobenzene to aniline, followed by partial oxidation of aniline to azoxybenzene. In contrast, one-pot strategies enabling the direct transformation of nitrobenzene into azoxybenzene via selective partial reduction have garnered considerable attention owing to their higher efficiency and operational simplicity [3,4,5]. A critical bottleneck remains in the design of catalysts capable of selectively reducing nitrobenzene to the requisite intermediates, while exhibiting cost-effectiveness and retaining catalytic activity over multiple reaction cycles [6,7].

Noble-metal catalysts such as Au, Pt, and Pd exhibit strong activity for nitrobenzene hydrogenation but frequently over-reduce to aniline, bypassing the azo intermediate [8,9,10,11,12]. To address this, strategies such as single-atom catalyst design, dual active-site synergistic catalysis, and efficient hydrogen transfer have been developed to achieve one-step nitrobenzene reduction for azobenzene synthesis [13,14,15]. Despite these advances, noble-metal catalysts still encounter significant barriers for scale-up, including high costs and poor stability.

Ni-based catalysts demonstrate mild and efficient hydrogenation with high selectivity for azoxybenzene under specific conditions [6,7,16]. For instance, Corma et al. [6] used Ni@C/CeO₂ nanocomposites in nitrobenzene catalytic reduction reactions, achieving ~100% nitrobenzene conversion and 94% azoxybenzene selectivity at 120 °C and 1.0 MPa H₂. Hou et al. [17] found that highly dispersed Ni-Ni^δ+ clusters formed on CeO₂ by controlling the reduction degree of NiO/CeO₂ in H₂, with strong metal–support interaction between Ni and CeO₂ inhibiting azoxybenzene hydrogenation to aniline. Shen et al. [18] prepared Ni@SiO_x catalysts to tune selectivity for azoxybenzene by adjusting the Ni⁰/Ni²⁺ ratio through variations in the reduction temperature. Ni⁰ species act as H₂ adsorption/activation sites, while Ni²⁺ species interact with the polar functional groups of the reactants. The low oxophilicity of Ni allows for horizontal adsorption on the Ni surface, which is crucial for high selectivity to fully hydrogenated products [19]. Electron-deficient metallic Ni species, especially Ni²⁺, are essential for nitrobenzene coupling reactions. And the Ni²⁺ ratio correlates positively with nitrobenzene adsorption amounts [20]. Nonetheless, maintaining Ni^δ+ stability under reduction conditions remains a big challenge.

Among various hydrogen donors evaluated for nitrobenzene hydrogenation, sodium borohydride (NaBH₄) demonstrates superior reaction mildness and enhanced hydrogen release efficiency compared to molecular hydrogen [21,22], hydrazine hydrate [23], isopropyl alcohol [20], and ammonia borane [24]. Despite these advantages, the application of Ni-based catalysts with NaBH₄ for the targeted synthesis of azo compounds remains scarcely documented. Teixeira et al. [7] developed Ni-PHI catalysts for nitrobenzene coupling reactions, achieving 92.0% nitrobenzene conversion and 87.0% azoxybenzene selectivity, with NaBH₄ as the hydrogen donor. Nevertheless, the intricate synthetic procedures required for these catalysts present substantial obstacles to their large-scale production. Consequently, the development of cost-effective and structurally simple nickel-based catalysts capable of retaining high catalytic activity under ambient reaction conditions has become a central research priority.

For conventional Ni/γ-Al₂O₃-supported catalysts, the stabilization of nickel species in Ni^δ+ remains challenging. Recent studies show that constructing the amorphous phase effectively enhances heterogeneous catalyst activity for hydrogenation reactions. Specially, Huo et al. [25] revealed that the uniform dispersion of amorphous AlO_x promotes electron transfer from metallic nickel, thereby stabilizing Ni^δ+ species. The introduction of heteroatoms can further suppress the crystalline phase transition, stabilizing the amorphous structure [26,27]. These disordered architectures may promote (i) superior dispersion of active components, (ii) strengthened metal–support interactions, and (iii) increased exposure of vacancy defects and unsaturated coordination sites—identified as catalytic active sites [28,29,30,31]. Owing to these synergistic effects, amorphous nanomaterials exhibit a superior catalytic performance relative to their crystalline counterparts [32,33].

This study aims to create the rational design of cost-effective, stable catalysts for the one-step reduction of nitrobenzene to azoxybenzene using NaBH₄ as the hydrogen donor. Low Ni-doped amorphous AlO_x catalysts were successfully constructed, wherein enhanced Ni-Al interactions, highly defective Ni-AlO_x interfaces, and direct electron transfers from Ni to AlO_x synergistically provide abundant active sites. Upon support loading, strengthened metal–support interactions further promote active Ni dispersion and catalytic efficacy. Catalytic performance in nitrobenzene hydrogenation and associated reaction kinetics was systematically investigated. Complementary characterization techniques, including XRD, SEM, H₂-TPR, XPS, and MAS NMR, were employed to elucidate structure–property relationships and identify critical structural features governing catalytic behavior.

2. Results

2.1. Characterization of Ni-Based Catalysts

The general physicochemical properties of the nickel-based catalysts were characterized. X-ray diffraction (XRD) patterns of Ni/γ-Al₂O₃, Ni₁Al₈₇O_x, and AlO_x catalysts are presented in Figure 1. As shown in Figure 1A, the Ni/γ-Al₂O₃ series samples exhibited diffraction peaks at 2θ of 37.6°, 45.8°, and 66.8°, assignable to the (311), (400), and (440) planes of γ-Al₂O₃ (PDF 29-0063). No diffraction peaks associated with Ni species were observed, even at calcination temperatures up to 600 °C. This is attributed to the low Ni loading and its high dispersion on the γ-Al₂O₃ surface [34]. Figure 1B displays the XRD patterns of Ni₁Al₈₇O_x catalysts calcined at different temperatures. The sample calcined at 400 °C shows a broad diffraction peak in the 2θ range of 15.0–35.0°, which may arise from residual carbonaceous species produced by the incomplete decomposition of citrate precursors. This peak gradually diminished and eventually disappeared at elevated calcination temperatures. No characteristic diffraction peaks corresponding to crystalline Al₂O₃ or NiO were detected for Ni₁Al₈₇O_x samples in the temperature range of 500–700 °C. In contrast, Al₂O₃ prepared via the same synthetic route exhibited distinct diffraction peaks of crystalline Al₂O₃ at 700 °C (Figure 1C). The absence of sharp diffraction peaks in the Ni₁Al₈₇O_x series confirms the preservation of an amorphous structure, implying that the incorporation of Ni modifies the alumina framework and inhibits the formation of crystalline phases [27,35].

Figure 1. XRD patterns of (A) n wt% Ni/γ-Al₂O₃-T, (B) Ni₁Al₈₇O_x-T, and (C) AlO_x-T catalysts.

SEM images and EDS mappings of the catalysts are presented in Figure 2. All samples possessed blocky structures with diverse particle sizes. Notably, Ni₁Al₈₇O_x displayed a more uniform morphology owing to the introduction of citric acid during gel formation, which may suppress the aggregation of large particles [36,37]. Meanwhile, the decomposition of citric acid during calcination also promoted the generation of an abundant porous structure [38]. EDS mappings verified the homogeneous distribution of Al (Figure 2A(c,d)) and Ni (Figure 2B(c,d)) species in both samples. N₂ adsorption–desorption analysis (Figure 3) showed that the specific surface areas of all samples ranged from 150 to 200 m²·g⁻¹. The 0.2 wt% Ni/γ-Al₂O₃ exhibits a typical-type IV N₂ adsorption isotherm with a total pore volume of 0.210 cm³/g, whereas the Ni₁Al₈₇O_x sample features a distinctive H3-type hysteresis loop, accompanied by an increase in pore volume to 0.431 cm³/g. Such differences in the hysteresis loop reflects divergent hysteresis behaviors in the mesoporous pores of the two catalysts [36,39]. The higher pore volume of Ni₁Al₈₇O_x may facilitate the intraparticle diffusion of reactant molecules. Pore size distribution analysis demonstrates that the pore diameters of both catalysts are centered at 2–4 nm, falling within the typical mesoporous range.

Figure 2. SEM and EDS mapping image of (A)-(a–d) 0.2 wt% Ni/γ-Al₂O₃-400 and (B)-(a–d) Ni₁Al₈₇O_x-500.

Figure 3. N₂ adsorption–desorption isotherms and pore size distributions of (A) 0.2 wt% Ni/γ-Al₂O₃-400 and (B) Ni₁Al₈₇O_x-500.

To elucidate the structural characteristics of the samples, ²⁷Al MAS NMR spectra were recorded. As shown in Figure 4, three peaks centered at around 7, 38, and 70 ppm are observed for both Ni₁Al₈₇O_x and Al₂O₃ samples synthesized via a citric acid-assisted sol–gel method. The two characteristic resonances at 7 and 70 ppm correspond to octahedrally coordinated Al_octa and tetrahedrally coordinated Al_tetra, respectively, while the peak at 38 ppm is assigned to coordinatively unsaturated Al_penta in pentahedral coordination [40]. In contrast, 0.2 wt% Ni/γ-Al₂O₃-400 exhibited only two distinct signals at 10 and 70 ppm. Al_penta represents the dominant species (43%) in the as-prepared Al₂O₃. However, its relative content decreases drastically from 43% to 3.4% as the calcination temperature is elevated from 500 to 700 °C. This observation confirms that Al_penta species gradually diminish during the phase transformation of alumina from an amorphous state to a crystalline phase upon calcination temperature. Conversely, for the amorphous Ni₁Al₈₇O_x catalyst, the effect of calcination temperature on the Al_penta content is far less pronounced, with its proportion declining only from 45% to 30%. Thermal stability analysis demonstrates that amorphous Al₂O₃ converts to γ-Al₂O₃ accompanied by the loss of Al_penta species upon heating to 700 °C, which is consistent with the aforementioned XRD results, whereas amorphous Ni₁Al₈₇O_x preserves its amorphous structure at both 600 and 700 °C, underscoring the role of Ni in suppressing the crystalline phase transformation.

Figure 4. ²⁷Al MAS NMR spectra of (A) Ni₁Al₈₇O_x-T, T = (a) 500, (b) 600, and (c) 700; (B) 0.2 wt% Ni/γ-Al₂O₃-400 (a) and AlO_x-T, T = (b) 500, (c) 600, and (d) 700.

It is well established that Al_penta sites possess a coordinatively unsaturated nature and can form strong chemical interactions with metal species through oxygen bridges [41,42,43]. The high abundance of pentahedrally coordinated Al species in amorphous Ni₁Al₈₇O_x originates from the defect-rich character of its amorphous phase. Furthermore, Ni species are more likely to anchor at these coordinatively unsaturated pentacoordinate Al centers via oxygen bridges rather than merely being dispersed on the Al₂O₃ surface.

FT-IR spectroscopy was employed to characterize the surface functional groups of Ni/γ-Al₂O₃ and Ni₁Al₈₇O_x catalysts calcined at different temperatures. As shown in Figure 5A, the broad band observed in the 3000–3600 cm⁻¹ region may come from the contribution of adsorbed water, while the peaks at 1395 cm⁻¹ and 1622 cm⁻¹ are assigned to the stretching vibrations of (Al)-OH and the bending vibrations of H-O-H in adsorbed water, respectively [44,45]. Peaks corresponding to metal–oxygen (M–O) bonds appear below 1000 cm⁻¹, with Ni-O and Al-O vibrations observed at 620 cm⁻¹ and 779 cm⁻¹, respectively [46]. For the amorphous Ni₁Al₈₇O_x catalysts, these M-O peaks shifted slightly to higher wavenumbers and merged into a broad overlapping peak below 1000 cm⁻¹, with a significantly different M-O/(Al)-OH intensity ratio compared to crystalline Al₂O₃. These observations indicate the substantial differences in M-O coordination environments, which may be induced by the strong interactions between Ni species and coordinatively unsaturated Al_penta sites on the Ni₁Al₈₇O_x surface.

Figure 5. (A) Infrared spectra. (B) The H₂-TPR profiles of Ni₁Al₈₇O_x-T catalysts and Ni/γ-Al₂O₃ catalysts.

H₂-TPR experiments investigated the reducibility of Ni species and the nature of metal–support interactions. As shown in Figure 5B, broad reduction peaks in the range of 300~800 °C were observed for all catalysts. The 1.0 wt% Ni/γ-Al₂O₃ catalyst exhibited a reduction peak at 420 °C, which is likely associated with surface-dispersed NiO particles with weak metal–support interactions. In contrast, 0.2 wt% Ni/γ-Al₂O₃ showed a higher reduction temperature, attributed to strongly interacting NiO nanoparticles that arise from low Ni loading and high dispersion on the γ-Al₂O₃ surface. The Ni₁Al₈₇O_x catalyst displayed reduction peaks ranging from 400 °C to 800 °C, with a slight shift toward higher temperatures after calcination. This shift indicates modifications in the coordination environment of Ni and an increase in the strength of M-O bonds [38,47,48,49,50,51]. These findings further confirm the presence of strong interactions between Ni species and coordinatively unsaturated Al_penta sites in the Ni₁Al₈₇O_x catalyst, which impede the reduction of nickel species and enhance the thermal stability of the amorphous Ni₁Al₈₇O_x phase [52,53].

It is well established that the electronic structures of active phases are closely correlated with their coordination environments, which in turn govern catalytic behavior. To explore the electronic states of Ni species in the amorphous Ni₁Al₈₇O_x phase, XPS measurements were performed. As shown in Figure 6, the Ni 2p^3/2 peaks at approximately 852.4 eV and 856.4 eV are assigned to Ni⁰ and Ni²⁺ species, respectively [25,39]. Notably, the majority of surface Ni species existed in oxidized states. For Ni₁Al₈₇O_x-500, the Ni 2p^3/2 peak shifted to a higher binding energy (856.7 eV vs. 856.4 eV) compared to 0.2 wt% Ni/γ-Al₂O₃-400. This is primarily attributed to the electron-donating nature of Ni species. Such electron transfer from Ni to Al further enhances the stabilization of Ni²⁺ species in the catalyst [54].

Figure 6. The XPS spectra of (A) Ni 2p and (B) O 1s, (a) 0.2 wt% Ni/γ-Al₂O₃-400 and (b) Ni₁Al₈₇O_x-500 catalysts.

The O 1s binding energy of Ni₁Al₈₇O_x-500 was a little higher than that of 0.2 wt% Ni/γ-Al₂O₃-400, accompanied by an increased surface oxygen content (63.2% vs. 53.7%). This enhancement is ascribed to the stronger Ni-O-Al electron transfer capability in the presence of coordinatively unsaturated Al sites. The elevated reactivity of surface oxygen atoms facilitated their departure from the lattice upon electron loss, thereby leading to the formation of oxygen vacancies [55].

2.2. Catalytic Performance of Ni-Doped Amorphous Al₂O₃ Catalysts

The catalytic performance of the Ni/γ-Al₂O₃-400 series catalysts and the Ni₁Al₈₇O_x-500 catalyst was evaluated for the catalytic reduction in nitrobenzene (NB) by NaBH₄ under identical reaction conditions. As shown in Figure 7, γ-Al₂O₃ alone exhibited negligible catalytic activity, achieving only 29.5% nitrobenzene conversion with azoxybenzene (AOB) as the dominant product. The introduction of Ni significantly enhanced the catalytic activity while maintaining azoxybenzene selectivity above 80% across the tested Ni content range (0.05–10 wt%). The Ni content directly influenced the catalyst performance. For 0.05 wt% Ni/γ-Al₂O₃-400, it achieved 42.6% nitrobenzene conversion, with 82.3% azoxybenzene selectivity and 17.7% azobenzene (AB) selectivity. When the Ni content increased to 0.5 wt%, aniline (AN), selectivity remained below 1.0%, suggesting that moderate Ni loading favors azoxybenzene formation while suppressing aniline production. The 1 wt% Ni/γ-Al₂O₃-400 catalyst achieved the highest nitrobenzene conversion (80.3%), with 3.5% aniline selectivity. However, further increasing the Ni content to 10 wt% resulted in a slight decrease in conversion, accompanied by a gradual decline in azoxybenzene selectivity and an increase in aniline selectivity to 5.1%, indicating that high Ni content promotes aniline formation. Nitrobenzene conversion increases gradually as Ni loading increases from 0.05 to 1.0 wt%, with optimal activity obtained at 1.0–2.0 wt% Ni. A plausible explanation is that 1.0 wt% Ni supplies adequate active sites and achieves proper metal–support interactions, whereas excessive Ni loading may cover and block accessible active sites, thereby lowering catalytic performance.

Figure 7. Catalytic performance of n wt% Ni/γ-Al₂O₃-400 and Ni₁Al₈₇O_X catalysts for the reduction in nitrobenzene (reaction condition: 0.1 g catalyst, 4 mL ethanol, 1 mmol nitrobenzene, 2 mmol NaBH₄, 60 °C, and 60 min).

Ni₁Al₈₇O_x-400 exhibited a superior catalytic performance compared to 0.2 wt% Ni/γ-Al₂O₃-400 in the nitrobenzene reduction reaction. The 0.2 wt% Ni/γ-Al₂O₃-400 catalyst achieved a nitrobenzene conversion of 53.3% with 88.1% azoxybenzene selectivity, whereas the Ni₁Al₈₇O_x-400 catalyst exhibited a much higher conversion of 82.9%, along with an improved azoxybenzene selectivity of 90.2%. Notably, even when compared to the 1 wt% Ni/γ-Al₂O₃-400, whose Ni content is five times higher, Ni₁Al₈₇O_x-400 still outperformed it. These pronounced differences highlight the critical role of robust Ni–Al interactions in enhancing the catalytic performance of Ni₁Al₈₇O_x, wherein Ni species are effectively stabilized by coordinatively unsaturated Al sites.

Figure 8 further investigates the influence of catalyst calcination temperature on the reaction performance. It was found that Ni₁Al₈₇O_x catalysts maintained a consistent catalytic performance across different calcination temperatures, whereas the activity of 1 wt% Ni/γ-Al₂O₃ catalysts decreased significantly after high-temperature treatment. This discrepancy in catalytic behavior is consistent with the structural characteristics of the catalysts. The density of coordinatively unsaturated Al_penta sites in γ-Al₂O₃ decreases progressively upon calcination temperature. And a similar reduction is observed in the 0.2 wt% Ni/γ-Al₂O₃-400 catalyst. Ni₁Al₈₇O_x catalyst features strengthened the interaction between electron-deficient Ni species and Al_penta sites, thereby promoting both the polarization of the N=O bond and the reductive activation of NaBH₄. Consequently, Ni₁Al₈₇O_x demonstrates superior catalytic activity for the selective hydrogenation of nitrobenzene to azoxybenzene relative to the Ni/γ-Al₂O₃ catalyst series.

Figure 8. Effect of calcination temperature on catalytic reduction performance of nitrobenzene over (A) 1 wt% Ni/γ-Al₂O₃-T and (B) Ni₁Al₈₇O_x-T catalysts (reaction condition: 0.1 g catalyst, 4 mL ethanol, 1 mmol nitrobenzene, 2 mmol NaBH₄, 60 °C, and 60 min).

The stability of Ni-based catalysts is critical to their practical application. The 0.2 wt% Ni/γ-Al₂O₃-400 underwent rapid deactivation, with nitrobenzene conversion declining from 53% to 40% within only two reaction cycles. In contrast, reusability evaluations revealed that nitrobenzene conversion over Ni₁Al₈₇O_x-500 remained above 80% even after five consecutive cycles (Figure 9). Notably, azoxybenzene selectivity remained nearly constant across all recycling tests. These results collectively demonstrate that the Ni₁Al₈₇O_x-500 catalyst exhibits excellent reusability.

Figure 9. Reusability tests for the reduction in nitrobenzene over (A) 0.2 wt% Ni/γ-Al₂O₃-400 and (B) Ni₁Al₈₇O_x-500 catalysts (reaction condition: 0.1 g catalysts, 4 mL ethanol, 1 mmol nitrobenzene, 2 mmol NaBH₄, 60 °C, and 60 min).

Table 1 summarizes the catalytic performance of Ni-based catalysts for the reduction of nitrobenzene to azoxybenzene. For amorphous Ni₁Al₈₇O_x and Ni₁Al₈₇O_x/support catalysts, azoxybenzene selectivity was maintained above 80.0%. The 20 wt% (Ni₁Al₈₇O_x)/ZSM-5 catalyst achieved the highest nitrobenzene conversion of 85.2%, followed by 75.2% for MCM-41-supported, and even the molar ratio of nickel to substrate was up to a ratio of 1:1867. This performance is also comparable to other Ni-based catalysts reported in the literature. As shown in Table 1, Ni₁Al₈₇O_x-500 exhibited a comparable performance to 1 wt% Ni-PHI but consumed only half the NaBH₄ [7]. It also outperformed Ni-CeO₂-based catalysts using hydrogen as the reducing agent [17]. Notably, supported Ni₁Al₈₇O_x catalysts achieved exceptional azoxybenzene formation rates, with 20 wt% (Ni₁Al₈₇O_x)/ZSM-5 reaching 9.02 mol·g-Ni⁻¹·h⁻¹. The high specific surface area of the support facilitates the dispersion of Ni₁Al₈₇O_x species. These results suggest that the synergistic interaction between Ni species and unsaturated coordinated Al in Ni₁Al₈₇O_x contributes to its activity in nitrobenzene conversion.

Table 1. Catalytic performance of Ni-based catalysts for the reduction of nitrobenzene to azoxybenzene.

Sample	nNi:nNB	Reaction Conditions	Conv.%	AOB Sel.%	AOB Formation Rate (mol·g-Ni−1·h−1)
0.2 wt% Ni/γ-Al2O3-400	1:373	2 mmol NaBH4, 4 mL ethanol, 60 °C, 60 min	53.3	88.1	1.17
0.2 wt% Ni1Al87Ox-500	1:373	2 mmol NaBH4, 4 mL ethanol, 60 °C, 60 min	88.0	86.1	1.89
20 wt% (Ni1Al87Ox)/ZSM-5	1:1867	2 mmol NaBH4, 4 mL ethanol, 60 °C, 60 min	85.2	84.7	9.02
20wt% (Ni1Al87Ox)/MCM-41	1:1867	2 mmol NaBH4, 4 mL ethanol, 60 °C, 60 min	75.2	87.0	8.18
20 wt% (Ni1Al87Ox)/C	1:1867	2 mmol NaBH4, 4 mL ethanol, 60 °C, 60 min	56.4	84.0	5.92
7.6 wt% Ni@C/CeO2 [6]	1:85	10 bar H2, 2 mL toluene, 120 °C, 50 min	100	94.1	0.65
1.94 wt% Ni/CeO2 [17]	1:96	2 mmol N2H4·H2O, 2 mL ethanol, Ar, 80 °C, 80 min	99.0	92.0	0.44
1 wt% Ni-PHI [7]	1:747	4 mmol NaBH4, 3 mL ethanol, 50 °C, 120 min	92.0	87.0	2.00

Based on the characterization results, the exceptionally high activities and sintering resistances of the amorphous Ni₁Al₈₇O_x catalysts are attributed to the Ni-doped AlO_x, which hosts more vacancy defects, and unsaturated Al_penta sites anchored the electron-deficient Ni species through their strong interactions. These defects promote the activation of the hydrodeoxygenation reaction, thereby contributing to enhanced nitrobenzene hydrogenation activity [56,57]. Meanwhile, the robust Ni-doped AlO_x inhibited the ordered growth of γ-Al₂O₃ crystallites and hindered the reduction in Ni^δ+ species, which play an important role in the dehydrogenation of NaBH₄. The existence of Ni^δ+ species may allow for a non-horizontal adsorption of NB on the Ni surface, which contributes to the high selectivity to partial hydrogenated products.

3. Materials and Methods

3.1. Catalyst Preparation

3.1.1. Preparation of Ni/γ-Al₂O₃-T Catalysts

Ni/γ-Al₂O₃ catalysts were synthesized via an impregnation method. An aqueous solution of Ni(NO₃)₂·6H₂O (Kemiou Chemical Reagent Co., Ltd., Tianjin, China) was first prepared by dissolving the precursor in 100 mL of distilled water. Al₂O₃ (Shanghai Macklin Biochemical Co., Ltd., Shanghai, China) was then added to the resulting solution, followed by stirring at room temperature for 4 h. The resulting mixture was concentrated using a rotary evaporator to remove excess water. The obtained solid residue was dried at 100 °C for 2 h and subsequently calcined in a muffle furnace at predetermined temperatures (400 °C, 500 °C, or 600 °C) for 3 h in static air. A series of catalysts with varying nickel loadings (0.05, 0.1, 0.2, 0.5, 0.75, 1.0, 2.0, and 10.0 wt%) were thus prepared. The final materials were labeled as n wt% Ni/γ-Al₂O₃-T, where n denotes the nominal nickel loading (in wt%), and T represents the calcination temperature (in °C).

3.1.2. Preparation of Ni₁Al₈₇O_x-T Catalysts

The Ni₁Al₈₇O_x-T catalysts, with 0.2 wt% Ni loading (equivalent to that in 0.2 wt% Ni/Al₂O₃.), were prepared using a citric acid-assisted sol–gel method. Precursor solutions were prepared by co-dissolution of Ni(NO₃)₂·6H₂O and Al(NO₃)₃·9H₂O (Guanghua Sci-Tech Co., Ltd., Shantou, China) in 50 mL of distilled water to achieve a Ni:Al atomic ratio of 1:87. Citric acid (Sinopharm Chemical Reagent Co., Ltd., Shanghai, China) was subsequently introduced at a metal-to-ligand molar ratio of 2:3. The mixture was held at 70 °C under constant stirring for 5 h to promote gelation, followed by aging at 110 °C for 12 h. The dried gel was calcined in a furnace at specific temperatures (400 °C, 500 °C, 600 °C, or 700 °C) for 3 h. The final catalysts were denoted as Ni₁Al₈₇O_x-T, with T indicating the calcination temperature.

3.1.3. Preparation of 20 wt% (Ni₁Al₈₇O_x)/Support Catalysts

A series of 20 wt% (Ni₁Al₈₇O_x)/support catalysts were synthesized following a procedure similar to that used for Ni₁Al₈₇O_x-T catalysts. Specifically, 0.5 g of various commercial catalyst supports was added to the precursor mixture containing nickel nitrate, aluminum nitrate, and citric acid. The loading of Ni₁Al₈₇O_x was maintained at 20 wt% relative to the support mass, yielding a nickel content of 0.04 wt% in the final catalyst. The resulting samples were labeled as 20 wt% (Ni₁Al₈₇O_x)/ZSM-5, 20 wt% (Ni₁Al₈₇O_x)/C, and 20 wt% (Ni₁Al₈₇O_x)/MCM-41 catalysts (ZSM-5 is from Nankai University Catalyst Co., Ltd., Tianjin, China, MCM-41 is from Zhuoran Environmental Technology Co., Ltd., Quzhou, China, and carbon is from Huzhou Jia Tan Company, Huzhou, China).

3.2. Characterization

The crystalline phases of the synthesized catalysts were characterized using a D8 Advance X-ray diffractometer (Bruker Corporation, Billerica, MA, USA) equipped with Cu Kα radiation (40 kV, 40 mA). Powder XRD patterns were collected over a 2θ range from 10° to 80° with a scanning rate of 10°/min.

Nitrogen adsorption–desorption isotherms were measured at −196 °C using an Autosorb-6B volumetric system (Quantachrome, Boynton Beach, FL, USA). The specific surface area and pore structure characteristics of the catalysts were determined using the BET method. Prior to analysis, samples were degassed under vacuum at 300 °C for 1 h. Cumulative pore volume was determined from the amount adsorbed at a relative pressure of 0.99, and pore size distributions were derived from the desorption branch of the isotherms using the Barrett–Joyner–Halenda (BJH) model.

The morphological and microstructural features of the catalysts were examined using a field-emission scanning electron microscope (FE-SEM, JEOL JSM-7500F, JEOL Ltd., Tokyo, Japan) operating at an accelerating voltage of 5–15 kV. Elemental composition and distribution were analyzed using energy-dispersive X-ray spectroscopy (EDX) integrated with the SEM system.

Surface functional groups of the synthesized materials were identified using Fourier-transform infrared (FTIR) spectroscopy. Spectra were recorded in the range of 4000–400 cm⁻¹ using a Bruker Tensor 27 spectrometer (Bruker Tensor, Billerica, MA, USA) with a spectral resolution of 4 cm⁻¹. Prior to measurement, samples were ground, mixed with spectroscopic-grade potassium bromide (KBr), and pressed into pellets under vacuum to ensure optimal transmittance.

The ²⁷Al MAS NMR spectra were obtained on the Bruker 600 MHz AVANCE III spectrometer (Bruker Tensor, Billerica, MA, USA) with a magnetic field strength of 14.09 T. The spin rate of the samples was 20 kHz, the π/12 pulse was 0.2 us, with a recycle delay of 0.5 s, and a total of 256 scans were performed to each sample.

The reduction behavior of the catalysts was evaluated using H₂-TPR experiments conducted on a GC-2060 instrument (Shimadzu Corporation, Yantai, China). Typically, 25 mg of catalyst was loaded into a quartz tube and heated from 50 °C to 900 °C at a rate of 10 °C/min in 5 vol% H₂/N₂ gas mixture.

The surface chemical composition and electronic states of the catalysts were analyzed using X-Ray Photoelectron Spectroscopy (Thermo Fisher ESCALAB 250xi, Thermo Fisher Scientific, USA) with monochromatized Al Kα X-rays (1486.6 eV). Binding energies were calibrated using the C 1s adventitious carbon signal at 284.8 eV.

3.3. Catalyst Performance Evaluation

Catalytic performance evaluations were carried out in a 100 mL three-neck flask equipped with a magnetic stirrer operating at 200 rpm and 60 °C. Reaction mixtures typically contained 40 mL of anhydrous ethanol, 1 mL of nitrobenzene, 1 mL of n-decane (internal standard), NaBH₄, and a predetermined amount of catalyst. After the reaction, the catalyst was separated via centrifugation, washed with anhydrous ethanol, and directly reused for subsequent reaction cycles. The liquid product composition, including nitrobenzene (NB), azoxybenzene (AOB), azobenzene (AB), and aniline (AN), as outlined in reaction scheme (Scheme 1), was analyzed using Agilent 6890 GC (Agilent Technologies, Santa Clara, CA, USA) equipped with a flame ionization detector (FID) and an HP-5 capillary column. Reaction conversion and product selectivity were calculated using the following formulas:

$$\mathrm{Conversion}\left(\%\right)={\displaystyle \frac{\mathrm{moles} \mathrm{of} \mathrm{reactant} \mathrm{converted}}{\mathrm{moles} \mathrm{of} \mathrm{initial} \mathrm{reactant}}} \times 100\%$$

$$\mathrm{Selectivity}\left(\%\right)={\displaystyle \frac{\mathrm{moles} \mathrm{of} \mathrm{product}}{\mathrm{moles} \mathrm{of} \mathrm{reactant} \mathrm{converted}}} \times 100\%$$

Scheme 1. Reaction scheme of nitrobenzene reduction.

The AOB formation rate was calculated using the following formula:

$$\mathrm{AOB} \mathrm{formation} \mathrm{rate} (\mathrm{mol}·{\mathrm{g-Ni}}^{-1}·{\mathrm{h}}^{-1}) = {\displaystyle \frac{{\mathrm{n}}_0 \times \mathrm{Conv}.\% \times \mathrm{Sel}.\%}{{\mathrm{m}}_{\mathrm{Ni}} \times \mathrm{t}}} \times 1/2$$

where n₀ denotes the reactant’s initial moles, conversion denotes the reactant’s conversion, m_Ni denotes the catalyst’s actual input in the process, and t indicates the actual reaction time.

4. Conclusions

Low Ni-doped amorphous Al₂O₃ catalysts were successfully prepared for one-step nitrobenzene reduction to azoxybenzene using NaBH₄ as a hydrogen donor under mild conditions. The Ni₁Al₈₇O_x catalyst achieved a highly efficient azoxybenzene production rate of 1.89 mol·g-Ni⁻¹·h⁻¹, significantly outperforming Ni/γ-Al₂O₃ counterparts. The enhanced catalytic performance is attributed to Ni incorporation, which stabilizes Ni^δ+ against reduction and introduces rich unsaturated pentacoordinate Al species with oxygen vacancies.

Notably, Ni₁Al₈₇O_x loaded on commercial supports (ZSM-5, MCM-41, and activated carbon) maintained excellent catalytic activity for nitrobenzene hydrogenation even at extremely low nickel-to-substrate molar ratios (up to 1:1867). Among these, the 20 wt% (Ni₁Al₈₇O_x)/ZSM-5 catalyst achieved an exceptional azoxybenzene yield of 9.02 mol·g-Ni⁻¹·h⁻¹, demonstrating potential for further industrial applications. This work provides valuable insights into designing efficient and durable non-noble-metal catalysts for catalytic hydrogenation reactions.