Three-Dimensional Network Pd-Ni/γ-Al2O3 Catalysts for Highly Active Catalytic Hydrogenation of Nitrobenzene to Aniline under Mild Conditions.

In view of the current situation of high cost and low catalytic efficiency of the commercial Pd-based catalysts, adding transition metals (Ni, Co, etc.) to form the Pd-M bimetallic catalyst not only reduces the consumption of Pd but also greatly improves the catalytic activity and stability, which has attracted increasing attention. In this work, the three-dimensional network Pd-Ni bimetallic catalysts were prepared successfully by a liquid-phase in situ reduction method with the hydroxylated γ-Al2O3 as the support. Through investigating the effects of the precursor salt amount, reducing agent concentration, stabilizer concentration, and reducing stirring time on the synthesis of the Pd-Ni nanocatalyst, the three-dimensional network Pd-Ni bimetallic nanostructures with four different atomic ratios were prepared under an optimal condition. The obtained wire-like Pd-Ni catalysts have a uniform diameter size of about 5 nm and length up to several microns. After closely combining with the hydroxylated γ-Al2O3, the supported Pd-Ni/γ-Al2O3 catalysts exhibit nearly 100% conversion rate and selectivity for the hydrogenation of nitrobenzene to aniline at low temperature and normal pressure. The stability testing of the supported Pd-Ni/γ-Al2O3 catalysts shows that the conversion rate still remained above 99% after 10 cycles. There is no doubt that the supported catalysts show significant catalytic efficiency and recyclability, which provides important theoretical basis and technical support for the preparation of low-cost, highly efficient catalysts for the hydrogenation of nitrobenzene to aniline.

Introduction

Aniline (AN) is an important chemican class="Chemical">al intermediate, which is widely used in the fields of dyes, pigments, pesticides, medicines, and other chemical industries. At present, AN is mainly obtained by reduction of nitrobenzene (NB) with alkali sulfide reduction, iron powder reduction, catalytic hydrogenation reduction, electrochemical reduction, or CO/H2O reduction.[1−5] The catalytic hydrogenation reduction method has attracted extensive attention due to its advantages such as the low reductive cost, high product quality, and green synthetic process.[6,7] However, high pressure and high temperature are usually required in the current catalytic hydrogenation reduction process. Therefore, it is of great significance to develop a new type of catalyst to react at normal pressure and low temperature for the improvement of industrial efficiency and environmental protection. The noble metal palladium (Pd) catalyst has been widely used in the field of hydrogenation NB because of its excellent catalytic activity, good electrical conductivity, high-temperature oxidation resistance, and corrosion resistance.[8−11] However, its rare resources, high price, low utilization rate, and poor stability are still big obstacles for further application of Pd. Therefore, it is currently urgent to develop high-efficiency catalysts with low cost and superior catalytic performance.

The catalytic activity and stability of the noble an class="Chemical">metal Pd depend on the size, morphology, and dispersion state of the catalysts. In order to stimulate the maximum catalytic potential of the catalyst, the following methods are usually adopted: (1) preparation of catalysts with specific morphology for increasing the specific surface area and exposing more active sites of the catalysts; (2) loading the catalysts on the support to improve its stability, avoid the agglomeration, and extend the service life of the catalysts;[12−15] and (3) adding the transition metal M (Ni, Co, Fe, Cu, Ag, etc.) to form the bimetal or multiple metal catalysts with Pd, thus reducing the consumption of noble metals, and some synergies will be achieved among the different metals that will substantially benefit the efficiency of the noble catalyst.[16−19] At present, most of the catalysts used for hydrogenation of NB to AN are supported catalysts, and the impregnation and precipitation are the commonly used methods. Mark et al.[20] used Al2O3 as a carrier to synthesize supported mono-(Pd or Ni) and bimetallic (Pd/Ni = 1:3, 1:1, and 3:1) catalysts by an equal volume impregnation method. Pd and Ni mainly exist in the particle state of 15–30 nm in the catalysts. Compared with the monometallic catalysts, the bimetallic catalysts delivered higher yield when used in the selective hydrogenation of p-chloronitrobenzene (p-CNB) to p-chloroaniline (p-CAN). Feng et al.[21] prepared a series of Pd/FeO catalysts with an average Pd particle size of 2.2 nm using a coprecipitation method by controlling the washing operation after coprecipitation of the catalyst precursors. The catalysts were tested by the hydrogenation reactions of NB, benzaldehyde, and styrene, and the yield was greater than 99% under certain conditions.

Although various Pd-based catalysts have been prepared and proved successfully in improving the hydrogenation performance, there are still some problems to be solved for accelerating the practical application of the hydrogenation of NB to AN.[22,23] First, the process of preparation was relatively complex, many factors need to be considered, and most would generate a large amount of waste water. Second, the synthesized catalysts mainly existed in the particle state, but the particle size was usually not well controlled. It was easy to agglomerate and deactivate in the reaction process, so the stability of the catalyst could not be guaranteed. Therefore, it is necessary to find new catalysts with controllable size, good stability, and high catalytic performance under the conditions of a simple preparation process and low cost.

In this work, the supported Pd-based bimetallic catalysts with a low productive cost and high hydrogenation catalytic performance were prepared by the in situ liquid-phase reduction method. Using Triton X-114 as the stabilizer and structure control agent, the three-dimensional network structured catalysts with uniform size and good dispersion were obtained. Compared with the particle catalysts prepared by the traditional method, the supported three-dimensional network catalysts exhibit a long and smooth adsorption surface, large force and strong stability with the support, and thus exhibit excellent catalytic activity in NB hydrogenation reaction under mild conditions. Meanwhile, by introducing Ni to replace part of Pd, the consumption of Pd and the production cost were reduced, and the hydrogenation performance of the catalyst was improved by utilizing the synergistic effect among multimetals. The selectivity of the supported Pd–Ni/γ-Al2O3 catalyst to AN was also enhanced. The conversion rate reached 100% when reacting at 25 min, and the turnover frequency (TOF) was 7.2 times higher than that of the commercial Pd/C. The stability testing result showed that the conversion rate of the supported Pd–Ni/γ-Al2O3 catalyst still remained above 99% after 10 cycles. All these excellent characteristics contributed from the three-dimensional network structure formed by the bimetal (Pd and Ni) catalysts. This work provides an effective strategy for the synthesis of supported catalysts with controllable morphology and high catalytic activity.

Results and Discussion

Material Characterization

The in situ liquid-phase reduction method was utilized to obtain the Pd–Ni/γ-an class="Chemical">Al2O3 catalysts under mild conditions. Scheme shows the preparation and hydrogenation process of the supported Pd–Ni/γ-Al2O3 catalysts. First, the amorphous commercial γ-Al2O3 nanoparticles were dispersed in the deionized water. When the solution was stirred and heated to a certain temperature, NH3·H2O was added and stirred for a certain time, and then, the hydroxylated γ-Al2O3 was obtained. Fourier transform infrared (FTIR) spectroscopy was used to demonstrate the effect of the NH3·H2O treatment on the surface of the γ-Al2O3. As shown in Figure S1 (Supporting Information), the broad absorbance band that peaked at 3310 cm–1 represents the stretching mode of surface −OH groups on the particles, indicating that the hydroxyl groups were successfully introduced onto the surface of γ-Al2O3 by NH3·H2O treatment.[24] Second, the modified γ-Al2O3 was added into the metal salt solution with Triton X-114 as the stabilizer and stirred for a certain time, so that metal cations were adsorbed on the surface of the support. The electrostatic adsorption mechanism was used in the preparation of the catalysts to allow the metal ions to be adsorbed on the surface of the support, and then, the catalyst was loaded onto the support in one step by means of in situ reduction. Finally, through reducing the mixed solution with NaBH4, the supported Pd–Ni/γ-Al2O3 catalysts with a three-dimensional network structure could be obtained. Undoubtedly, the linear network structure has the advantages of long segment smoothness, stable structure, being self-supporting, and so forth. Compared with the particles, the Pd–Ni bimetallic catalysts with a linear structure have greater force with the γ-Al2O3 support, thus leading to strong stability, which is favorable for improving the hydrogenation performance and stability of the catalyst. The preparation process is simple, and the conditions are mild. To find out the optimal synthesis conditions, the influencing factors including the precursor salt amount, reducing agent concentration, stabilizer concentration, and reducing stirring time were explored. Transmission electron microscopy (TEM) images of samples obtained under different conditions are shown in Figures S2–S5. Results show that when the precursor salt amount is 0.6 mmol, the mass fraction of the stabilizer Triton X-114 is 0.5%, the concentration of the reducing agent NaBH4 is 2 mg/mL, and the reducing stirring time is 1 min, the obtained Pd–Ni nanocatalysts exhibit a well-controllable linear network structure, uniform size, and good dispersion.

Scheme 1

Preparation and Hydrogenation Process of the Supported Pd–Ni/γ-Al2O3 Catalysts

The morphology, composition, and microstructure features of samples were investigated by TEM analysis. Figure shows TEM, high resolution TEM (HRTEM), selected area electron diffraction (SAED), and energy-dispersive spectrometry (EDS) images of the Pd–Ni catalysts with different atomic ratios. When the molar ratio of Pd(NO3)2 to Ni(NO3)2 is 3:1, 2:1, and 1:1, the obtained samples exhibit a nanowire structure with a uniform diameter of around 4–6 nm (Figure a–c). These nanowires intertwine with each other and form a three-dimensional network. When the molar ratio of Pd(NO3)2 to Ni(NO3)2 decreases to 1:2, the sheet structure instead of nanowires could be found in Figure d.

Figure 1

TEM images of Pd–Ni catalysts with different atomic ratios of Pd/Ni. (a) Pd/Ni = 3:1; (b) Pd/Ni = 2:1; (c) Pd/Ni = 1:1; (d) Pd/Ni = 1:2; (e) HRTEM image, (f) SAED pattern, and (g) EDS spectrum of Pd–Ni catalysts; TEM images of the (h) γ-Al2O3 support and (i) Pd–Ni/γ-Al2O3 catalysts.

HRTEM image (Figure e) shows the clear lattice fringe of the Pd–Ni bimetallic catalysts with an atomic ratio of 1:1, and the spacing distance is about 0.231 nm, which is between Pd(111) and Ni(111) crystal plane spacing, indicating the formation of the Pd–Ni alloy. The SAED diagram (Figure f) taken from the nanowire in Figure c clearly shows the diffraction rings. From the inside to the outside, they correspond to the (111), (200), (220), (311), and (222) crystal planes, indicating that the Pd–Ni bimetal catalysts exist in the polycrystalline state. The energy-dispersive X-ray spectra further prove that the product is composed of Pd and Ni elements, and the atomic ratio is close to 1:1, of which C and Cu mainly come from the copper net and carbon film.

The morphological features of the γ-Al2O3 support and the obtained Pd–Ni/γ-Al2O3 catalysts were investigated by transmission electron microscope analysis, as shown in Figure h,i. The γ-Al2O3 support (Figure h) exhibits sheet-like morphology and almost uniform size of about 20 nm, which could play a good supporting role. Figure i shows the morphology of the Pd–Ni/γ-Al2O3 catalyst, in which the nanowires are intertwined on the γ-Al2O3 support and closely combined with the support to form the supported three-dimensional network catalyst.

The phase composition and crystal structure of the samples were tested with X-ray diffraction (XRD). Figure a,b shows the representative XRD patterns of the an class="Chemical">Pd–Ni nanocatalysts and Pd–Ni/γ-Al2O3 catalysts with different Pd/Ni atomic ratios. The obvious diffraction peaks of the nickel or nickel oxide are not observed in the XRD spectrum in Figure a, indicating that the Ni atoms enter into the crystal structure of Pd, and some Pd atoms are replaced by Ni atoms, thus making the interplanar spacing of Pd become narrowed.[25] The position of the diffraction peak shifts to a high angle with respect to the standard peak position of the Pd (standard card PDF#46-1046). The larger the Ni ratio, the shift of the Pd(111) diffraction peak is more obvious, indicating that the Pd–Ni three-dimensional network nanostructures exist mainly as bimetal solid solutions. Figure b shows that the γ-Al2O3 support has no effect on the phase change of the catalysts. Compared with the standard Pd, it also could be found that the diffraction peaks shift to higher angles at 40.1, 46.6, 68.1, and 82.1°, indicating that the active component part of the supported Pd–Ni/γ-Al2O3 catalysts exist as solid solution. In addition, the presence of Pd and Ni oxide characteristic peaks can be seen from the XRD image. The possible reason is that the Pd and Ni elements mainly exist at the 0-valent state in the center, while existing mainly in the form of oxides on the surface, which might be oxidized to oxide when exposed to oxygen and moisture in air.

Figure 2

XRD patterns of (a) Pd–Ni nanocatalysts and (b) Pd–Ni/γ-Al2O3 with different Pd/Ni atomic ratios; high-resolution XPS spectra of (c) Pd 3d and (d) Ni 2p of the synthesized Pd–Ni bimetallic nanocatalysts; N2 adsorption–desorption isotherms curves of the (e) γ-Al2O3 support and (f) Pd–Ni/γ-Al2O3 catalyst (inset: pore-size distribution curves).

X-ray photoelectron spectroscopy (XPS) was tested to explore the valence state of each element in the three-dimensional network Pd–Ni bimetallic catalyst. The presence of Pd 3d and Ni 2p peaks confirms the coexistence of Pd and Ni in the bimetallic catalysts. The Pd 3d XPS spectrum presents two important forms of Pd species (Figure c), namely, Pd(0) and Pd(II). The binding energy of the Pd 3d5/2 level at 335.8 eV and the Pd 3d3/2 level at 340.9 eV could be ascribed to Pd(0), while the binding energy of the Pd 3d5/2 level at 336.2 eV and the Pd 3d3/2 level at 341.4 eV could be ascribed to Pd(II).[26−28] By comparing the intensities of the peaks of Pd(0) and Pd(II), it is found that Pd(0) is the predominant form in the Pd–Ni nanocatalyst. The Ni 2p XPS spectra (Figure d) show that the peaks at the bond energies of 854.5 and 872.1 eV are stronger, which are assigned to Ni 2p3/2 and Ni 2p1/2, respectively. It is proved that the Ni element mainly exists in the metallic and metal oxide state.[29] Owing to the fact that the addition of the Ni element could change the electronic structure of Pd, the Pd 3d5/2 and 3d3/2 peaks of the bimetallic catalyst are both moving toward the higher binding energy compared with that of the monometallic Pd/Al2O3 catalyst.[30,31] The positive shift is associated with the upshift d-band center of Pd and indicates a change in the electronic structure of Pd, possibly due to the electron energy relaxation of Pd levels caused by the presence of Ni atoms in Pd. As a result, it is assumed that there is a close contact between Pd and Ni and the Pd–Ni bimetal exists mainly as the solid solution. This observation is consistent with the literature report.[32]

The specific surface area and pore size distribution of the samples were investigated by N2 adsorption–desorption isotherm curve analysis. Figure e,f shows the nitrogen isothermal desorption and pore size distribution curves of the γ-Al2O3 support and the Pd–Ni/γ-Al2O3 catalysts, respectively. For the γ-Al2O3 support, the adsorption–desorption curve has a smaller slope at the beginning and a clear platform in the low-pressure area. With the increase in pressure, the curve rises rapidly, and the specific surface area is calculated to be about 148.8 m2/g. The adsorption–desorption curve of the Pd–Ni/γ-Al2O3 catalyst in Figure f has the same trend as the curve of the γ-Al2O3 support, and the specific surface area is calculated to be 214.4 m2/g. There is an obvious peak at 11.5 nm in the pore size distribution curve, which indicates that the supported Pd–Ni/γ-Al2O3 catalyst has a porous structure. The presence of the active component significantly increases the specific surface area of the catalyst, thus leading to more active sites, which is beneficial to improve the hydrogenation performance.

Catalytic Performance

The catalytic activity of the samples for an class="Chemical">NB hydrogenation to AN under atmospheric pressure was investigated, and the optimum conditions for hydrogenation were explored by adjusting the experimental conditions. As analyzed above, the supported Pd–Ni bimetallic nanocatalysts have a larger specific surface area; meanwhile, the addition of Ni metal would change the electronic structure and metal configuration of Pd metal, which is beneficial to increase the active sites exposed by Pd. The proper atomic ratios of Pd–Ni will help give full play to the bimetallic synergistic effect, which is helpful to show a better catalytic effect. Figure a and 3b shows the effect of four different atomic ratios (3:1, 2:1, 1:1, and 1:2) of the supported Pd–Ni/γ-Al2O3 catalysts on the hydrogenation performance of NB. With the increase in the content of Ni metal, the reaction rate shows a trend from slow to fast and then slow, in Figure a. When the molar ratio of palladium nitrate and nickel nitrate is 1:1, the hydrogenation rate of NB is the fastest. The NB can be completely converted at 25 min, and the selectivity of the AN is close to 100%, as shown in Figure b.

Figure 3

Variation of concentration over time under different conditions. (a) NB with different atomic ratios of the Pd/Ni catalysts; (b) AN with different atomic ratios of the Pd/Ni catalysts; (c) NB at different temperatures; (d) AN at different temperatures; (e) NB in different solvents; (f) AN in different solvents.

The reaction temperature has an important influence on the hydrogenation reaction rate of an class="Chemical">NB. To a certain extent, increasing the temperature can effectively enhance the diffusion of the reactant molecule, which could improve the reaction speed.[33] However, a large amount of AN molecules generated instantaneously cannot effectively detach from the surface of the catalyst. Therefore, the catalysts will be fully occupied and no active site is available, thus leading to the decrease in the reaction rate.[34]Figure c,d shows the effect of the different temperatures on the conversion of the NB to AN. It can be seen that the NB conversion rate reaches the fastest when the temperature is 40 °C, and the conversion rate of NB changes from fast to slow with the increase in temperature (Figure c). Also, the change in the conversion rate is consistent with the trend of the concentration, as shown in Figure d. The calculated amount of NB is equal to the amount of AN generated, and the conversion rate of AN is 100%. It shows that the optimum reaction temperature is 40 °C.

The reaction of NB hydrogenation involves three phases of gas, liquid, and solid, so the diffusion rate of the reagent and product can be increased by choosing the suitable solvent, which is helpful for the improvement of the reaction rate.[35,36] There is a certain requirement for the selection of the solvent. On the one hand, the solubility of the solvent to the reactant and the product should be great, so that the reactant molecules can be fully exposed to the catalyst and the AN molecule can diffuse in time. On the other hand, the solvent should not be chemically reacted under the experimental conditions, and the catalyst is insoluble in the reaction solvent and is easily separated. Figure e,f shows the effect of the different solvents on the conversion of the NB and AN. Due to the high yield of AN under neutral conditions, the effect of methanol, ethanol, and n-hexane on the hydrogenation performance of NB was investigated. It can be concluded that when n-hexane is used as the solvent, the conversion of NB can reach 100% for 25 min, indicating that n-hexane as the solvent is more conducive to the diffusion of the reactants and the separation of the AN. The choice of neutral solvents makes the selectivity of the AN reach 100%.

In order to further explore the NB hydrogenation rate of the Pd–Ni/γ-Al2O3 catalyst (precursor salt mole ratio 1:1), the catalytic hydrogenation performance was tested at 40 °C, and the reaction solvent was n-hexane. For comparison, the commercial Pd/C catalyst was also studied. Meanwhile, the monometallic Ni nanoparticles prepared by NaBH4 reduction of the nickel precursor were used for the hydrogenation of NB under the same conditions as the present Pd–Ni/γ-Al2O3 bimetallic catalyst for comparison. However, the Ni nanoparticles were proved to have no activity at all even though five times the amount of Ni was used. Similar situations were also observed for Raney Ni, a typical Ni catalyst. Figure shows the curve of the conversion of NB and AN over time. The conversion rate could reach 100% at 25 min for the Pd–Ni/γ-Al2O3 catalyst, while the Pd/C catalyst needs a longer time of 90 min for complete conversion. The TOF is calculated on the basis of the conversion rates of NB (in mol min–1, determined according to the measured NB conversion) and the amount of catalyst. The TOF of the supported Pd–Ni/γ-Al2O3 catalyst is 940.4 h–1, which is almost 8.3 times higher than that of the commercial Pd/C catalyst (130.6 h–1). Figure b shows that the reduction amount of NB is equal to the amount of AN produced. The selectivity of AN is calculated to be 100%, and there is no change after the AN production rate reaches 100%, indicating that no other byproducts are generated.

Figure 4

Comparison of the catalytic hydrogenation performance. Concentration of (a) NB and (b) AN over time.

In order to investigate the service life of the supported Pd–Ni/γ-Al2O3 catalyst, the stability test was carried out. The Pd–Ni/γ-Al2O3 catalyst (precursor salt mole ratio 1:1) with the highest catalytic activity was selected as the test object. The reaction temperature was 40 °C, and n-hexane was used as the reaction solvent for the hydrogenation of NB. Samples were taken for every continuous reaction of 25 min for testing. As can be seen from Figure a, the catalytic activity still remains above 99% and even the supported Pd–Ni/γ-Al2O3 catalyst is recycled 10 times. In order to observe the morphology change in the catalyst, TEM investigation was carried out for the Pd–Ni/γ-Al2O3 catalyst after 10 cycles (Figure b). Compared with the catalyst before the hydrogenation process shown in Figure i, the size of the supported Pd–Ni catalysts has no significant change, and there is almost no agglomeration appearing after 10 cycles. All these confirm an excellent stability for the synthesized three-dimensional network Pd–Ni/γ-Al2O3 catalysts.

Figure 5

(a) Stability test chart of the Pd–Ni/γ-Al2O3 catalyst and (b) TEM image of Pd–Ni/γ-Al2O3 after 10 cycles.

To better understand the performance of the synthesized three-dimensional network Pd–Ni/Al2O3 catalyst, we compared the synthesis methods, hydrogenation reaction conditions, and catalytic performance with some previously reported Pd-based catalysts.[37−43] As presented in Table , the synthesis method in our work is simple,[44−49] and the Pd–Ni/Al2O3 catalysts exhibit high hydrogenation catalytic performance under relatively mild reaction conditions, which reaches nearly 100% conversion rate for the hydrogenation of NB to AN at a low temperature of 40 °C and normal pressure, demonstrating important applications in catalyst design.[50−55]

Table 1

Comparison of the Synthesis Method and Hydrogenation Catalytic Performance of the Synthesized Pd–Ni/Al2O3 Catalysts with Previous Reports

			reaction conditions
samples		synthesis methods	P H2 (MPa)	T (°C)	catalytic performance conversion (%)	refs
Pd–6Ni–N–C60			1.0	70	98.0	(37)
Pd/LDH(Mg, Al)		ion-exchange method	1.0	50	close to 100	(38)
Pd3Au0.5/SiC		reduction of Pd(NO3)2 and HAuCl4 in SiC suspension in succession.	in ambient H2 flow	25	100	(39)
			under Xe-lamp irradiation (400–800 nm, 0.8 W/cm2)
UiO-66-NH2@COP@(2.34%)Pd		reverse double-solvent approach	0.1	25	99.9	(40)
Pd/MIL-101		colloidal deposition method	0.6	120	100	(41)
Pd@mSiO2	Pd@m14SiO2	facile one-pot method	2.0	110	97.57	(42)
	Pd@m16SiO2				98.97
	Pd@m18SiO2				99.01
hierarchical Pd@Ni		combining the hydrothermal synthesis method with spontaneously galvanic replacement reaction	0.1	25	93	(43)
Pd–Ni/γ-Al2O3		a liquid phase in situ reduction method	in ambient H2 flow	40	nearly 100	this work

Conclusions

In summary, the in situ liquid-phase reduction method was utilized to obtain the Pd–Ni/γ-an class="Chemical">Al2O3 catalysts under mild conditions. The metal ions were adsorbed on the surface of the support by the electrostatic adsorption mechanism and then loaded onto the support in one step by means of in situ reduction. The linear structure for the synthesized Pd–Ni catalysts has unique advantages, such as long segment smoothness, stable structure, being self-supporting, and the great force between the Pd–Ni catalysts with the γ-Al2O3 support. Meanwhile, the addition of Ni element can not only reduce the cost of the catalysts but also improve the catalytic efficiency through the synergistic effect of multimetals. All these features contribute to improve the hydrogenation performance and stability of the catalyst. The conversion of NB to AN is much higher than that of the commercial Pd/C catalyst, and it also exhibits great selectivity and stability. Based on the liquid-phase in situ reduction method, the prepared supported catalysts have a controllable morphology and achieve high catalytic efficiency under low-cost conditions, which provides a potential research direction for the controllable synthesis of efficient catalysts for hydrogenation of NB to AN.

Experimental Section

Synthesis of the Hydroxylated γ-Al2O3 Support

γ-Al2O3 (1.7 g) and ammonia (1.7 mL) were dissolved in 100 mL of distilled water in the three-necked flask and sonicated for 30 min, and then, the solution was stirred at 90 °C for 2 h. By cooling down to room temperature after completion of the reaction, the white powders were filtered out and washed with distilled water three times. At last, the modified γ-Al2O3 support was successfully prepared.

Synthesis of Supported Pd–Ni/γ-Al2O3 Catalysts

Aqueous solution of Triton X-114 with a mass concentration of 0.5% was prepared by ultrasonic dispersion. Metal salts of 0.6 mmol were placed in a beaker with nickel nitrate (Ni(NO3)2) and palladium nitrate (Pd(NO3)2) of different molar ratios (2:1, 1:1, 1:2, and 1:3). Triton X-114 solution (50 mL) was added to each beaker and dispersed evenly with ultrasound. The prepared modified γ-Al2O3 was added to the abovementioned beaker and stirred magnetically for 30 min at room temperature for mixing uniformly with the solution. The fresh NaBH4 aqueous solution was quickly poured into the metal solution, with stirring for 1 min. The mixed solution was then set for 10 min, washed by centrifugation with ethanol and water five to six times, and dried at 70 °C; finally, the supported Pd–Ni bimetallic catalysts could be obtained.

The crystal phase of the catalysts was analyzed with the Philips X-ray diffractometer using Cu Kα radiation in the 2θ range from 35 to 90°. TEM and HRTEM images were taken from the FEI Tecnai F30 transmission electron microscope with a field emission gun operating at 200 kV for examining the size and microstructure of the samples. The Brunauer–Emmett–Teller (BET) surface area and the adsorption/desorption isotherms of the samples after thermal treatment and H2 reduction were determined via N2 adsorption at 77 K, using the Micrometrics ASAP-2020 M automatic specific surface area and porous physical adsorption analyzer. The electronic structure of samples was investigated by a Thermo Scientific ESCALAB 250Xi XPS instrument using Mg Kα as the exciting source (1253.6 eV).

Hydrogenation of NB

The hydrogenation performance of NB was tested in a three-necked flask. First, the mixed solution was prepared with 1:999 volume ratio of NB and the solvent and ultrasonicated to make the dispersion even. Second, 0.0798 g of the Pd–Ni/γ-Al2O3 catalysts (mass percentage of Pd-M is 1%) was added to the mixed solution (15 mL) and transferred to the 100 mL three-necked flask. The gas was replaced by H2 to remove the air in the reaction apparatus four to five times and then heated up to a certain temperature. It was required to record the start and sample on time. Finally, 30 μL of the supernatant was pipetted with a micropipette and diluted with the solvent to 500 μL in a small sample vial for being tested.

The reaction products were qualitatively and quantitatively analyzed using the gas chromatograph of the Aligent 7890A with high purity nitrogen as carrier gas and the hydrogen fire source ionization detector (FID detector). The type of capillary column was SE-30 (30 m × 0.5 mm × 0.32 μm) to analyze the component content in the product. The conversion of NB (XNB) and the selectivity to AN (SAN) were calculated according to eqs and 2, respectively.

In the equations, CNB,0 is the initial NB concentration in the gas feed and CNB and CAN are the concentrations of NB and AN monitored by the online GC during the hydrogenation reaction, respectively.