Highly Dispersed and Stable Ni/SBA-15 Catalyst for Reverse Water-Gas Shift Reaction

: A 1%Ni/SBA-15(P) catalyst was synthesized with a P123-assisted impregnation method, which exhibited high CO 2 conversion and stability in the reverse water-gas shift reaction. For the 1%Ni/SBA-15(P) catalyst, TEM and TPR characterizations demonstrated that the highly dispersed NiO particles at about 3 nm strongly interacted with the SiO 2 support. During reverse water-gas shift reaction, the 1%Ni/SBA-15(P) catalyst exhibited higher CO 2 conversion than the 1%Ni/SBA-15 catalyst prepared by the conventional impregnation method without P123. The CO 2 conversion of the 1%Ni/SBA-15(P) catalyst at 700 ◦ C was 33.7%, which was three times that of the 1%Ni/SBA-15 catalyst. Moreover, the former catalyst was stable at 700 ◦ C within 1000 min. The good activity and stability of the 1%Ni/SBA-15(P) catalyst was owing to small Ni particles that strongly interacted with SBA-15.


Introduction
CO 2 may be utilized through the dry reforming of methane [1][2][3] and reverse watergas shift reaction (RWGS) [4][5][6], which are promising models for CO production. CO can be used as a co-reactant for petrochemical manufacturing (syn-gas for Fischer-Tropsch synthesis, hydroformylation, carbonylation) [7,8]. The RWGS can produce CO from the greenhouse gas carbon dioxide and green hydrogen produced by renewable energy. Nickelbased catalysts show good activity in RWGS [9][10][11][12]. However, the side reactions of CO 2 methanation are prone to occur on nickel catalysts [10,12,13]. Highly dispersed small Ni particles are favorable for RWGS, while large Ni particles easily generate methane [14]. The endothermic RWGS reaction has a higher equilibrium conversion at high temperature. In addition, because methanation reaction is exothermic, high temperature is beneficial to inhibit methane formation and improve CO selectivity. However, as highly dispersed Ni is prone to sintering deactivation in the high temperature reaction, it is desired to prepare stable Ni-based RWGS catalysts with high Ni dispersion.
In this work, the 1%Ni/SBA-15(P) catalyst was synthesized with the P123-assisted impregnation method. Compared with the conventional 1%Ni/SBA-15 catalyst, the

Catalyst Preparation
The SBA-15 was synthesized according to the literature [27]. The 1%Ni/SBA-15(P) catalyst was synthesized by the P123-assisted impregnation method [28,29]. Briefly, 0.050 g nickel nitrate hexahydrate and 0.010 g P123 (n(P123):n(Ni) = 1:100) were added to 20 mL deionized water. After stirring the solution for 1 h, 1 g of SBA was added and then stirred for 12 h. The solution was heated at 60 • C for 3 h and 110 • C for 2 h. Finally, the 1%Ni/SBA-15(P) catalyst was prepared by the calcination the sample at 600 • C for 4 h.

Catalyst Characterization
Nitrogen adsorption-desorption analysis was performed on a Quantachrome Autosorb-iQ analyzer at −196 • C (Quantachrome Corporation, Boynton Beach, FL, USA). The sample was degassed at 300 • C for 10 h before analysis. The pore size distribution was obtained using the BJH model. The morphology of the samples was determined by a Tecnai G2 F20 (FEI company, Hillsboro, OR, USA) transmission electron microscope (TEM). The powder X-ray diffraction (XRD) patterns were collected on a DX-2700 X-ray diffractometer (Haoyuan Instrument Co., Ltd., Dandong, China) with Cu Kα radiation at 40 kV and 30 mA. H 2 temperature programmed reduction (H 2 -TPR) measurements were performed on TP-5080 equipment from room temperature to 900 • C (10 • C /min) in 5% H 2 /Ar (30 mL/min).

Catalytic Test
Hydrogenation of CO 2 was conducted under atmospheric pressure in a fixed bed reactor. Prior to each test, 0.005 g of catalyst and 0.010 g quartz sand were packed in the quartz reactor, and the catalyst was reduced to 650 • C for 2 h under 70% H 2 /N 2 (70 mL/min) flow. When the temperature dropped to 500 • C, the mixed gases (CO 2 /H 2 = 1:1, 100 mL/min) were fed into the reactor for the reaction. The gases product was analyzed by a gas chromatography (GC-7900, Techcomp, Kowloon, Hong Kong). Only CO, CH 4 , and H 2 O were produced during the catalytic reaction, and no C2+ hydrocarbon was detected by the GC TCD detector, which was also confirmed by a GC FID detector. Therefore, the conversion and selectivity were estimated as follows: CO selectivity = F out, CO F out, CO +F out, CH 4 × 100% where F in and F out stand for the input and output flow rates.

Catalyst Characterization
All the N 2 adsorption-desorption isotherms in Figure 1a were type IV isotherms with a hysteresis loop, which confirmed the uniform mesoporous structure of the samples. As shown in Figure 1b, the pore sizes of all the samples were similar, indicating that the mesoporous structure did not change after adding Ni to SBA-15. Table 1 shows that the BET specific surface area decreased with the addition of Ni into SBA-15, while the pore volume was almost unchanged.   Transmission electron microscope (TEM) images in Figure 2 show that both catalysts had the ordered mesoporous structures. As shown in Figure 2a  The small-angle XRD patterns of the fresh and reduced catalysts in Figure 3a showed three typical peaks, which indexed to a hexagonal lattice of mesoporous SBA-15 [30]. The  Transmission electron microscope (TEM) images in Figure 2 show that both catalysts had the ordered mesoporous structures. As shown in Figure 2a   Transmission electron microscope (TEM) images in Figure 2 show that both catalysts had the ordered mesoporous structures. As shown in Figure 2a  The small-angle XRD patterns of the fresh and reduced catalysts in Figure 3a showed three typical peaks, which indexed to a hexagonal lattice of mesoporous SBA-15 [30]. The The small-angle XRD patterns of the fresh and reduced catalysts in Figure 3a showed three typical peaks, which indexed to a hexagonal lattice of mesoporous SBA-15 [30]. The results indicate that all the catalysts possessed the ordered mesoporous structure. After reduction, angles of all reflection planes were moved toward higher angle, implying that d-spacing of SBA-15 support must have been contracted. peaks were observed in the fresh and reduced 1%Ni/SBA-15 samples, respectively. On the contrary, there was no peak assigned to NiO in the 1%Ni/SBA-15(P), indicating that the NiO particles were highly dispersed in the 1%Ni/SBA-15(P) [28]. No peak assigned to Ni in the reduced 1%Ni/SBA-15(P) indicates that no large Ni particle was formed, even after reduction at 650 °C.  Figure 4 shows the H2-TPR profiles of the two catalysts. The 1%Ni/SBA-15 had a sharp peak mainly located at 420 °C, assigned to the reduction peak of large particles of NiO [31,32]. The broad reduction peak of the 1%Ni/SBA-15(P) catalyst was between 300-650 °C, and the reduction peak of more than 500 °C was attributed to the reduction of highly dispersed NiO, which had a strong interaction with the SiO2 [31,[33][34][35][36]. The TPR results show that Ni was strongly interacted with the support of SiO2 in the 1%Ni/SBA-15(P) [21,37].   As shown in Figure 3b, the broad peak at about 22 • was assigned to the SiO 2 in SBA-15 [24]. The reduced catalysts were treated with 70%H 2 /N 2 at 650 • C for 2 h. NiO and Ni peaks were observed in the fresh and reduced 1%Ni/SBA-15 samples, respectively. On the contrary, there was no peak assigned to NiO in the 1%Ni/SBA-15(P), indicating that the NiO particles were highly dispersed in the 1%Ni/SBA-15(P) [28]. No peak assigned to Ni in the reduced 1%Ni/SBA-15(P) indicates that no large Ni particle was formed, even after reduction at 650 • C. Figure 4 shows the H 2 -TPR profiles of the two catalysts. The 1%Ni/SBA-15 had a sharp peak mainly located at 420 • C, assigned to the reduction peak of large particles of NiO [31,32]. The broad reduction peak of the 1%Ni/SBA-15(P) catalyst was between 300-650 • C, and the reduction peak of more than 500 • C was attributed to the reduction of highly dispersed NiO, which had a strong interaction with the SiO 2 [31,[33][34][35][36]. The TPR results show that Ni was strongly interacted with the support of SiO 2 in the 1%Ni/SBA-15(P) [21,37]. results indicate that all the catalysts possessed the ordered mesoporous structure. After reduction, angles of all reflection planes were moved toward higher angle, implying that d-spacing of SBA-15 support must have been contracted.

Catalytic Performance
As shown in Figure 3b, the broad peak at about 22° was assigned to the SiO2 in SBA-15 [24]. The reduced catalysts were treated with 70%H2/N2 at 650 °C for 2 h. NiO and Ni peaks were observed in the fresh and reduced 1%Ni/SBA-15 samples, respectively. On the contrary, there was no peak assigned to NiO in the 1%Ni/SBA-15(P), indicating that the NiO particles were highly dispersed in the 1%Ni/SBA-15(P) [28]. No peak assigned to Ni in the reduced 1%Ni/SBA-15(P) indicates that no large Ni particle was formed, even after reduction at 650 °C.  Figure 4 shows the H2-TPR profiles of the two catalysts. The 1%Ni/SBA-15 had a sharp peak mainly located at 420 °C, assigned to the reduction peak of large particles of NiO [31,32]. The broad reduction peak of the 1%Ni/SBA-15(P) catalyst was between 300-650 °C, and the reduction peak of more than 500 °C was attributed to the reduction of highly dispersed NiO, which had a strong interaction with the SiO2 [31,[33][34][35][36]. The TPR results show that Ni was strongly interacted with the support of SiO2 in the 1%Ni/SBA-15(P) [21,37].

Catalytic Performance
The catalytic results are shown in Figure 5. The CO 2 conversion and CO selectivity of the two catalysts increased gradually with the increase of temperature. The 1%Ni/SBA-15(P) catalyst showed higher CO 2 conversions than the 1%Ni/SBA-15 catalyst at the same temperature. The CO 2 conversion of the 1%Ni/SBA-15(P) catalyst at 700 • C was 33.7%, which was three times that of the 1%Ni/SBA-15 catalyst. Because methanation is an exothermic reaction, low temperatures are more conducive to producing methane on the basis of thermodynamics. The CO selectivity of the 1%Ni/SBA-15(P) catalyst was 96.7% at Crystals 2021, 11, 790 5 of 7 500 • C, and it increased to 99.8% at 700 • C. Combined with characterization results, the high activity of 1%Ni/SBA-15(P) was due to the small nickel particles in the catalyst. the two catalysts increased gradually with the increase of temperature. The 1%Ni/SBA-15(P) catalyst showed higher CO2 conversions than the 1%Ni/SBA-15 catalyst at the same temperature. The CO2 conversion of the 1%Ni/SBA-15(P) catalyst at 700 °C was 33.7%, which was three times that of the 1%Ni/SBA-15 catalyst. Because methanation is an exothermic reaction, low temperatures are more conducive to producing methane on the basis of thermodynamics. The CO selectivity of the 1%Ni/SBA-15(P) catalyst was 96.7% at 500 °C, and it increased to 99.8% at 700 °C. Combined with characterization results, the high activity of 1%Ni/SBA-15(P) was due to the small nickel particles in the catalyst. The stability study of the 1%Ni/SBA-15(P) catalyst is shown in Figure 6. The CO2 conversion was about 35%, and the CO selectivity was higher than 99.8% during 1000 min. The catalytic results show that the 1%Ni/SBA-15(P) catalyst was stable in the RWGS reaction at 700 °C. Combined with characterization results, the excellent stability should be attributed to the confined effect of SBA-15. The mesoporous confinement and metal-support interaction [38] can enhance the thermal stability of the nickel particles and inhibit sintering.   The stability study of the 1%Ni/SBA-15(P) catalyst is shown in Figure 6. The CO 2 conversion was about 35%, and the CO selectivity was higher than 99.8% during 1000 min. The catalytic results show that the 1%Ni/SBA-15(P) catalyst was stable in the RWGS reaction at 700 • C. Combined with characterization results, the excellent stability should be attributed to the confined effect of SBA-15. The mesoporous confinement and metal-support interaction [38] can enhance the thermal stability of the nickel particles and inhibit sintering. 500 °C, and it increased to 99.8% at 700 °C. Combined with characterization results, the high activity of 1%Ni/SBA-15(P) was due to the small nickel particles in the catalyst. The stability study of the 1%Ni/SBA-15(P) catalyst is shown in Figure 6. The CO2 conversion was about 35%, and the CO selectivity was higher than 99.8% during 1000 min. The catalytic results show that the 1%Ni/SBA-15(P) catalyst was stable in the RWGS reaction at 700 °C. Combined with characterization results, the excellent stability should be attributed to the confined effect of SBA-15. The mesoporous confinement and metal-support interaction [38] can enhance the thermal stability of the nickel particles and inhibit sintering.      [40] In summary, the 1%Ni/SBA-15(P) was active and stable during the RWGS at hightemperature. The characteristic results proved that the NiO particles were highly dispersed Crystals 2021, 11, 790 6 of 7 and confined in mesoporous SiO 2 channels for the 1%Ni/SBA-15(P) catalyst, and the nickel strongly interacted with SiO 2 support. Based on the characterization and catalytic test results, the good activity and stability of the 1%Ni/SBA-15(P) catalyst was due to the small Ni particle and enhanced interaction between Ni and SiO 2 .

Conclusions
A highly dispersed 1%Ni/SBA-15(P) catalyst, with Ni particle size at about 3 nm, was synthesized, which was highly active and stable during the high-temperature RWGS reaction. Based on the characterizations, it was proposed that the excellent catalytic performance was due to the small Ni particles and the metal-support interaction.