Dry Reforming of Methane over Ni-Supported SBA-15 Prepared with Physical Mixing Method by Complexing with Citric Acid

: Ni-supported SBA-15 catalysts were prepared by physical mixing of Ni(NO 3 ) 2 · 6H 2 O and SBA-15 (Ni/SBA-15-M) and in the presence of citric acid as the complexing agent (Ni/SBA-15-M-C). Moreover, an Ni-supported SBA-15 catalyst was also prepared by the conventional incipient impregnation method (Ni/SBA-15-I). All the catalysts were systematically evaluated for carbon dioxide reforming of methane (CDR) at CO 2 /CH 4 = 1.0, gas hourly space velocity of 60,000 mL · g − 1 · h − 1 , and reaction temperature of 700 ◦ C. The characterization results show that the Ni particle size of Ni/SBA-15-M-C is signiﬁcantly smaller than that of Ni/SBA-15-M due to the coordination effect of citric acid and Ni 2+ . Consequently, the Ni/SBA-15-M-C exhibits superior anti-coking and anti-sintering during the CDR-operated period because of the higher Ni dispersion and stronger Ni–support interaction. Compared to the Ni/SBA-15-I, the physical mixing of nickel salt and mesoporous material for preparing of Ni-based catalyst is easy to operate, although the crystal size and catalytic performance of Ni/SBA-15-C are very similar to that of Ni/SBA-15-M-I. Thus, the efﬁcient and easily controlled catalyst structure makes the physical mixing strategy very promising for preparing highly active and stable CDR catalysts.


Introduction
In recent years, with the development of industry, global climate change has become a major issue of public concern. Meanwhile, the proposal of the "Dual Carbon Goal" indicates the importance and urgency of carbon capture and utilization. Thus, the efficient conversion and utilization technology of CO 2 generated in the process of fossil energy utilization is significant for CO 2 emission reduction [1][2][3]. The carbon dioxide reforming of methane (CDR) technology has received great attention and has been extensively studied and reported [3][4][5][6]. The CDR reaction is an efficient process for utilizing the major greenhouse gases CO 2 and CH 4 simultaneously, which is of great significance for reducing greenhouse gas emissions. Moreover, the H 2 /CO of the produced syngas is ≤1, which can be used as a feed gas for the synthesis of carbonyl and organic oxygen-containing compounds, as well as for the synthesis of long-chain hydrocarbons through Fischer-Tropsch (FT) synthesis reaction [7,8]. More importantly, compared with other CO 2 conversion and utilization technologies, CDR is expected to be directly applied to the reforming of CH 4 and CO 2 in flue gas without the need for pre-separation of CO 2 [9][10][11]. Therefore, accelerating the industrialization process of CDR reaction is important for achieving CO 2 emission reduction and efficient utilization. exhibited higher Ni dispersion and improved stability, but the visible growth of Ni particles was shown after 50 h for CDR reaction [25,36].
In this work, the Ni-based catalysts were prepared by the physical mixing of nickel salt and mesoporous material. By adding a complexing agent, the particle size of the active component Ni has been reduced and the dispersion has been improved, thereby obtaining a high-performance CDR reaction Ni-based catalyst. Compared with the traditional incipient impregnation method, the physical grinding method is easy to operate and control. Moreover, all the Ni-based catalysts were comparatively investigated for CDR under severe conditions of P = 1.0 atm, T = 700 • C, CH 4 /CO 2 = 1.0, and GHSV = 60,000 mL·g −1 ·h −1 . Compared to the Ni/SBA-15 catalyst prepared by the physical mixing method in the absence of citric acid (Ni/SBA-15-M), the Ni/SBA-15 catalyst prepared by adding citric acid as complexing agent (Ni/SBA-15-M-C) exhibits higher activity and stability in CDR reaction. Moreover, although the catalytic performance of Ni/SBA-15-M-C and Ni/SBA-15 prepared by using the conventional incipient impregnation methods (Ni/SBA-15-I) is very similar, the easy operation makes the physical mixing strategy very promising for preparing highly active and stable CDR catalysts.

Structural Properties of Ni-Based Catalysts
All the Ni-based catalysts were characterized by X-ray diffraction (XRD) to display the crystal structure and phase analysis, and the results are shown in Figure 1. The position of sharp diffraction peaks for the three catalysts is identical, which indicates that the crystal phase of Ni species is independent of the preparation method. The broad diffraction peak at about 23 • is attributed to the amorphous silica of SBA-15 [37,38]. The diffraction peaks at about 37 • , 43 • , 63 • , and 76 • are attributed to the cubic phase NiO [3,18]. When the intensity of the NiO diffraction peaks is given more attention, a significant difference can be observed, i.e., stronger NiO diffraction peaks were shown over Ni/SBA-15-M, while weaker NiO diffraction peaks were shown over Ni/SBA-15-M-C and Ni/SBA-15-I. Based on the Scherrer formula and the XRD diffraction at 43.3 • , the crystal size of NiO over different catalysts was calculated, and the results were summarized in Table 1. Bigger NiO particle sizes were shown over Ni/SBA-15-M (16.1 nm), and smaller NiO particle sizes were shown over Ni/SBA-15-M-C (6.5 nm) and Ni/SBA-15-I (6.1 nm). Compared to the Ni/SBA-15-M, the smaller particle sizes of Ni over the Ni/SBA-15-M-C can be attributed to the formation of the complex by the Ni 2+ and citric acid, which effectively inhibited the agglomeration of Ni particles over SBA-15 in the preparation process. For the Ni/SBA-15-I, the smaller Ni particle size indicates that the Ni 2+ was infiltrated and uniformly dispersed into the internal surface of mesopores by using the incipient wetness impregnation method. Moreover, as reported in the literature [39], the Ni/SBA-15 catalyst with Ni loading of 10% was prepared by using the citric acid-assisted impregnation method, and the NiO particle size was about 8.1 nm, which was slightly larger than that of Ni/SBA-15-M-C (6.5 nm) in this work. This result indicated that the physical mixing of Ni(NO 3 ) 2 ·6H 2 O and SBA-15 in the presence of citric acid as a complexing agent is an efficient strategy for preparing the Ni-based catalyst with higher dispersion.

Structural Properties and Crystal Size of Reduced Ni-Based Catalysts
The metal Ni was the active site for the CDR reaction, the Ni-based catalysts were reduced at the conditions of T = 700 °C for 120 min at the atmospheric pressure, and the XRD characterization results are shown in Figure 2. All the catalysts showed the similar diffraction pa erns. The broad peak at 23° indicates that the structure of SiO2 was unchanged at the reduction condition. Moreover, the peaks at 44°, 53°, and 76° were a ributed to the characteristic diffraction peak of metallic Ni [18,25]. Importantly, there was no diffraction peak of any NiO species, suggesting that the nickel oxide was successfully reduced to metallic nickel at the reduction conditions. Moreover, the peak intensity of metallic Ni exhibited the same trends as the NiO over the Ni-based catalyst. Based on the Scherrer formula and the diffraction peak at 44.5°, the crystal size of Ni was calculated, and the results were summarized in Table 1. The Ni particle size of Ni/SBA-15-M (17.2 nm) was clearly larger than that of Ni/SBA-15-M-C (8.1 nm) and Ni/SBA-15-I (7.8 nm). As reported in the literature [40], Ni/SBA-15 catalysts with Ni loading of 7% were prepared using the citric acid-assisted incipient wetness impregnation method. The authors found that the Ni particle size estimated by H2 pulse chemisorption was in the range of 7.3-8.2 nm, which is dependent on the concentration of the citric acid used. The Ni particle size of Ni/SBA-15 reported in the above-mentioned literature was similar to that of Ni/SBA-15-M-C (8.1 nm) prepared by simply physical mixing method in this work. Thus, the similar Ni particle size over Ni/SBA-15 can be obtained through the physical mixing and co-impregnation method, respectively.
To confirm the particle size and the distribution of Ni, the reduced Ni-based catalysts were subjected to transmission electron microscopy (TEM) characterization, and the

Structural Properties and Crystal Size of Reduced Ni-Based Catalysts
The metal Ni was the active site for the CDR reaction, the Ni-based catalysts were reduced at the conditions of T = 700 • C for 120 min at the atmospheric pressure, and the XRD characterization results are shown in Figure 2. All the catalysts showed the similar diffraction patterns. The broad peak at 23 • indicates that the structure of SiO 2 was unchanged at the reduction condition. Moreover, the peaks at 44 • , 53 • , and 76 • were attributed to the characteristic diffraction peak of metallic Ni [18,25]. Importantly, there was no diffraction peak of any NiO species, suggesting that the nickel oxide was successfully reduced to metallic nickel at the reduction conditions. Moreover, the peak intensity of metallic Ni exhibited the same trends as the NiO over the Ni-based catalyst. Based on the Scherrer formula and the diffraction peak at 44.5 • , the crystal size of Ni was calculated, and the results were summarized in Table 1. The Ni particle size of Ni/SBA-15-M (17.2 nm) was clearly larger than that of Ni/SBA-15-M-C (8.1 nm) and Ni/SBA-15-I (7.8 nm). As reported in the literature [40], Ni/SBA-15 catalysts with Ni loading of 7% were prepared using the citric acid-assisted incipient wetness impregnation method. The authors found that the Ni particle size estimated by H 2 pulse chemisorption was in the range of 7.3-8.2 nm, which is dependent on the concentration of the citric acid used. The Ni particle size of Ni/SBA-15 reported in the above-mentioned literature was similar to that of Ni/SBA-15-M-C (8.1 nm) prepared by simply physical mixing method in this work. Thus, the similar Ni particle size over Ni/SBA-15 can be obtained through the physical mixing and co-impregnation method, respectively.
To confirm the particle size and the distribution of Ni, the reduced Ni-based catalysts were subjected to transmission electron microscopy (TEM) characterization, and the results are shown in Figure 3. It was clear that a larger Ni particle size and broad distribution of Ni particles were obtained over the reduced Ni/SBA-15-M, while smaller Ni particle size and a narrower distribution of Ni particles were observed over the reduced Ni/SBA-15-M-C and Ni/SBA-15-I, which were consistent with the XRD results ( Figure 2). Compared to Ni/SBA-15-M, Ni/SBA-15-M-C with a higher dispersion of Ni was obtained by using physical mixing method with adding citric acid as complexing agent. Moreover, Ni/SBA-15-I also exhibited smaller Ni particle size and a narrower distribution of Ni particles, which can be explained by the larger surface area of the support SBA-15 [41].
Ni particle size and a narrower distribution of Ni particles were observed over the reduced Ni/SBA-15-M-C and Ni/SBA-15-I, which were consistent with the XRD results ( Figure 2). Compared to Ni/SBA-15-M, Ni/SBA-15-M-C with a higher dispersion of Ni was obtained by using physical mixing method with adding citric acid as complexing agent. Moreover, Ni/SBA-15-I also exhibited smaller Ni particle size and a narrower distribution of Ni particles, which can be explained by the larger surface area of the support SBA-15 [41].

Textural Properties of the Ni-Based Catalysts
The textural properties of the Ni/SBA-15-M, Ni/SBA-15-M-C, and Ni/SBA-15-I calculated from the N2 adsorption-desorption isotherms are also summarized in Table 1. For comparation, SBA-15 was also subjected by the N2 adsorption-desorption analyzed, and the results are also shown in Table 1. The BET surface area, pore volumes, and pore diameter were 470.7 m 2 ·g −1 , 0.86 cm 3 ·g −1 , and 7.32 nm for the SBA-15, respectively. Compared to SBA-15, the BET surface area, pore volume, and pore diameter of the Ni/SBA-15-M, Ni/SBA-15-M-C, and Ni/SBA-15-I were clearly decreased. This can be reasonably a ributed to the deposition of the NiO in the internal surface of mesopores of the SBA-15, leading to the decrease in available surface area, pore volume, and pore diameter. Moreover, the BET surface area, pore volume, and pore diameter were unobvious change at about 435~439 m 2 ·g −1 , 0.72~0.77 m 3 ·g −1 , and 6.76~6.89 nm for the Ni supported catalysts. After careful comparison, it was found that Ni/SBA-15-M-C showed Ni/SBA-15-M-C and Ni/SBA-15-I, which were consistent with the XRD results ( Figure 2). Compared to Ni/SBA-15-M, Ni/SBA-15-M-C with a higher dispersion of Ni was obtained by using physical mixing method with adding citric acid as complexing agent. Moreover, Ni/SBA-15-I also exhibited smaller Ni particle size and a narrower distribution of Ni particles, which can be explained by the larger surface area of the support SBA-15 [41].

Textural Properties of the Ni-Based Catalysts
The textural properties of the Ni/SBA-15-M, Ni/SBA-15-M-C, and Ni/SBA-15-I calculated from the N2 adsorption-desorption isotherms are also summarized in Table 1. For comparation, SBA-15 was also subjected by the N2 adsorption-desorption analyzed, and the results are also shown in Table 1. The BET surface area, pore volumes, and pore diameter were 470.7 m 2 ·g −1 , 0.86 cm 3 ·g −1 , and 7.32 nm for the SBA-15, respectively. Compared to SBA-15, the BET surface area, pore volume, and pore diameter of the Ni/SBA-15-M, Ni/SBA-15-M-C, and Ni/SBA-15-I were clearly decreased. This can be reasonably a ributed to the deposition of the NiO in the internal surface of mesopores of the SBA-15, leading to the decrease in available surface area, pore volume, and pore diameter. Moreover, the BET surface area, pore volume, and pore diameter were unobvious change at about 435~439 m 2 ·g −1 , 0.72~0.77 m 3 ·g −1 , and 6.76~6.89 nm for the Ni supported catalysts. After careful comparison, it was found that Ni/SBA-15-M-C showed

Textural Properties of the Ni-Based Catalysts
The textural properties of the Ni/SBA-15-M, Ni/SBA-15-M-C, and Ni/SBA-15-I calculated from the N 2 adsorption-desorption isotherms are also summarized in Table 1. For comparation, SBA-15 was also subjected by the N 2 adsorption-desorption analyzed, and the results are also shown in Table 1. The BET surface area, pore volumes, and pore diameter were 470.7 m 2 ·g −1 , 0.86 cm 3 ·g −1 , and 7.32 nm for the SBA-15, respectively. Compared to SBA-15, the BET surface area, pore volume, and pore diameter of the Ni/SBA-15-M, Ni/SBA-15-M-C, and Ni/SBA-15-I were clearly decreased. This can be reasonably attributed to the deposition of the NiO in the internal surface of mesopores of the SBA-15, leading to the decrease in available surface area, pore volume, and pore diameter. Moreover, the BET surface area, pore volume, and pore diameter were unobvious change at about 435~439 m 2 ·g −1 , 0.72~0.77 m 3 ·g −1 , and 6.76~6.89 nm for the Ni supported catalysts. After careful comparison, it was found that Ni/SBA-15-M-C showed the bigger BET surface area, pore volumes, and pore diameter than the Ni/SBA-15-M and Ni/SBA-15-I. In the case of Ni/SBA-15-M-C, the Ni 2+ was complexed with citric acid to form the complex, and the porous structure was formed after calcination, leading to the higher BET surface area and pore volume for Ni/SBA-15-M-C [34,35,41].

Reduction Behavior of the Ni-Based Catalysts
The reducibility of the Ni-based catalysts was evaluated by H 2 temperature programmed reaction (H 2 -TPR), and the results were presented in Figure 4. Only one significant reduction peak centered at about 349 • C was displayed over the Ni/SBA-15-M. However, except for the peak centered at about 349 • C, a clear reduction peak centered at about 534 • C was also observed over the Ni/SBA-15-M-C and Ni/SBA-I. It is well known that Ni 2+ is directly reduced to the metallic Ni [42][43][44]. Thus, the reduction peaks of hydrogen consumption at varied temperature regions can be reasonably attributed to the reduction of different kinds of Ni 2+ species [41,45]. Generally, the reduction peak of larger NiO particles was less than 400 • C, while the reduction peak of relatively smaller NiO particles or the NiO having strong interactions with the support was higher than 450 • C [46,47]. Thus, the reduction peaks centered at about 349 • C were ascribed to the reduction of larger NiO particles, and the peaks appearing at about 534 • C can be assigned to either the reduction of smaller NiO particles or the NiO having strong interactions with the support [41,46,47]. Thus, only one reduction peak of 349 • C was observed over the Ni/SBA-15-M, suggesting the formation of bigger NiO particles over Ni/SBA-15-M. Compared to Ni/SBA-15-M, except for the reduction peak at about 349 • C, a higher reduction peak at about 534 • C was also observed over Ni/SBA-15-I and Ni/SBA-15-M-C. The lower reduction peak was also ascribed to the reduction of the bigger NiO particles, while the higher reduction peak at about 530 • C was ascribed to the reduction of the smaller NiO particles and NiO having stronger interactions with support. These results can be explained by the growth of Ni particle sizes being significantly inhibited through the complexation of metallic Ni and citric acid in the physical mixing process. These results were consistent with the results of XRD and TEM.

Reduction Behavior of the Ni-Based Catalysts
The reducibility of the Ni-based catalysts was evaluated by H2 temperature programmed reaction (H2-TPR), and the results were presented in Figure 4. Only one significant reduction peak centered at about 349 °C was displayed over the Ni/SBA-15-M. However, except for the peak centered at about 349 °C, a clear reduction peak centered at about 534 °C was also observed over the Ni/SBA-15-M-C and Ni/SBA-I. It is well known that Ni 2+ is directly reduced to the metallic Ni [42][43][44]. Thus, the reduction peaks of hydrogen consumption at varied temperature regions can be reasonably a ributed to the reduction of different kinds of Ni 2+ species [41,45]. Generally, the reduction peak of larger NiO particles was less than 400 °C, while the reduction peak of relatively smaller NiO particles or the NiO having strong interactions with the support was higher than 450 °C [46,47]. Thus, the reduction peaks centered at about 349 °C were ascribed to the reduction of larger NiO particles, and the peaks appearing at about 534 °C can be assigned to either the reduction of smaller NiO particles or the NiO having strong interactions with the support [41,46,47]. Thus, only one reduction peak of 349 °C was observed over the Ni/SBA-15-M, suggesting the formation of bigger NiO particles over Ni/SBA-15-M. Compared to Ni/SBA-15-M, except for the reduction peak at about 349 °C, a higher reduction peak at about 534 °C was also observed over Ni/SBA-15-I and Ni/SBA-15-M-C. The lower reduction peak was also ascribed to the reduction of the bigger NiO particles, while the higher reduction peak at about 530 °C was ascribed to the reduction of the smaller NiO particles and NiO having stronger interactions with support. These results can be explained by the growth of Ni particle sizes being significantly inhibited through the complexation of metallic Ni and citric acid in the physical mixing process. These results were consistent with the results of XRD and TEM.

Catalytic Performance Tests
Commonly, the higher reaction temperature (>750 • C) is beneficial to the carbon elimination for Ni-based catalysts in CDR reaction. To reflect the variation of catalytic stability of the catalysts in this work, severe reaction condition, i.e., lower reaction temperature (700 • C), was chosen for the catalyst's evaluation. Thus, CDR performance, i.e., the CH 4 and CO 2 conversions, H 2 /CO, and H 2 yield, for all Ni-based catalysts was carried out at P = 1.0 atm, T = 700 • C, CO 2 /CH 4 = 1.0, and GHSV = 60,000 mL·g −1 ·h −1 , and the results are presented in Figure 5. The CDR performance was significantly affected by the preparation methods and the addition of citric acid. All the catalysts showed similar initial CH 4 and CO 2 conversion (Figure 5a,b), i.e., about 72% and 81%, respectively. However, the stability was significantly different over the Ni-based catalysts. For the Ni/SBA-15-M catalyst, the conversions of CH 4 and CO 2 were linearly decreased to about 20% and 28% after 2.5 h CDR testing, respectively. The conversions of CH 4 and CO 2 were continuously decreased to 13% and 24% at the time on stream of 8 h, respectively. On the contrary, the Ni/SBA15-M-C and Ni//SBA-15-I exhibited higher stability in CRD reaction. The conversions of CH 4 and CO 2 were almost unchanged during 20 h of CDR testing. These results can be explained based on the varied particle sizes of Ni. The bigger particle sizes of Ni were observed over Ni/SBA-15-M, which made it easy to form the coking deposited. Therefore, the active site of Ni over the Ni/SBA-15-M was covered by the coking deposition, which resulted in the decreased conversions of CH 4 and CO 2 . Importantly, the smaller particle sizes of Ni and stronger interaction between Ni and support were obtained over Ni/SBA-15-M-C and Ni/SBA-15-I, which effectively inhibit the sintering of Ni particles and the formation of carbon deposition at high temperatures. All the results were consistent with the previous results of XRD, TEM, and H 2 -TPR results. Moreover, as reported in the literature [40], the Ni/SBA-15 prepared by the acid-assisted co-impregnation method exhibited excellent stability within the testing time of 30 h, which was similar to the results in this work. The conversion of CH 4 and CO 2 in the above-mentioned literature was dependent on the concentration of citric acid used for the preparation of Ni/SBA-15. The conversion of CH 4 and CO 2 over the optimal catalyst was higher than that in this work. Therefore, optimizing the mole ratio of citric and Ni for the preparation of Ni/SBA15-M-C in this work is required in the future.

Characterization of Used Catalysts
Coke deposition is one of the main reasons for the deactivation of Ni-based catalysts for CDR. Thus, the coke deposition of the used catalysts after the CDR reaction was subjected to thermogravimetric analysis (TG), and the results are shown in Figure 6. As expected, all the catalysts showed significant weight losses, which was consistent with the CDR performance. As shown in Figure 6, a visible weight increase of around 250-600 °C was observed for all the catalysts, which can be a ributed to the oxidation of metal Ni to NiO in the air atmosphere [48,49]. The weight increase in the Ni/SBA-15-M, Ni/SBA-15-M-C, and Ni/SBA-15-I are 0.7%, 1.5%, and 2.5%, respectively. The lowest weight increase in the Ni/SBA-15-M should be a ributed to the coating of carbon over the Ni particles, which restrains the oxidation of metal Ni. This explanation can be certified by the obvious Moreover, all the Ni-based catalysts showed an H 2 /CO ratio <1 (Figure 5c), which was explained by the unavoidable reversed water gas shift (RWGS) reaction in CDR, i.e., the CO and H 2 O was formed by the reaction of CO 2 and H 2 . The H 2 /CO ratio for Ni/SBA-15-M-C and Ni/SBA-15-I was far higher than that of Ni/SBA-15-M catalysts. Meanwhile, the H 2 yield showed the same trend with CH 4 conversion for all catalysts (Figure 5a), while it was slightly lower than the CH 4 conversion shown in Figure 5d, which was also caused by the RWGS reaction in CDR. All these results certify the importance of Ni particle size in CDR reaction, which is dependent on the preparation method and the addition of citric acid.

Characterization of Used Catalysts
Coke deposition is one of the main reasons for the deactivation of Ni-based catalysts for CDR. Thus, the coke deposition of the used catalysts after the CDR reaction was subjected to thermogravimetric analysis (TG), and the results are shown in Figure 6. As expected, all the catalysts showed significant weight losses, which was consistent with the CDR performance. As shown in Figure 6, a visible weight increase of around 250-600 • C was observed for all the catalysts, which can be attributed to the oxidation of metal Ni to NiO in the air atmosphere [48,49]. The weight increase in the Ni/SBA-15-M, Ni/SBA-15-M-C, and Ni/SBA-15-I are 0.7%, 1.5%, and 2.5%, respectively. The lowest weight increase in the Ni/SBA-15-M should be attributed to the coating of carbon over the Ni particles, which restrains the oxidation of metal Ni. This explanation can be certified by the obvious weight loss (about 9.5%) in the range of 550-800 • C over Ni/SBA-15-M, which was ascribed to the gasification of the coke deposition. In contrast, the amount of the deposited coke on Ni/SBA-15-M-C and Ni/SBA-15-I was less than 1%, indicating very small coke deposition. These results can be explained by the Ni particle size of the Ni-based catalysts. The amount of coke deposition over Ni/SBA-15-M-C and Ni/SBA-15-I with smaller Ni particle size was lower, which leads to its excellent stability within 20 h CDR testing ( Figure 5). The highest weight loss for used Ni/SBA-15-M was consistent with the literature that the large Ni particle sizes of the catalyst tend to the coke deposition during the CDR, which results in the rapid deactivation of the catalyst [50,51]. Thus, physical grinding of Ni(NO 3 ) 2 ·6H 2 O and SBA-15 in the presence of citric acid as a complexing agent is an effective method to obtain smaller Ni particles, which led to the decrease in coke deposition and higher stability of the catalysts for CDR reaction.
Catalysts 2023, 13, x 9 of 13 effective method to obtain smaller Ni particles, which led to the decrease in coke deposition and higher stability of the catalysts for CDR reaction.

Catalysts Preparation
Before impregnation, the commercial SBA-15 (purchased from Nanjing Ji Cang Nano Technology Co., Ltd., Nanjing, China) was treated at 700 °C for 4 h. Ni-based catalyst was prepared through a simple physical mixing method by adding citric acid as the complexing agent. Ni(NO3)2·6H2O was used as the precursor of Ni, commercial SBA-15

Catalysts Preparation
Before impregnation, the commercial SBA-15 (purchased from Nanjing Ji Cang Nano Technology Co., Ltd., Nanjing, China) was treated at 700 • C for 4 h. Ni-based catalyst was prepared through a simple physical mixing method by adding citric acid as the complexing agent. Ni(NO 3 ) 2 ·6H 2 O was used as the precursor of Ni, commercial SBA-15 was used as the support, and the load of nickel on catalysts was 10 wt%. A certain amount of Ni(NO 3 ) 2 ·6H 2 O, SBA-15, and citric acid was mixed in a agate mortar, which was fully ground for 30 min. Then, the solid was transferred to a muffle furnace and calcined at 500 • C for 4 h, and the obtained sample was denoted as Ni/SBA-15-M-C. The Ni/SBA-15-M was synthesized by the same process but without the addition of citric acid. For comparison, Ni/SBA-15-I was prepared with the traditional incipient wetness impregnation method. The detailed preparation of catalysts is as follows: Ni(NO 3 ) 2 ·6H 2 O was dissolved in distilled water, which is the same volume as the pore volume of SBA-15. SBA-15 was added to the nickel nitrate solution and shaken until it was evenly mixed. After impregnation, the sample was dried at 80 • C for 12 h and calcined at 500 • C for 4 h.

Characterization Technique of the Catalysts
The crystal structure and particle size of Ni species over the catalysts were obtained using Bruker powder X-ray diffractometer (D8 advanced, Karlsruhe, Germany). The sample was tested at the conditions as follows: Cu Kα, λ = 1.5418 Å, 40 KV, 40 mA. The particle size of NiO and Ni was qualitatively calculated according to the half peak width of NiO and Ni diffractions peaks over the fresh and reduced catalyst, respectively, based on the formula of Scherrer.
The textural properties of the sample, such as BET surface areas, pore volumes, and pore sizes distribution, were characterized by a physisorption analyzer (BelSorp-Max, MicrotracBEL Corp., Osaka, Japan) at −196 • C. Before the testing, all samples were pretreated at 300 • C for 12 h to desorb adsorbed gas and moisture. The BET method was used to estimate the BET surface areas, while the pore volumes were calculated by using the adsorption capacity at the relative pressure (P/P 0 ) = 0.99.
The H 2 -TPR testing was carried out on a Micromeritics Autochem apparatus (Autochem II 2920, Norcross, GA, USA). Prior to the measurement, the sample (about 0.0500 g) was treated at 300 • C for 60 min under Ar gas flow. Then, the samples were cooled to 100 • C. H 2 -TPR pattern was recorded from 100 to 1000 • C at the heating rate of 10 • C min −1 in 10 vol% H 2 /Ar gas flow.
TEM images of the reduced sample were acquired using a HITACHI transmission electron microscope (HT7800, Hitachi, Tokyo, Japan). The sample was dispersed in ethanol using ultrasonic for 30 min. The suspension was dripped onto a carbon-enhanced copper grid and dried in air.
The coke deposition on the used catalyst was measured using a thermoanalyzer system of TA instruments (Q1000DSC+LNCS+FACS Q600SDT, New Castle, DE, USA) at an air flow with a heating rate of 10 • C/min from 100 to 1000 • C.

Activity Evaluation of the Catalysts
CDR performance was evaluated on a fixed bed reactor (quartz tubular, i.d. = 8 mm). Typically, the catalyst (about 0.10 g) diluted with quartz sand (0.9 g) was loaded and sandwiched by two quartz layers. Before evaluation, the catalyst was reduced with 20% H 2 /N 2 of 50 mL/min at 700 • C for 2.5 h and then N 2 was fed into the reactor. Then, CO 2 and CH 4 with a molar ratio of 1.0 were injected into the reactor. The reaction was performed at 700 • C, and GHSV of 60,000 mL·g −1 ·h −1 . The products were analyzed online by gas chromatography (GC9790II, Zhejiang Fuli chromatographic analysis Co., Ltd., Wenling, China) after cooling to room temperature and condensing the water. The composition of exhaust was detected by a thermal conductivity detector (TCD) with the Molecular Sieve 5A and Porapak Q capillary columns by using Ar as carrier gas. Because the reactant (CH 4 , CO 2 ) and the primary product (CO, H 2 ) can be simultaneously detected and quantified by using a TCD detector, the TCD was used for composition detection to reduce system complexity in this work. Typically, the temperatures of the sample injection system, column, and detector are 100, 80, and 100 • C. The calibration factor of CH 4 , CO 2 , H 2 , and CO was obtained with the external standard method by using a standard gas mixture. The conversion of CH 4 and CO 2 (abbreviated as X i ), the molar ratio of H 2 to CO (H 2 /CO), and H 2 yield (Y) were calculated as below:

Conclusions
The Ni/SBA-15-M and Ni/SBA-15-M-C catalysts were successfully prepared by physical grinding of nickel salt and mesoporous material SBA-15. Compared to Ni/SBA-15-M, Ni/SBA-15-M-C prepared by using citric acid as the complexing agent exhibited a significantly smaller Ni particle size and narrower Ni distribution due to the coordination effect of citric acid and Ni 2+ . The Ni/SBA-15-M-C exhibits high CDR activity and stability during the CDR reaction because of the improved resistance of Ni sintering and coke deposition. Compared to the Ni/SBA-15-I prepared by conventional incipient impregnation method, the physical mixing of nickel salt and mesoporous material for preparing of Ni-based catalyst is easy to operate, although the crystal size and catalytic performance of Ni/SBA-15-C is very similar to that of Ni/SBA-15-M-I. Thus, the efficient and easily controlled catalyst structure make the physical mixing strategy very promising for preparing highly active and stable CDR catalysts.