Influence of Supports on the Catalytic Activity and Coke Resistance of Ni Catalyst in Dry Reforming of Methane

The dependence of the catalytic activity and coke resistance of Ni-based catalysts on the support type was investigated in the dry reforming of methane (DRM). Catalysts were prepared using incipient wetness impregnation and analyzed using ICP-OES, BET-BJH, XRD, H2chemisorption, H2-TPR, and CO2-TPD. DRM was performed at 600–750 ◦C at 144,000 mL/gcat·h of GHSV (CH4/CO2/N2 = 1/1/1). Ni/Al2O3 and Ni/MgO catalysts formed NiAl2O4 and NiO-MgO solid solutions, respectively, owing to strong binding between the metal and support. In contrast, MgO-Al2O3 and MgAl2O4 supports suppressed NiAl2O4 and NiO-MgO solid solution formation, due to Mg addition, with high metal dispersions of 4.6 and 6.6%, respectively. In the DRM reaction, the Ni/MgO-Al2O3 and Ni/MgAl2O4 catalysts showed high CH4 conversions of 78.1 and 76.8%, respectively, compared with Ni/Al2O3 and Ni/MgO at 750 ◦C. A stability test was performed at 600 ◦C for 20 h. A coke study of the spent catalysts was performed using SEM and TGA. Alkaline-earth metal-containing catalysts Ni/MgO-Al2O3 and Ni/MgAl2O4 with strong CO2 adsorption properties showed 20 wt% reduction in carbon deposition compared to commercial catalysts. Therefore, the support and basic properties of the catalyst significantly influenced the catalyst performance and coke resistance in the DRM.


Introduction
Recently, there has been an increased focus on problems related to climate change and the need for reducing carbon dioxide (CO 2 ) emissions. The development of carbon dioxide utilization (CDU) technologies is necessary to reduce CO 2 emissions over the long term. Among CDU technologies, the dry reforming of methane (DRM) reaction is an important technique that could be used to reduce the total annual anthropogenic greenhouse gas emissions (multi-Gt CO2 /year) [1,2]. An advantage of the DRM reaction is its low operational cost, as it does not require complicated gas separation [3]. The DRM reaction (Equation (1)) produces the beneficial syngas (H 2 + CO) from CH 4 and CO 2 [4]. Syngas is a raw material for high value-added chemicals, such as oxo-alcohol, formaldehyde, and methanol. Notably, the product H 2 /CO ratio of the DRM reaction is 1, which renders the produced syngas suitable for the synthesis of methanol and dimethyl ether or Fischer-Tropsch synthesis [5]. The DRM reaction is strongly endothermic [6] and proceeds at a relatively high temperature (650-850 • C) [7]. The reverse water gas shift (RWGS) reaction (Equation (2)) occurs as a side reaction up to 820 • C under the DRM reaction conditions, due to the endothermic nature of the reaction. Since the RWGS reaction produces CO and H 2 O using CO 2 and H 2 , the H 2 /CO ratio is lower than 1. CH 4 + CO 2 → 2H 2 + 2CO (∆H 298K = +247 kJ/mol) (1) CH 4 → 2H 2 + C ( H 298K = +74.9 kJ/mol) 2CO → CO 2 + C ( H 298K = −172.4 kJ/mol) Ni-based catalysts are the most widely used in the DRM reaction because of their high activity and low cost [8]. However, they can cause catalyst deactivation due to processes such as carbon deposition and sintering. Moreover, they have been reported to have three phases: Ni metallic state, Ni oxide, and Ni carbide phase, which influence carbon deposition. Among these, Ni metal atoms of the Ni carbide phase bind strongly with carbon; thus, easily causing carbon deposition [9]. In addition, at high reaction temperatures the active metal forms aggregates, owing to the sintering effect. The agglomeration of particles occurs randomly, causing low metal dispersion and low specific surface area at high temperatures. Moreover, the DRM reaction is more vulnerable to carbon deposition because of the higher carbon ratio in the reactants compared with other methane reforming reactions. Carbon deposition mainly occurs through CH 4 decomposition (Equation (3)) and CO disproportionation (Equation (4)) reactions. As CH 4 decomposition is endothermic, it mainly occurs above 557 • C, whereas the CO disproportionation reaction occurs below 700 • C, due to the exothermic nature of the reaction. Therefore, carbon deposition occurs in the temperature range of 557-700 • C [10]. The morphology of carbon exhibits variations in properties, which cause differences in the activity of the catalyst. In particular, filamentous carbon influences catalyst deactivation during the DRM reaction. Filamentous carbon is formed through diffusion, nucleation, and growth through Ni particles at temperatures > 450 • C. While filamentous carbon does not block the active sites of the Ni metal, it causes catalyst destruction and a pressure drop, due to pore plugging [11]. For this reason, the DRM reaction has limited commercial viability. In this context, the development of catalysts with high catalytic activity and coke resistance in the DRM reaction is essential.
In the preparation of DRM reaction catalysts, the type of support used has a significant influence on the catalytic activity. Among the various supports used in the DRM reaction, Zhang et al. reported that a ZrO 2 support had a low metal dispersion, owing to its low specific surface area of 30 m 2 /g, and also had a weak interaction between NiO and the support [12]. Whereas, amorphous ZrO 2 showed a high specific surface area of 200-300 m 2 /g and had weak thermal stability. However, a tetragonal phase was formed by calcination at temperatures above 450 • C, which remained stable at high temperatures, owing to its strong thermal stability [13]. Depending on the calcination temperature used, Al 2 O 3 supports are divided into α-Al 2 O 3 , θ-Al 2 O 3 , and γ-Al 2 O 3 . In comparison to other Al 2 O 3 supports, the γ-Al 2 O 3 support has high catalytic activity and anti-coking properties, owing to its smaller particle size [14]. In the Ni/γ-Al 2 O 3 catalyst, relatively fine Ni particles are formed, owing to the high specific surface area of γ-Al 2 O 3 . However, NiAl 2 O 4 forms a spinel structure during the heat treatment process. The high thermal stability of NiAl 2 O 4 spinel results in an increase in the reduction temperature and a reduction in the number of active sites, owing to the low metal dispersion [12]. The MgO support has a low specific surface area and forms a NiO-MgO solid solution in the presence of the Ni catalyst, which limits reduction and causes low metal dispersion [15]. In contrast, the MgO support enhances the adsorption of acidic CO 2 gas, owing to the basic properties of MgO. As the adsorbed CO 2 releases oxygen species upon dissociation, the MgO support enhances the coke resistance by promoting the oxidation of carbon formed on the catalyst surface [16,17]. In addition, owing to the basic properties of the Mg-Al mixed oxide support, the Mg-Al mixed oxide support improves the amount of CO 2 adsorption, compared with the Al 2 O 3 support [18]. A Mg-Al mixed oxide has a layered double hydroxide (LDH) structure, in which ion exchange between the cation Catalysts 2022, 12, 216 3 of 13 and anion layers is possible. The Mg 2+ and Al 3+ ions are uniformly dispersed between the hydroxide layers, therefore the LDH structure can effectively disperse the nanoparticles, without particle agglomeration. In addition, the MgO-Al 2 O 3 (LDH) support is transformed into a spinel structure during the heat treatment process and has high thermal stability. In particular, MgO-Al 2 O 3 as a support for a Ni catalyst suppresses the formation of the NiAl 2 O 4 and NiO-MgO solid solution, which is advantageous for high metal dispersion, owing to its large specific surface area [12,17]. Moreover, highly dispersed active metals on the MgO-Al 2 O 3 support have strong metal-support interactions (SMSIs). Consequently, the sintering effect is inhibited [19,20].
Ni-based catalysts have a low coke resistance, but can improve the catalytic activity and durability by dispersing the active metals or enhancing their basic properties [21]. The γ-Al 2 O 3 support has a large specific surface and forms fine particles, while the MgO support has excellent basic properties. The bimetallic MgO-Al 2 O 3 and MgAl 2 O 4 supports can, thus, overcome the disadvantages of the γ-Al 2 O 3 and MgO supports. Therefore, in this study, the influence of various supports: γ-Al 2 O 3 , MgO, MgO-Al 2 O 3 , and MgAl 2 O 4 , was evaluated to improve the catalytic activity and coke resistance of Ni-based catalysts for the DRM reaction. In addition, the prepared catalysts were compared with commercial catalysts (Ni/Al 2 O 3 and Ni/CaAl 2 O 4 ) in the dry reforming of methane (DRM) reaction.

Characteristics
The characteristics of the catalysts are listed in Table 1. The Ni metal content of the catalysts was confirmed using ICP-OES analysis. The loading amounts of the prepared and commercial catalysts were approximately 10 and 15 wt%, respectively. Ni-based catalysts have a reduced specific surface area compared to that of the fresh support, due to the pore-blocking during the process of preparing the supported Ni metals. In particular, the Ni/MgO catalyst exhibited a significant decrease in the specific surface area of 17.7 m 2 /g after the Ni metal was supported. Zanganeh et al. reported that the crystallite size increased and the specific surface area decreased with increasing calcination temperature when NiO-MgO was calcined at 600-800 • C [22]. The Ni/MgAl 2 O 4 catalyst maintained a relatively high specific surface area of 113.3 m 2 /g. Since the Ni/MgO-Al 2 O 3 catalyst exhibited changes in the textural properties of the MgO-Al 2 O 3 support during the heat treatment, its specific surface area was lower by 108.7 m 2 /g, compared to that of the Ni/MgAl 2 O 4 catalyst. This was attributed to the lack of calcination of the MgO-Al 2 O 3 (LDH) support used for the Ni catalyst preparation, which contained MgO and Al 2 O 3 in a 30:70 (wt%) ratio. Among the prepared Ni-based catalysts, the Ni/Al 2 O 3 catalyst had the lowest metal dispersion of 2.1%, and the metal dispersion of the Ni/MgO catalyst could not be measured because of the high reduction temperature. The Ni/MgO-Al 2 O 3 catalyst, which exhibited changes in the textural properties of the support during heat treatment, showed a lower metal dispersion (4.6%) than the Ni/MgAl 2 O 4 catalyst (6.6%). In addition, the commercial catalysts had very low metal dispersions, of less than 0.5%, compared those of the prepared catalysts. The composition and crystalline phases of the catalysts were confirmed by XRD analysis. XRD patterns are shown in Figure 1  The composition and crystalline phases of the catalysts were confirmed by XRD analysis. XRD patterns are shown in Figure 1. The Ni/Al2O3 catalyst formed a NiAl2O4 spinel crystalline phase. Nurunnabi et al. reported that the NiO peak shifted from 2θ = 43.26° to 2θ = 42.90-43.00° with the formation of a NiO-MgO solid solution [23]. The NiO peak of the prepared Ni/MgO catalyst also shifted to 2θ = 42.76°, owing to the formation of the NiO-MgO solid solution. In particular, Ni/MgO-Al2O3, Ni/MgAl2O4 catalysts displayed peaks corresponding to MgAl2O4 and NiO, and those corresponding to the NiAl2O4 and NiO-MgO solid solutions were not detected. The catalysts without alkaline earth metals presented peaks of NiAl2O4, but the Ca-containing commercial catalyst B showed a reduced formation of NiAl2O4 compared with catalysts using Al2O3 as a support. Therefore, the addition of alkaline earth metals effectively suppressed the formation of NiAl2O4 and NiO-MgO solid solutions.  H 2 -TPR analysis was conducted to analyze the reducibility of the catalysts ( Figure 2). Pure NiO (unsupported) can be reduced in the 300-550 • C temperature range, while pure NiAl 2 O 4 (unsupported) was reduced in the 730-900 • C temperature range [24,25]. In addition, a reduction peak associated with surface NiO weak interactions was formed in the temperature range of 500-550 • C, while the reduction peak formed in the temperature range of 700-800 • C was attributed to strong interactions between NiO and the support [26,27]. As bulk NiO has a weak interaction with the support, bulk NiO was mainly reduced at 250-500 • C [28]. A peak corresponding to the reduction of the commercial catalyst A was observed at 300-500 • C. The reduction peak corresponding to the commercial catalyst B was observed at 400-800 • C. Therefore, commercial catalysts were confirmed to have NiO with relatively weak interactions and bulk NiO. As confirmed by the XRD results, the Ni/Al 2 O 3 catalyst, which was difficult to reduce due to the formation of NiAl 2 O 4 , underwent reduction at temperatures above 900 • C. In particular, because the Ni H2-TPR analysis was conducted to analyze the reducibility of the catalysts (Figure 2). Pure NiO (unsupported) can be reduced in the 300-550 °C temperature range, while pure NiAl2O4 (unsupported) was reduced in the 730-900 °C temperature range [24,25]. In addition, a reduction peak associated with surface NiO weak interactions was formed in the temperature range of 500-550 °C, while the reduction peak formed in the temperature range of 700-800 °C was attributed to strong interactions between NiO and the support [26,27]. As bulk NiO has a weak interaction with the support, bulk NiO was mainly reduced at 250-500 °C [28]. A peak corresponding to the reduction of the commercial catalyst A was observed at 300-500 °C. The reduction peak corresponding to the commercial catalyst B was observed at 400-800 °C. Therefore, commercial catalysts were confirmed to have NiO with relatively weak interactions and bulk NiO. As confirmed by the XRD results, the Ni/Al2O3 catalyst, which was difficult to reduce due to the formation of NiAl2O4, underwent reduction at temperatures above 900 °C. In particular, because the Ni  CO2-TPD analysis was conducted to determine the basic properties of the catalysts (Figure 3). In general, the catalytic CO2 adsorption significantly influenced the catalytic activity, as well as the oxygen species formed by the dissociation of the CO2 molecules, and suppressed carbon deposition [16]. The Ni/Al2O3 catalyst employing the acidic Al2O3 support showed low CO2 desorption characteristics of 26 μmol/g ( Table 1). The Ni/MgO catalyst presented a CO2 desorption peak in the 100-700 °C temperature range, and exhibited the highest CO2 desorption amount (170 μmol/g), owing to the basic properties of the MgO support. In addition, because Ni/MgO-Al2O3 and Ni/MgAl2O4 catalysts contained the alkaline earth metal Mg, they exhibited improved CO2 adsorption properties compared to the Ni/Al2O3 catalyst. In particular, Ni/MgAl2O4, which had a high metal dispersion and large specific surface area, exhibited a higher CO2 desorption amount (125 CO 2 -TPD analysis was conducted to determine the basic properties of the catalysts (Figure 3). In general, the catalytic CO 2 adsorption significantly influenced the catalytic activity, as well as the oxygen species formed by the dissociation of the CO 2 molecules, and suppressed carbon deposition [16]. The Ni/Al 2 O 3 catalyst employing the acidic Al 2 O 3 support showed low CO 2 desorption characteristics of 26 µmol/g ( Table 1). The Ni/MgO catalyst presented a CO 2 desorption peak in the 100-700 • C temperature range, and exhibited the highest CO 2 desorption amount (170 µmol/g), owing to the basic properties of the MgO support. In addition, because Ni/MgO-Al 2 O 3 and Ni/MgAl 2 O 4 catalysts contained the alkaline earth metal Mg, they exhibited improved CO 2 adsorption properties compared to the Ni/Al 2 O 3 catalyst. In particular, Ni/MgAl 2 O 4 , which had a high metal dispersion and large specific surface area, exhibited a higher CO 2 desorption amount (125 µmol/g) than the Ni/MgO-Al 2 O 3 catalyst (110 µmol/g). Commercial catalyst A, which employed the acidic Al 2 O 3 support showed poor CO 2 adsorption properties, due to the formation of bulk NiO. The commercial catalyst B presented CO 2 desorption peaks at 200 and 500 • C, and an increased CO 2 desorption amount of 81 µmol/g compared to the catalysts that employed Al 2 O 3 as a support, due to the presence of the alkaline earth metal Ca.

Catalytic Test in DRM
The catalytic activity of the DRM reaction was evaluated in the temperature range of 600-750 °C, and the CH4 and CO2 conversions are presented in Figure 4. Since the DRM is an endothermic reaction, it exhibited high CH4 and CO2 conversions with increasing reaction temperature. The Ni/MgO catalyst, which formed a NiO-MgO solid solution, was reduced at temperatures above 1000 °C; therefore, the Ni/MgO catalyst did not show catalytic activity in the DRM reaction. Since the Ni/Al2O3 catalyst also formed a NiAl2O4 spinel, this resulted in low metal dispersion and low reducibility of the catalyst. For this reason, the Ni/Al2O3 catalyst showed a low CH4 conversion (66%) at 750 °C, whereas, Ni/MgO-Al2O3 and Ni/MgAl2O4 catalysts exhibited high CH4 conversions of 78 and 77%, respectively, compared with Ni/Al2O3. This is because of the high dispersion of the Ni metals due to the suppression of the formed NiAl2O4 and NiO-MgO solid solutions. On the other hand, commercial catalysts A and B showed low CH4 conversions, due to the formation of bulk NiO, despite the presence of 15 wt% Ni metal. However, because the alkaline earth metal-containing commercial catalyst B formed a lower amount of NiAl2O4 compared with the commercial catalyst A, the catalyst B presented relatively high CH4 and CO2 conversions. This result indicates that the catalytic characteristics (such as metal dispersion and crystalline phase), which depend on the type of support employed, played an important role in determining the catalytic activity.

Catalytic Test in DRM
The catalytic activity of the DRM reaction was evaluated in the temperature range of 600-750 • C, and the CH 4 and CO 2 conversions are presented in Figure 4. Since the DRM is an endothermic reaction, it exhibited high CH 4 and CO 2 conversions with increasing reaction temperature. The Ni/MgO catalyst, which formed a NiO-MgO solid solution, was reduced at temperatures above 1000 • C; therefore, the Ni/MgO catalyst did not show catalytic activity in the DRM reaction. Since the Ni/Al 2 O 3 catalyst also formed a NiAl 2 O 4 spinel, this resulted in low metal dispersion and low reducibility of the catalyst. For this reason, the Ni/Al 2 O 3 catalyst showed a low CH 4 conversion (66%) at 750 • C, whereas, Ni/MgO-Al 2 O 3 and Ni/MgAl 2 O 4 catalysts exhibited high CH 4 conversions of 78 and 77%, respectively, compared with Ni/Al 2 O 3 . This is because of the high dispersion of the Ni metals due to the suppression of the formed NiAl 2 O 4 and NiO-MgO solid solutions. On the other hand, commercial catalysts A and B showed low CH 4 conversions, due to the formation of bulk NiO, despite the presence of 15 wt% Ni metal. However, because the alkaline earth metal-containing commercial catalyst B formed a lower amount of NiAl 2 O 4 compared with the commercial catalyst A, the catalyst B presented relatively high CH 4 and CO 2 conversions. This result indicates that the catalytic characteristics (such as metal dispersion and crystalline phase), which depend on the type of support employed, played an important role in determining the catalytic activity.
H 2 and CO yields and H 2 /CO ratio are presented in Figure 4 and Table 2, respectively. The CO yield (56-86%) was higher than the H 2 yield (34-74%). In addition, the H 2 /CO ratio of the produced syngas after the DRM reaction was 0.6-0.9. These results are attributed to the RWGS reaction (Equation (2)) as a side reaction, as mentioned above. Theoretically, the ratio of the H 2 /CO produced in the DRM reaction (Equation (1)) is 1. However, the H 2 /CO ratio was lower than 1, because the RWGS reaction occurs as a side reaction under the reaction conditions of the DRM process [29]. Among the tested catalysts, Ni/MgO-Al 2 O 3 and Ni/MgAl 2 O 4 presented ratios close to the equilibrium H 2 /CO ratio of 0.9.   H2 and CO yields and H2/CO ratio are presented in Figure 4 and Table 2, respectively. The CO yield (56-86%) was higher than the H2 yield (34-74%). In addition, the H2/CO ratio of the produced syngas after the DRM reaction was 0.6-0.9. These results are attributed to the RWGS reaction (Equation (2)) as a side reaction, as mentioned above. Theoretically, the ratio of the H2/CO produced in the DRM reaction (Equation (1)) is 1. However, the H2/CO ratio was lower than 1, because the RWGS reaction occurs as a side reaction under the reaction conditions of the DRM process [29]. Among the tested catalysts, Ni/MgO-Al2O3 and Ni/MgAl2O4 presented ratios close to the equilibrium H2/CO ratio of 0.9.

Coke Formation Study
To investigate the stability of the reaction and coke resistance, a stability test was conducted in the absence of the diluent ( Figure 5). Coke formation reactions of the CH 4 decomposition reaction (Equation (3)) and CO disproportionation reaction (Equation (4)) occur simultaneously in the temperature range of 557-700 • C [10]. Thus, stability tests were performed at 600 • C and a TOS of 20 h. As the Ni/MgO catalyst did not exhibit catalytic activity in the catalytic test, in accordance with the reaction temperature, a stability test was conducted, except for the Ni were performed at 600 °C and a TOS of 20 h. As the Ni/MgO catalyst did not exhibit catalytic activity in the catalytic test, in accordance with the reaction temperature, a stability test was conducted, except for the Ni/MgO catalyst. The Ni/Al2O3 catalyst tended toward constantly reduced CH4 and CO2 conversions over 20 h. The Ni/Al2O3 catalyst had 21% of the initial CH4 conversion, the Ni/Al2O3 catalyst decreased to 16% of CH4 conversion after 20 h. Ni/MgO-Al2O3 and Ni/MgAl2O4 catalysts, which added the Mg alkaline earth metal, had high CH4 and CO2 conversions compared to the Ni/Al2O3 catalyst, owing to the improved reducibility of the catalyst and high metal dispersion. In particular, the Ni/MgO-Al2O3 and Ni/MgAl2O4 catalysts showed higher CH4 and CO2 conversions than the commercial catalyst A. On the other hand, the commercial catalyst B exhibited a CH4 conversion of 20-22%, and had a higher CH4 conversion than the catalyst that employed Al2O3 as a support. To investigate the carbon formed on the surface of the catalyst, SEM and TGA analyses were performed on the recovered spent catalysts. The morphologies of the formed carbon were analyzed using SEM images, as shown in Figure 6. Sehested reported that three types of carbon were formed during reforming reactions: pyrolytic carbon, encapsulating carbon, and filamentous carbon [30]. Pyrolytic carbon is formed at temperatures higher than 600 °C, owing to thermal decomposition of hydrocarbon compounds. Since the encapsulating carbon covered particles of the catalyst, it caused immediate catalyst deactivation. Filamentous carbon is formed and grown between the Ni particles and the support, which causes catalyst pore plugging. For this reason, filamentous carbon results in catalyst deactivation, causing catalyst destruction and a pressure drop. The formation of filamentous carbon was also confirmed on the surface of the spent catalysts after the stability test, and SEM images of all the analyzed catalysts indicated the formation of a large amount of filamentous carbon on the catalyst surface. To investigate the carbon formed on the surface of the catalyst, SEM and TGA analyses were performed on the recovered spent catalysts. The morphologies of the formed carbon were analyzed using SEM images, as shown in Figure 6. Sehested reported that three types of carbon were formed during reforming reactions: pyrolytic carbon, encapsulating carbon, and filamentous carbon [30]. Pyrolytic carbon is formed at temperatures higher than 600 • C, owing to thermal decomposition of hydrocarbon compounds. Since the encapsulating carbon covered particles of the catalyst, it caused immediate catalyst deactivation. Filamentous carbon is formed and grown between the Ni particles and the support, which causes catalyst pore plugging. For this reason, filamentous carbon results in catalyst deactivation, causing catalyst destruction and a pressure drop. The formation of filamentous carbon was also confirmed on the surface of the spent catalysts after the stability test, and SEM images of all the analyzed catalysts indicated the formation of a large amount of filamentous carbon on the catalyst surface.
Quantitative analysis of the amount of deposited carbon was performed through weight loss analysis using TGA up to 800 • C. The TGA results are shown in Figure 7 and Table 3. The weight loss of the spent catalyst was classified according to the temperature range; desorption of H 2 O and CO 2 occurred in the temperature range of 250-300 • C, and the oxidation of carbon due to the release of CO and CO 2 occurred in the temperature range of 480-640 • C [31]. In addition, graphitic filamentous carbon is mainly oxidized at a high temperature of 530 • C [32]. Quantitative analysis of the formed carbon was performed using the recovered spent catalysts containing alkaline earth metals, which showed a relatively high catalytic activity during the stability test. Results from the TG profiles indicated that most of the weight loss of the spent catalysts occurred at temperatures above 500 • C. The amount of deposited carbon was calculated from the weight loss, in the temperature range of 50-800 • C ( Table 3). The amount of deposited carbon on the Ni/MgO-Al 2 O 3 and Ni/MgAl 2 O 4 catalysts was 21.4 and 29.3 wt%, respectively. In comparison, the commercial catalyst B contained twice the amount of deposited carbon (47.1 wt%). These results are related to the CO 2 adsorption characteristics of the catalysts [33]. Since CO 2 is an acidic gas, CO 2 adsorption was enhanced by the addition of basic materials, such as Mg and Ca.
In addition, oxygen species produced by the dissociation of CO 2 molecules are used in the oxidation of deposited carbon on the catalyst surface. Therefore, increasing the basic properties of the DRM catalyst could improve catalyst durability. Therefore, commercial catalyst B, which showed relatively low CO 2 adsorption properties (81 µmol/g), contained a large amount of carbon compared to the Ni/MgO-Al 2 O 3 and Ni/MgAl 2 O 4 catalysts, which had high CO 2 adsorption properties, of 110 and 125 µmol/g, respectively.  ecules are used in the oxidation of deposited carbon on the catalyst surface. Therefore, increasing the basic properties of the DRM catalyst could improve catalyst durability. Therefore, commercial catalyst B, which showed relatively low CO2 adsorption properties (81 μmol/g), contained a large amount of carbon compared to the Ni/MgO-Al2O3 and Ni/MgAl2O4 catalysts, which had high CO2 adsorption properties, of 110 and 125 μmol/g, respectively.    analysis was performed to investigate the reducibility of the catalyst and the interaction between the active metal and support. The analysis was performed by increasing the temperature up to 1000 • C (10 • C/min) under 10% H 2 /Ar flow using a 50 mg sample. The CO 2 adsorption properties were analyzed through CO 2 -temperature programmed desorption (CO 2 -TPD, BELCAT-B, MicrotracBEL Corp., Osaka, Japan) analysis. The sample (50 mg) was reduced at 700 • C for 2 h under 10% H 2 /Ar flow. The CO 2 was adsorbed at 50 • C using 10% CO 2 /He gas and desorbed by increasing the temperature to 800 • C (10 • C/min). In addition, the morphology of the formed carbon in the spent catalysts was analyzed using scanning electron microscope (SEM, Regulus 8220, HITACHI, Tokyo, Japan) analysis. Quantitative analysis of carbon on the spent catalysts was performed using thermogravimetric analysis (TGA, TGA 92-18, Setaram, Caluire-et-Cuire, France). The weight loss was measured by loading 7 mg of the sample. The sample was oxidized by increasing the temperature to 800 • C at 10 • C/min in air.

Catalytic Tests
Catalytic activity was evaluated in a fixed bed reactor system. The 3/8 quartz reactor was filled with a catalyst (50 mg, 150-250 µm) and 1100 • C pre-calcined MgAl 2 O 4 as a diluent (500 mg, 150-250 µm) in a ratio of 1:10. Before the catalytic test, the catalyst was reduced at 700 • C for 2 h with 10 vol.% H 2 /N 2 gas. The evaluation of the catalytic activity was performed in a temperature range of 600-750 • C. The feeds of CH 4 , CO 2 , and N 2 were injected to equal ratio. The total flow rate was 120 mL/min, and GHSV corresponded to 144,000 mL/g cat ·h. The reaction temperature was controlled using a K-type thermocouple (TC), the TC was located inside the catalyst layer. The produced gas was detected using a micro-gas chromatograph (3000A Micro GC, Agilent, CA, USA) equipped with a thermal conductivity detector (TCD) using MolSeive 5A and PlotU columns.

Conclusions
DRM reaction over Ni-based catalysts was studied as a function of the type of catalyst support used (Al 2 O 3 , MgO, MgO-Al 2 O 3 (LDH), and MgAl 2 O 4 (spinel)). The type of support significantly influenced the reducibility and metal dispersion of the Ni-based catalysts. The Ni/Al 2 O 3 and Ni/MgO catalysts had low metal dispersion, due to the formation of NiAl 2 O 4 and NiO-MgO solid solutions, respectively. In addition, these catalysts had a strong interaction between the Ni active metal and support, indicated by the high temperature (more than 900 • C) required for reduction. In contrast, Ni catalysts showed enhanced coke resistance compared to commercial catalyst B. Therefore, the metal dispersion, basic properties, and metal-support interaction of the catalyst in accordance with the type of support were confirmed to significantly influence the catalyst performance in the DRM reaction.