Improving the Stability of Ru-Doped Ni-Based Catalysts for Steam Methane Reforming during Daily Startup and Shutdown Operation

: In this study, a Ru-doped Ni pellet-type catalyst was prepared to produce hydrogen via steam methane reforming (SMR). A small amount of Ru addition on the Ni catalyst improved Ni dispersion, thus affording a higher catalytic activity than that of the Ni catalyst. During the daily startup and shutdown (DSS) operations, the CH 4 conversion of Ni catalysts signiﬁcantly decreased because of Ni metal oxidation to NiAl 2 O 4 , which is not reduced completely at 700 ◦ C. Conversely, the oxidized Ni species in the Ru–Ni catalyst can be reduced under SMR conditions because of H 2 spillover from the surface of Ru onto the surface of Ni. Consequently, the addition of a small quantity of Ru to the Ni catalyst can improve the catalytic activity and stability during the DSS operation.


Introduction
Hydrogen is generally accepted as an attractive alternative to support energy consumption while reducing the environmental impact. Among the energy-generating devices using hydrogen, proton-exchange membrane fuel cells (PEMFC) are appealing devices for residential power generation applications due to their high efficiency and low emission tendency. The reforming of methane is the most common method to produce hydrogen. There are several types of methane reforming, such as steam reforming (SR), dry reforming (DR) and partial oxidation (POx). Among the methane reforming technologies, steam methane reforming (SMR) (Equation (1)) is the most widely used process to produce hydrogen [1][2][3][4][5][6]. SMR has a high hydrogen yield efficiency (−74%), yielding 4 mol H 2 and 1 mol CO 2 for each methane group (Equations (1) and (2)).
CO + H 2 O → CO 2 + H 2 , ∆H 298k = −41 kJ mol −1 The various noble metal and transition metal-based catalysts for the SMR reaction have been reported [7][8][9][10][11][12][13][14][15][16][17][18][19][20][21][22]. Although noble metals (e.g., Ru, Rh and Pt) show excellent catalytic activity and high coke resistance [7,8], they are too expensive to be applied in Figure 1 shows external surface and cross-sectional images of Ni/Al 2 O 3 and Ru-Ni/Al 2 O 3 catalysts using a digital microscope. The external surface of the Ni/Al 2 O 3 catalyst exhibited a light blue color, whereas a black color was observed in Ru-Ni/Al 2 O 3 . The cross-sectional images of both catalysts exhibited a light blue color. The black color was only detected on the outermost surface of Ru-Ni/Al 2 O 3 . It can be confirmed through the distribution of nickel and ruthenium in the cross-section of both catalysts from the edge to the center, measured via SEM-EDS ( Figure S1). For the Ni/Al 2 O 3 catalyst, a very high concentration of Ni was detected on the external catalyst surface, and the nickel concentration gradually decreased inside the catalyst ( Figure S1a). As shown in Figure S1b, a similar trend was also observed for the Ru-Ni/Al 2 O 3 catalyst, but the Ni concentration inside the catalyst was slightly higher than that of the Ni catalyst. It seems that the Ni particles on the external surface of pellets penetrate inside the pellet in the second impregnation step. Ru in the Ru-Ni/Al 2 O 3 catalyst was supported only on the external pellet surface due to strong Ru precursor-support interaction, as seen in the cross-sectional image.
Catalysts 2023, 13, x FOR PEER REVIEW 3 of 12 Recently, we reported that the Ru-based eggshell-type catalysts exhibited high activity and stability during the DSS operation [28]. Nevertheless, several problems exist with Ru-based catalysts, such as cost and availability on an industrial scale. In this study, we developed Ru-doped Ni catalysts on pellet-type alumina to investigate activity and stability during the DSS operation. The effects of Ru addition on SMR activity and stability during the DSS operation were investigated. The catalysts were analyzed using scanning electron microscopy (SEM) with energy-dispersive X-ray spectroscopy (EDS), a high-resolution 2D X-ray diffractometer (HR-2D XRD), temperature-programmed reduction with H2 (H2-TPR), Brunauer-Emmett-Teller (BET) analysis, X-ray photoelectron spectroscopy (XPS), and thermogravimetric analysis with H2 (H2-TGA). Figure 1 shows external surface and cross-sectional images of Ni/Al2O3 and Ru-Ni/Al2O3 catalysts using a digital microscope. The external surface of the Ni/Al2O3 catalyst exhibited a light blue color, whereas a black color was observed in Ru-Ni/Al2O3. The cross-sectional images of both catalysts exhibited a light blue color. The black color was only detected on the outermost surface of Ru-Ni/Al2O3. It can be confirmed through the distribution of nickel and ruthenium in the cross-section of both catalysts from the edge to the center, measured via SEM-EDS ( Figure S1). For the Ni/Al2O3 catalyst, a very high concentration of Ni was detected on the external catalyst surface, and the nickel concentration gradually decreased inside the catalyst ( Figure S1a). As shown in Figure S1b, a similar trend was also observed for the Ru-Ni/Al2O3 catalyst, but the Ni concentration inside the catalyst was slightly higher than that of the Ni catalyst. It seems that the Ni particles on the external surface of pellets penetrate inside the pellet in the second impregnation step. Ru in the Ru-Ni/Al2O3 catalyst was supported only on the external pellet surface due to strong Ru precursor-support interaction, as seen in the cross-sectional image. High-resolution two-dimensional X-ray diffraction (HR-2D XRD) was used to study the structural properties on the surface of Ni/Al2O3 and Ru-Ni/Al2O3 catalysts in fresh and reduced states ( Figure 2). Figure 2a shows the XRD patterns of Ni/Al2O3 catalysts in the fresh and reduced states. For the Ni/Al2O3 catalyst, peaks of NiAl2O4 (JCPDS No. 10-0339) and γ-Al2O3 (JCPDS No. 79-1558) were observed in the fresh state without peaks of NiO due to high calcination conditions. After the H2 reduction treatment at 800 °C for 3 h, peaks of metallic Ni (JCPDS No. 70-1849) were observed. Additionally, the overlapped peaks of NiAl2O4 and γ-Al2O3 shifted to peaks of γ-Al2O3, indicating that the NiAl2O4 phase was reduced to the metallic Ni phase. For the Ni-Ru/Al2O3 catalyst (Figure 2b), peaks of RuO2 (JCPDS No. 88-0286) existed with peaks of NiAl2O4 and γ-Al2O3. After the High-resolution two-dimensional X-ray diffraction (HR-2D XRD) was used to study the structural properties on the surface of Ni/Al 2 O 3 and Ru-Ni/Al 2 O 3 catalysts in fresh and reduced states ( Figure 2). Figure 2a shows   The H2-TPR profiles of the Ni-based catalysts were obtained under 10% H2 at a temperature range of 100 °C to 900 °C with a temperature ramp rate of 10 °C min −1 after pretreatment with N2 at 200 °C for 1 h to desorb the adsorbed gases ( Figure 3). The H2-TPR profiles showed no peaks at temperatures below 500 °C, only peaks at approximately 800 °C due to the reduction of NiAl2O4: NiAl2O4(s) + H2(g) ⟷ Ni(s) + Al2O3(s) + H2O(g) [18]. Notably, the reduction temperature of NiAl2O4 shifted to lower temperatures in the presence of Ru. Jeong et al. reported that the formation of a Ru-Ni alloy resulted in an easy reduction of Ni on the Ru-Ni/Al2O3 catalyst [23]. Ni 2+ particles were reduced by the hydrogen spillover from metallic Ru to form a Ru-Ni alloy on the surface of the metallic Ni particles, indicating that Ru not only increased the dispersion of Ni but also improved its reducibility. The H 2 -TPR profiles of the Ni-based catalysts were obtained under 10% H 2 at a temperature range of 100 • C to 900 • C with a temperature ramp rate of 10 • C min −1 after pretreatment with N 2 at 200 • C for 1 h to desorb the adsorbed gases ( Figure 3). The H 2 -TPR profiles showed no peaks at temperatures below 500 • C, only peaks at approximately 800 • C due to the reduction of NiAl 2 O 4 : NiAl 2 O 4 (s) + H 2 (g) ←→ Ni(s) + Al 2 O 3 (s) + H 2 O(g) [18]. Notably, the reduction temperature of NiAl 2 O 4 shifted to lower temperatures in the presence of Ru. Jeong et al. reported that the formation of a Ru-Ni alloy resulted in an easy reduction of Ni on the Ru-Ni/Al 2 O 3 catalyst [23]. Ni 2+ particles were reduced by the hydrogen spillover from metallic Ru to form a Ru-Ni alloy on the surface of the metallic Ni particles, indicating that Ru not only increased the dispersion of Ni but also improved its reducibility. Intensity (a.u.) The H2-TPR profiles of the Ni-based catalysts were obtained under 10% H2 at a temperature range of 100 °C to 900 °C with a temperature ramp rate of 10 °C min −1 after pretreatment with N2 at 200 °C for 1 h to desorb the adsorbed gases ( Figure 3). The H2-TPR profiles showed no peaks at temperatures below 500 °C, only peaks at approximately 800 °C due to the reduction of NiAl2O4: NiAl2O4(s) + H2(g) ⟷ Ni(s) + Al2O3(s) + H2O(g) [18]. Notably, the reduction temperature of NiAl2O4 shifted to lower temperatures in the presence of Ru. Jeong et al. reported that the formation of a Ru-Ni alloy resulted in an easy reduction of Ni on the Ru-Ni/Al2O3 catalyst [23]. Ni 2+ particles were reduced by the hydrogen spillover from metallic Ru to form a Ru-Ni alloy on the surface of the metallic Ni particles, indicating that Ru not only increased the dispersion of Ni but also improved its reducibility.   4 conversions of each catalyst changed dramatically as the WHSV increased from 12,000 mL/g/h to 48,000 mL/g/h. As the WHSV increased, the CH 4 conversion of both catalysts decreased due to the reduction in contact time between the reactant and active site. At low WHSV, the CH 4 conversion of both catalysts was comparable regardless of Ru addition. However, at the highest WHSV (48,000 mL/g/h) under our experimental conditions, the Ru-Ni/Al 2 O 3 catalyst showed about 6% higher CH 4 conversion than that of the Ni/Al 2 O 3 catalyst. Considering the very low CH 4 conversion over Ru/Al 2 O 3 catalyst (0.1 wt% Ru), the increase in Ni dispersion by Ru addition facilitated the formation of CH x − * and O − *, resulting in an improvement in CH 4 conversion [31]. As shown in Figure  S2, at the WHSV (12,000 mL/g/h), the Ni/Al 2 O 3 catalyst showed the selectivity of H 2 , CO and CO 2 to be 78%, 13% and 9%, respectively. Additionally, the selectivity of Ru-Ni/Al 2 O 3 was almost identical to that of the Ni/Al 2 O 3 catalyst. It might be that adding Ru to Ni/Al 2 O 3 does not affect selectivity during the SMR reaction. As the WHSV increased, the CO 2 selectivity of both catalysts slightly increased because the lower CH 4 conversion increased the H 2 O/CO ratio, which caused the WGS reaction [28].

Characterization of the Ni-Based Catalysts
Catalysts 2023, 13, x FOR PEER REVIEW 5 of 12

Steam Reforming Reaction with Ni-Based Catalysts
SMR reactions were performed on both catalysts at various WHSVs to investigate the effect of Ru addition on the catalytic activity. Figure 4 and Figure S2 show the CH4 conversion and selectivity of the Ni/Al2O3 and Ru-Ni/Al2O3 catalysts at various WHSVs. The CH4 conversions of each catalyst changed dramatically as the WHSV increased from 12,000 mL/g/h to 48,000 mL/g/h. As the WHSV increased, the CH4 conversion of both catalysts decreased due to the reduction in contact time between the reactant and active site. At low WHSV, the CH4 conversion of both catalysts was comparable regardless of Ru addition. However, at the highest WHSV (48,000 mL/g/h) under our experimental conditions, the Ru-Ni/Al2O3 catalyst showed about 6% higher CH4 conversion than that of the Ni/Al2O3 catalyst. Considering the very low CH4 conversion over Ru/Al2O3 catalyst (0.1 wt% Ru), the increase in Ni dispersion by Ru addition facilitated the formation of CHx − * and O − *, resulting in an improvement in CH4 conversion [31]. As shown in Figure S2, at the WHSV (12,000 mL/g/h), the Ni/Al2O3 catalyst showed the selectivity of H2, CO and CO2 to be 78%, 13% and 9%, respectively. Additionally, the selectivity of Ru-Ni/Al2O3 was almost identical to that of the Ni/Al2O3 catalyst. It might be that adding Ru to Ni/Al2O3 does not affect selectivity during the SMR reaction. As the WHSV increased, the CO2 selectivity of both catalysts slightly increased because the lower CH4 conversion increased the H2O/CO ratio, which caused the WGS reaction [28].  The DSS operations were performed to investigate the long-term stability of Ni/Al2O3 and Ru-Ni/Al2O3 catalysts for the application of PEMFC. After the pre-reduction, the reaction started in a CH4/H2O/N2 (10/30/60 mL/min) for 90 min at 700 °C, then the temperature was cooled to 200 °C under steam purging with H2O/N2 (30/60 mL/min), in which only the CH4 supply was stopped under reaction conditions. After the reactor temperature was maintained at 200 °C for 30 min, it was then heated to 700 °C under purge conditions. CH4 gas was added to start the reaction; the reaction was held for 90 min at 700 °C. To investigate the long-term stability of the DSS operation, the SMR reaction and cooling were repeated. Figure 5 shows the CH4 conversion and selectivity of catalysts during the DSS operation. At the first SMR reaction, the Ni/Al2O3 catalyst showed 99% CH4 conversion, and the selectivity of H2, CO and CO2 was 77%, 13% and 9%, respectively. Ru-Ni/Al2O3 showed similar values of CH4 conversion and selectivity compared to Ni/Al2O3 The DSS operations were performed to investigate the long-term stability of Ni/Al 2 O 3 and Ru-Ni/Al 2 O 3 catalysts for the application of PEMFC. After the pre-reduction, the reaction started in a CH 4 /H 2 O/N 2 (10/30/60 mL/min) for 90 min at 700 • C, then the temperature was cooled to 200 • C under steam purging with H 2 O/N 2 (30/60 mL/min), in which only the CH 4 supply was stopped under reaction conditions. After the reactor temperature was maintained at 200 • C for 30 min, it was then heated to 700 • C under purge conditions. CH 4 gas was added to start the reaction; the reaction was held for 90 min at 700 • C. To investigate the long-term stability of the DSS operation, the SMR reaction and cooling were repeated. Figure 5 shows the CH 4 conversion and selectivity of catalysts during the DSS operation. At the first SMR reaction, the Ni/Al 2 O 3 catalyst showed 99% CH 4 conversion, and the selectivity of H 2 , CO and CO 2 was 77%, 13% and 9%, respectively. Ru-Ni/Al 2 O 3 showed similar values of CH 4 conversion and selectivity compared to Ni/Al 2 O 3 catalysts. The Ni/Al 2 O 3 catalyst showed a slight decrease in CH 4 conversion after the first DSS operation and significant deactivation after three cycles. In addition, the selectivity of CO and CO 2 changed drastically during DSS operation. The CO selectivity tended to decrease and the CO 2 selectivity to increase because the low CH 4 conversion induced the WGS reaction as the H 2 O/CO ratio increased. Conversely, the CH 4 conversion and selectivity of the Ru-Ni catalyst were maintained during DSS operation. Considering that Ru/Al 2 O 3 showed very poor activity and deactivation ( Figure S3), a small loading of Ru in Ru-Ni/Al 2 O 3 as a promoter could prevent deactivation during the steam purge. As shown in Table 1, in a previous report, the 2 wt%Ru/Al 2 O 3 showed a high CH 4 conversion of 99.0% with stability during DSS operation [28]. Compared to 2%Ru/Al 2 O 3 , the Ru-Ni/Al 2 O 3 catalyst (0.1 wt% Ru/10 wt%) also exhibited high CH 4 conversion and stability for SMR during DSS operation. High amounts of H 2 production with cyclic stability can be achieved over Ru-Ni/Al 2 O 3 catalysts in DSS operation during SMR by reducing the amount of Ru by 95% (2 to 0.1 wt%). Considering the price of Ru is much higher than Ni, the amount of Ru is the key factor in determining the production costs of catalysts. Therefore, the reduction of the Ru loading amount in the catalyst is expected to improve economic feasibility.

WHSV (ml/g/h)
Catalysts 2023, 13, x FOR PEER REVIEW 6 of 12 catalysts. The Ni/Al2O3 catalyst showed a slight decrease in CH4 conversion after the first DSS operation and significant deactivation after three cycles. In addition, the selectivity of CO and CO2 changed drastically during DSS operation. The CO selectivity tended to decrease and the CO2 selectivity to increase because the low CH4 conversion induced the WGS reaction as the H2O/CO ratio increased. Conversely, the CH4 conversion and selectivity of the Ru-Ni catalyst were maintained during DSS operation. Considering that Ru/Al2O3 showed very poor activity and deactivation ( Figure S3), a small loading of Ru in Ru-Ni/Al2O3 as a promoter could prevent deactivation during the steam purge. As shown in Table 1, in a previous report, the 2wt%Ru/Al2O3 showed a high CH4 conversion of 99.0% with stability during DSS operation [28]. Compared to 2%Ru/Al2O3, the Ru-Ni/Al2O3 catalyst (0.1 wt% Ru/10wt%) also exhibited high CH4 conversion and stability for SMR during DSS operation. High amounts of H2 production with cyclic stability can be achieved over Ru-Ni/Al2O3 catalysts in DSS operation during SMR by reducing the amount of Ru by 95% (2 to 0.1 wt%). Considering the price of Ru is much higher than Ni, the amount of Ru is the key factor in determining the production costs of catalysts. Therefore, the reduction of the Ru loading amount in the catalyst is expected to improve economic feasibility.  Both catalysts were characterized using N2 adsorption−desorption instrumentation, H2-TGA, 2D-XRD and XPS to investigate the reasons for the difference in catalyst deactivation during DSS operation. The N2 adsorption−desorption isotherms and the pore size distributions of catalysts are shown in Figure S4 and S5, respectively. Both catalysts exhibited a type IV isotherm with a hysteresis loop, which is associated with mesoporous structures. After the DSS operation, the pore size distribution of catalysts gradually  Both catalysts were characterized using N 2 adsorption−desorption instrumentation, H 2 -TGA, 2D-XRD and XPS to investigate the reasons for the difference in catalyst deactivation during DSS operation. The N 2 adsorption−desorption isotherms and the pore size distributions of catalysts are shown in Figures S4 and S5, respectively. Both catalysts exhibited a type IV isotherm with a hysteresis loop, which is associated with mesoporous structures. After the DSS operation, the pore size distribution of catalysts gradually shifted toward larger pores. This change may suggest that catalysts suffered agglomeration and migration during DSS operation. Moreover, the textural properties of the catalysts are listed in Table 2. The BET surface area, pore volume and pore size of both catalysts indicated similar decreasing trends during DSS operation. It is known that the reduction of textural properties such as BET surface area and volume leads to a decrease in catalytic activity.
Despite the decrease in textural properties of both catalysts, the Ru-Ni/Al 2 O 3 catalyst showed stable activity during DSS operation. Therefore, given this trend, it is considered that other factors probably have a greater impact on catalyst deactivation. High-resolution two-dimensional X-ray diffraction (HR-2D XRD) was analyzed after the DSS operation cycle (Figure 6a,b). For the Ni/Al 2 O 3 catalyst, peaks of metallic Ni, NiO, NiAl 2 O 3 and γ-Al 2 O 3 were observed after the 1 st DDS, indicating oxidation of metallic Ni under the steam purge step. The peaks of metallic Ni remained after the 2nd SMR reaction. However, peaks of metallic Ni were not observed after the 5th DSS and 6th reactions. It suggested that the DSS operation step could gradually oxidize the metallic Ni to NiO on the Ni/Al 2 O 3 catalyst, and NiO was not reduced under SRM conditions. Therefore, a decrease in CH 4 conversion on the Ni catalyst seems to be due to Ni oxidation during DSS operation and the poor reducibility of Ni species under SRM conditions. On the other hand, for the Ni-Ru/Al 2 O 3 catalyst, peaks of metallic Ni, metallic Ru, NiAl 2 O 3 and γ-Al 2 O 3 were observed without peaks of NiO after the 1DDS. The sharp peak of metallic Ni was still observed after the 2nd reaction. After the 5th DSS, the peaks of NiO on the Ru-Ni/Al 2 O 3 catalyst were detected, indicating metallic Ni was oxidized because of repeated exposure to steam during DSS operation. However, it was observed that NiO can be reduced to metallic Ni during the 6th SMR reaction.
shifted toward larger pores. This change may suggest that catalysts suffered agglomeration and migration during DSS operation. Moreover, the textural properties of the catalysts are listed in Table 2. The BET surface area, pore volume and pore size of both catalysts indicated similar decreasing trends during DSS operation. It is known that the reduction of textural properties such as BET surface area and volume leads to a decrease in catalytic activity. Despite the decrease in textural properties of both catalysts, the Ru-Ni/Al2O3 catalyst showed stable activity during DSS operation. Therefore, given this trend, it is considered that other factors probably have a greater impact on catalyst deactivation. High-resolution two-dimensional X-ray diffraction (HR-2D XRD) was analyzed after the DSS operation cycle (Figure 6a,b). For the Ni/Al2O3 catalyst, peaks of metallic Ni, NiO, NiAl2O3 and γ-Al2O3 were observed after the 1 st DDS, indicating oxidation of metallic Ni under the steam purge step. The peaks of metallic Ni remained after the 2nd SMR reaction. However, peaks of metallic Ni were not observed after the 5th DSS and 6th reactions. It suggested that the DSS operation step could gradually oxidize the metallic Ni to NiO on the Ni/Al2O3 catalyst, and NiO was not reduced under SRM conditions. Therefore, a decrease in CH4 conversion on the Ni catalyst seems to be due to Ni oxidation during DSS operation and the poor reducibility of Ni species under SRM conditions. On the other hand, for the Ni-Ru/Al2O3 catalyst, peaks of metallic Ni, metallic Ru, NiAl2O3 and γ-Al2O3 were observed without peaks of NiO after the 1DDS. The sharp peak of metallic Ni was still observed after the 2nd reaction. After the 5th DSS, the peaks of NiO on the Ru-Ni/Al2O3 catalyst were detected, indicating metallic Ni was oxidized because of repeated exposure to steam during DSS operation. However, it was observed that NiO can be reduced to metallic Ni during the 6th SMR reaction.   Figure 7 shows the Ni 2p 3/2 spectra of Ni/Al 2 O 3 and Ru-Ni/Al 2 O 3 catalysts after reduction and DSS operation. For the reduced catalysts, the three peaks of the Ni 2p 3/2 spectra ranging from 873 to 847 eV were observed. The first peak at 852.7 eV corresponds to the Ni 0 , the peak at 855.5 eV corresponds to the Ni 2+ , and the last peak at 861.5 eV corresponds to the satellite peak [32]. For the Ru-Ni/Al 2 O 3 catalysts, the peak intensity of Ni 0 was much higher compared to Ni/Al 2 O 3 , due to the better reducibility. After the 5th DSS, the peak of Ni 0 was not observed on the Ni/Al 2 O 3 catalyst, whereas the Ni 0 peak was still observed on the Ru-Ni/Al 2 O 3 catalyst, suggesting that the addition of Ru to the Ni/Al 2 O 3 catalyst could suppress the oxidation of the Ni species after continuous DSS operation.
XPS analyses of Ni/Al2O3 and Ru-Ni/Al2O3 catalysts were conducted to determine the valence states of the Ni species after DSS operation. Figure 7 shows the Ni 2p3/2 spectra of Ni/Al2O3 and Ru-Ni/Al2O3 catalysts after reduction and DSS operation. For the reduced catalysts, the three peaks of the Ni 2p3/2 spectra ranging from 873 to 847 eV were observed. The first peak at 852.7 eV corresponds to the Ni 0 , the peak at 855.5 eV corresponds to the Ni 2+ , and the last peak at 861.5 eV corresponds to the satellite peak [32]. For the Ru-Ni/Al2O3 catalysts, the peak intensity of Ni 0 was much higher compared to Ni/Al2O3, due to the better reducibility. After the 5th DSS, the peak of Ni 0 was not observed on the Ni/Al2O3 catalyst, whereas the Ni 0 peak was still observed on the Ru-Ni/Al2O3 catalyst, suggesting that the addition of Ru to the Ni/Al2O3 catalyst could suppress the oxidation of the Ni species after continuous DSS operation.  In addition, H 2 -TGA was conducted to investigate the differences in the reducibility of both catalysts. Firstly, the fresh and spent catalysts were heated to 800 • C and 700 • C, respectively, under N 2 gas. Then, while maintaining the temperature, the N 2 gas was changed to H 2 gas to reduce the catalysts for 2 h. As shown in Table 2, fresh Ni/Al 2 O 3 and Ru-Ni/Al 2 O 3 catalysts showed H 2 consumption amounts of 1.515 and 1.578 mmol/g, respectively, which are similar to the theoretical maximum value in the 10 wt% Ni catalyst (1.704 mmol/g). After the 1st DDS, Ni/Al 2 O 3 and Ru-Ni/Al 2 O 3 catalysts exhibited low H 2 consumption amounts of 0.189 and 0.252 mmol/g, respectively, compared to fresh samples. It might be because the metallic Ni on catalysts was not oxidized significantly in the first steam purging step. For the Ni/Al 2 O 3 catalyst, H 2 consumption after the 5th DSS was identical to that after the 1st DSS, regardless of the Ni oxidation state. On the other hand, the Ru-Ni/Al 2 O 3 catalyst exhibited a relatively high amount of H 2 consumption (0.884 mmol/g) to reduce NiO formed during continuous DSS operation, compared to the Ni/Al 2 O 3 catalyst. As mentioned in H 2 -TPR, Ru further assisted the Ni reduction in the Ru-Ni/Al 2 O 3 catalyst due to easy H 2 dissociation on Ru, followed by a spillover of hydrogen to Ni. Thus, it is worth noting that adding Ru to the Ni/Al 2 O 3 catalyst can suppress the oxidation of Ni during steam purge conditions and enhance reducibility, resulting in cyclic stability during DSS operation.

Catalyst Synthesis
All catalysts were synthesized via impregnation methods. Before preparation, the pellet was dried at 150 • C for 2 h. The concentration of Ni metal was 10 wt% based on Ni/Al 2 O 3 . The desired amounts of pellets and Ni(NO 3 ) 2 ·6H 2 O were dispersed in deionized (DI) water, and the mixture was continuously stirred for 1 h. The mixed solution was evaporated in a drying oven at 150 • C for 5 h and calcined in a muffle furnace under air at 700 • C for 4 h with a temperature ramping rate of 10 • C/min. Ru-added Ni catalyst was prepared by a two-step impregnation method. The desired amounts of calcined Ni/Al 2 O 3 and Ru(NO 3 ) 3 (NO) were dispersed in DI water and stirred for 1 h. The concentration of Ru metal was 0.1 wt% based on Ru-Ni/Al 2 O 3 . The Ru-Ni/Al 2 O 3 catalyst was dried at 150 • C for 5 h and calcined at 700 • C for 4 h. 10 wt% Ni catalyst, 0.1 wt% Ru-10 wt% Ni catalyst, and 0.1 wt% Ru catalyst are designated as Ni/Al 2 O 3 , Ru-Ni/Al 2 O 3 and Ru/Al 2 O 3 , respectively. The steps of the catalyst preparation are presented in Figure S6.

Characterization
The cross-sectional images of catalysts after preparation were observed using a Dino-Lite Premier Digital Microscope (AM3113T, ANMO Electronics Corp., New Taipei City, Taiwan). The temperature-programmed reduction (TPR) profiles of the Ni-based catalysts were obtained under 10% H 2 treatments from 100 • C to 900 • C at a temperature ramping rate of 10 • C/min. The outlet gases were recorded via gas chromatography (GC) equipped with thermal conductivity detectors (TCD). The change in the catalytically active components inside the alumina pellet was measured via field-emission scanning electron microscopy (FE-SEM, SU-8230, Hitachi, Tokyo, Japan) with EDS (Oxford Instruments). HR-2D XRD measurements were performed using a Bruker D8 Discover High-Resolution X-ray diffractometer (Korea Basic Science Institute, Daegu, Republic of Korea) with a VANTEC500 (2D detector) and Cu Kα radiation filtered by a Montel mirror operated at 40 kV and 40 mA. The surface area, pore volume and pore size were measured by the BET method using ASAP2020 (Micromeritics Instrument Co., Norcross, GA, USA). H 2 -TGA was conducted at 700 • C or 800 • C to determine the H 2 consumption of the fresh and spent catalysts, respectively, and the results were recorded using the SDT Q600. X-ray photoelectron spectroscopy (XPS) was used to obtain surface chemical states on the surface of catalysts in an X-ray photoelectron spectrometer using a NEXSA (ThermoFisher, Waltham, MA, USA) equipped with an Al Kα source.

Steam-Methane Reforming Tests
The steam-methane reforming tests were determined by monitoring their concentrations via GC. The Ni-based pellet-type catalysts (0.5 g) were packed into a fixed-bed reactor with a diameter of 1 2 an inch and placed in an electric furnace at atmospheric pressure. A thermocouple was placed at the center of the catalyst bed. Prior to the steam-methane reforming, 10 vol% H 2 and balanced N 2 at 100 mL/min (WHSV: 12,000 mL/g/h) were passed through the packed bed at 800 • C for 3 h to reduce the catalysts. After H 2 pretreatment, the gas composition was changed to pure N 2 to purge H 2 from the reactor for 2 min. During the SMR test, the reaction temperatures were maintained at 700 • C, and the gas composition was changed to 30 vol% H 2 O, 10 vol% CH 4 (steam/carbon ratio: 3), and balance N 2 at 100 mL/min. In the DSS operation (Figure 8), the catalysts were cooled to 200 • C at a rate of 10 • C/min under purge conditions after a 90-min reaction. The purge condition consisted of H 2 O/N 2 (30/60 mL/min), wherein only the CH 4 supply was stopped under reaction conditions. The catalyst bed temperature was maintained at 200 • C for 30 min and later increased to 700 • C under purge conditions. CH 4 gas was added to start the reaction, and the reaction was performed again for 90 min. To perform the DSS operation, this reaction was repeated five times. condition consisted of H2O/N2 (30/60 mL/min), wherein only the CH4 supply was stopped under reaction conditions. The catalyst bed temperature was maintained at 200 °C for 30 min and later increased to 700 °C under purge conditions. CH4 gas was added to start the reaction, and the reaction was performed again for 90 min. To perform the DSS operation, this reaction was repeated five times. The inlet and outlet lines of the reactor were maintained at 120 °C using a heating tape to prevent the condensation of water vapor. Furthermore, the reactor outlet stream was passed through a condenser to capture water before entering the reactor and GC column. The dried gases were analyzed using a gas chromatograph (Agilent 6890) equipped with two TCDs. The first Carboxen 1000 packed column was connected to a TCD for analysis of N2, CO, CH4 and CO2 gases, and the second one was connected to the other TCD for H2 gas analysis.

Conclusions
In this study, Ni/Al2O3 and Ru-Ni/Al2O3 catalysts were prepared to produce hydrogen via steam-methane reforming and tested in DSS operation. The prepared catalysts were characterized by SEM-EDS, HR-2D XRD, H2-TPR, N2-physisorption, XPS and H2-TGA. Compared to Ni/Al2O3 catalyst, Ru-Ni/Al2O3 catalyst showed high Ni dispersion and enhanced reducibility due to the interaction between Ni and Ru metal and hydrogen spillover from Ru. Therefore, the Ru-Ni/Al2O3 catalyst showed higher catalytic activity than the Ni/Al2O3 catalyst at high WHSV (48,000 mL/g/h). As the DSS cycle progressed, The inlet and outlet lines of the reactor were maintained at 120 • C using a heating tape to prevent the condensation of water vapor. Furthermore, the reactor outlet stream was passed through a condenser to capture water before entering the reactor and GC column. The dried gases were analyzed using a gas chromatograph (Agilent 6890) equipped with two TCDs. The first Carboxen 1000 packed column was connected to a TCD for analysis of N 2 , CO, CH 4 and CO 2 gases, and the second one was connected to the other TCD for H 2 gas analysis.
The conversion of CH 4 (X CH4 ) and selectivity of the products (CO, CO 2 and H 2 ) were calculated using Equations (3)-(6) as follows:

Conclusions
In this study, Ni/Al 2 O 3 and Ru-Ni/Al 2 O 3 catalysts were prepared to produce hydrogen via steam-methane reforming and tested in DSS operation. The prepared catalysts were characterized by SEM-EDS, HR-2D XRD, H 2 -TPR, N 2 -physisorption, XPS and H 2 -TGA. Compared to Ni/Al 2 O 3 catalyst, Ru-Ni/Al 2 O 3 catalyst showed high Ni dispersion and enhanced reducibility due to the interaction between Ni and Ru metal and hydrogen spillover from Ru. Therefore, the Ru-Ni/Al 2 O 3 catalyst showed higher catalytic activity than the Ni/Al 2 O 3 catalyst at high WHSV (48,000 mL/g/h). As the DSS cycle progressed, the methane conversion of Ni catalysts significantly decreased, whereas Ru-Ni/Al 2 O 3 exhibited sustainable activity. In fact, the repeated DSS operation turned out to partially oxidize the Ni metal and reduce the texture properties of Ni/Al 2 O 3 and Ru-Ni/Al 2 O 3 catalysts. Although Ni metal on the Ru-Ni/Al 2 O 3 catalyst was oxidized, oxidized nickel can be reduced better because Ru can spillover hydrogen into nearby Ni sites. Consequently, adding Ru to the Ni catalyst could be beneficial for the stability of the catalyst by enhancing the reducibility, resulting in the re-reduction of NiO on the catalyst surface.

Data Availability Statement:
The data presented in this study are available on request from the corresponding author. Data are contained within the article or Supplementary Materials.

Conflicts of Interest:
The authors declare no conflict of interest.