Effect of Adding Gadolinium Oxide Promoter on Nickel Catalyst over Yttrium-Zirconium Oxide Support for Dry Reforming of Methane

The dry reforming of methane (DRM) was studied for seven hours at 800 °C and 42 L/(g·h) gas hourly space velocity over Ni-based catalysts, promoted with various amounts of gadolinium oxide (x = 0.0, 1.0, 2.0, 3.0, 4.0, and 5.0 wt.%) and supported on mesoporous yttrium-zirconium oxide (YZr). The best catalyst was found to have 4.0 wt.% of gadolinium, which resulted in ∼80% and ∼86% conversions of CH4 and CO2, respectively, and a mole ratio of ∼0.90 H2/CO. The addition of Gd2O3 shifted the diffraction peaks of the support to higher angles, indicating the incorporation of the promoter into the unit cell of the YZr support. The Gd2O3 promoter improved the catalyst basicity and the interaction of NiO with support, which were reflected in the coke resistance (6.0 wt.% carbon deposit on 5Ni+4Gd/YZr; 19.0 wt.% carbon deposit on 5Ni/YZr) and the stability of our catalysts. The Gd2O3 is believed to react with carbon dioxide to form oxycarbonate species and helps to gasify the surface of the catalysts. In addition, the Gd2O3 enhanced the activation of CH4 and its conversion on the metallic nickel sites.


Introduction
Energy is fundamental to modern economies, and it is anticipated that its demand will continue to rise for many years to come [1][2][3][4]. The bulk of this energy is derived from fossil fuel, which emits the two most common greenhouse gases (CH 4 and CO 2 ). For centuries, the burning of hydrocarbons has increased the atmospheric CO 2 concentration. Methane remains in the atmosphere for much less time than carbon dioxide, but it is more potent in the greenhouse effect [5][6][7]. The concentration of these two greenhouse gases is steadily rising, causing global warming, which harms biodiversity and the ecosystem.
Developing active, selective, energy-efficient heterogeneous catalytic processes is key to a sustainable future because heterogeneous catalysis is at the center of the chemicals and energy industries. The increased urgency of catalyst development for key processes, for instance, biomass upgrading, CO 2 reduction, water splitting, and light alkanes activation, is due to the soaring demand for energy, chemical products, and food and the rise in anthropogenic CO 2 emissions worldwide. As a result, several investigators have performed deep research on various processes. Zada et al. investigated photocatalytic H 2 generation and pollutant withdrawal using g-C 3 N 4 with SnO 2 [8]. Alternatively, Hayat et al. explored Both the nickel and gadolinium nitrates were loaded over meso-YZr support by the dry impregnation method. The required amount of nickel nitrate hexahydrate to give 5.0 wt.% of Ni and the required amount of gadolinium nitrate hexahydrate to give 0.0, 1.0, 2.0, 3.0, 4.0, or 5.0 wt.% of Gd 2 O 3 were mixed and ground with the support, followed by the addition of drops of ultrapure water to obtain a green paste. Upon mechanical stirring, this paste was dried and ground. Mechanical stirring and water addition were repeated three times. The mixtures were calcined for three hours at 600 • C with a temperature ramp of 3.0 • C/min.

Catalyst Activity
DRM experiments were performed at 800 • C under ambient pressure. A tubular stainless-steel reactor (i.d. = 0.009 m; length = 0.3 m) was used. An amount of 0.1 g of the catalyst was used for catalytic testing. The temperature was measured using a sheathed stainless-steel K-type thermocouple, which was placed axially at the center of the catalyst bed. The catalyst was reduced for one hour at 700 • C with a H 2 flow prior to the reaction. During the experiments, methane, carbon dioxide, and nitrogen gases were mixed in a 3:3:1 volume ratio. This mixture was used as a reactant feed with a space velocity of 42 L/h/g cat . The effluent gas was connected to an online GC, which was equipped with a thermal conductivity detector (TCD) to analyze its composition. CH 4

Catalyst Characterization
X-ray diffraction (XRD) patterns of the catalysts were recorded on a Thermo Fisher diffractometer equipped with Cu Kα X-ray radiation and operated at 40 mA and 40 kV. The isotherms of nitrogen physisorption, depending on the Brunauer-Emmett-Teller method (BET), were determined using a Micromeritics Tristar II 3020 surface area and porosity analyzer at −196 • C after outgassing the samples at 200 • C for three hours to remove any adsorbed gases or vapors. The distributions of pore size of the samples were analyzed from the adsorption of isotherms using the Barrett-Joyner-Halenda (BJH) model. Hydrogen temperature-programmed reduction (H 2 -TPR) and carbon dioxide temperatureprogrammed desorption (CO 2 -TPD) analyses of the freshly synthesized catalysts were performed on a Micromeritics Auto Chem II 2920. The analyses were conducted over a temperature range of 50-800 • C and 40 mL/min flow of 10% H 2 /Ar mixture for the TPR analysis and 10% CO 2 /He mixture for CO 2 -TPD measurement, respectively. The coke formation and the amount of carbon deposit on the surface of the spent catalysts were assessed by a thermogravimetric analyzer (Shimadzu-TGA). The deposited carbon was burned in an air atmosphere by heating the samples up to 1000 • C at a rate of 10 • C/min and recording the weight loss. The morphology of the catalysts was examined using a high-resolution transmission electron microscope (HRTEM model: JEM-2100 F, JEOL; Akishima, Tokyo, Japan) and a field emission scanning electron microscope (FE-SEM, 7100F (JEOL; Tokyo, Japan) equipped with energy-dispersive X-ray spectroscopy (EDX) for surface elemental analysis.

Nitrogen Physisorption Analysis
The nitrogen isotherms were of type IV, as shown in Figure 1. Table 1 shows the BET surface area (S BET ), pore volume (P v ), and average pore diameter (P d ) for all the catalysts. The surface area of the unprompted catalyst (5Ni/YZr) displayed the highest specific surface area. After the addition of the Gd 2 O 3 promoter, the specific surface area decreased slightly to a range of 26.0-27.0 m 2 /g. For all catalysts, an increase of the relative pressure at 0.8 was observed. H3 hysteresis loops were exhibited in all the catalysts, indicating the presence of aggregates of plate-like particles that resulted in slit-shaped pores. Table 1 discloses that the surface area of the Gd 2 O 3 -promoted catalysts was primarily unaffected by the variation of Gd 2 O 3 promoter loadings, implying that the Gd 2 O 3 particles diffused inside the pores of the support.
surface area (SBET), pore volume (Pv), and average pore diameter (Pd) for all the catalysts. The surface area of the unprompted catalyst (5Ni/YZr) displayed the highest specific surface area. After the addition of the Gd2O3 promoter, the specific surface area decreased slightly to a range of 26.0-27.0 m 2 /g. For all catalysts, an increase of the relative pressure at 0.8 was observed. H3 hysteresis loops were exhibited in all the catalysts, indicating the presence of aggregates of plate-like particles that resulted in slit-shaped pores. Table 1 discloses that the surface area of the Gd2O3-promoted catalysts was primarily unaffected by the variation of Gd2O3 promoter loadings, implying that the Gd2O3 particles diffused inside the pores of the support.

Hydrogen Temperature-Programmed Reduction (H 2 -TPR)
The TPR profile of the 5Ni/YZr catalyst is shown in the inset Figure 2 (With its very low intensity, this TPR profile diminished when combined with the other profiles, and thus, it is shown in the inset). The reduction peak appeared at a moderate-temperature region of around 300-500 • C with a broad peak and three maxima, corresponding to the reduction of bulk NiO. Generally, this kind of reduction is a characteristic feature of stoichiometric NiO [30]. The appearance of these reduction peaks in the moderate-temperature region indicates a good interaction between the support and the NiO. The absence of reduction peaks below 300 • C indicates that the unprompted catalysts had neither free NiO species nor weakly interacted NiO species with the support. Figure 2 depicts the TPR profiles of the Gd 2 O 3 -promoted catalysts, where the reduction peaks in the temperature range of 300-500 • C, which indicates relatively easy and high reducibility of the NiO phases. With the increase in Gd 2 O 3 loading, the broad peak in the temperature range of 300-500 • C shifted progressively toward a lower temperature, and a small peak below 300 • C emerged, owing to the formation of alloy at the NiO/Gd 2 O 3 interfaces [35]. In addition, for the Gd 2 O 3 -promoted catalysts, the TPR profiles showed broad, low-intensity peaks between 600 and 800 • C, which could be attributed either to the reduction of Gd 2 O 3 [36] or to the reduction of strongly interacted NiO species with the meso-YZr support because of the presence of the promoter [37].

Hydrogen Temperature-Programmed Reduction (H2-TPR)
The TPR profile of the 5Ni/YZr catalyst is shown in the inset Figure 2 (With its very low intensity, this TPR profile diminished when combined with the other profiles, and thus, it is shown in the inset). The reduction peak appeared at a moderate-temperature region of around 300-500 °C with a broad peak and three maxima, corresponding to the reduction of bulk NiO. Generally, this kind of reduction is a characteristic feature of stoichiometric NiO [30]. The appearance of these reduction peaks in the moderate-temperature region indicates a good interaction between the support and the NiO. The absence of reduction peaks below 300 °C indicates that the unprompted catalysts had neither free NiO species nor weakly interacted NiO species with the support. Figure 2 depicts the TPR profiles of the Gd2O3-promoted catalysts, where the reduction peaks in the temperature range of 300-500 °C, which indicates relatively easy and high reducibility of the NiO phases. With the increase in Gd2O3 loading, the broad peak in the temperature range of 300-500 °C shifted progressively toward a lower temperature, and a small peak below 300 °C emerged, owing to the formation of alloy at the NiO/Gd2O3 interfaces [35]. In addition, for the Gd2O3-promoted catalysts, the TPR profiles showed broad, low-intensity peaks between 600 and 800 °C, which could be attributed either to the reduction of Gd2O3 [36] or to the reduction of strongly interacted NiO species with the meso-YZr support because of the presence of the promoter [37].

Carbon Dioxide Temperature-Programmed Desorption (CO2-TPD)
Because the acidic support increases the coke deposition, the researchers focused on nickel catalysts, supported or promoted by metal oxides with strong Lewis basicity. The basicity of the un-promoted and Gd2O3-promoted fresh catalysts was estimated by CO2-TPD experiments, as illustrated in Figure 3. For Gd2O3-promoted catalysts, desorption peaks with maxima centered around 200-300 °C were associated with both weak Bronsted basic sites, such as surface OH − groups, and medium-strength Lewis base sites, while the un-promoted catalyst showed desorption peaks at maxima centered around 100-200 °C associated only to weak Bronsted basic sites. In general, the CO2-TPD profiles had similar

Carbon Dioxide Temperature-Programmed Desorption (CO 2 -TPD)
Because the acidic support increases the coke deposition, the researchers focused on nickel catalysts, supported or promoted by metal oxides with strong Lewis basicity. The basicity of the un-promoted and Gd 2 O 3 -promoted fresh catalysts was estimated by CO 2 -TPD experiments, as illustrated in Figure 3. For Gd 2 O 3 -promoted catalysts, desorption peaks with maxima centered around 200-300 • C were associated with both weak Bronsted basic sites, such as surface OH − groups, and medium-strength Lewis base sites, while the un-promoted catalyst showed desorption peaks at maxima centered around 100-200 • C associated only to weak Bronsted basic sites. In general, the CO 2 -TPD profiles had similar temporal features. However, when increasing the Gd 2 O 3 loading, the intensity of the peaks in CO 2 -TPD profiles increased, implying the increase of catalysts' basicity. The basicity of the Gd 2 O 3 -promoted catalysts was moderate because of the appearance of the desorption peaks in the moderate-temperature region between 200 and 500 • C. The relatively high basicity of the 5Ni+4Gd/YZr catalyst favored efficient CO 2 adsorption and dissociation, which helped to reduce carbon deposits and catalyst deactivation [24][25][26]. temporal features. However, when increasing the Gd2O3 loading, the intensity of the peaks in CO2-TPD profiles increased, implying the increase of catalysts' basicity. The basicity of the Gd2O3-promoted catalysts was moderate because of the appearance of the desorption peaks in the moderate-temperature region between 200 and 500 °C. The relatively high basicity of the 5Ni+4Gd/YZr catalyst favored efficient CO2 adsorption and dissociation, which helped to reduce carbon deposits and catalyst deactivation [24][25][26].

XRD Analysis
The XRD patterns of the fresh 5Ni+xGd/YZr (x = 0, 1, 2, 3, 4, or 5) catalysts are shown in Figure 4A. The XRD patterns of 5Ni/YZr and Gd 2 O 3 -promoted catalysts displayed peaks at 2θ of~30,~35,~50,~60,~63,~74,~82,~84, and~94 • , which refer, respectively, to the (111), (200), (220), (311), (222), (400), (331), (420), and (422) crystallographic planes of the cubic phase of yttria-stabilized zirconia (JCPDS No. 49-1642). Furthermore, the peak at 2θ of~43 • (200 crystallographic plane) could be ascribed to the cubic phase of nickel oxide (PDF 00-044-1159). The peak at 2θ~28 • could be ascribed to the cubic phase of gadolinium oxide (JCPDS No. 12-0797) for the crystallographic phase with Miller indices (222). The addition of the Gd 2 O 3 promoter shifted the peaks of YZr support slightly to a higher 2θ angle; i.e., it caused a slight reduction in the d-spacing parameter, implying the incorporation of Gd 2 O 3 in the lattice of the YZr support, as shown in Table 2. The XRD patterns of the spent 5Ni+xGd/YZr (x = 0, 1, 2, 3, 4, or 5) catalysts are shown in Figure 4B. The XRD patterns of 5Ni/YZr and Gd 2 O 3 -promoted catalysts displayed peaks at 2θ of~30,~35,~50,~60,~63,~74,~82,~84, and~94 • , which refer, respectively, to the (111) Table 3, the intensity and the broadness of the peaks were reduced in comparison to those of the fresh catalysts. This observation could be attributed to the deposition of carbon, where a higher amount of deposited carbon yielded a larger d-spacing, as confirmed by TGA results. Moreover, the absence of NiO diffraction peak in the patterns of the spent catalysts might be due to its reduction to metallic Ni, which was incorporated in the multi-walled carbon nanotubes. The disappearance of the Gd 2 O 3 diffraction peak would be ascribed to its conversion to Gd 2 O 2 CO 3 , as illustrated in the plausible mechanism section below. Table 3. The shift in the 2θ angle and change in the d-spacing of (111) and (220) crystallographic planes of cubic yttria-stabilized zirconia phase for the spent catalysts.      Scherrer's equation was utilized to assess the crystallite size: where D p is the crystallite size in nanometers; λ is the X-ray wavelength (0.15406 nm); β is the full width at half maximum of the diffraction peak of the sample; K is the shape factor, which is 0.94; and θ is the diffraction angle in degrees.

Catalytic Activity
The performance of the catalysts is presented in Figure 6. The reaction was performed at 800 °C and 1.0 atm and for a duration of seven hours. The catalyst activity was expressed in terms of CH4 and CO2 conversions and the H2/CO mole ratio. The general trend of results showed a decrease in the conversions along with TOS due to deactivation by carbon deposition, as confirmed by the TGA results. The 5Ni+4Gd/YZr catalyst was found to be the best in DRM, where it resulted in ∼80% and ∼86% conversions of CH4 and CO2, respectively, and a mole ratio of ∼0.90 H2/CO. The methane conversion profile of the catalysts (5Ni+xGd/YZr, x = 0, 1, 2, 3, 4, or 5) showed an increasing trend from 0.0 wt.% Gd2O3 up to 4.0 wt.% Gd2O3 and then tended to decline at 5.0 wt.% Gd2O3. For all catalysts, the CH4 conversion was lower than that of CO2. This observation could be linked to the reverse water gas shift (RWGS) reaction, which consumed CO2 alongside the main reaction. The H2/CO ratio profile showed a declining tendency, which could be ascribed to the occurrence of the RWGS reaction, where the produced H2 was consumed by the CO2 in the feed to generate CO and thus caused a drop in H2/CO mole ratio. A comparison between

Catalytic Activity
The performance of the catalysts is presented in Figure 6. The reaction was performed at 800 • C and 1.0 atm and for a duration of seven hours. The catalyst activity was expressed in terms of CH 4 and CO 2 conversions and the H 2 /CO mole ratio. The general trend of results showed a decrease in the conversions along with TOS due to deactivation by carbon deposition, as confirmed by the TGA results. The 5Ni+4Gd/YZr catalyst was found to be the best in DRM, where it resulted in ∼80% and ∼86% conversions of CH 4 and CO 2 , respectively, and a mole ratio of ∼0.90 H 2 /CO. The methane conversion profile of the catalysts (5Ni+xGd/YZr, x = 0, 1, 2, 3, 4, or 5) showed an increasing trend from 0.0 wt.% Gd 2 O 3 up to 4.0 wt.% Gd 2 O 3 and then tended to decline at 5.0 wt.% Gd 2 O 3 . For all catalysts, the CH 4 conversion was lower than that of CO 2 . This observation could be linked to the reverse water gas shift (RWGS) reaction, which consumed CO 2 alongside the main reaction. The H 2 /CO ratio profile showed a declining tendency, which could be ascribed to the occurrence of the RWGS reaction, where the produced H 2 was consumed by the CO 2 in the feed to generate CO and thus caused a drop in H 2 /CO mole ratio. A comparison between our best catalyst, 5Ni+4Gd/YZr, and other similar catalysts in terms of the conversions of CH 4 and CO 2 , as well as H 2 /CO mole ratio, is shown in Table 4

Transmission Electron Microscope (TEM)
TEM images of both fresh and spent 5Ni+4Gd/YZr catalysts are shown in Figure 7. The fresh and spent catalyst particles were agglomerated. The TEM images do not show the filamentous carbon deposits on the spent 5Ni+4Gd/YZr catalyst. This observation could be attributed to the even distribution of carbon deposits on the surface of the catalyst and to the carbon-resistance feature of our catalyst, as discussed later on the "plausible mechanism". Moreover, the TEM images indicate no significant change in particle size of the spent catalyst in comparison to the fresh one, indicating the sintering resistance feature of our catalyst. Such observation is consistent with the crystallite size determined from the XRD pattern.

Transmission Electron Microscope (TEM)
TEM images of both fresh and spent 5Ni+4Gd/YZr catalysts are shown in Figure 7. The fresh and spent catalyst particles were agglomerated. The TEM images do not show the filamentous carbon deposits on the spent 5Ni+4Gd/YZr catalyst. This observation could be attributed to the even distribution of carbon deposits on the surface of the catalyst and to the carbon-resistance feature of our catalyst, as discussed later on the "plausible mechanism". Moreover, the TEM images indicate no significant change in particle size of the spent catalyst in comparison to the fresh one, indicating the sintering resistance feature of our catalyst. Such observation is consistent with the crystallite size determined from the XRD pattern.  Table 5, where the calculated d-spacing values were similar to those of the bulk YZr support. These results also indicated the good crystallinity of our support without affecting the d-spacing values for the 400 and 411 crystallographic planes by loading NiO and Gd2O3, as well as the carbon deposition.   The high-resolution TEM (HRTEM) of the fresh and spent 5N+4Gd/YZr catalysts showed parallel lattice plane fringes of the YZr support. The d-spacing value was calculated from the corresponding distance between the lattice plane fringes, as shown in Figure 8 and Table 5, where the calculated d-spacing values were similar to those of the bulk YZr support. These results also indicated the good crystallinity of our support without affecting the d-spacing values for the 400 and 411 crystallographic planes by loading NiO and Gd 2 O 3, as well as the carbon deposition. The high-resolution TEM (HRTEM) of the fresh and spent 5N+4Gd/YZr catalysts showed parallel lattice plane fringes of the YZr support. The d-spacing value was calculated from the corresponding distance between the lattice plane fringes, as shown in Figure 8 and Table 5, where the calculated d-spacing values were similar to those of the bulk YZr support. These results also indicated the good crystallinity of our support without affecting the d-spacing values for the 400 and 411 crystallographic planes by loading NiO and Gd2O3, as well as the carbon deposition.

Scanning Electron Microscope (SEM)
The SEM technique was used to investigate the morphology of the catalysts. Figure 9 shows the SEM image of the fresh sample of the best catalyst (5Ni+4Gd/YZr), where agglomerated particles were observed. The SEM technique was used to investigate the morphology of the catalysts. Figure  9 shows the SEM image of the fresh sample of the best catalyst (5Ni+4Gd/YZr), where agglomerated particles were observed.     The SEM technique was used to investigate the morphology of the catalysts. Figure  9 shows the SEM image of the fresh sample of the best catalyst (5Ni+4Gd/YZr), where agglomerated particles were observed.  Figure 10 displays the EDX analysis of the best catalyst, where all the elements expected to be on the surface were detected qualitatively, implying the success of our preparation method. Figure 10. The EDX spectrum of the fresh 5Ni+4Gd/YZr catalyst. Figure 10. The EDX spectrum of the fresh 5Ni+4Gd/YZr catalyst.

Thermogravimetric Analysis (TGA) of the Spent Catalyst
Based on the TGA plot (Figure 11), all the spent catalysts displayed weight loss in the temperature range of 500-1000 • C. The 5Ni/YZr catalyst contained~19 wt.% of carbon deposit, while the Gd 2 O 3 -promoted catalysts had~6-14 wt.% of carbon deposit, depending on Gd 2 O 3 loading. The 5Ni+4Gd/YZr catalyst produced the least carbon deposition of 6 wt.%. This observation indicates that Gd 2 O 3 not only promoted the reaction performance of the catalysts but also contributed to increasing the coke resistance of the catalysts. The amount of carbon deposit on the Gd 2 O 3 -promoted catalysts (1.0, 2.0, 3.0, and 5.0 Gd 2 O 3 wt.%) was relatively close to each other. The promotional effect and coke resistance of the Gd 2 O 3 -promoted catalyst could be explained by the increased adsorption and activation of CO 2 on Gd 2 O 3 sites as carbonates (Gd 2 O 2 CO 3 ) [40]. Furthermore, Gd 2 O 3 also increased the CH 4 activation and conversion on the metallic nickel sites.

Thermogravimetric Analysis (TGA) of the Spent Catalyst
Based on the TGA plot ( Figure 11), all the spent catalysts displayed weight loss in the temperature range of 500-1000 °C. The 5Ni/YZr catalyst contained ~19 wt.% of carbon deposit, while the Gd2O3-promoted catalysts had ~6-14 wt.% of carbon deposit, depending on Gd2O3 loading. The 5Ni+4Gd/YZr catalyst produced the least carbon deposition of ~6 wt.%. This observation indicates that Gd2O3 not only promoted the reaction performance of the catalysts but also contributed to increasing the coke resistance of the catalysts. The amount of carbon deposit on the Gd2O3-promoted catalysts (1.0, 2.0, 3.0, and 5.0 Gd2O3 wt.%) was relatively close to each other. The promotional effect and coke resistance of the Gd2O3-promoted catalyst could be explained by the increased adsorption and activation of CO2 on Gd2O3 sites as carbonates (Gd2O2CO3) [40]. Furthermore, Gd2O3 also increased the CH4 activation and conversion on the metallic nickel sites.

Plausible Mechanism
It is well-known that zirconium oxide (ZrO2) facilitates the decomposition of carbon dioxide into carbon monoxide and oxygen radical, owing to the oxygen vacancy in ZrO2 support, as shown in Equation (5): where and O are for oxygen vacancy and oxygen on the surface of ZrO2 support, respectively.
Moreover, carbon monoxide could be created by the decomposition of the bicarbonate intermediate, as displayed in Equations (6) and (7): HCO  + + * → CO  + OH + O where CO  and OH are, respectively, adsorbed carbon dioxide and hydroxyl species on the ZrO2 surface. Yttria-stabilized ZrO2 has more oxygen vacancies and possesses basic sites, and therefore, it enhances the decomposition of carbon dioxide according to Equation (5) [37].
Incorporating of Gd2O3 promoter into the catalyst could facilitate the formation of Gd2O2CO3 on the catalyst surface, owing to the interaction of basic Gd2O3 with the acidic CO2 as per Equation (8)

Plausible Mechanism
It is well-known that zirconium oxide (ZrO 2 ) facilitates the decomposition of carbon dioxide into carbon monoxide and oxygen radical, owing to the oxygen vacancy in ZrO 2 support, as shown in Equation (5): where Zr and O Zr are for oxygen vacancy and oxygen on the surface of ZrO 2 support, respectively. Moreover, carbon monoxide could be created by the decomposition of the bicarbonate intermediate, as displayed in Equations (6) and (7): where CO 2 | Zr and OH Zr are, respectively, adsorbed carbon dioxide and hydroxyl species on the ZrO 2 surface. Yttria-stabilized ZrO 2 has more oxygen vacancies and possesses basic sites, and therefore, it enhances the decomposition of carbon dioxide according to Equation (5) [37].
Incorporating of Gd 2 O 3 promoter into the catalyst could facilitate the formation of Gd 2 O 2 CO 3 on the catalyst surface, owing to the interaction of basic Gd 2 O 3 with the acidic CO 2 as per Equation (8) [39]: The Gd 2 O 2 CO 3 surface species would react promptly with the adsorbed CH x species, which emerged from the dissociation of methane on nickel metallic active sites, as illustrated in Equation (9): In addition, the formed Gd 2 O 2 CO 3 participated in the gasification of the deposited carbon, resulting from the decomposition of methane, and hence, in the regeneration of the metallic nickel active sites for hydrogen production, as shown in Equation (10): The Gd 2 O 2 CO 3 species also has the capability to react with the adsorbed hydrogen atom, produced via methane decomposition, to generate the Gd 2 O 3 , CO, and an active surface hydroxyl group, as displayed in Equation (11): Thus, as per the above-suggested reaction scenario, we think that the interfacial areas among Gd 2 O 2 CO 3 and metallic nickel would be the most active sites for DRM, where carbon dioxide activation is improved by the formation of carbonate species, which, in turn, urges and accelerates the decomposition of methane [39]. Moreover, we cannot exclude the role of the oxygen radical produced from the dissociation of the adsorbed carbon dioxide over the oxygen vacancies of the yttria-stabilized ZrO 2 support from participating in the regeneration of the active metallic nickel sites via the reaction with the deposited carbon and adsorbed hydrogen atom produced by the methane decomposition [40].
On the surface of Ni particles, the abstraction of hydrogen from methane takes place, as illustrated in Equations (12) CH * + Ni → 4H/Ni + C * 4H/Ni → 2H 2 + C/Ni The deposited C can be gasified and removed from the Ni surface according to Equation (10) or Equation (17) [41,42].

Conclusions
The study of N 2 -physisorption analysis displayed a type-IV isotherm with H3 hysteresis. The surface areas of Gd 2 O 3 -promoted catalysts were independent of their loading.
In the TPR investigation, the reduction peaks appeared at the moderate-temperature regions, indicating a good interaction between the support and NiO. The addition of Gd 2 O 3 shifted the diffraction peaks of the support to higher angles, implying the incorporation of the promoter into the unit cell of the YZr support. This fact of Gd 2 O 3 incorporation was supported by the crystallite size determination of fresh catalysts via Scherrer's equation, where the un-promoted catalyst presented the lowest value of 16.80 nm, while the crystallite size increased to 17.3 nm for 5Ni+4Gd/YZr and 5Ni+5Gd/YZr and to 17.4 nm for 5Ni+1Gd/YZr, 5Ni+2Gd/YZr, and 5Ni+3Gd/YZr. In the CO 2 -TPD analysis, Gd 2 O 3promoted catalysts presented both weak Bronsted basic sites and medium-strength Lewis base sites, unlike the un-promoted one, which exhibited low-intensity weak basic sites.
The Gd 2 O 3 promotion enhanced catalyst stability by lowering carbon deposition, owing to the dissociative adsorption of methane. The relative carbon resistance of our catalysts could be linked to their basicity endowed by the Gd 2 O 3 promoter and its ability to sweep off the carbon deposit by gasification reaction via oxycarbonate species. The 5Ni/YZr system showed lower activity than the Gd 2 O 3 -promoted catalysts. The 5Ni+4Gd/YZr catalyst was found to have the highest methane conversion (∼80%), CO 2 conversion (~86%), and H 2 /CO mole ratio (∼0.90) and the lowest carbon deposit (6.0 wt.%), suggesting that 4.0 wt.% loading was the optimum for the Gd 2 O 3 promoter. This inspection indicates that Gd 2 O 3 not only promoted the reaction performance of the catalysts but also contributed to increasing the coke resistance of the catalysts.