Effect of the Addition of Alkaline Earth and Lanthanide Metals for the Modification of the Alumina Support in Ni and Ru Catalysts in CO₂ Methanation

Abstract

In order to reduce greenhouse gas emissions, which are reaching alarming levels in the atmosphere, capture, recovery, and transformation of carbon dioxide emitted to methane is considered a potentially profitable process. This transformation, known as methanation, is a catalytic reaction that mainly uses catalysts based on noble metals such as Ru and, although with less efficiency, on transition metals such as Ni. In order to improve the efficiency of these conventional catalysts, the effect of adding alkaline earth metals (Ba, Ca, or Mg at 10 wt%) and lanthanides (La or Ce at 14 wt%) to nickel (13 wt%), ruthenium (1 wt%), or both-based catalysts has been studied at temperatures between 498 and 773 K and 10 bar pressure. The deactivation resistance in presence of H₂S was also monitored. The incorporation of La into the catalyst produces interactions between active metal Ni, Ru, or Ru-Ni and the alumina support, as determined by the characterization. This fact results in an improvement in the catalytic activity of the 13Ni/Al₂O₃ catalyst, which achieves a methane yield of 82% at 680 K for 13Ni/14La-Al₂O₃, in addition to an increase in H₂S deactivation resistance. Furthermore, 89% was achieved for 1Ru-13Ni/14La-Al₂O₃ at 651 K, but it showed to be more vulnerable to H₂S presence.

1. Introduction

Greenhouse gases (GHGs) constitute part of the Earth’s atmosphere naturally, maintaining the optimal conditions for life [1]. The ability of GHG to conserve part of the energy from solar radiation allows the planet’s temperature to be moderate (17 °C global average) [2]. However, with the advent of the industrial revolution, new energy sources based on the combustion of fossil fuels appeared. With the increasing industrialization of the last century and a half, energy demand has been going up to gratify the growing food demand caused by the increase in the world’s population. Moreover, large-scale agriculture has increased, this entails great deforestation to enlarge arable land. All this has meant that GHGs levels have reached levels never seen before. Carbon dioxide (CO₂) is especially important in composing two thirds of GHGs and whose emission is related to three quarters of the combustion of fossil fuels [3].

In the recent climatic summit of the United Nations of 2019, member countries were instigated to mobilize the implementation of several of the sustainable development goals for 2030, complying with the measures established by the 2015 Paris agreement. The objectives set include the reduction of greenhouse gas emissions up to 45% in 2030 and eliminating them in 2050 to limit the global warming to below 2 °C [4]. The strategies established by the European Union (EU) to achieve the agreed objectives consist of betting on energy efficiency, renewable energy, reducing emissions in transport and industry, and carbon capture and storage [5]. There are several possible ways to achieve the objectives in reducing GHGs emissions. The first way is the minimization of emissions by the capture and storage of gases, a greater use of waste materials, an improvement of the efficiency of combustion systems, the implementation of more biomass-based energy systems, replacing traditional fuels with biofuels, or the proper management of arable land and cultivation techniques. However, at present it is strongly believed that the best way to achieve these objectives is the capture, transformation, and recovery of emitted gases, especially carbon dioxide. These gases would become other compounds that can be used to obtain fuels (methane, ethane), plastics (ethylene), medicines (formaldehyde), solvents (methanol, ethanol), or other uses (carbon monoxide, formic acid) [6]. The methanation of CO₂, studied by Sabatier (Equation (1) [7]), proposes a sustainable way to transform residual CO₂ into synthetic methane.

$${\mathrm{CO}}_2+4{\mathrm{H}}_2\rightleftarrows 2{\mathrm{H}}_2\mathrm{O}+{\mathrm{CH}}_4 \Delta \mathrm{G}=-114 \mathrm{kJ}·{\mathrm{mol}}^{-1}$$

(1)

This reaction constitutes the second stage of a larger process known as power-to-gas [8]. In this process, hydrogen is obtained from water electrolysis (Equation (2)) using energy from renewable sources, whose intermittent production would cause problems in the operability of the electricity grid [9].

$${\mathrm{H}}_2\mathrm{O}\to {\mathrm{H}}_2+\frac{1}{2}{\mathrm{O}}_2 \Delta \mathrm{G}=237.1 \mathrm{kJ}·{\mathrm{mol}}^{-1}$$

(2)

Obtaining methane by the Sabatier reaction is a favorable process from a thermodynamic point of view, but with kinetic limitations for the complete reduction of CO₂ to methane, due to its exothermic nature [10]. For this reason, various catalysts have been studied in order to overcome these limitations for methane production. Some of the metals that have proven to be active in methane production are noble metals such as ruthenium (Ru), rhodium (Rh), and palladium (Pd), but also nickel (Ni), and their combinations [10,11,12,13,14,15,16,17,18,19,20,21,22,23,24]. The most commonly used supports in the literature for the dispersion of these metals are metal oxides such as Al₂O₃, SiO₂, TiO₂, CeO₂, zeolites (among others) and even, in recent works, reduced graphene oxide (rGO) [12,13,14,15,16,17,18,19,20,21,22]. Nickel-based catalysts are common in the literature due to their competitive activity and low price, being proposed as a viable alternative to noble metals [10,13,17,18,19,21,23]. However, in order to verify the effect that the noble metal has on Ni, Mihet M. and Lazar M.D. [23] studied the addition of small amounts (0.5 wt%) of Pd, platinum (Pt), and Rh on Ni (10 wt%) catalysts in the methanation reaction. They demonstrated that doping Ni with Pd or Pt improves activity at lower temperatures, contrary to what happens when Rh is added. Renda et al. [25] studied the effect of incorporating different amounts of platinum and ruthenium (0.5, 1, and 3 wt%) on a catalyst based on ceria as support and the compared with nickel (10 wt%) as promoter. It was observed that the most suitable amount of ruthenium that improves the activity of the nickel catalyst is 1 wt%, the best value being 0.5 wt% in the case of platinum, reaching a high conversion at a lower temperature. A similar conclusion was found by Bustinza et al. [26], by incorporating ruthenium in different proportions and order on a nickel (12.7 wt%) catalyst supported in alumina-coated monolith. In this case, the best conversion in the methanation reaction was achieved for catalysts with 1 wt% of Ru.

Alumina is the most common catalyst support, due to its high surface area which favors the dispersion of the metal atoms, facilitating the access of reaction gases to the active center [13,14,15,19,20,21,22,23]. Namely, γ-Al₂O₃ structure favors metallic catalyst stability, activity and selectivity within the operation temperature range for this reaction, pointing to this support as one of the most suitable [27]. It has been shown in recent works that alumina support modification in Ni catalysts for the methane reaction increases activity, especially at low temperatures [19,28,29]. Liang et al. described an increment in the conversion of CO₂ at low temperatures after the addition of magnesium (Mg), calcium (Ca), strontium (Sr), and barium (Ba) to the Ni/alumina catalysts, especially the last two [28]. In the work of Tsiotsias et al. (2020) [30], it was shown that the incorporation of alkaline and alkaline earth metals increases the alkalinity of the support, facilitates the chemisorption of CO₂, increases the population of oxygen vacancies on the surface of the metal oxide, and improves the dispersion of the metal on the support. Jomjaree et al. [31] prepared Ni/CeO₂ nanostructured catalysts with tunable CeO₂ morphology/structure determining the influence in increasing oxygen vacancies and oxygen storage capacity, and their influence on catalytic activity enhancement between 473 and 773 K under atmospheric pressure. Cardenas-Arenas et al. [29] prepared a 3D-ordered macroporous structure of NiO-CeO₂ mixed oxide (NiO-CeO₂ (np)) to achieve high CO₂ methanation conversion due to its high specific surface area if compared with a reference catalyst without size control (NiO-CeO₂ (Ref)). In addition, García-García et al. [19] studied the addition of cerium (Ce) to modified alumina loading 4 and 6 wt% of Ce and 3 wt% zirconium (Zr) increasing the yield to methane at low temperature [19]. However, there were small amounts of Ce and Zr and it was observed that at least a 6 wt% of Ce is needed to improve the catalyst activity towards methanation. Moreover, Ni-Ce-Zr catalyst increasing surface oxygen species implies the formation of intermediates enhancing the CO₂ conversion even at very low temperatures (473 K) [32]. Furthermore, the effect of modifying alumina support with different amounts of La (0, 4, 14, and 37 wt%) has also been studied by Garbarino et al. [29]. They concluded that the methane yield increases at low temperatures, as the La amount added increases to a maximum of 14 wt%. Finally, the modification of alumina supports with alkaline earth metals or lanthanides has been studied in other reactions (Fischer–Tropsch or dry reforming), showing an improvement in reaction performance [33,34,35,36].

In the present work, according to the previous observations of recent works, the effect of modifying the support of the 13Ni/γ-Al₂O₃ catalyst was studied. To that end, alkaline earth metals (Ba, Ca, and Mg) and lanthanides (La and Ce) were added to the bare support and their effect in catalyst containing Ru, Ni, and Ni-Ru as active metal on the CO₂ methanation reaction was examined. The Ni content used as promoter was 13 wt%, and that of Ru 1 wt%, based on previous studies already done by the research group, i.e., [37]. In the case of Ba, Ca, and Mg, it was 10 wt%, and in that of La and Ce, 14 wt%, justified by the above studies available in the bibliography.

2. Materials and Methods

2.1. Catalyst Preparation

The modified catalysts were prepared by wet impregnation methods in two successive stages. Determined amounts of γ-Al₂O₃ (Merck) and support modifying metals (barium nitrate 99.999% Sigma-Aldrich, calcium nitrate tetrahydrate 99% Sigma-Aldrich, magnesium nitrate hexahydrate 99% Sigma-Aldrich, lanthanum (III) nitrate hydrate 99.9% Sigma-Aldrich, and Cerium (III) nitrate hexahydrate 99.999% Sigma-Aldrich), were mixed in appropriate proportions (10 wt% Ba, Ca, or Mg; or 14 wt% La or Ce). Twenty milliliters of distilled water were added to the mixed salt supports and the pH value of γ-Al₂O₃ (7.5) was fitted over point of zero charge (pzc) at 8.5 [27,38]. The pH was adjusted adding ammonium (Panreac) to increase or nitric acid (Scharlau) to decrease the pH value. The solution was stirred overnight, in a rotary evaporator (Heidolph Laborota 4000) and the solvent was evaporated, heating at 338 K, and reducing slowly the pressure until 40–100 mbar helped by a vacuum pump. The support obtained was dried in an oven at 373 K during 2 h and calcined at 673 K in presence of air for 2 h, with a ramp of 1 K/min.

The catalysts were prepared by adding the metal salt of nickel (II) nitrate hexahydrate 99.999% (Sigma Aldrich) and/or ruthenium (III) chloride hydrate 99.98% (Sigma-Aldrich) to the prepared support and mixing with 20 mL of distilled water to obtain catalysts with 13 wt% of nickel, 1 wt% Ru, or both. The procedure was like the support preparation, using the wet impregnation method: pH control at 8.5, overnight stirring, solvent evaporation, drying, and calcining.

Finally, the calcined catalysts were pressed and sieved to the desired particle size: 0.42 mm < d_p < 0.50 mm. This particle size (d_p) was chosen in order to avoid reagents bypassing near the wall, by keeping an internal pipe diameter-to-particle size ratio higher than 10 [39].

The catalysts prepared were named, according to their nominal composition, as follows: 13Ni/10Ba-Al₂O₃, 13Ni/10Ca-Al₂O₃, 13Ni/10Mg-Al₂O₃, 13Ni/14Ce-Al₂O₃, 13Ni/14La-Al₂O₃, 1Ru/14La-Al₂O₃, 1Ru-13Ni/14La-Al₂O₃, 1Ru-13Ni/Al₂O₃, and 1Ru/Al₂O₃.

2.2. Catalyst Characterization

The N₂ adsorption-desorption method was employed to evaluate the textural properties of the catalysts in the Autosorb 1C-TCD, after outgassing the samples at 573 K for 3 h. The specific surface area was calculated using Brunauer–Emmett–Teller (BET) equation and pore volume and pore sizes were calculated by the Barrett–Joyner–Halenda (BJH) method.

• The real composition of the catalysts was determined by inductively coupled plasma optical emission spectroscopy (ICP-OES) on a Perkin–Elmer Optima 3300DV. Moreover, after disaggregating the catalyst in acid solution (mixture of HCl 3:1 and HNO₃), the composition and chemical element were identified by characteristic wavelength emission intensity.
• The crystalline species and an approximation of the average crystal size (estimated by XRD patterns and Scherrer equation [40]) of the catalysts were carried out on a PANalytical X’Pert Pro diffractometer using a CuK_α radiation (λ = 0.15418 nm) at 40 kV and 30 mA and a scanning angle (2θ) of 20° to 90°.
• X-ray photoelectron spectroscopy (XPS) was obtained on a SPECS (Berlin, Germany) system equipped with a Phoibos 150 1D-DLD analyzer and an Al Kα (1486.6 eV) monochromatic radiation source with electrons output angle of 90°. XPS is a technique that allows the study of the species present on the surface of the catalyst and their chemical state.
• The temperature-programmed reduction of H₂ (H₂-TPR) and temperature programmed desorption of ammonia (NH₃-TPD) or of CO₂ (CO₂-TPD) were carried out in a Micromeritics AutoChem II RS232, equipped with a thermal conductivity detector (TCD). The TPR profiles were recorded heating 50 mg of the sample from 323 K to 973 K, at a ramp rate of 10 K/min. Previously, the sample was heated from room temperature to 473 K for 1 h in Ar stream (30 Nml/min). Later, the sample was cooled down to 323 K and the Ar was replaced by 5 vol.%, H₂/Ar (45 Nml/min) stream and analyzed. The TPD profiles were recorded after reducing the sample with 5 vol.%, H₂/Ar (45 Nml/min) stream, and heating to 723 K. Then, the sample was cooled with He, replacing this stream by 10 vol.% NH₃/He or CO₂ stream during 30 min. Recovering the He stream, NH₃ desorbed per unit time and catalyst mass was determined heating the sample from 423 K to 1173 K, at a ramp rate of 10 K/min.

2.3. Activity Test

The CO₂ methanation was carried out in a stainless steel fixed bed reactor of 32 cm of length, with an outside diameter of ¼”, placed inside a bench-scale plant (PID Eng&Tech), at 10 bar and temperature range from 498 to 773 K. The reactor was loaded with 60 mg of catalyst, and diluted with inert SiC, in order to minimize thermal gradients in the catalytic bed (weight_catalyst/weight_SiC = 1:4.5). Furnace temperature was adjusted to maintain the catalyst bed under isothermal condition using four thermocouples. The first step was the catalyst activation, reducing the catalyst with a mixture of N₂ (99.999%, Air Liquide) and H₂ (99.999%, Air Liquide) with a ratio of H₂/N₂ equal to 3:1 (65 Nml/min of H₂) at 673 K during 4 h. The reduction temperature of the catalysts was determined from the reduction profiles obtained by the TPR-H₂, finding 673 K as a suitable temperature for activating the catalysts. Even more, this temperature was the same for all of them so the results could be comparable. Once the catalyst was activated, the reaction gases, H₂ and CO₂ (99.98%, Air Liquide), were fed in a ratio H₂/CO₂ of 4:1 such that the weight hourly space velocity (WHSV) of the flow fed was 38.30, 60.80, or 128.75 g_feed/(g_cat·h), depending on the amount of catalyst used 200, 125, or 60 mg respectively.

The stoichiometry proportion for the methanation reaction was employed to obtain methane and water (Equation (3)). The methanation reaction is the combination of Equation (4) and (5), where CO could be generated as an intermediate product, or as a final product in combination with methane [21].

CO₂ + 4 H₂ ⇄ CH₄ + 2 H₂O

(3)

CO₂ + H₂ ⇄ CO + H₂O

(4)

CO + 3 H₂ ⇄ CH₄ + H₂O

(5)

The reactor was heated at a rate of 10 K/min under N₂ flow, until the reaction temperature was reached. Afterwards, the N₂ flow was replaced by the mixture of reactants. The catalyst activity was studied at reaction temperatures from 498 K to 573 each 25 K and from 573 K to 773 K each 50 K. The products were separated in a Peltier condenser (condensed water and gas products), and then analyzed on-line using a Varian CP-4900 MicroGC equipped with a high sensitivity TCD and two columns (10 m Molecular Sieve 5, 10 m Poraplot Q). The reaction was stabilized at each temperature for 120 min, recording the output flow and composition of gases every 10 min. The water produced was collected and weighted in this time. The gas phase composition was on-line analyzed. The CH₄ yield was calculated using the following equation:

$${\mathsf{\eta }}_{{\mathrm{CH}}_4}={\mathrm{mol} \mathrm{CH}}_4^{\mathrm{out}}/{\mathrm{mol} \mathrm{CH}}_4^{\mathrm{out}}\mathrm{stoichiometric}·100$$

(6)

In the Equation (6), ${\mathsf{\eta }}_{{\mathrm{CH}}_4}$ represents the methane yield, mol CH₄^out represents the flow rate of the methane out of the reactor, and mol CH₄^out stoichiometric represents the theorical flow rate of the methane out of the reactor, based on the stoichiometry of the Sabatier reaction. Thus, if all CO₂ is converted to CH₄, the value of ${\mathsf{\eta }}_{{\mathrm{CH}}_4}$ would be 100%.

All the experiments were repeated to ensure reproducibility.

3. Results and Discussion

3.1. Catalyst Characterization

3.1.1. ICP Analysis

The chemical composition of the calcined catalysts employed in the CO₂ methanation was measured by ICP-OES analyses. The catalysts were prepared to have a target nominal content of 13 wt% for the Ni, 1 wt% for Ru, 10 wt% for the alkaline earth Ba, Ca, and Mg, and 14 wt% for the lanthanides La and Ce. Catalysts that only contain Ni showed a Ni content slightly higher than the target, as reported by García-García I. et al. [19] for a Ni catalyst supported on γ-Al₂O₃. However, the experimental nominal content of Ni was higher or lower than the target composition depending on the metal modifying the support. In the presence of La, the amount of Ni determined was greater than expected, as it was observed experimentally and confirmed in the literature [41,42]. In the case of Ca and Mg, a considerable decrease in its value was observed with respect to 13Ni/Al₂O₃. Moreover, a decrease in the proportion of Ni with respect to Ba was produced by the presence of Ba modifying the support. The catalyst, whose support was modified with Ce, slightly reduced the Ni content. In contrast, the catalyst 1Ru-13Ni/Al₂O₃ showed slightly higher nickel and lower Ru contents respective to the expected values. The chemical composition measured by ICP of the prepared catalysts and their target nominal contents (in brackets) are shown in the

Table 1. Summary of the contents (wt%) of Ni, Ru, and modifier metal (M), textural properties, Ni crystal size.

Catalyst	Nominal Content			Measured Content by ICP			SBET a (m2/g)	VP b (cm3/g)	DP c (nm)	SXRD d1 (nm)	SXRD d2 (nm)
	Ni	Ru	M	Ni	Ru	M
γ-Al2O3	−	−	−	−	−	−	202	0.81	7.7	−	−
13Ni/Al2O3	13	−	−	13.9	−	−	180	0.55	7.2	5	10
13Ni/10Ba-Al2O3	13	−	10	8.83	−	14.17	93	0.12	6.9	5	15
13Ni/10Ca-Al2O3	13	−	10	15.73	−	7.27	74	0.08	6.1	10	25
13Ni/10Mg-Al2O3	13	−	10	17.52	−	5.48	109	0.15	7.3	5	15
13Ni/14Ce-Al2O3	13	−	14	12.16	−	14.84	86	0.12	7.2	5	15
13Ni/14La-Al2O3	13	−	14	15.41	−	11.59	79	0.11	6.5	5	15
1Ru/14La-Al2O3	−	1	14	−	1.19	13.81	129	0.17	7.2	−	−
1Ru-13Ni/14La-Al2O3	13	1	14	14.79	0.52	12.69	93	0.12	7.1	5	15
1Ru-13Ni/Al2O3	13	1	−	13.73	0.27	−	163	0.22	7.5	5	15

^a The surface area was calculated by the Brunauer–Emmett–Teller (BET) method; ^b Barrett–Joyner–Halenda (BJH) desorption pore volume; ^c BJH desorption average pore diameter; ^d S_XRD ^d1 (after reduction) and S_XRD ^d2 (after reaction) represent Ni crystallite size approximation calculated from Ni (111) plane using Scherrer equation.

.

3.1.2. BET Measurements

In Table 1, the surface area, pore volume, and pore size are represented in order to analyze the textural properties of supports and prepared catalysts. The isotherms obtained for the supports and catalysts are IV-type curves, with a remarkable H₂ hysteresis loop, representative of mesoporous structure and due to the presence of “ink-bottle” or cylindrical channels [43].

The addition of Ru barely reduced the surface area of the alumina support (202 m²/g), due to its small amount added. Contrary to what happened with the addition of Ni, a significant reduction of the surface area and pore volume were detected, probably due to the partial blockage of the small pores of the support [44]. The addition of a second metal contributed to the pore blocking effect depending on the metal and its interaction with the support. In the case of Mg, the surface area (and pore volume and size) reduction was lower than in the cases of Ca and Ba due to the creation of porous structures during the catalyst calcination stage [28]. The addition of Ba in amounts greater than 5 wt% produced the higher occupation by the metal, causing the reduction of the total surface area [28]. Ca promoted the coalescence of small pores to obtain larger pores during the calcination step, being increased with the amount of metal added [28]. The incorporation of La or Ce as support modifiers has an effect similar to that observed for alkaline earths, so that as metal was added to the catalyst, the sintering increased, producing a blockage of the small pores of the supports and the increase of pore size via merge of the catalysts.

3.1.3. H₂-TPR Studies

The effect of the alkaline earth and lanthanides on the reducibility and the metal-support interaction of the catalysts was evaluated using the H₂-TPR. Measurements and results are collected in the Figure (Figure 1 and Figure 2). According to the reducibility and the strength of interaction Ni-support, the peaks represented in the H₂-TPR show four reducible species of nickel oxides. In the range of 523 to 623 K, reducible species correspond to nickel oxide with weak interaction with the support. The NiO with weak-medium interaction with the support corresponds to the reducible species between 623 K and 773 K. Stronger interaction (medium-strong) with the support for the Ni²⁺ occurs in the range of 773 to 1023 K for the non-stoichiometric nickel aluminate (NiO-Al₂O₃), the stoichiometric nickel aluminate species (NiAl₂O₄) being harder to reduce. These are reducible from 1073 K having a strong interaction with the support [28,45,46,47,48].

The addition of noble metals, such as Ru, changes the reduction profiles of the Ni-catalyst as reported in the literature [19,49,50]. The reduction profile of the Ru catalyst showed a maximum located at 416 K corresponding to the reduction of supported RuO₂ into metallic Ru and a shoulder at 348 K according to the reduction of well-dispersed RuO_x species [50]. The catalyst of Ru and Ni exhibit mainly two peaks at 410 and 623 K and two shoulders at 376 and 562 K, respectively. The peak and shoulder reported at 410 and 376 K correspond to those determined for the monometallic catalyst of Ru. The main reduction peak reported at 623 K represents NiO reduction, and the shoulder at 562 K represents the reduction of RuO₂ and NiO on the alumina surface, mainly forming Ni-Ru particles [49].

TPR patterns of La modified supports show a splitting of the peaks at temperatures between 500 and 700 K. The new peak at 505 K corresponds to free NiO species in the alumina support, which increases as the La content of the catalyst increases [29,45,51]. Based on the results and the literature observations, the stable La-alumina structure weakens the interaction between NiO and support as La content increases, by partial destruction of metastable Ni-Al mixed oxide phase [45,51].

As reported in literature, the addition of Ce to an alumina support results in hydrogen consumption due to different interaction of ceria with the support [52]. The reduction peaks were found at 598, 798, and 993 K corresponding to non-stoichiometric CeO₂, surface reducible metal centers of CeO₂, and bulk reducible centers of CeO₂, respectively [52]. Attending to the TPR profile obtained, the participation of the support in the reducibility of the nickel catalyst was observed at low reduction temperatures (561 K), corresponding the reduction peak at medium temperature (715 K) with the interaction of NiO with the support. At low temperature, the reduction peak represents the contribution in the reduction of Ni and Ce⁴⁺ at the surface of CeO₂ [52].

The TPR profiles of the supports of Ni catalysts modified with alkaline earth define a different spectrum, depending on the metal. Barium in contact with the alumina support produces reducible species as BaAl₂O₄ that at 763 K show a reduction peak. In combination with nickel, the greatest contribution in the TPR profile corresponds to the reduction of nickel species (with reduction peaks at 583, 903, and 1053 K), being facilitated the interaction of Ni and alumina by the presence of Ba. The contribution of Ba was mainly observed in the increase of the first two catalyst reduction peaks and decrease in the third, due to the interaction of BaO with the support, forming other structures (as spinel) [53]. The experimental profile presented three peaks and one shoulder at 498, 650, 682, and 561 K, respectively, as the literature shows [28,53]. The contribution of Ca is due to the presence of CaO in the catalyst, which is related to reduction at low temperatures, strengthening the interaction with NiO and improving the reducibility of Ni [28]. This produces a splitting of the low temperature reduction peak corresponding with the reduction of NiO, obtaining two peaks at 595 and 615 K. The first one increases with respect to the second, as the amount of Ca added increases [54]. It was observed that the addition of Mg (presented in form of MgO) decreased the reduction peaks corresponding to NiO and Ni aluminate [55]. In the TPR profile obtained for Ni/Mg-Alumina, three peaks at 513, 600, and 890 K were determined, with lower hydrogen consumption regarding to other catalysts analyzed. Some causes attributed by Tan et al. [55] are the effect of coverage of MgO on the NiO species; a strong interaction between NiO and MgO; a formation of larger Ni(Mg)O; or the combination of some of them [28]. NiO-MgO catalysts usually form NiO-MgO solid solution, which makes it very difficult to reduce, thus, it is reduced at higher temperatures. Its reduction usually gives very well dispersed crystals, as observed in the Requies work [56]. In general, according to the above, the addition of Ba facilitates the reduction of the catalyst against the use of Ca and Mg that makes it more difficult.

Based on the results obtained and the literature, the catalysts activated by the reduction at 673 K should be completely reduced at the reaction conditions between 498 and 773 K.

3.1.4. NH₃ and CO₂-TPD Studies

The acidity and basicity of the catalysts was studied by NH₃-TPD and CO₂-TPD desorption profiles, determining the quantity and strength of catalyst acid and basic sites, respectively. Desorption temperature determines the strength of the bonds, so that higher desorption temperature means stronger bond.

According to their nature, acid centers of different strength and adsorbed NH₃ reaction towards NO_x formation, define three regions: up to 523 K, weak acid sites (Lewis nature); from 523 to 673 K, middle acid sites (Lewis); and strong acid sites (Brönsted nature) from 673 K [41,49,57]. Depending on the metal modifying the alumina support of the catalyst and according to the metal active phase, the NH₃-TPD profiles will be different. These profiles are shown in Figure 3 and Figure 4. The main peaks are shown at around 520 and 541 K corresponding to Lewis acid sites (middle acidity) and 599 K, 825 K, and 1093 K conforming to Brönsted acid sites (strong acidity). The total amount of acid centers, and their distribution, related to ammonia desorption, are presented in

Table 2. Distribution of the strength of acidity and basicity by NH₃-TPD and CO₂-TPD, respectively, of the different catalysts.

Catalyst	Acidity Amount μmol NH3/gcat			Basicity Amount μmol CO2/gcat
	Weak	Media	Strong	Weak	Media	Strong
Temperature	<523	523 < T < 673	>673	<423	423 < T < 723	>723
13Ni/Al2O3	127.3	337.9	1146.2	11.0	56.7	100.8
13Ni/10Ba-Al2O3	85.7	197.3	1407.3	12.7	64.6	127.4
13Ni/10Ca-Al2O3	83.7	260.2	1091.1	18.1	70.6	124.5
13Ni/10Mg-Al2O3	119.6	251.2	1116.5	15.6	79.3	143.0
13Ni/14Ce-Al2O3	143.1	383.2	1274.8	14.2	62.0	105.2
13Ni/14La-Al2O3	166.8	345.3	1268.7	15.2	70.8	118.5
1Ru-13Ni/14La-Al2O3	167.4	378.8	1253.1	13.4	65.8	108.5
1Ru/14La-Al2O3	165.5	316.2	1180.7	10.8	64.9	112.3
1Ru-13Ni/Al2O3	169.1	470.7	1167.0	12.7	55.8	91.9
1Ru/Al2O3	153.6	411.3	974.6

for the studied range, calculated as the area under the concentration-time curve of the TPD profiles, according to regions. The Lewis acid sites of alumina are coordinated with NH₃, due to the electron deficiency of trivalent aluminum atoms, presenting a maximum centered at 548 K [57]. Furthermore, the peak centered at 1073 K (strong acid region) was the most significant in the catalysts studied and corresponds with Brönsted acid sites where the hydrogen atoms may act as proton donor [58].

The 13Ni/Al₂O₃ catalyst is characterized by having a small peak at 460 K in weak acid region; two peaks in the medium acid region, with a main peak at 599 K and a weaker peak at 552 K; and another two wider peaks in the strong acid region at 825 K and 1093 K. In the case of the 1Ru/Al₂O₃ catalyst, the peaks present in the region of medium acidity are more significant, taking into account the profile of Figure 3. These peaks appear at 541 and 599 K. This catalyst shows a quantitatively greater number of weak and medium acid centers than 13Ni/Al₂O₃ catalyst. However, the peak present in strong acid regions is significantly lower for ruthenium. Therefore, the combination of both metals, Ru and Ni, produces an additive effect on the number of weak and medium acid centers. Thus, a significant increase in these centers is observed with respect to the Ni catalyst (and that of Ru), exhibiting two peaks similar to those of Ru, slightly displaced at lower temperatures. These peaks appear at 520 K and at 485 K.

Ba produces a significant decrease in the number of weak and medium acid centers, compared to the sharp increase in strong acid centers, due to the presence of two peaks, one smaller at 825 K and the other more noticeable at 1093 K. It is observed in 13Ni/10Ca-Al₂O₃ catalyst, that the presence of Ca produces a migration of the density of active centers towards a medium interaction strength, close to a low interaction, at 558 K and to the region of strong centers at 915 K. However, despite its concentration, the density of acid centers in this case decreased for all strengths. Something similar happens to 13Ni/Mg-Al₂O₃ catalyst, concentrating the centers mainly in the region of strong acid centers with a wide peak at 845 K and with a significant reduction in the density of centers of medium strength. The Ce-incorporating catalyst shows little significant difference from 13Ni/Al₂O₃ spectrum, except for a slight increase in the density of active centers in all regions.

Finally, the addition of La to the catalysts containing Ru, Ni, or both demonstrates its contribution to the significant increase in the density of acid centers of higher strength. This effect is justified in the literature, referring to an interaction that takes place between La³⁺-O²⁻ pairs with the strongest acid and basic centers of alumina. This results in the reduction of Lewis acidity, as was more easily observed in 1Ru/Al₂O₃ catalyst, but providing stronger acid-basic couples [41].

Figure 3 and Figure 4 show the CO₂-TPD profiles defined by the catalysts studied. The curves defined in this analysis define the distribution of basic catalyst centers as a function of temperature. In this way, as temperature increases, interaction between CO₂ and catalyst is stronger and, therefore, the basic centers are stronger. According to the strength of the basic centers of the catalyst, three regions of the curve are distinguished: the region of weak basic centers (between 313 and 423 K), moderate basic centers (between 423 and 723 K), and strong basic centers (between 723 and 823 K). Total basicity in each region are shown in Table 2.

The incorporation of alkaline earth elements and lanthanides to the catalyst structure contributes to the increase of basic centers that facilitate the adsorption of CO₂ for its reaction with H₂. The low electronegativity of these metals results in the metal oxide formed having a basic behavior [59]. In this way, the presence of these metals increases the number of basic centers of 13Ni/Al₂O₃ catalyst, as observed in the table for the three regions, especially for 13Ni/10Mg-Al₂O₃ catalyst.

The CO₂-TPD profiles are similar in the catalysts analyzed, having in common the presence of three peaks that appear at temperatures of 350, 473, and 750 K. The first, in the region of weak centers, corresponds to the formation of carbonates on the catalyst surface. Second, the peak appearing in the moderate region, at 473 K, is due to bicarbonate formation on the catalyst surface. Finally, when the interaction between CO₂ and catalyst is stronger (750 K), the formation of monodentate bicarbonate species takes place [60]. In the 1Ru-13Ni/14La-Al₂O₃, 13Ni/14Ce-Al₂O₃, and 13Ni/10Ca-Al₂O₃ catalysts, the formation of bicarbonate species is facilitated at lower temperatures, with the corresponding peak appearing at a temperature of approximately 423 K. The incorporation of Ba and Ca to the 13Ni/Al₂O₃ catalyst produces an increase in strong basic centers (although without reaching the one that produces 13Ni/10Mg-Al₂O₃), shifting the peak corresponding to the formation of monodentate bicarbonate species to 808 and 846 K, respectively. The number of strong basic centers is also remarkable. This demonstrates the strong character of the interactions of these catalysts with CO₂, interacting with the monodentate species of the bicarbonate species. In ceria-supported catalysts, the presence of medium-strength basic sites significantly improved the catalytic activity, as demonstrated by Liu et al. [61]. Alumina is known for its acidic character, but the intermediate strength basic sites led to an increase in the CO₂ adsorption capacity enable.

3.1.5. XPS

The XPS technique allowed us to determine the species present on the surface of the catalyst, their oxidation state and their quantification. The main species identified on the surface correspond to the support, identifying Al 2p at 74.2 eV and O 1s at 531.0–531.8 eV that make up the alumina and the oxides of the rest of the metals. Nickel is the main active metal of the catalysts studied, showing a multiple split peak at 854–854.5, 855.7–857.2, and 860.8–862.6 eV [62] according to Ni⁰, Ni²⁺ from NiAl₂O₄, and Ni²⁺ satellite respectively for fresh catalyst [62], as observed in Figure 5 and Figure 6. Used catalyst (after activity test and H₂S exposure) also showed three peaks in 13Ni/14La-Al₂O₃ and 1Ru-13Ni/14La-Al₂O₃, at 855.1, 857.5 and 862.8 eV, corresponding to Ni²⁺ from NiO, Ni²⁺ from NiAl₂O₄, and Ni²⁺ satellite respectively, and only two in the others at 855.8–856.6 and 862.0–863.0 eV, corresponding to Ni²⁺ from NiAl₂O₄, and Ni²⁺ satellite, respectively, as observed in Figure 5 and Figure 6. Specifically, in fresh catalysts, a greater intensity of the metallic nickel peak than of the oxide is observed, due to the in situ reduction stage in the reduction cell incorporated in the equipment. However, in the case of the used catalysts, without reducing treatment, and due to the aggressive effect of temperature, the effect of hydrogen sulfide poisoning and sintering, only the presence of nickel aluminate is determined in most catalysts, identifying nickel oxide only in catalysts composed of Ni-La. The disappearance of metallic nickel in its transformation into nickel oxide or nickel aluminate can cause loss of activity of the catalyst.

Ru 3p 3/2 at 459.5–461.9 eV, used as an active metal or mainly as a promoter, was difficult to identify in some of the catalysts due to its low concentration, which approximates its spectrum to that of noise. The rest of the metals incorporated in the support were also identified. Ba 3d 5/2 with doublet at 781.1–781.9 and 796.4–797.2 eV with overlap in the Auger Ba MNN line that can lead to an underestimation of %Ni. Ca 2p at 346.7–347.5 eV. Mg 2p at 49.5–49.6 eV. Ce with quintuplet 3d 5/2 and 3d 3/2 in positions 880.0–880.6 and 885.0–885.5 eV for Ce³⁺ duplicated at 898.3–898.9 and 903.5–904.1 eV. In addition, at 882.1–882.9, 888.1–889.0 and 897.2–898.2 eV for Ce⁴⁺ duplicated at 900.4–901.2, 906.9–907.6, and 915.8–916.5 eV. Thus, in the less energetic positions, there is a slight inclusion of the Ce spectrum in that of Ni, which hinders the identification of nickel 2p ½ and underestimate %Ni.

Finally, La 3d 5/2 and its satellite were located at 834.9–836 and 838.5–839.7 eV, and 3d 3/2 and its satellite at 852.5–853 and 856.5–856.9 eV. However, it was not possible to identify La 3d 3/2 in those catalysts whose active metal is Ni, in the same way that the Ni content cannot be quantified. This is due to the superposition with the Auger line of Ni 2p 3/2 and La 3d 3/2 in the case of the catalyst 13Ni/14La-Al₂O₃ and 1Ru-13Ni/14La-Al₂O₃. Nonetheless, the Ni 2p 3/2 corresponding triplet moves toward lower bond energy, while the La 3d 3/2 doublet increases its bond energy, producing the overlapping of the corresponding peaks, confirming the interaction that occurs between La³⁺ and Ni²⁺ [29].

According to the results shown in the spectra represented in Figure 7, a lower proportion of Ni⁰ than of NiO is observed in 13Ni/10Mg-Al₂O₃ catalyst, after being reduced in situ. This can be justified by a higher ease for the oxidation of this catalyst, compared to the rest of the analyzed catalysts. The introduced magnesium interacts with the nickel forming a solid solution difficult to reduce [56]. After its use in the reaction, the complete oxidation of the nickel present in the catalysts is observed, raising the proportion of oxygen present, as shown in

Table 3. Surface atomic values obtained by XPS analysis of the different catalysts studied.

	Fresh Catalyst					Used Catalyst
	Al (%)	O (%)	Ni (%)	Ru (%)	M (%)	Al (%)	O (%)	Ni (%)	Ru (%)	M (%)
13Ni/Al2O3	40.4	56.5	2.7	−	−	37.7	59.1	2.6	−	−
1Ru-13Ni/Al2O3	36.9	58.3	3.2	−	−	35.9	60.6	2.7	0.3	−
1Ru-13Ni/14La-Al2O3	38.6	57.0	*	−	4.4	24.5	69.8	3.9	−	1.8
13Ni/14La-Al2O3	35.7	60.3	*	−	3.4	26.7	64.9	5.3	−	3.1
13Ni/10Ba-Al2O3	32.7	59.2	2.1	−	6.0	28.9	62.9	2.9	−	5.3
13Ni/10Ca-Al2O3	27.7	55.2	3.8	−	13.4	21.2	65.5	6.5	−	6.8
13Ni/10Mg-Al2O3	33.3	55.6	4.8	−	6.3	30.6	58.1	5.2	−	6.1
13Ni/14Ce-Al2O3	32.3	60.8	1.9	−	5.1	23.7	67.5	4.2	−	4.5
1Ru/14La-Al2O3	38.1	57.3	−	0.6	3.5	28.5	67.0	−	0.7	3.1
1Ru/Al2O3	42.1	55.9	−	0.5	−	40.2	58.2	−	1.1	−

* Presence of Ni cannot be quantified (2p line of Ni overlapped with 3d line of La).

. This increment is especially significant in catalysts with Ca and Ru-La in their composition, assuming an increment of more than 10%. This increment in the oxygen amount is compensated by a reduction in the same proportion of the Al content in 1Ru-13Ni/14La-Al₂O₃ and 1Ru/14La-Al₂O₃ catalysts, and a significant reduction in the proportion of the secondary metal in 13Ni/10Ca-Al₂O₃ and 1Ru-13Ni/14La-Al₂O₃ catalysts (Figure 8). This is due to metal sintering, which produces an agglomeration in deeper layers of the catalyst. On the other hand, although the increment in the proportion of oxygen is not as marked in 13Ni/14Ce-Al₂O₃ and 13Ni/14La-Al₂O₃ catalysts, the reduction in the proportion of Al is up to 9%. Finally, it is necessary to highlight the increase that occurs in the proportion of Ni after its use in the reaction, in 13Ni/14Ce-Al₂O₃ and 13Ni/10Ca-Al₂O₃ catalysts, raising the proportion to double its value. Something similar happens with Ru in 1Ru/Al₂O₃ catalyst.

The XPS results show that the Al present in used catalysts is found primarily as aluminate. The aluminate is attached to the active metal in its oxidized form, giving rise to a strong interaction. Due to the effect of high temperature and the action of H₂S, it is possible that the migration of the strongly oxygen-bound metal from the aluminate to the surface occur. Despite the increased exposure of the metal, its interaction with the support is likely to block its catalytic capacity. Due to the strong structural change, despite the removal of the H₂S, the catalyst will probably not recover its activity.

3.1.6. XRD

The identification of the main crystalline species was possible through the study of the XRD patterns presented in Figure 9 and Figure 10, corresponding to the XRD patterns of the fresh and used catalysts studied, respectively. The γ-Al₂O₃ support exhibits its diffraction peaks at 37.5°, 44.5°, and 67.4° mainly, corresponding to the crystal planes of (331), (400) and (440), respectively [63,64]. Peaks at 19.7° (111), 34° (220), 57° (422), 61° (511), and 85° (444) can also be used for the identification of the alumina, although they are more difficult to detect specially in the presence of other metals [63,64]. The diffraction peaks of the support observed in the XRD pattern after using the catalysts in the methane reaction up to 773 K, with a current containing H₂S, do not show significant variations.

The diffraction peaks obtained before and after the reaction stage coincide in the same position and without variation of the determined crystal size. In the case of metal promoters, Ni and Ru are in their metallic form both in the reduced fresh and used catalysts, their oxidized form not being detected. Ni peaks are observed at 44.5° (111), 51.6° (200), and 76.7° (220) [17]. In the case of Ru, the main peak is shown at 44.2° (101), the peaks of positions 38.5° (100), 42.3° (002), 78.6° (103), and 84.9° (004) also being part of its pattern [65,66]. Due to the low amount of Ru added to the catalysts, the peaks of this metal disappear [67] in the presence of other promoters, such as Ni in this case. In the presence of the support, it is possible to identify some peaks overlapped by those of the support, especially at 44.2°. This is the only peak located in the presence of La as a support modifier. The lanthanum detected in the XRD pattern is the associated one with the support, in the form of Al₂₂La₂O₃₆ and La₃Al, the La₂O₃ phase cannot be identified, which implies its high dispersion on the surface of the support [35]. After the reaction step, the La₂O₃ dispersed into the 13Ni/14La-Al₂O₃ catalyst is transformed into the La₂O₂CO₃ species. In the rest of the catalysts other La compounds interacting with the support were not identified, remaining as La₂O₃ dispersed in the support.

The addition of La, Ce, and Ba to the Ni/Al₂O₃ catalysts produces the damping of the Ni reflection peaks, due to the dispersion of the metal in the support [17,28,29]. The catalyst with Ce has a similar XRD pattern before and after participating in the reaction, identifying the diffraction peaks of CeO₂ in both spectra.

The XRD patterns of the alkaline earth studied in the reduced catalysts show the metals in their metallic or oxidized form: Ba, CaO, and MgNiO. After the reaction, the spectrum of the Mg hardly changes, identifying Mg₀.₄Ni₀.₆O present in fresh catalyst in the form of MgO in used catalyst. Calcium remains as CaO, which, as in the catalyst before the reaction, is difficult to identify, due to the coincidence of the peaks that allow its identification with those of the support, much more abundant. However, it is possible to identify in this case the nickel carbide that appears, with the reduction in the intensity of the Ni peaks, as a result of the disappearance of part of the Ni to yield NiCx.

Compared to the previous ones, barium is transformed in the reaction stage into two different compounds: BaCO₃ and BaAl₂O₄. These species exhibit certain peaks represented in Figure 10, with low intensity, which implies a moderate formation of these elements.

The 2θ position of the diffraction peaks of the XRD patterns determined by X-ray diffraction for the fresh and reduced catalysts (Figure 9), and for the used catalysts (Figure 10) are collected in

Table 4. XRD peaks for the metallic species of the catalysts analyzed, determined graphically and contrasted with the bibliography.

Metallic Species	JCPDS Code	Value (2θ)	Bibliography
Al2O3	077-0396	19.7, 34, 37.5, 44.5, 57, 61, 67.4, 85	[63,64]
Ni	087-0712	44.5, 51.6, 76.7	[17]
Ru	088-1734	38.5, 42.3, 44.2, 78.6, 84.9	[65,66]
Ba	006-0235	24.1, 34.5	[28,53]
CaO	086-0402	32.2, 37.3, 53.8, 64.1, 67.3	[34,68]
Mg0.4Ni0.6O	034-0410	37.4, 43.5, 63, 75.5, 79.4	[69,70]
MgO	087-0653	37.5, 43.5, 63, 75.7, 79.2	[69,70]
CeO2	075-0076	28.9, 33.4, 47.7, 56.6, 76.8	[65,71]
Al22La2O36	028-0502	25.8, 34.1, 37.5, 44.5, 45.9, 51.6	[29]
LaAlO3	031-0022	25.3, 30.8, 44.1, 51.3, 56.7	[29]
BaAl2O4	017-0306	19.7, 22.2, 28.4, 34.5, 37.5, 40.3, 42.2, 45.1, 45.9, 54.7, 58, 61.5, 67.2, 76.4	[34,53,72]
BaCO3	005-0378	19.7, 24, 24.4, 34.5, 40.3, 42.2, 44.6, 45.1, 54.7, 61.5, 76.4	[34,72]
La2O2CO3	025-0424	26, 30.5, 33.74, 44.6, 51.82	[73]

and contrasted with the bibliography.

The average crystallite size of species present in the catalysts was estimated employing the Scherrer equation and a significant peak of the XRD patterns. The peaks used and the crystal size determined are collected in Table 1 and

Table 5. Crystal size determined by Scherrer equation and a diffraction peak of XRD pattern for the species present in the reduced and used in reaction catalysts.

Species	Peak Position (°)	Crystal Size Approximation of Reduced Catalyst (nm)	Crystal Size Approximation of Used Catalyst (nm)
Ru	44.2	20	5 (20 for the 1Ru/Al2O3)
Ni	51.9	<5	15

. After the reaction stage, an increase in crystal size is observed in all the species analyzed, with the exception of Ru. The increase in Ni crystal size is a consequence of the sintering that takes place in the catalyst.

3.2. Activity Tests

The catalytic activity was measured as the CH₄ yield obtained for each temperature analyzed in the CO₂ methanation. The activity study was carried out primarily for Ni-based catalysts supported on alumina, and modified with lanthanides (La or Ce, 14 wt%), and alkaline earth metals (Ba, Ca, or Mg, 10 wt%). This study was done for temperatures between 573 and 773 K, with a quantity of catalyst of 200 mg, corresponding to a weight hourly space velocity (WSHV) of 38.3 h⁻¹. The results obtained for the methane yield are reported in Figure 11.

All the catalysts studied reaches 100% CH₄ selectivity at the experimental conditions, detecting only trace amounts of CO as a by-product. The maximum methane yield value was reached and established by the thermodynamic equilibrium in the temperature range. In the case of 13Ni/10Mg-Al₂O₃ catalyst, it needed a higher temperature to reach the equilibrium value, 700 K. This was followed by 13Ni/14Ce-Al₂O₃ and 13Ni/10Ba-Al₂O₃ catalysts, which reached it at 639 K and 598 K, respectively. The rest of the catalysts started from the thermodynamic equilibrium value. To facilitate the comparison, the amount of catalyst was reduced to 125 mg to increase the WSHV value to 60.8 h⁻¹.

On the other hand, the resistance of the catalysts to H₂S poisoning was previously evaluated. For this, 50 ppm of H₂S were introduced with the reaction gases, determining the time necessary for the complete deactivation of the catalysts, results are represented in Figure 12.

The catalysts with the greatest resistance to deactivation were those modified with lanthanides (13Ni/14La-Al₂O₃ and 13Ni/14Ce-Al₂O₃), requiring up to 110 min for complete deactivation. On the contrary, those catalysts modified with alkaline earth deactivated the first ones, starting with 13Ni/10Ba-Al₂O₃ at 40 min, followed by 13Ni/10Mg-Al₂O₃ at 50 min and finally 13Ni/10Ca-Al₂O₃ at 60 min. The 13Ni/Al₂O₃ catalyst showed an intermediate resistance, lower than those modified with lanthanides, deactivating at 70 min.

The results obtained using 125 mg of catalyst, and a temperature range between 498 and 573 K, are shown in Figure 13.

As expected, the reduction in the amount of catalyst (or increase in WSHV) increased the temperature necessary to achieve the maximum performance established by thermodynamic equilibrium. In the case of supports modified with alkaline earth metals (13Ni/10Ba-Al₂O₃, 13Ni/10Ca-Al₂O₃, and 13Ni/10Mg-Al₂O₃) and the reference catalyst (13Ni/Al₂O₃), the activity increased progressively until reaching the thermodynamic equilibrium at the highest temperature studied. 13Ni/Al₂O₃ and 13Ni/10Mg-Al₂O₃ catalysts reached a yield of 92 and 89% at 668 and 665 K respectively, while 13Ni/10Ba-Al₂O₃ and 13Ni/10Ca-Al₂O₃ catalysts reached 86 and 90% at 683 and 688 K, respectively. Moreover, according to the ICP characterization results, these catalysts presented the lower Ni content. The catalysts modified with lanthanide metals achieved the equilibrium yield at lower temperatures. 13Ni/14Ce-Al₂O₃ reached a methane yield of 96% at 618 K. However, catalyst 13Ni/14La-Al₂O₃ at 543 K yielded a 90%, reaching up to 98% at 572 K with a medium content of Ni measured by ICP. As a previous conclusion, cerium and specially lanthanum addition enhanced the activity to methane with respect to Ce or La-free and alkaline earth modified catalysts.

The high yields of catalyst 13Ni/14La-Al₂O₃ at low temperatures, would allow the methanation of a greater flow of reaction gases or, the use of a smaller amount of catalyst. Therefore, the feasibility of this hypothesis is compared in the case of catalyst 13Ni/Al₂O₃, for different amounts of catalyst. Figure 14 compares the methane yield of catalyst 13Ni/Al₂O₃ for 200 mg of catalyst and 38.3 h⁻¹ of WSHV; 125 mg and 60.8 h⁻¹; and 60 mg and 128.75 h⁻¹.

The decrease in the amount of catalyst used resulted in an increase in the temperature at which the thermodynamic equilibrium is reached, as previously observed. With a load of 200 mg of catalyst in the reactor, the equilibrium temperature was reached at a temperature of 572 K, while for 125 mg, the temperature needs to be raised to 668 K. Finally, for an amount of 60 mg the equilibrium temperature is not reached until 745 K was reached.

Thus, 60 mg of catalyst (128.75 h⁻¹), and a temperature range between 598 and 773 K, was used to compare the improvement in the activity of catalysts based on Ru-, Ni-, and La-supported on alumina. Results are shown in the Figure 15.

Catalyst 13Ni/Al₂O₃, as seen in Figure 15, achieved a maximum methane yield of 90% at 745 K. The addition of La reduced the value of the temperature necessary to reach a yield of 82% to 680 K (and 89% at 727 K). Substituting nickel for a noble metal, such as Ru, at a nominal content of 1 wt% with La-modified support, it reached 80% yield at 694 K. The difference between the 1Ru/14La-Al₂O₃ and 13Ni/14La-Al₂O₃ catalysts is quite small, the Ni-containing catalyst being slightly better. The incorporation of Ru into the 13Ni/14La-Al₂O₃ catalyst produced a clear improvement in the methanation rate with respect to the other catalysts, despite the decrease in the amount of Ru incorporated as observed in the ICP results. At 581 K, the methane yield reached by this catalyst was 74%, reaching a maximum value of 89% at 651 K. Therefore, although the contribution of Ru appears to be similar to that of Ni, the effect of Ni is greater than that of Ru, especially in combination.

Once the activity of the catalysts in the methanation reaction has been studied, as for catalysts modified with alkaline earth metals, the resistance of the catalysts to poisoning in the presence of 50 ppm of H₂S is determined. The results obtained for the deactivation of the catalysts are presented in Figure 16.

The addition of 50 ppm of H₂S to the reaction gases reduced the activity of catalysts containing Ru faster than those containing only Ni. Thus, 1Ru/14La-Al₂O₃ and 1Ru-13Ni/14La-Al₂O₃ were deactivated in 40 min. The addition of La to Ni, in addition to improving the methane activity of the catalysts, improved its resistance to H₂S poisoning, lengthening the deactivation from 70 min of catalyst 13Ni/Al₂O₃ to 110 min of catalyst 13Ni/14La-Al₂O₃.

After deactivation of the catalysts in the presence of H₂S, the methane yield of the catalysts was studied by eliminating H₂S from the feed at 773 K. In all the cases analyzed, the catalysts did not recover the activity.

Finally, catalysts were regenerated, employing a mixture of 3% O₂ and N₂, in order to eliminate the coke or the sulfur that may block the catalytic action of the catalysts. Then, the catalysts were reactivated by reduction with H₂. The recovered catalyst was used in a reaction stage, at 773 K. The catalyst activity was observed as methane yield was not recovered [37,74]. During the reaction or deactivation stage, CO could have appeared as a by-product of the reverse water gas shift reaction, but it could also appear by the partial oxidation of the coke in the regeneration. In spite of everything, the presence of this gas was not detected.

3.3. Discussion

The use of noble metals as catalyst in the methanation reaction has been extensively studied in the literature, demonstrating high efficiency with greater thermal stability, as observed in Garbarino’s work [13]. Despite showing that the performance achieved for nickel-based catalysts is not as high as that obtained for ruthenium, the performance is high and with easier activation [10,13,50]. In order to complement the advantages of nickel and ruthenium as active metals in the methanation reaction, it was co-impregnated on the support. Thus, it was verified that the addition of ruthenium and nickel improved the performance of the catalyst, improving thermal stability and facilitating its activation, as determined in Liu’s work [75].

The incorporation of alkaline earth metals and lanthanides that modified the support has been shown to reach higher activity, improving the interaction between support and metal [28,29]. The incorporation of alkaline earth metals has improved the activity of the nickel catalyst, but less significantly than adding lanthanide metals. The alkaline earth metals could increase the alkalinity of the catalyst, resulting in a reduction in the number of acid centers with low and medium strength and an increase in the basic centers with medium and high strength. This results in less interaction of nickel with alumina than observed for lanthanide-modified catalysts, which favors catalyst activity. Taking into account the obtained results for the alkaline earth metals catalyst, the general properties (as textural properties) analyzed are similar to those of the monometallic catalyst, concluding with a similar activity. However, it is necessary to specify that according to H₂-TPR results, the addition of barium facilitated the activation of the catalyst, contrary to what was observed for calcium and magnesium. These results are in a good agreement with those obtained in Liang’s work [28], in which little improvement was observed in the catalytic activity of the nickel catalyst in the presence of Ca or Mg. The methane yield determined for catalysts of Ni and Ni with 5% of Ca or Mg were around 50% in these cases at 623 K. However, the addition of Ba (from 1 to 7.5%) showed a reduction in the temperatures at which the catalyst is active for the methanation reaction, reaching a yield of 80% at 623 K, incorporating in that case 7.5% of Ba. The differences in performance obtained in our catalysts using 125 mg of catalyst instead of 500 mg are less significant, reaching a methane yield close to 50% in all of them.

A significant improvement in the activity of the nickel catalyst was produced by the addition of lanthanide metals, especially with La. Ce was incorporated as non-stoichiometric CeO₂, surface reducible metal centers of CeO₂, and bulk reducible metal centers of CeO₂ [52]. Ce⁴⁺ from CeO₂ interacted with Ni, as observed in the H₂-TPR analysis, improving its reducibility which facilitated its activation. There is a strong interaction with Ni, determined by XRD, which contributes to the strengthening of the interaction with the support and improved the dispersion of the metal on its surface, increasing the catalytic activity of the catalyst. Previous studies from our laboratory [19] have shown that the addition of cerium promotes the formation of active species, of nickel-aluminate spinel-like structure, which improves the yield of methane up to 75% at 647 K for the 13Ni/6Ce-Al₂O₃ catalyst, a temperature for the Ni catalyst of 723 K being necessary to achieve the same performance.

The addition of lanthanum, as well as cerium, produced an improvement in the reducibility of the catalyst that facilitated its activation, increasing the amount of free NiO dispersed on the surface of the alumina [29,45,51]. This was caused due to a weakening of the interaction between NiO and the support by destroying partially metastable Ni-Al mixed oxide phase, caused by La incorporation [45,51]. On the other hand, La interacted with the strongest acid and basic centers of alumina, through La³⁺-O²⁻ pairs, as was observed by NH₃-TPD, increasing the density of acid centers of higher strength [41]. La contribution was especially important in Ru and Ni catalysts, in which the number of acid centers increased more significantly. It implied a higher dispersion of the active metals on the surface of the support, as verified in the results obtained by XRD. In this analysis, La was dispersed, interacting with the support, forming species such as Al₂₂La₂O₃₆ and La₃Al. The Ni and/or Ru centers, facilitated by the structures made up of La and alumina, were dispersed over the surface. It helped to increase the activity of the catalyst in the methanation reaction, as it was deduced from the previous analyses and it was verified in the activity results. The increase in the yield to methane was observed from 34% for 13Ni/Al₂O₃ to 65% for 13Ni/14La-Al₂O₃ at 650 K, just like in Garbarino’s work [29], in which this important increase in activity was observed by the addition of lanthanum to the nickel catalyst, raising the yield from 72 to 90% to 625 K.

The most outstanding observation deduced from our work was the importance of the incorporation of Ru as a promoter, together with Ni, for the improvement of the activity of the catalyst in the reaction. An increase in yield from 65% for 13Ni/14La-Al₂O₃ to 90% for 1Ru-13Ni/14La-Al₂O₃ was observed at 650 K. This increase was probably due to the synergistic effect that occurs between Ru and Ni, due to the interaction that takes place between said metals as observed in H₂-TPR and verified with the Liu’s work [75]. Ru dispersed as RuO₂ and RuO_x across the surface of the support in contact with the NiO present, being able to produce Ru-Ni bimetallic particles [50]. Ru decreased the reduction temperature of Ni, favoring its activation and facilitating its dispersion, which led to increased activity as shown, and contributing to this effect La.

Regarding the resistance of the catalyst to poisoning with H₂S, it was observed that the incorporation of alkaline earth metals exposed to a greater extent the active centers of the metal to the H₂S, facilitating its deactivation. The incorporation of lanthanide metals improved resistance to H₂S poisoning, due to the strong interaction between nickel and Ce or Ni and La, which makes it difficult to approach and block these centers, by reducing their affinity. Ni is more resistant to poisoning than Ru, which, in the presence of H₂S, was rapidly deactivated due to its greater affinity for this element (more electronegative) than for H₂.

Once the catalyst was deactivated, the activity was not recovered, due to the morphological change that follows in the catalyst, in view of the analyzed XPS and XRD results. An increase in the amount of oxygen and nickel at the catalyst surface was observed, along with a reduction for aluminum. This could involve the formation of an aluminate structure that covers the nickel, blocking the access of the reagents and their interaction, as the presence of sulfur in their content was not determined. Under these conditions, the catalyst could completely lose its activity, preventing its recovery.

4. Conclusions

The conclusions summarized below have been deduced from the development of this work.

The addition of Ba, Ca, or Mg in a proportion of 10% does not improve the activity of the nickel catalyst in a significant way, probably due to the slight differences obtained in the physico-chemical characterization.

The addition of 14% of Ce improves the reducibility and dispersion of nickel, as suggested in the characterization, by facilitating new anchor centers on the surface of the support, which improves its activity at lower temperatures.

La (incorporated in a proportion of 14%) shows a strong interaction with alumina, which facilitates the incorporation of Ni. Greater dispersion and higher density of active centers were obtained, composed of easily reducible NiO species, which probably were responsible of the improvement in activity at lower temperatures.

La-doped Ni/Al₂O₃ catalyst is competitive with Ru/Al₂O₃ catalyst. The incorporation of ruthenium as a promoter produces a strong interaction with nickel, increasing its catalytic activity, reinforced by lanthanum, which increases dispersion and reducibility reaching a 90% methane yield at 650 K for a WSHV of 128.75 h⁻¹.

The strong interaction between La or Ce with Ni hinders the exposure and access of the H₂S to the active centers of the metal.

The high temperatures (773 K) and the action of the H₂S produced the complete deactivation of the catalysts studied. A possible change in the morphology of the catalyst was determined, whereby the oxygenated structures of the support and the metals they incorporate, block access to the metal centers of the reagents.