CO2 methanation over Ca doped ordered mesoporous Ni-Al composite oxide catalysts: The promoting effect of basic modifier

Journal of CO₂ Utilization 21 (2017) 200–210

Contents lists available at ScienceDirect

Journal of CO2 Utilization journal homepage: www.elsevier.com/locate/jcou

CO2 methanation over Ca doped ordered mesoporous Ni-Al composite oxide catalysts: The promoting effect of basic modifier ⁎

MARK

⁎

Leilei Xua, , Haoming Yanga, Mindong Chena, , Fagen Wangb, Dongyang Niea, Lu Qia, Xinbo Liana, Hanxiang Chena, Mei Wuc a Collaborative Innovation Center of the Atmospheric Environment and Equipment Technology, School of Environmental Science and Engineering, Nanjing University of Information Science & Technology, Jiangsu Key Laboratory of Atmospheric Environment Monitoring and Pollution Control, 210044, Nanjing, PR China b School of Chemistry and Chemical Engineering, Jiangsu University, 301 Xuefu Road, Zhenjiang 212013, PR China c Key Laboratory for Palygorskite Science and Applied Technology of Jiangsu Province, Faculty of Chemical Engineering, Huaiyin Institute of Technology, Huai’an 223003, PR China

A R T I C L E I N F O

A B S T R A C T

Keywords: Ca modification CO2 activation Ni catalyst Low-temperature activity CO2 methanation

Ni is usually used as the catalyst for the CO2 methanation to generate the substitute natural gas (SNG) due to the low cost. However, it often possesses relatively worse low-temperature catalytic activity than the noble metal counterparts. In order to solve this problem, a series of Ca doped Ni based ordered mesoporous materials had been facilely fabricated via the one-pot evaporation induced self-assembly (EISA) strategy and directly used as the catalysts for CO2 methanation. These materials had been systematically characterized by XRD, N2 adsorption-desorption, TEM, SAED, EDS, XPS, H2-TPR, and CO2-TPD measurements. It was discovered that the large specific surface areas (210.7–240.7 m2/g), big pore volumes (0.37–0.46 cm3/g), and narrow pore size distributions (8.3–9.5 nm) of these materials had been successfully retained after 700 °C calcination. In these materials, Ni species were homogenously dispersed among the Al2O3 matrix via the one-pot fabrication strategy and the strong interaction between Ni and mesoporous framework had been formed. Thus, the seriously thermal sintering of the metallic Ni nanoparticles could be successfully inhibited. Hence, they displayed no deactivation after 50 h stability test toward CO2 methanation. More importantly, the doping of the Ca greatly enhanced the surface basicity, which could favor the chemisorption and activation of CO2. As a result, the apparent activation energies of CO2 could be remarkably decreased from 75.2 to 53.6 kJ/mol and the low-temperature catalytic activity had been significantly promoted. Therefore, these Ca doped Ni based ordered mesoporous materials promised potential catalysts for CO2 methanation.

1. Introduction In recent years, the anthropogenic emission of CO2 drastically increases because of the combustion of the fossil fuels and subsequently causes the global climate change via its greenhouse effect [1,2]. Thus, it is of great urgency as well as significance to reduce the CO2 concentration in the atmosphere. The transformation of CO2 into valueadded fuels and chemicals is of great importance from the perspectives of recycling the CO2 resource and processing the environmental issues [1,3,4]. The hydrogenation of CO2 by sustainable hydrogen source to generate methane (CO2 methanation), also known as Sabatier reaction, has been considered as one of the most promising CO2 recycling utilization routes [5–8]. By means of this approach, CH4 can be theoretically generated under mild conditions due to its thermodynamically favorable feature ((CO2(g) + 4H2(g) = CH4(g) + 2H2O(g),

⁎

Corresponding authors. E-mail addresses: [email protected] (L. Xu), [email protected] (M. Chen).

http://dx.doi.org/10.1016/j.jcou.2017.07.014 Received 31 May 2017; Received in revised form 7 July 2017; Accepted 17 July 2017 Available online 15 September 2017 2212-9820/ © 2017 Elsevier Ltd. All rights reserved.

ΔG = −173.1 + 0.1983T kJ/mol)) [5,7]. Nevertheless, the full reduction of the CO2 (+4) into CH4 (-4) is an eight-electron process with greatly kinetic barrier, which requires an efficient catalyst to achieve an acceptable CO2 conversion, especially at low reaction temperature [5]. Therefore, the development of effective and stable catalysts is an important concern. Although the noble metal catalysts, such as Ru [9–12], Rh [13,14], and Pd [15,16], possess outstanding low-temperature catalytic activity as well as stability toward CO2 methanation, their largescale industrial application will be restricted due to the high cost. Therefore, it is more practical to design and develop Ni based non-noble transition catalysts with outstanding low-temperature activity [17–20], which has gradually attracted increasing attention for their comparable catalytic activity and availability. Compared with the noble catalysts, the Ni based catalysts usually perform poorer low-temperature catalytic activity. Great efforts have

Journal of CO₂ Utilization 21 (2017) 200–210

L. Xu et al.

would intensify the capacity of CO2 chemisorption. As a result, compared with the pristine Ni-Al reference catalyst, the low-temperature catalytic activities of the Ca modified catalysts had been noticeably promoted. Furthermore, the kinetic investigation indicated that the doping of Ca could greatly decrease the apparent activation energies. Besides, these ordered mesoporous catalysts displayed excellent stability up to 50 h without deactivation due to the outstanding anti-sintering property of the metallic Ni via the confinement effect of mesoporous framework. Generally, the cooperative effect between the Ni active site and Ca basic promotor synergistically contributed to the enhanced catalytic performance.

been devoted to exploring the effective and stable Ni catalysts. For CO2 methanation, it has been widely believed that the activation of the stable carbon dioxide molecule could finally determine the reaction rate of the whole reaction [1,5]. Therefore, the catalytic activity at low temperature was desirable to be improved through enhancing the process of CO2 activation. Thus, the basic species, such as alkaline-earth and rare earth elements, have been doped into the catalysts by intensifying the chemisorption of the CO2, which will further decrease the activation energy of CO2 [21–23]. The influences of the basic categories and quantities have been extensively investigated. It has been found that the promoting effect of the basic modifier on the catalytic performance is greatly determined by its surface properties, such as intensity, type, etc [23–26]. Pioneering literatures confirmed that the basic modifier could initiate the methanation reaction by binding a CO2 molecule to form the carbonate species by ambient-pressure XPS, in situ DRIFT characterizations, and DFT calculations, which would be followed by hydrogenation with dissociated hydrogen to generate methane [9,23,27]. For example, Liu et al. reported that the Nix/Mg2-xAlMMO (mixed metal oxides, MMO) catalysts with hydrotalcite structure exhibited enhanced low-temperature catalytic activities toward CO2 methanation [23]. They found that the incorporation of Mg alkaline promotor could remarkably intensify the medium-strong basic sites, which served as the active sites for converting CO2 into carbonate/ hydrocarbonate intermediates. This could decrease the energy barrier for CO2 activation, finally accounting for enhanced low-temperature activity. The activation of the H2 molecule is also another important concern for enhancing the catalytic activity and CH4 selectivity toward CO2 methanation. It is widely believed that the H2 molecule is dissociated over the surface of metallic Ni active sites [19,28], which is greatly influenced by the metal dispersion. The high dispersion of the metallic Ni is beneficial to the facile dissociation of the H2 by exposing sufficient active sites [19]. In order to improve the dispersion of Ni based catalysts, the porous materials are commonly selected as the candidates of catalytic supports. For instance, Du et al. reported that the excellent catalytic performance (96.0% CH4 selectivity, 91.4 g/(kg h) space-time yield) comparable to Ru/SiO2 catalyst could be achieved over 1%Ni/ MCM-41 catalyst (1140 m2/g) at a space velocity of 5760/(kg h) owing to the high dispersion metallic Ni up to 100% [29]. Zhen et al. reported the CO2 methanation over Ni/MOF-5 (2961 m2/g) and Ni/MIL-101 (3297 m2/g), on which 41.8% and 42.3% metal dispersion could be achieved [30,31]. As a result, the highly dispersed Ni nanoparticles confined in metal-organic frameworks of MOF-5 and MIL-101 endowed these catalysts with enhanced low-temperature catalytic activities (e.g. 47.2% CO2 conversion at 280 °C over Ni/MOF-5) and excellent stabilities as long as 100 h. Therefore, the high dispersion of Ni can be facilely obtained by employing porous materials as supports, which will subsequently contribute to the promotion of the catalytic performances. Based on this viewpoint, the ordered mesoporous Ni-Al composite oxide catalyst for CO2 methanation has already been designed and synthesized via one-pot evaporation induced self-assembly (EISA) strategy in our previous study [32], where the Ni active sites were homogenously embedded among the Al2O3 matrix. The tiny metallic Ni nanoparticles could be obtained by the in-situ reduction and stabilized by the confinement effect of the mesoporous framework. Therefore, compared with the conventional Ni/γ-Al2O3 supported catalyst, the NiO-Al2O3 mesoporous catalysts displayed much better catalytic activity and stability. In this work, so as to further improve the low-temperature catalytic activity of ordered mesoporous Ni-Al catalyst, the Ca basic promoter has been successfully doped into the mesoporous framework via the one-pot EISA strategy. The final Ca doped ordered mesoporous Ni-Al composite oxides with different Ca/Al molar ratios (0–10%) were carefully characterized by various techniques and directly used as the catalysts for CO2 methanation. The introduction of Ca could dramatically increase the surface basicity based on CO2-TPD analysis, which

2. Experimental 2.1. Materials The anhydrous C2H5OH (Sinopharm Chemical Reagent Co. Ltd.), (EO)20(PO)70(EO)20 triblock copolymer (Pluronic P123, Mn = 5800, Sigma-Aldrich), 67% HNO3 (Sinopharm Chemical Reagent Co. Ltd.), aluminum isopropoxide (C9H21AlO3, 98+%, Sigma-Aldrich), Ni (NO3)2·6H2O (Sigma-Aldrich), Ca(NO3)2·4H2O (Sigma-Aldrich) were employed as the solvent, structure directing agent (SDA), acid modifier, and precursors, respectively. All of the chemicals were directly employed without further purification. 2.2. The synthesis of the Ca doped ordered mesoporous Ni-Al composite metal oxides The Ca doped ordered mesoporous Ni-Al composite oxides with various Ca/Al molar ratios were fabricated by improved one-pot EISA strategy reported elsewhere [33–35]. In these materials, the Ni/Al molar ratios were fixed at 10% and the Ca/Al molar ratios were controlled at x% (x = 0, 1, 3, 5, 8, and 10), respectively. These mesoporous materials were denoted as OMA-10NixCa, where the “OMA”, “10”, and “x” referred to the “mesoporous alumina matrix”, “Ni/Al molar ratio”, “Ca/Al molar ratio”, respectively. In a specific synthesis procedure, 1.0 g P123 was completely dissolved in 20.0 mL anhydrous ethanol with vigorous agitation. 1.6 mL 67% HNO3, 10 mmol Al(iso-OPr)3, 1 mmol Ni(NO3)2·6H2O, and certain amounts of Ca(NO3)2·4H2O were sequentially added into the above P123-ethanol solution and vigorously stirred for 5 h. The obtained transparent solution was transferred into a 60 °C convection oven with low relative humidity (< 50%) to conduct the EISA process for 48 h. The obtained xerogels were further calcined at 700 °C for 5 h with 1 °C/min ramping rate in Muffle furnace under static air atmosphere. Finally, these OMA-10NixCa materials would be directly investigated as the catalysts for CO2 methanation. 2.3. Catalysts characterization Small-angle (0.5–5.0°) and wide-angle (20–80°) powder X-ray diffraction (XRD) patterns of the samples were obtained on a PANalytical X’pert Pro multipurpose diffractometer, using Ni-filtered Cu Kα radiation (λ = 0.15046 nm) at 40 kV voltage, 40 mA current, and a step size of 0.02°/s. N2 adsorption-desorption analyses were carried out using a Quantachrome NOVA 2200e instrument with two analysis stations. Prior to N2 adsorption, the samples were outgassed at 200 °C for 4 h to desorb moisture and impurity on the surface of the sample. Metal elements analysis was carried out using an Optima 7300DV (Perkin Elmer) inductively coupled plasma–atomic emission spectrometer (ICP−AES). Transmission electron microscopy (TEM) observation, selected area electron diffraction (SAED), and energy-dispersive spectroscopy (EDS) were carried out on a JEOL 2010F transmission electron microscope under a working voltage of 200 kV. X-ray photoelectron spectroscopy (XPS) analyses of the catalysts were carried out on a VG ESCALAB 210 (Thermo Scientific) spectrometer. The fresh catalyst powder was loaded on sample holder with the conductive adhesive tape. The binding 201

Journal of CO₂ Utilization 21 (2017) 200–210

L. Xu et al.

energies were calibrated using the C 1 s line at 284.8 eV as the reference. Hydrogen temperature programmed reduction (H2-TPR) measurements were conducted in a U-type quartz tube reactor on a Quantachrome ChemBET Pulsar TPR/TPD analyzer with a thermal conductivity detector (TCD). In a typical H2-TPR process, 0.1 g of a sample placed in a quartz tube reactor was first purged under flowing He (50 mL at 300 °C) for 1 h and then cooled down to room temperature. Then, a gaseous mixture of 5 vol%H2-95 vol%He was fed to the reactor at 80 mL/min. The temperature was raised to 1100 °C at a ramping rate of 10 °C/min and the amount of H2 consumption was recorded by the TCD. CO2 temperature-programmed desorption (CO2TPD) measurements were carried out on the Micromeritics AutoChem II 2920 Chemisorption analyzer. For the typical CO2-TPD process, 0.1 g of the sample was first pretreated in a reactor under a He stream (50 mL/ min) at 300 °C for 1 h. After cooling down to 25 °C, the pretreated sample was exposed to CO2 atmosphere (50 mL/min) for 1 h to make the sample surface saturated with the chemisorbed CO2. Then, the sample was purged with He stream to remove the physically absorbed CO2 until the baseline was steady. Finally, the CO2-TPD was carried out with a ramping rate of 20 °C/min to 900 °C under He stream (50 mL/ min) and the amount of CO2 desorption was also recorded by the TCD. 2.4. Catalytic evaluation The catalytic performances of these Ca doped Ni-based ordered mesoporous Ni-Al catalysts were evaluated under atmospheric pressure in a fixed-bed quartz reactor with an interior diameter of 10 mm. The gas flows of the H2 and CO2 were separately controlled by two mass flow controllers. The reaction temperature was monitored and controlled by a thermocouple located at the center of the catalyst bed. Prior to the regular evaluation tests, the blank test had been carried out in the range of 200–450 °C, performing negligible catalytic activity. For the catalytic performance evaluation test, 0.1 g of the catalyst was firstly in situ reduced in a gaseous mixture of H2/N2 (20/10 mL/min) for 2 h at 800 °C with a ramping rate of 1 °C/min, and then cooled to 200 °C in nitrogen stream to remove chemisorbed H2. Subsequently, a mixture of H2/CO2 with a molar ratio of H2/CO2 = 4/1 was introduced into the reactor without any dilution at a gas hourly space velocity (GHSV) of 15,000 mL/(g h). 50 h lab-scale stability tests were conducted at 400 °C with the GHSV of 15,000 mL/(g h) under atmospheric pressure. Before being introduced into the on-line GC system, the effluent product gases should flow through a cold trap to completely remove the water and thoroughly mix with the N2 (5 mL/min), which acted as an internal standard gas for calculation. The outlet gases was injected via the online automatic six-port valve and analyzed using an Agilent 7890 B gas chromatograph equipped with TCD and FID detectors. Nevertheless, except for methane, there was no other organic species detected by FID detector. The CO2 conversion (denoted as CCO2) and CH4 selectivity (denoted as SCH4) were calculated based on the below formulas. CCO2 = (FCO2,inlet − FCO2,outlet)/FCO2,inlet × 100%

(1)

SCH4 = FCH4,outlet/(FCH4,outlet + FCO,outlet) × 100%

(2)

Fig. 1. (1) Small-angle and (2) wide-angle X-ray diffraction of the as-prepared OMA10NixCa catalysts.

3. Results and discussion 3.1. Structural and morphological study of the fresh OMA-10NixCa catalysts 3.1.1. XRD analysis The small-angle XRD patterns of the fresh OMA-10NixCa catalysts were displayed in Fig. 1(1). As can be observed, all the OMA-10NixCa materials with various Ca/Al molar ratios presented a strong (1 0 0) peak around 0.8° together with a weak (1 1 0) peak around 1.5°, revealing that the two-dimensional p6mm hexagonal ordered mesoporous structures had been successfully constructed among the frameworks of OMA-10NixCa after the process of EISA [33]. Furthermore, it was worth noting that the doping of Ca did not damage the orderliness of the mesoporous structure. The d-spacing of the (1 0 0) direction for these materials was calculated based on the Bragg’s law and their values were summarized in Table 1. It was noticeable that all the samples displayed their d(100) values in the range of 11.0–11.3 nm, which were attributable to the mesopore according to the definition of the IUPAC [36]. As regards their wide-angle XRD patterns in Fig. 1(2), there was no apparent NiO, CaO, and Al2O3 diffraction peaks observed. This phenomenon indicated that these species were highly as well as homogenously dispersed among the ordered mesoporous framework due to the one-pot EISA method.

In all formulas, the Fx,inlet and Fx,outlet represented the flow rate of the x species, which flowed into and out of the reactor, respectively. Furthermore, the kinetic experiments were also conducted in the same fixed-bed reactor at atmosphere pressure. To minimize the effects of transport and water inhibition, the conversion of CO2 were restricted to less than c.a. 30% by controlling the reaction temperature in the range of 200–250 °C. The apparent activation energies (Ea) of CO2 over different catalysts were calculated according to the below Arrhenius equation. lnk = −Ea/RT + lnA

3.1.2. N2 adsorption-desorption analysis Fig. 2(1) showed the N2 adsorption-desorption isotherms of the fresh OMA-10NixCa catalysts. It could be observed that all the samples irrespective of the Ca/Al molar ratios exhibited the representative IV(a)

(3)

Here, k, R, and A referred to the rate constant, gas constant, and preexponential factor, respectively. 202

Journal of CO₂ Utilization 21 (2017) 200–210

L. Xu et al.

Table 1 Textual properties of the as-prepared and 50 h spent catalysts based on the XRD, N2 adsorption-desorption, ICP-AES, and kinetic analyses. Samples

SBETa(m2/g)

VBJHb (cm3/g)

APDc (nm)

d(100)e (nm)

WTf(nm)

Ni/Alg

Ca/Alg

Eah (kJ/mol)

Isotherm Type

OMA-10Ni OMA-10Ni1Ca OMA-10Ni3Ca OMA-10Ni5Ca OMA-10Ni8Ca OMA-10Ni10Ca SPd-OMA-10Ni SP-OMA-10Ni3Ca SP-OMA-10Ni8Ca

232.8 216.9 239.7 240.7 210.7 214.6 139.0 156.2 138.7

0.42 0.40 0.45 0.46 0.37 0.38 0.27 0.28 0.30

9.5 8.3 8.4 8.4 8.4 8.3 8.5 7.5 7.5

11.3 11.0 11.3 11.0 11.3 11.3 – – –

3.5 4.4 4.6 4.3 4.6 4.5 – – –

0.097 0.096 0.098 0.095 0.097 0.094 0.098 0.095 0.094

– 0.009 0.029 0.051 0.078 0.096 – 0.028 0.075

75.2 66.4 62.0 58.5 53.6 56.2 – – –

IV(a) IV(a) IV(a) IV(a) IV(a) IV(a) IV(a) IV(a) IV(a)

H1 H1 H1 H1 H1 H1 H1 H1 H1

a

SBET stands for the specific area calculated based on Brunauer-Emmett-Teller theory. VBJH stands for the pore volume calculated based on Barrett-Joyner-Halenda theory. APD stands for average pore diameter. d SP stands for 50 h spent catalyst. e d(100) stands for the d-spacing of the (1 0 0) direction calculated by the Bragg’s law: d = nλ/2sin θ, where λ is the wavelength of the X-ray wave (0.15406 nm). f WT stands for the wall thickness (calculated by 2 × d(100)/ 3 –pore diameter). g Ni/Al and Ca/Al molar ratios were obtained by ICP-AES analyses. h Ea stood for the apparent activation energy of CO2. b c

was worth noting that the samples doped with Ca displayed a bit smaller pore diameter than pristine OMA-10Ni. This indicated that the Ca2+ cation could regulate the pore diameter by taking part in the selfassembly process. In addition, the specific values of their specific structural properties were summarized in Table 1. It was notable that the large surface areas (210.7–240.7 m2/g) and big pore volumes (0.37–0.46 cm3/g) over all the OMA-10NixCa materials were successfully maintained after calcination at 700 °C. The effect of the Ca doping on the surface areas and pore volumes was almost negligible. As regards the wall thicknesses of the OMA-10NixCa materials, it could be observed that their values were located in the range of 3.5–4.6 nm. The Ca doped materials displayed relatively larger wall thicknesses than pristine OMA-10Ni, which would be in favor of improving the thermal stability of these materials. 3.1.3. TEM, SAED, and EDS analyses The TEM images of the fresh OMA-10NixCa were displayed in Fig. 3. The ordered mesopores viewed along the [1 1 0] direction (see Fig. 3(a, c, d, e, f, h)) and the [0 0 1] direction (see Fig. 3(b, g)) could be clearly observed over all the OMA-10NixCa materials with different Ca/ Al molar ratios, indicating the presence of ordered mesopores with twodimensional p6mm hexagonal symmetry. This was well consistent with the small-angle XRD and N2 adsorption-desorption measurements as discussed above. For the SAED patterns (insets of Fig. 3(a, c, d, e, f, h)), the fuzzy diffraction rings suggested the poor crystallinity of their mesoporous framework. This had been already verified by the wideangle XRD characterization in Fig. 1(2). Besides, the EDS profile of OMA-10Ni10Ca was displayed in Fig. 3(i) and the characteristic peaks of Ni, Ca, Al, O, and Cu elements could be clearly observed in the region of ordered mesopore. For the Cu characteristic peak, they should be originated from the copper sample holder. The simultaneous appearance of Ni, Ca, Al, and O characteristic peaks indicated that they had been successfully introduced into the framework of OMA-10Ni10Ca via the one-pot EISA fabrication strategy. Overall, the doping of the Ca basic modifier into the mesoporous framework had not greatly influenced the orderliness of the mesostructure.

Fig. 2. (1) Nitrogen adsorption-desorption isotherms and (2) pore size distribution curves of the as-prepared OMA-10NixCa catalysts.

type curve with H1 shaped hysteresis loops according to the latest IUPAC definition in 2015 [37], which were the typical characteristics of the mesoporous materials with cylindrical channels. The capillary condensation step of the hysteresis loops was steep in the range of 0.60–0.80 P/Po. This implied that the mesopores of these materials were uniform in pore diameter. The corresponding pore size distributions (PSD) of these samples were obtained based on the Barrett-JoynerHalenda (BJH) calculation by using the adsorption isotherm branch and their PSD curves were shown in Fig. 2(2). It was noticeable that all the samples were provided with relatively narrow PSD in the range of 8.3–9.5 nm, suggesting the uniform size of the mesopore. Besides, it

3.1.4. XPS and H2-TPR analyses The XPS characterization was carried out to collect additional information related with the surface chemical state of the materials. The Ni 2p XPS spectra of the fresh OMA-10NixCa catalysts with different Ca/Al molar ratios were displayed in Fig. 4(1). For these samples, they all exhibited Ni 2p3/2 peaks at 856.0 ± 0.1 eV together with satellite peaks around 862.0 eV regardless of the Ca/Al molar ratios, which was totally different with the Ni 2P3/2 peak (854.4 eV) for pure NiO [38]. 203

Journal of CO₂ Utilization 21 (2017) 200–210

L. Xu et al.

Fig. 3. TEM, SAED, and EDS measurements of the asprepared OMA-10NixCa catalysts: (a, b) OMA-10Ni, (c) OMA-10Ni1Ca, (d) OMA-10Ni3Ca; (e) OMA10Ni5Ca; (f, g) OMA-10Ni8Ca; (h, i) OMA10Ni10Ca.

These could be ascribed to the Ni2+ cation in the NiAl2O4 spinel species according to previous reports [39–41]. Thus, the doping of the Ca modifier into the mesoporous framework did not greatly change the chemical coordination environment of the Ni2+ cation. For the OMA10NixCa catalysts, it was supposed that the Ni species were uniformly distributed among the mesoporous framework owing to the unique merit of the one-pot EISA fabrication strategy. Therefore, similar to the surface Ni species, the Ni species located in the bulk part of the mesoporous framework might be also in the state of NiAl2O4. The absence of the NiAl2O4 diffraction peak in Fig. 1(2) might be derived from its high dispersion among the amorphous alumina framework. The H2-TPR measurements of the fresh OMA-10NixCa catalysts were carried out to study the interaction between the Ni species and mesoporous framework and their profiles were displayed in Fig. 4(2). It was greatly interesting to observe that all these catalysts only demonstrated one distinct hydrogen reduction peak in the range of 765.0–825.0 °C. There was no reduction peak observed in the range of 300–500 °C, suggesting the inexistence of dissociated NiO species weakly bonded with the mesoporous framework [42,43]. This indicated that the strong interaction between the Ni species and the mesoporous framework had been formed. The only one pronounced reduction peak suggested that only one sort of Ni species existed among these mesoporous materials due to the distinctive characteristic of one-pot EISA synthesis strategy. As mentioned above, the Ni species ought to be homogenously dispersed among the ordered mesoporous framework, promising the identical reducibility of Ni2+ owing to the same coordination environment. This phenomenon had been encountered in other sorts of materials fabricated via one-pot method, such as nickel aluminate mesoporous materials [44,45], Ni based hydrotalcite-like compounds [46,47], NixMg2-xAl-LDH compounds [23], etc, which also

presented only one remarkable reduction peak in H2-TPR profile. Together with the results of XPS in Fig. 4(1), it could be confirmed that almost all the Ni species had been successfully incorporated into the mesoporous Al2O3 matrix in the form of NiAl2O4 spinel, which commonly demanded the high reduction temperature around 800 °C as reported in previous literatures [32,43,44]. Besides, compared with the pristine OMA-10Ni catalyst (823.4 °C) without modification, the Ca doped catalysts displayed reduction peaks in relatively lower region (780.0–795.0 °C). This indicated that incorporating the Ca modifier into the mesoporous framework could make the reduction of Ni species become easier. The presence of Ca might also interact with the mesoporous Al2O3 framework by forming calcium aluminate (CaAl2O4) phase (CaO + Al2O3 → CaAl2O4), which would compete with the formation process of nickel aluminate (NiAl2O4) phase (NiO + Al2O3 → NiAl2O4) based on previous reports [35,48]. The formation of the calcium aluminate would reduce the possibility of subsequent formation of nickel aluminate. Due to this reason, the interaction between Ni species and ordered mesoporous alumina framework had been weakened to some degree. 3.1.5. CO2-TPD analysis CO2-TPD analysis was carried out to characterize the surface basicity by employing CO2 as the probe molecule and their profiles were summarized in Fig. 5. Commonly, it was well known that the CO2 chemically adsorbed on the weak basic sites could be desorbed under low temperature and that adsorbed on the strong basic sites could be desorbed at relatively high temperature according to the pioneer reports [35,49,50]. It could be interestingly found that most of the CO2TPD profiles of OMA-10NixCa samples with different Ca/Al molar ratios (0%–10.0%) resembled each other in shape. Specifically, they 204

Journal of CO₂ Utilization 21 (2017) 200–210

L. Xu et al.

Overall, the doping of the Ca species could remarkably increase the surface basicity, which would be conductive to the chemisorption of CO2. As a result, the apparent activation energy of the CO2 methanation could be decreased and the catalytic activity at low reaction temperature could be promoted. 3.2. Evaluation of catalytic behavior 3.2.1. Effect of Ca basic modification The effect of the Ca basic promotor on the catalytic activities had been carefully investigated at different reaction temperatures in the range of 200–450 °C at the interval of 50 °C. The CO2 conversion as a function of Ca/Al molar ratio had been summarized in Fig. 6. As can be observed, the CO2 conversion rapidly increased and finally reached the maximum point at 8% when the Ca/Al molar ratio gradually increased from 0% to 8%. Nevertheless, the CO2 conversion suffered some decline when the Ca/Al molar ratio further increased to 10%. This indicated that only the optimum amount of Ca dopant could maximally enhance the catalytic activity. The above CO2-TPD characterization (see Fig. 5) indicated that the incorporation of the Ca had already obviously intensified the surface basicity, which was conductive to the chemisorption of CO2 and decreased the reaction activation energy during the process of CO2 methanation. Accordingly, the catalytic activity, especially at low temperature, could be observably improved. Nevertheless, the superfluous Ca doping amount might make the basic sites cover the neighboring metallic Ni active sites, which would block the following H2 activation process. As a result, the CO2 conversion suffered some decline over OMA-10Ni10Ca catalyst. In order to more visually describe the role of the Ca basic modifier in promoting the catalytic activity at different temperatures, the K value was defined as “K = C/Co”, where the C and Co were the CO2 conversions of the Ca modified catalyst and pristine OMA-10Ni reference catalyst, respectively. The relationship between K value (K1Ca, K3Ca, K5Ca, K8Ca, and K10Ca) with reaction temperature over the basic modified catalysts with different Ca/Al molar ratios was depicted in Fig. 7(1). As can be observed, the K values over these Ca doped catalysts gradually decreased and nearly approached 1.0 when the temperatures progressively increased from 200 to 450 °C, suggesting that the difference between C and Co became smaller. The roles of the Ca basic modifier in promoting the catalytic activity seemed to be more effective at low temperature. The reason for this might be that the CO2 methanation reaction was a kinetically controlled process. The CO2 activating capacity of the catalyst could greatly affect and determine the reaction rate of the reaction. The CO2 activation ability in low temperature region (200–250 °C) had been significantly improved owing to the presence of the Ca promotor by intensifying the processes of CO2 chemisorption and activation. Furthermore, it was also of great interest to find that the K values of the OMA-10NixCa catalysts with different Ca/Al molar ratios followed the sequence: K8Ca > K10Ca > K5Ca > K3Ca > K1Ca. Therefore, the OMA-10NixCo with 8% Ca/Al molar ratio performed the best low-temperature catalytic activity due to its optimum basicity. The kinetic study related with the role of the Ca basic modifier in promoting the low-temperature catalytic activities was systematically investigated over OMA-10NixCa catalysts and their Arrhenius plots were presented in Fig. 7(2). It was interestingly found that the slope value of the OMA-10Ni plot was a bit bigger than those of the Ca doped catalysts. This implied that OMA-10Ni exhibited larger apparent activation energy than other catalysts. Specifically, as summarized in Table 1, the Ea values over OMA-10NixCa catalysts followed the below sequence: Eablank > Ea1Ca > Ea3Ca > Ea5Ca > Ea10Ca > Ea8Ca. These kinetic results suggested that the presence of the Ca basic modifier had efficiently decreased the apparent activation energy of the CO2 methanation from 75.2 kJ/mol to 53.6 kJ/mol, accounting for the enhanced low-temperature catalytic activities of Ca doped catalysts.

Fig. 4. (1) H2-TPR and (2) Ni 2p XPS profiles of the as-prepared OMA-10NixCa catalysts.

Fig. 5. CO2-TPD profiles of the as-prepared OMA-10NixCa catalysts.

displayed two groups of desorption peaks around 85.0 °C and 460.0 °C, which could be assigned to the weak and strong basic sites, respectively. Besides, it was noticeable that the intensities of the CO2-TPD peaks gradually increased and the peak positions progressively shifted to higher temperatures (typically, from 80.3 to 89.3 °C and from 457.6 to 468.2 °C) with the increase of the Ca/Al molar ratio from 0% to 10%. These phenomena suggested that the increase of the Ca/Al molar ratio could increase the number as well as intensity of the basic sites at the same time. For the CO2-TPD profiles of OMA-10Ni8Ca and OMA10Ni10Ca samples, the new small peaks centered at c.a. 576.2, 614.5, and 800.0 °C could be observed, indicating the appearance of new sorts of strong basic sites with the increase of Ca/Al molar ratio up to 8%.

3.2.2. The influence of the reaction temperature The influence of the reaction temperature on the catalytic activity 205

Journal of CO₂ Utilization 21 (2017) 200–210

L. Xu et al.

Fig. 6. The curve of the CO2 conversion versus Ca/Al molar ratio at different reaction temperatures over the OMA-10NixCa catalysts; reaction condition: H2/CO2 = 4, GHSV = 15000 mL/(g h), 1 atm.

Therefore, in the viewpoint of chemical equilibrium (Ln K = −ΔG/RT), the lower reaction temperature would be beneficial to higher CO2 conversion and CH4 selectivity. Although the low reaction temperature could create favorable conditions for high catalytic activity, the actual activities over these OMA-10NixCa catalysts were relatively poor. The reason for this ought to be attributed to the great kinetic barrier for the full reduction of CO2 (+4) into CH4 (−4), an eight-electron process, which usually required high activation energy. Herein, the incorporation of the Ca could effectively decrease the kinetic barrier for CO2 activation. Therefore, the OMA-10Ni3Ca and OMA-10Ni8Ca with Ca modification displayed much higher CO2 conversions than the pristine OMA-10Ni reference catalyst, especially at low temperature. For example, the CO2 conversions at 250 °C over OMA-10Ni8Ca (26.9%) and OMA-10Ni3Ca (14.3%) were 3.7 and 1.9 times of that over OMA-10Ni

and CH4 selectivity had been investigated over these OMA-10NixCa catalysts toward CO2 methanation in the region of 200–450 °C. The OMA-10Ni, OMA-10Ni3Ca and OMA-10Ni8Ca were selected as the representative catalysts. For these catalysts, as can be seen in Fig. 8(1), with the increase of the reaction temperature from 200 to 400 °C, the CO2 conversions rapidly increased; nevertheless, further increasing the temperature up to 450 °C caused the decrease of the CO2 conversions over these catalysts. As a comparison, the equilibrium CO2 conversion monotonously declined with the increase of the reaction temperature due to the thermodynamic feature of the CO2 methanation reaction ((CO2(g) + 4H2(g) = CH4(g) + 2H2O(g), ΔG = −173.1 + 0.1983T kJ/mol)) [5,51,52]. The Gibbs free energy would increase with the increase of the reaction temperature, which would make the reaction gradually shift towards the reactants. 206

Journal of CO₂ Utilization 21 (2017) 200–210

L. Xu et al.

Fig. 7. (1) The curves of the K values versus reaction temperature over OMA-10NixCa catalysts; (2) Arrhenius plots for CO2 reaction rate over OMA-10NixCa catalysts; reaction condition: H2/CO2 = 4, GHSV = 15000 mL/(g h), 1 atm.

Fig. 8. The curves of the (1) CH4 conversion and (2) CH4 selectivity versus reaction temperature over different OMA-10NixCa catalysts; reaction condition: H2/CO2 = 4, GHSV = 15000 mL/(g h), 1 atm.

(7.3%), demonstrating the critical roles of Ca dopant in promoting the low-temperature catalytic activity. Fig. 8(2) exhibited the effect of the reaction temperature on the CH4 selectivity over these catalysts. It could be noticeable that the equilibrium CH4 selectivity suffered gradual decline as the reaction temperature increased, which ought to be attributed to the presence of the reverse water-gas shift (RWGS) side reaction. The CO byproduct could be generated in RWGS reaction, subsequently decreasing the CH4 selectivity. As contrast, the actual values of CH4 selectivity over these catalysts were a bit lower than the theoretical equilibrium value. All the catalysts performed their respective maximum values at 350 °C. Besides, compared with the OMA-10Ni reference catalyst, the OMA10Ni3Ca and OMA-10Ni8Ca with basic modification displayed higher CH4 selectivity. This suggested that the doping of Ca basic modifier could enhance the CH4 selectivity by strengthening the CO2 chemisorption.

H2-TPR measurement in Fig. 4(2), which would restrict the free movement of the metallic Ni active sites during the processes of reduction at 800 °C and CO2 methanation at 400 °C. As a result, the seriously thermal sintering of the metallic Ni nanoparticles had been successfully suppressed, accounting for no deactivation during 50 h stability tests. Besides, the OMA-10Ni3Ca and OMA-10Ni8Ca catalysts also displayed higher catalytic activities than the OMA-10Ni reference catalyst during the whole 50 h stability tests, once again confirming the promoting effect of the Ca basic modification. Besides, the catalytic performance related with the CH4 selectivity during the 50 h stability tests were reflected in Fig. 9(2). It was observable that the CH4 selectivity over all these mesoporous catalysts was very steady during the whole 50 h time on stream, also implying that the seriously thermal sintering of the metallic Ni active sites did not take place. It was widely believed that the H2 was usually activated and dissociated over the metallic Ni active sites during the CO2 methanation reaction [5,8,32]. Thus, the selectivity of CH4 was closely related with the size of the Ni nanoparticles and the higher CH4 selectivity could be much easier to obtain over the smaller Ni nanoparticles [53]. If the thermal sintering of the metallic Ni nanoparticles occurred, the dissociation of the H2 into H atom over the surface Ni active sites would be severely affected and the subsequent hydrogenation of the reaction intermediates (CO or formate) would be inhibited, finally leading to the decrease in CH4 selectivity because of the shortage of the H source [26,32]. Furthermore, the CH4 selectivity values over OMA-10Ni3Ca and OMA-10Ni8Ca catalysts were also a bit higher than that over OMA-10Ni reference catalyst, once again illustrating the enhancement effect of the Ca basic modifier. Overall, both the low-temperature catalytic activity and CH4 selectivity could be greatly promoted over OMA-10NixCa by doping Ca

3.2.3. Stability tests For the Ni based catalysts, the thermal sintering of the metallic Ni active sites during the CO2 methanation usually caused the rapid deactivation of the catalysts due to its exothermic feature. In order to investigate the catalytic stability of these OMA-10NixCa mesoporous catalysts, 50 h stability tests were conducted over OMA-10Ni3Ca, OMA10Ni8Ca, and OMA-10Ni catalysts under specific conditions: H2/ CO2 = 4, 400 °C, GHSV = 15,000 mL/(g h), 1 atm. As shown in Fig. 9(1), the CO2 conversions over these catalysts did not suffer serious deactivation after 50 h stability tests. This could be attributable to the outstanding anti-sintering properties of the metallic Ni active sites via the confinement effect of mesoporous framework. The strong interaction between metal and framework had been formed according to the 207

Journal of CO₂ Utilization 21 (2017) 200–210

L. Xu et al.

Fig. 9. 50 h long-term stability tests over OMA-10NixCa catalysts: (1) CO2 conversion and (2) CH4 selectivity; reaction conditions: H2/CO2 = 4, GHSV = 15000 mL/(g h), 400 °C, 1 atm.

Fig. 11. (1) Nitrogen adsorption-desorption isotherms and (2) pore size distribution curves of the 50 h spent (SP-) OMA-10NixCa catalysts.

basic modifier. As for the Ni-Al based catalysts for CO2 methanation, the Ni-Al hydrotalcite type catalysts gradually attracted lots of research interests due to their unique structural characteristics, where the Ni active sites were well confined and uniformly distributed among the layers of hydrotalcite crystalline structures [54,55]. Abate et al. reported that the equilibrium CO2 conversion could be obtained at the temperature above 300 °C over Ni-Al hydrotalcite catalysts (75–80% wt % NiO) with diluted feed gases under high pressure (H2/CO2/N2 = 10/ 2.5/87.5, GHSV = 20,000 h−1, 5 bar) [54]. Wierzbicki et al. reported that ca. 46.0% CO2 conversion could be achieved at 250 °C over the (Ni, La, Mg, Al)-hydrotalcite derived catalysts (15 wt% Ni) with medium

strength basic sites by La dopant under diluted feed gas stream (H2/ CO2/Ar = 12/3/5, GHSV = 12,000 h−1) [55]. Although these catalysts exhibited a bit higher low-temperature CO2 conversions than the present OMA-10NixCa catalysts (ca. 10.0 wt% Ni), the Ni loading amounts of these catalysts were much larger than OMA-10NixCa and the evaluation of the catalytic performances were conducted under different conditions, such as diluted feed gases and/or high pressure. Therefore, these results could not be directly compared with each other. The common advantages of the Ni-Al hydrotalcite and ordered mesoporous Ni-Al catalysts were that the Ni active sites were effectively confined and the thermal sintering of the metallic Ni active sites during CO2 methanation could be effectively suppressed, accounting for excellent catalytic stabilities. Compared with the Ni-Al hydrotalcite catalysts, the unique advantage of the current ordered mesoporous Ni-Al catalysts lied in the excellent textural properties, such as large surface areas, big pore volumes, unblocked mesoporous channels, etc., which greatly favored the mass transfer of the feed gases during the CO2 methanation reaction. Besides, compared with the noble metal (e.g. Ru, Rh) based catalysts, the low-temperature catalytic activities over OMA10NixCa catalysts still had plenty of room for improvement [56,57]. For example, Tada et al. reported that the equilibrium CO2 conversion (ca. 93.0%) and almost 100% CH4 selectivity could be achieved at 350 °C over Ru/CeO2/Al2O3 catalyst, which was much better than the present results [56]. However, as Zhen et al. reported [58], 0.5 wt% of Ru doping could remarkably promote the low-temperature catalytic activity of Ni/γ-Al2O3 catalyst. In the view of the high cost of the noble metal catalysts, it was more practical to dope tiny amount of noble metals in the current ordered mesoporous Ni-Al catalysts to further improve the low-temperature activities in the future research.

Fig. 10. X-ray diffraction patterns of the 50 h spent (SP-) OMA-10NixCa catalysts.

208

Journal of CO₂ Utilization 21 (2017) 200–210

L. Xu et al.

Fig. 12. TEM and SAED images of the 50 h spent (SP-) catalysts: (a, b) SPOMA-10Ni, (c, d) SP-OMA-10Ni3Ca, (e, f) SP-OMA-10Ni8Ca.

3.3.2. N2 adsorption-desorption analysis The nitrogen adsorption-desorption analyses were also conducted over the 50 h spent catalysts. Their isotherms and pore size distribution curves were displayed in Fig. 11. As shown in Fig. 11(1), identical to their corresponding fresh samples, the 50 h spent OMA-10Ni3Ca, OMA10Ni8Ca, and OMA-10Ni catalysts also displayed IV(a) type isotherms with steep H1-shaped hysteresis loops in the range of 0.6–0.8 P/Po. This suggested that the ordered mesostructure of these materials had been successfully retained after 50 h stability tests. Besides, the corresponding pore size distribution (PSD) curves of these 50 h spent catalysts were depicted in Fig. 11(2). As can be observed, they exhibited greatly narrow PSD in the range of 7.5–8.5 nm, which was a bit smaller than their corresponding fresh samples because of the thermal shrinkage of the mesoporous framework. As shown in Table 1, like the average pore diameters, the values of the surface areas and pore volumes of these spent catalysts were also a bit smaller than the fresh catalysts. However, the Ni/Al and Ca/Al molar ratios (see Table 1) of the spent catalysts were comparable to their corresponding fresh catalysts, implying that the loss of the metallic Ni active sites and Ca basic sites during the CO2 methanation was negligible.

3.3. Characterizations of the spent catalysts 3.3.1. XRD analysis The XRD characterizations of the 50 h spent (denoted as SP-) OMA10Ni3Ca, OMA-10Ni8Ca, and OMA-10Ni catalysts were carried out to confirm the anti-sintering properties of the metallic Ni active sites and their XRD patterns were summarized in Fig. 10. As can be observed, all the spent mesoporous catalysts only displayed tiny γ-Al2O3 (PDF-#-100425) [59,60] and metallic Ni (PDF-#-45-1027) [61,62] diffraction peaks after 50 h stability tests. It was difficult to precisely calculate the crystalline sizes of these metallic Ni nanoparticles by the Scherrer formula because of the diffraction peak broadening. Therefore, the seriously thermal sintering of the metallic Ni active sites had been successfully avoided. For these mesoporous catalysts, it had been confirmed that both the confinement effect of the mesoporous framework and the strong metal-framework interaction could contribute to the stabilization of the Ni nanoparticles. Consequently, there was no obvious deactivation in catalytic activity and CH4 selectivity after 50 h stability tests.

209

Journal of CO₂ Utilization 21 (2017) 200–210

L. Xu et al.

[11] M.S. Duyar, A. Ramachandran, C. Wang, R.J. Farrauto, J. CO2 Util. 12 (2015) 27–33. [12] S. Toemen, W.A.W. Abu Bakar, R. Ali, J. CO2 Util. 13 (2016) 38–49. [13] A. Beuls, C. Swalus, M. Jacquemin, G. Heyen, A. Karelovic, P. Ruiz, Appl. Catal. B: Environ. 113 (2012) 2–10. [14] A. Karelovic, P. Ruiz, J. Catal. 301 (2013) 141–153. [15] A. Karelovic, P. Ruiz, ACS Catal. 3 (2013) 2799–2812. [16] J.-N. Park, E.W. McFarland, J. Catal. 266 (2009) 92–97. [17] J. Ren, X. Qin, J.-Z. Yang, Z.-F. Qin, H.-L. Guo, J.-Y. Lin, Z. Li, Fuel Process. Technol. 137 (2015) 204–211. [18] Q. Pan, J. Peng, T. Sun, D. Gao, S. Wang, S. Wang, Fuel Process. Technol. 123 (2014) 166–171. [19] J. Liu, C. Li, F. Wang, S. He, H. Chen, Y. Zhao, M. Wei, D.G. Evans, X. Duan, Catal. Sci. Technol. 3 (2013) 2627–2633. [20] K.C. Zhao, W.H. Wang, Z.H. Li, J. CO2 Util. 16 (2016) 236–244. [21] Q. Pan, J. Peng, T. Sun, S. Wang, S. Wang, Catal. Commun. 45 (2014) 74–78. [22] Y. Li, G. Lu, J. Ma, RSC Adv. 4 (2014) 17420–17428. [23] J. Liu, W. Bing, X. Xue, F. Wang, B. Wang, S. He, Y. Zhang, M. Wei, Catal. Sci. Technol. 6 (2016) 3976–3983. [24] B. Veldurthy, J.-M. Clacens, F. Figueras, J. Catal. 229 (2005) 237–242. [25] M. Climent, A. Corma, S. Iborra, K. Epping, A. Velty, J. Catal. 225 (2004) 316–326. [26] M. Guo, G. Lu, RSC Adv. 4 (2014) 58171–58177. [27] C. Heine, B.A. Lechner, H. Bluhm, M. Salmeron, J. Am. Chem. Soc. 138 (2016) 13246–13252. [28] C. Swalus, M. Jacquemin, C. Poleunis, P. Bertrand, P. Ruiz, Appl. Catal. B: Environ. 125 (2012) 41–50. [29] G. Du, S. Lim, Y. Yang, C. Wang, L. Pfefferle, G.L. Haller, J. Catal. 249 (2007) 370–379. [30] W. Zhen, B. Li, G. Lu, J. Ma, Chem. Commun. 51 (2015) 1728–1731. [31] W. Zhen, F. Gao, B. Tian, P. Ding, Y. Deng, Z. Li, H. Gao, G. Lu, J. Catal. 348 (2017) 200–211. [32] L. Xu, F. Wang, M. Chen, J. Zhang, K. Yuan, L. Wang, K. Wu, G. Xu, W. Chen, RSC Adv. 6 (2016) 28489–28499. [33] Q. Yuan, A.X. Yin, C. Luo, L.D. Sun, Y.W. Zhang, W.T. Duan, H.C. Liu, C.H. Yan, J. Am. Chem. Soc. 130 (2008) 3465–3472. [34] S.M. Morris, P.F. Fulvio, M. Jaroniec, J. Am. Chem. Soc. 130 (2008) 15210–15216. [35] L. Xu, H. Song, L. Chou, ACS Catal. 2 (2012) 1331–1342. [36] A. Corma, Chem. Rev. 97 (1997) 2373–2420. [37] M. Thommes, K. Kaneko, A.V. Neimark, J.P. Olivier, F. Rodriguez-Reinoso, J. Rouquerol, K.S. Sing, Pure Appl. Chem. 87 (2015) 1051–1069. [38] Y. Kathiraser, Z. Wang, M.L. Ang, L. Mo, Z. Li, U. Oemar, S. Kawi, J. CO2 Util. 19 (2017) 284–295. [39] I. Murthy, C. Swamy, J. Mater. Sci. 28 (1993) 1194–1198. [40] G. Gavalas, C. Phichitkul, G. Voecks, J. Catal. 88 (1984) 54–64. [41] Y.S. Oh, H.S. Roh, K.W. Jun, Y.S. Baek, Int. J. Hydrogen Energy 28 (2003) 1387–1392. [42] B. Mile, D. Stirling, M.A. Zammitt, A. Lovell, M. Webb, J. Catal. 114 (1988) 217–229. [43] C. Li, Y.W. Chen, Thermochim. Acta 256 (1995) 457–465. [44] X. Huang, G. Xue, C. Wang, N. Zhao, N. Sun, W. Wei, Y. Sun, Catal. Sci. Technol. 6 (2016) 449–459. [45] L. Zhang, X. Wang, C. Chen, X. Zou, W. Ding, X. Lu, Int. J. Hydrogen Energy 42 (2017) 11333–11345. [46] K. Sato, N. Abe, T. Kawagoe, S.-i. Miyahara, K. Honda, K. Nagaoka, Int. J. Hydrogen Energy 42 (2017) 6610–6617. [47] S. Liu, D. Chen, K. Zhang, J. Li, N. Zhao, Int. J. Hydrogen Energy 33 (2008) 3736–3747. [48] J. Dias, J. Assaf, Catal. Today 85 (2003) 59–68. [49] V. Choudhary, V. Rane, J. Catal. 130 (1991) 411–422. [50] L. Pino, A. Vita, F. Cipitì, M. Laganà, V. Recupero, Appl. Catal. B: Environ. 104 (2011) 64–73. [51] J. Gao, Q. Liu, F. Gu, B. Liu, Z. Zhong, F. Su, RSC Adv. 5 (2015) 22759–22776. [52] X. Su, J. Xu, B. Liang, H. Duan, B. Hou, Y. Huang, J. Energy Chem. 25 (2016) 553–565. [53] G. Garbarino, P. Riani, L. Magistri, G. Busca, Int. J. Hydrogen Energy 39 (2014) 11557–11565. [54] S. Abate, K. Barbera, E. Giglio, F. Deorsola, S. Bensaid, S. Perathoner, R. Pirone, G. Centi, Ind. Eng. Chem. Res. 55 (2016) 8299–8308. [55] D. Wierzbicki, R. Debek, M. Motak, T. Grzybek, M.E. Gálvez, P. Da Costa, Catal. Commun. 83 (2016) 5–8. [56] S. Tada, O.J. Ochieng, R. Kikuchi, T. Haneda, H. Kameyama, Int. J. Hydrogen Energy 39 (2014) 10090–10100. [57] A. Karelovic, P. Ruiz, J. Catal. 301 (2013) 141–153. [58] W. Zhen, B. Li, G. Lu, J. Ma, RSC Adv. 4 (2014) 16472–16479. [59] S.C. Lee, Y.M. Kwon, H.J. Chae, S.Y. Jung, J.B. Lee, C.K. Ryu, C.K. Yi, J.C. Kim, Fuel 104 (2013) 882–885. [60] T. Wang, L. Qi, H. Lu, M. Ji, J. CO2 Util. 17 (2017) 162–169. [61] T.R. Reina, P. Yeletsky, J.M. Bermúdez, P. Arcelus-Arrillaga, V.A. Yakovlev, M. Millan, Fuel 182 (2016) 740–748. [62] E. Ruiz, D. Cillero, P.J. Martínez, Á. Morales, G. San Vicente, G. de Diego, J.M. Sánchez, J. CO2 Util. 8 (2014) 1–20.

3.3.3. TEM analysis TEM images of the 50 h spent catalysts had been taken to further verify the anti-sintering performance of the metallic Ni active sites and the thermal stability of the ordered mesoporous channels. As displayed in Fig. 12, the ordered mesoporous channels along the [1 1 0] direction with cylindrical shape could be observed over the 50 h spent OMA10Ni3Ca, OMA-10Ni8Ca, and OMA-10Ni catalysts. The presence of the ordered mesostructure further confirmed the prediction based on the results of N2 adsorption-desorption analyses in Fig. 11(1). Furthermore, it was of great interest to find that no obvious metallic Ni sintering cluster had been formed over these 50 h spent catalysts. This was in good agreement with the results of the XRD measurement in Fig. 10, which only displayed tiny metallic Ni diffraction peaks. Generally, these OMA-10NixCa catalysts performed excellent thermal stability and anti-sintering properties, accounting for no deactivation after 50 h longterm catalytic stability tests. 4. Conclusions In conclusion, a series of Ca doped ordered mesoporous Ni-Al composite oxides had been fabricated by the one-pot EISA strategy and investigated as the catalysts for CO2 methanation. The doping of Ca basic modifier into mesoporous framework remarkably intensified the surface basicity of the catalysts. This was conductive to the CO2 chemisorption and decreased the apparent activation energy toward CO2 methanation from 75.2 to 53.6 kJ/mol. As a result, the low-temperature catalytic activity and CH4 selectivity over Ca doped catalysts had been significantly promoted. It was also found that only the appropriate Ca/ Al molar ratio (8%) could maximally enhance the catalytic activity. Besides, the strong interaction between Ni active sites and mesoporous framework had been formed by homogenously distributing the Ni species among the mesoporous Al2O3 matrix via one-pot synthesis strategy. The thermal sintering of the metallic Ni nanoparticles could be effectively inhibited via the confinement effect of the mesoporous framework. Consequently, there was no evident deactivation over these mesoporous catalysts during the 50 h stability tests. Owing to these advantages, the present Ca doped ordered mesoporous Ni-Al composite oxides could be considered as the potential catalyst candidates for CO2 methanation with enhanced low-temperature activities. Acknowledgements The authors sincerely acknowledge the financial support from National Natural Science Foundation of China (Grant No.21503113, 21577065, 21503142, and 91543115), International ST Cooperation Program of China (2014DFA90780), Key Projects in the National Science & Technology Pillar Program of Jiangsu Province (BE2014602, SBE2014070928), a Project Funded by the Priority Academic Program Development of Jiangsu Higher Education Institutions, and Colleges Natural Science Foundation of the Jiangsu province (15KJB530003). References [1] [2] [3] [4] [5] [6] [7] [8] [9] [10]

W. Wang, S. Wang, X. Ma, J. Gong, Chem. Soc. Rev. 40 (2011) 3703–3727. T.R. Karl, K.E. Trenberth, Science 302 (2003) 1719–1723. X. Xiaoding, J. Moulijn, Energy Fuels 10 (1996) 305–325. F.D. Meylan, V. Moreau, S. Erkman, J. CO2 Util. 12 (2015) 101–108. W. Wei, G. Jinlong, Front. Chem. Sci. Eng. 5 (2011) 2–10. M. Younas, L. Loong Kong, M.J. Bashir, H. Nadeem, A. Shehzad, S. Sethupathi, Energy Fuels 30 (2016) 8815–8831. K. Ghaib, K. Nitz, F.Z. Ben‐Fares, ChemBioEng Rev. 3 (2016) 266–275. M. Aziz, A. Jalil, S. Triwahyono, A. Ahmad, Green Chem. 17 (2015) 2647–2663. F. Wang, S. He, H. Chen, B. Wang, L. Zheng, M. Wei, D.G. Evans, X. Duan, J. Am. Chem. Soc. 138 (2016) 6298–6305. J. Xu, X. Su, H. Duan, B. Hou, Q. Lin, X. Liu, X. Pan, G. Pei, H. Geng, Y. Huang, J. Catal. 333 (2016) 227–237.

210