1 Introduction

Carbon dioxide as an important component of greenhouse gas and C1 resource has been widely investigated for its capture, storage and utilization [1,2,3]. CO2 methanation (known as Sabatier reaction) which would reduce CO2 emissions and produce natural gas, is considered to be one of the most effective and practical technologies for CO2 recycling [4]. As a volume-reduce and exothermic reaction, CO2 methanation is favored at elevated pressures and low temperatures, whereas kinetic limitation reduces conversion efficiency of CO2 to methane [5].

Noble metals, especially Ru and Rh, are very active and selective for CO2 methanation at low temperatures [6]. Lin et al. [7] investigated the effect of TiO2 structure on the dispersion of Ru nanoparticles and found that the high interaction between RuO2 and rutile-TiO2 promotes the dispersion of active sites and prevent their aggregation. Therefore, Ru/rutile-TiO2 shows high thermal stability and catalytic activity with a CO2 conversion of 65% at 300 °C. Karelovic et al. [8] reported that methane selectivity was 100% over the Rh/Al2O3 catalyst at temperatures between 185 and 200 °C. However, the turnover frequency (TOF) for CH4 formation was found to be dependent on Rh particle size. Larger Rh particles are up to four times more active than smaller particles at low temperature (135–150 °C), whereas at higher temperatures (200 °C) TOFs are similar for all particle sizes.

Ni-based catalysts have also been widely studied for CO2 methanation due to its low cost and high activity [5, 9,10,11]. SiO2, MgO, Al2O3, TiO2, CeO2, ZrO2 and many other oxides have been exploited as supports for Ni catalysts in CO2 methanation [12,13,14]. Liu et al. [15] prepared a well-dispersed Ni/TiO2 catalyst with small Ni particle size (2.2 nm) and obtained high CO2 conversion (96%) at 260 °C. He et al. [16] reported a Ni–Al hydrotalcite-derived catalyst exhibited narrow Ni particle-size, which reached 82.5% CO2 conversion at 350 °C. Tada et al. [14] studied the effect of various supports (CeO2, α-Al2O3, TiO2 and MgO) on Ni catalysts for CO2 methanation and found that Ni/CeO2 catalyst showed high CO2 conversion, especially at low temperatures (350 °C).

However, the available results are discordant about how the support affects the activity of CO2 methanation. Vogt et al. [17] reported that CO2 methanation over Ni catalyst is structure-sensitive. The support plays key roles not only in the dispersion of Ni catalysts but also in the promotion of CO2 adsorption, activation and conversion. Aldana et al. [18] examined the CO2 mthanation over Ni-ceria-zirconia catalysts and revealed that the high catalytic activity was attributed to the weak basic sites of support for the adsorption of CO2. Lin et al. [19] found that incorporation of ZrO2 into Ni/Al2O3 weakened the Ni-Al2O3 interaction and increased the amount of active metallic sites and oxygen vacancies, obviously improving the lower temperature catalytic activity. Many studies on the improvement of CO2 methanation at low temperature are based on the dispersion of active metals or basic sites of supports [20,21,22].

Moreover, the reaction mechanism (e.g., the reaction intermediate and route) is highly correlated with the structure of the active sites and basic sites [23,24,25]. Wu et al. [26] studied CO2 methanation on both 0.5 wt% and 10 wt% Ni/SiO2 catalysts. The results indicated that the reaction pathways depend on the Ni particle size. The m-HCOO intermediate is intricately involved in CO2 hydrogenation over both Ni/SiO2 catalysts, regardless of the Ni loading and particle size. CO2 hydrogenation likely follows a consecutive pathway on the 0.5 wt% Ni/SiO2 catalyst with small Ni particles, forming CO and CH4. However, the low H2 coverage leads to the quick formation of CO from the m-HCOO intermediate. This process led to high selectivity for CO formation on the 0.5 wt% Ni/SiO2 catalyst. When the Ni loading was increased to 10 wt%, the reaction proceeds through the mixed consecutive and parallel pathways and the selectivity switched to favor CH4 formation.

Nevertheless, Beierlein et al. [27] prepared highly loaded Ni-Al2O3 catalysts and found that CO2 methanation on Ni-Al2O3 catalysts is a structure-insensitive reaction and the TOF does not depend on metal-support interactions, the metal-support interface or the particle size. The Ni surface area is the sole microscopic property which determines the CO2 conversion. Therefore, the catalysts with the highest Ni surface areas achieve the highest weight time yields.

In this work, four Ni catalysts supported on γ-Al2O3, ZrO2, TiO2 and CeO2, respectively were prepared and characterized to understand the interrelationship between the structure and catalytic performance for CO2 methanation. The physic-chemical properties of the catalysts were analyzed by BET, XRD, H2-TPR, H2-TPD and CO2-TPD. In particular, the cooperation between active metal and basic support are discussed in depth.

2 Experimental

2.1 Catalyst Preparation

CeO2 as a support was prepared by precipitation and hydrothermal treatment. The aqueous solution of Ce(NO3)3·6H2O and 7 mol/L NaOH were firstly added dropwise into a reaction vessel maintained at 40 °C under continuous mechanical stirring. Then, the suspending liquid was transferred to a Teflon-lined stainless steel autoclave that was in turn heated to 180 °C and maintained for 24 h. The precipitate was collected by filtration with thoroughly washing with distilled water. CeO2 was finally obtained by drying at 60 °C for 10 h and calcined at 500 °C for 4 h.

ZrO2 as a support was synthesized by precipitation. The aqueous solution of Zr–(NO3)4·5H2O and 1 mL/L NaOH were firstly added dropwise into a reaction vessel maintained at 60 °C under continuous mechanical stirring. The rates of adding the solution and precipitant were controlled to keep the reaction mixture for pH 10. The formed precipitate was aged at 60 °C for 10 h and collected by filtration with thoroughly washing using distilled water. ZrO2 was finally obtained by drying at 60 °C for 12 h and calcined at 500 °C for 4 h.

γ-Al2O3 as a support was prepared by calcination of pseudo boehmite (from Aluminum Corporation of China) at 500 °C for 4 h and TiO2 was purchased from Aladdin Industrial Corporation of China.

Four nickel-based catalysts were prepared by deposition precipitation. The aqueous solution of Ni(NO3)2·6H2O and the support were firstly mixed at 60 °C for 2 h. Then, 1 mol/L NaOH was added dropwise until the pH 10. The suspension was further stirred and aged at 60 °C for 10 h. The formed precipitate was aged at 60 °C for another 10 h and collected by filtration with thoroughly washing using distilled water. The catalyst was finally obtained by drying at 60 °C for 12 h and calcined at 500 °C for 4 h. Theoretically, the nickel loading in the catalysts was fixed at 20 wt%.

2.2 Catalyst Characterization and Analysis

N2 adsorption–desorption isotherms at − 196 °C were obtained on a Micrometrics ASAP 2020 HD88 analyzer. Before measurement, the samples were degassed under vacuum at 200 °C for 12 h. The crystal structure of the prepared catalysts was analyzed with X-ray power diffractometry (XRD, X’Pert MPD Pro, PANalytical) at its Cu Kα radiation of λ = 0.154 nm. The patterns were recorded with a scan angle range 5–90° at a scanning speed of 8°/min. The Ni loading on the supports was determined by X-ray fluorescence (XRF, AXIOX, PANalytical).

Temperature programmed reduction (TPR) and H2 or CO2 temperature programmed desorption (H2-TPD or CO2-TPD) of the catalysts were carried out in Auto Chem II2920 (Micrometrics) coupling with MS (TILON, US). Prior to H2-TPR tests, 0.1 g of sample was heated from room temperature to 200 °C at 10 °C/min and maintained for 1 h under He flow. After that, the sample was cooled to 50 °C and then heated to 900 °C at 10 °C/min under a binary gas (10 vol. % H2/Ar). For H2-TPD, 0.1 g of oxide catalyst was firstly reduced in situ under H2/Ar flow at 600 °C for 2 h and then cooled to 50 °C and saturated with H2 for 1 h. After removing the physically adsorbed H2 by purging with He, the sample was heated to 900 °C at a ramping rate of 10 °C/min under He flow. The desorbed H2 was detected simultaneously by a thermal conductivity detector (TCD) and MS. CO2-TPD of pre-reduced samples (0.1 g) was performed in flowing He from 50 to 700 °C with a heating rate of 10 °C/min after adsorption of 10 vol% CO2/Ar at 50 °C for 2 h.

2.3 Methanation Test

CO2 methanation was carried out in a quartz fixed-bed reactor (I.D. = 16 mm) at atmospheric pressure. Before the reaction, 500 mg of catalyst (75–109 μm) were reduced at 600 °C for 4 h under 10 vol% H2/N2 stream (50 mL/min). After reduction, the system was cooled down to reaction temperature under N2 flow (50 mL/min), and the mixture gas of H2/N2/CO2 with volume ratio of 4/1/1 was introduced into the reactor and the space velocity (SV) was 120,000 mL/g/h. The product gas was analyzed with a micro gas chromatography (Micro3000, Agilent) equipped with TCD. The reaction temperature was monitored by a thermocouple near the bottom of the catalyst bed. The flow rates of H2, N2 and CO2 were controlled by mass flow meters, and N2 was used as an internal standard to calculate the volume flow of each component in the product. The CO2 conversion and CH4 selectivity were calculated with the following equations:

$${\text{X}}_{{{{\rm CO}}_{2} }} { = }\frac{{{\text{f}}_{{\rm in}} {\text{y}}_{{{{\rm CO}}_{ 2} , {\text{in}}}} - {\text{f}}_{{\rm out}} {\text{y}}_{{{{\rm CO}}_{ 2} , {\text{out}}}} }}{{{\text{f}}_{{\rm in}} {\text{y}}_{{{{\rm CO}}_{ 2} , {\text{in}}}} }} \times 100\%$$
(1)
$${\text{S}}_{{{{\rm CH}}_{ 4} }} { = }\frac{{{\text{f}}_{{\rm out}} {\text{y}}_{{{{\rm CH}}_{ 4} , {\text{out}}}} }}{{{\text{f}}_{{\rm in}} {\text{y}}_{{{{\rm CO}}_{ 2} , {\text{in }}}} - {\text{f}}_{{\rm out}} {\text{y}}_{{{{\rm CO}}_{ 2} , {\text{out}}}} }} \times 100\%$$
(2)

where the XCO2 and SCH4 is the CO2 conversion and CH4 selectivity; fin and fout is the molar feed rate of import and export flow in the reactor; yCO2,in, yCO2,out and yCH4,out is the volume fraction of import and export of CO2 and CH4 in the reactor.

3 Results and Discussion

3.1 Catalytic Performance

Figure 1 shows the CO2 conversion and CH4 selectivity for CO2 methanation at different temperatures over the Ni/Al2O3, Ni/ZrO2, Ni/TiO2 and Ni/CeO2 catalysts, respectively. The experimental CO2 conversion and selectivity to CH4 were also compared with their thermodynamic equilibrium values calculated using the HSC Chemistry that considers reactions of methanation and reverse water–gas-shift (RWGS). Figure 1a exhibits that for the Ni/Al2O3, Ni/TiO2 and Ni/CeO2 catalysts, the CO2 conversion increased drastically as the temperature increased from 250 to 400 °C, then reached the maximum and decreased gradually with temperature increasing. For the Ni/ZrO2 catalyst, the CO2 conversion increased as the temperature increased from 250 to 550 °C. These results indicate that for the Ni/Al2O3, Ni/TiO2 and Ni/CeO2 catalysts, the CO2 methanation was subject to kinetic control at lower temperatures but to thermodynamics dominance above 400 °C when the equilibrium coefficient becomes smaller at higher temperatures. From Fig. 1b, it can be seen that controlled by the thermodynamics, the selectivity to CH4 decreased with increasing the reaction temperature for all the four catalysts [28, 29].

Fig. 1
figure 1

a CO2 conversion and b CH4 selectivity at different temperatures over the Ni-based catalysts with different supports

Full size image

Moreover, Fig. 1a also demonstrates that the realized CO2 conversion followed an order of Ni/CeO2 > Ni/Al2O3 > Ni/TiO2 > Ni/ZrO2 and the Ni/CeO2 catalyst showed the highest CO2 conversion at low temperatures, especially at temperatures lower than 350 °C. The CO2 conversion for the Ni/CeO2 catalyst was 60.1% at 300 °C while that for other three catalysts was all lower than 20%.

3.2 Catalyst Characterization

The N2 adsorption–desorption isotherms for the four catalysts were all of type IV with a hysteresis loop to characterize the mesoporous structure. The calculated surface area, pore volume and average pore size for the four catalysts are presented in Table 1. The Ni/Al2O3 catalyst exhibited the highest surface area, and then the Ni/CeO2 and Ni/ZrO2. The Ni/TiO2 catalyst possessed the surface area of only 37 m2/g, and the largest pore volume and pore diameter.

Table 1 Physic-chemical properties of the Ni-based catalysts with different supports
Full size table

Figure 2 shows the XRD patterns of the calcined (Fig. 2a) and reduced (Fig. 2b) Ni-based catalysts. The NiO diffraction peaks at 2θ = 37.3°, 43.3°, and 62.8° corresponding to the (111), (200), and (220) planes of NiO were observed in Fig. 2a. The NiAl2O4, and γ-Al2O3 generally co-presented at 2θ = 45.9° and 66.9° in Ni/Al2O3 catalyst and overlapped to be difficultly distinguished.

Fig. 2
figure 2

XRD patterns of a calcined and b reduced Ni-based catalysts with different supports

Full size image

After reduced at 600 °C, the four catalysts all showed the Ni peaks at 2θ = 44.6°, 51.9° and 76.5° corresponding to its (111), (200) and (220) planes in Fig. 2b, respectively. This shows that the NiO species were completely reduced for these catalysts. Meanwhile, the intensity of the Ni peaks for Ni/Al2O3 and Ni/CeO2 catalysts are much weaker, which reveals that active sites were well dispersed on Al2O3 and CeO2 support. The crystal sizes of metallic Ni calculated using the Scherrer equation for the (200) plane were 6.8, 8.2, 21.9 and 33.5 nm for Ni/Al2O3, Ni/CeO2, Ni/TiO2 and Ni/ZrO2, respectively.

The H2-TPR measurements were carried out to clarify the interaction between nickel species and support. Figure 3a shows the TPR profiles of Ni-based catalysts on different supports. The Ni/Al2O3, Ni/TiO2 and Ni/ZrO2 catalysts exhibited only one reduction band and the peak temperature followed an order of Ni/ZrO2 < Ni/TiO2 < Ni/Al2O3 indicating the highly dispersed Ni particles and stronger interaction between NiO and Al2O3 in Ni/Al2O3 catalyst. For Ni/CeO2 catalyst, three reduction peaks were observed. The first peak at around 248 °C belonged to the reduction of bulk NiO species. The second peak at 355 °C was attributed to the reduction of NiO species which interacted with the support. Finally, the reduction peak around 832 °C was basically identical to the reduction of the CeO2 support [30, 31]. Before methanation tests, the four catalyst were pre-reduced at 600 °C for 4 h under 10 vol% H2/N2 stream, so the NiO species in these catalysts were all completely reduced to metallic nickel in accordance with the result of XRD in Fig. 2b.

Fig. 3
figure 3

a H2-TPR and b H2-TPD profiles of the Ni-based catalysts with different supports

Full size image

The H2 chemsorption properties of catalysts depending on nickel amount and dispersion were measured by H2-TPD. The irregular shapes in Fig. 3b for all the catalysts could be a consequence of partial overlapping of H2 desorption peaks from the metals with distinct diameters [32], surfaces [33] or positions [34]. Gaussian multi-peak fitting results show that there are several peaks at lower temperatures (< 400 °C) for all the four catalysts. These peaks can be assigned to desorption of H2 that is weakly chemisorbed on the surface with highly dispersed Ni and a high density of defects which could act as traps during surface hydrogen diffusion and thus reduce the activation energy of H2 dissociation. Meanwhile, the higher-temperature (> 400 °C) peaks for Ni/CeO2 and Ni/Al2O3 can be due to H2 that is strongly chemisorbed on the catalyst surface. It may also be due to desorption of hydrogen strongly bonded to the substrate of the catalyst or spillovered hydrogen, which would enhance the storage capacity for H2 [35]. Integrated peak areas shown in Table 2 demonstrate that high amounts (242 and 172 μmol/gcat respectively) of total H2 adsorption were exhibited for Ni/Al2O3 and Ni/CeO2, while only 73 and 61 μmol/gcat for Ni/TiO2 and Ni/ZrO2.

Table 2 Chemisorption results for the reduced Ni-based catalysts with different supports
Full size table

By assuming a stoichiometry ratio of H/Ni = 1, the active surface area, nickel dispersion, and average nickel diameter calculated from the H2-TPD profiles are displayed in Table 2. The dispersion of active sites in the Ni-based catalysts followed an order of Ni/Al2O3 > Ni/CeO2 > Ni/TiO2 > Ni/ZrO2 and smaller nickel crystallites present in the Ni/Al2O3 and Ni/CeO2 catalysts, which were in accordance with the results of XRD patterns in Fig. 2b. This is because these two catalysts have larger surface areas and stronger interactions between NiO and support. Furthermore, the strongest interaction between NiO and Al2O3 resulted in the high dispersion of the metallic Ni on the reduced Ni/Al2O3 catalyst.

The adsorption of CO2 on the catalyst surface also plays an important role in maintaining the catalytic activity for CO2 methanation and the results are shown in Fig. 4. The profiles of Ni/ZrO2 and Ni/TiO2 exhibited only a small amount of desorbed CO2, while a large amount of desorbed CO2 was detected at 50–300 °C for Ni/CeO2 indicating the existence of weak basic sites. The CO2 desorption was observed in a wide temperature range from 50 °C to 600 °C for Ni/Al2O3, especially the CO2 desorption peaks at 300–600 °C attributing to the strong basic sites was more remarkable for Ni/Al2O3 than for the other samples [11]. The amount of CO2 adsorbed was also calculated, referring to the H2 uptake, to be 24, 63, 13 and 10 μmol/gcat for Ni/Al2O3, Ni/CeO2, Ni/TiO2 and Ni/ZrO2, respectively.

Fig. 4
figure 4

CO2-TPD profiles of the Ni-based catalysts with different supports

Full size image

3.3 Discussion

To further reveal the structure–activity relationship, the CO2 conversions with H2 and CO2 uptakes at different temperatures over the Ni-based catalysts are co-presented in Fig. 5. Four catalysts all showed low CO2 conversion at 250 °C, because the reaction was controlled by kinetics [36]. With temperature rising to 300 °C, the reaction rate increased and the adsorbed H2 and CO2 were activated, then CO2 conversion increased over Ni/CeO2 catalyst. For other three catalysts, only 1/3 amount of the CO2 was adsorbed and the CO2 conversions were lower. Elevating the temperature up to 400 °C continuously accelerate the reaction rate and then the CO2 conversion for Ni/ZrO2, Ni/TiO2 and Ni/Al2O3 catalysts increased rapidly from 10–20% at 300 °C to 50–70% at 400 °C. However, the uptake ratio of H2/CO2 for Ni/CeO2 was only 2 which is much lower than the stoichiometric factor, which would affect the methanation rate. Stangeland [29] also reported that high CO2 conversion was difficult to be achieved under 400 °C which was mainly associated with the difficulty of CO2 activation and slow kinetics of methanation.

Fig. 5
figure 5

Correlation between CO2 conversion and the H2 and CO2 uptakes for the Ni-based catalysts at different temperatures a 250 °C b 300 °C c 400 °C

Full size image

It is reported [16, 37] that the metal sites on the catalyst adsorb the H2 and provide reactive H-species and the support acts as the active site for CO2 coverage and activation. Therefore, a cooperative catalytic mechanism that enough H2 and CO2 adsorbed on the metallic nickel and support, respectively, followed by subsequent activation and reaction to yield methane, is proposed.

Furthermore, the peak for CO2 desorption from CeO2 lower than 230 °C is assigned essentially to bridged bidentate carbonates species that formed from CO2 absorbed on Ce3+ site or oxygen vacancy of the CeO2 surface [38], while the next peak lower than 430 °C is attributed to bidentate and polydentate carbonates on Ce4+ support. The highest peak between 430 and 630 °C is assigned to inorganic carboxylate and monodentate carbonate [39]. Therefore, it can be seen from Fig. 4 that CO2 adsorbed on Ni/CeO2 catalyst mainly formed bridged bidentate carbonates species and a part of bidentate or polydentate carbonates because the CeO2 support could be partially reduced to Ce3+ by pre-reduction and H2 during reaction. Bridged carbonates site can lower the activation energy for the formation of formate. The dissociated hydrogen on metallic Ni can react with the weakly adsorbed bridged carbonates on the Ce3+ site to produce methane and reduced ceria can be oxidized by CO2 at this temperature [40]. The Ni/CeO2 catalyst also provide stronger H adsorption at this temperature, leading to the enhancement of H2 coverage and in the likelihood of hydrogenation of the bridged bidentate carbonates species to CH4.

High temperatures, such as 400 °C, start to be high enough to dissociate CO2 and may tend toward rapid formation of CO then hydrogenate to the formation of methane. Due to the low H2 surface coverage, the selectivity to CH4 was lower than the other three catalysts for Ni/ZrO2 catalyst.

4 Conclusions

Four Ni catalysts supported on Al2O3, ZrO2, TiO2 and CeO2 were prepared and investigated for CO2 methanation performance. The Ni/CeO2 catalyst exhibited the highest CO2 conversion at temperatures lower than 400 °C. The CO2 conversion for the Ni/CeO2 catalyst was 60.1% at 300 °C while that for other three catalysts was all lower than 20%, while the CH4 selectivity all approached to the equilibrium value. The high interaction between nickel and support in the Ni/Al2O3 catalyst resulted in small metallic nickel well dispersed on the support and the H2 uptake was high as 242 μmol/gcat. However, more amount of CO2 was adsorbed on the strong basic sites. Therefore, at low reaction temperatures the CO2 conversion was lower than that over the Ni/CeO2 catalyst, in which Ni nanoparticles and basic support serve as H2 and CO2 active centers respectively and cooperatively catalyze CO2 methanation, resulting in low-temperature reaction activity.