Ni nanoparticles highly dispersed on ZrO₂ and modified with La₂O₃ for CO methanation

Abstract

To improve the anti-sintering and anti-carbon deposition ability of the supported metallic nano catalysts, a new scheme for designing and preparing catalysts for CO methanation is presented in this work. In the scheme, a series of xLaNiO₃/ZrO₂ (x = 15%, 20%, 25%) catalysts were prepared according to the citrate complexing method. The catalysts were characterized by using BET, XRD, H₂-chemisorption, H₂-TPR, and TEM technologies. After reduction, the LaNiO₃/ZrO₂ catalyst precursor preferred to form Ni nanoparticles highly dispersed on ZrO₂ and modified with La₂O₃. Compared with the catalyst prepared by the traditional impregnation method, the Ni/La₂O₃–ZrO₂ derived from LaNiO₃/ZrO₂ showed a higher dispersion of Ni on ZrO₂ and exhibited higher CO conversion and CH₄ selectivity. Meanwhile, the LaNiO₃/ZrO₂ catalyst showed excellent stability owing to the significant improvement in both anti-carbon deposition and anti-sintering. The catalyst prepared according to this scheme intensified the interaction between Ni and La₂O₃, thus favoring the synergistic effect of La₂O₃ and Ni, which led to the high dispersion of Ni nanoparticles as well as the very good anti-sintering and anti-carbon deposition ability.

---

1. Introduction

Natural gas (mainly CH₄) is a clean fuel, however, in China the reserve of natural gas is rather low. And due to the gradually increasing demand for natural gas, conversion of syngas to synthetic natural gas (SNG) via the methanation reaction has attracted much attention in the last decades.

During the production of synthetic natural gas from CO methanation, the following reaction occurs:

CO + 3H₂ → CH₄ + H₂O ΔH^°_{298 K} = −206 kJ mol⁻¹ (1)

CO + H₂O → CO₂ + H₂ ΔH^°_{298 K} = −41.2 kJ mol⁻¹ (2)

CH₄ → C + 2H₂ ΔH^°_{298 K} = +74.8 kJ mol⁻¹ (3)

The CO methanation reaction (eqn (1)) and the water gas shift reaction (WGSR, eqn (2)) are both thermodynamically favorable, and usually the heat of the reaction can hardly be removed. Consequently a hot spot is easily formed and sintering of the catalysts is serious.¹ Moreover, when the reaction temperature is high, the product CH₄ will be dissociated on the active component as shown in eqn (3),² which will result in the formation of a carbon deposition on the surface of the catalyst. The carbon will cover the active sites on the surface of the catalysts, leading to their deactivation. Therefore, developing an anti-sintering and/or anti-carbon deposition catalyst for the CO methanation reaction would be of great importance.

Currently, the catalysts reported for CO methanation include noble metal based catalysts and base metal catalysts, and the latter include mainly Fe, Co or Ni-based catalysts. Noble metal based catalysts such as Ru- and Rh-based catalysts are considered to be most active for syngas methanation.^3–5 However the high cost and limited reserves restrict their large-scale industrial applications. Fe-based catalysts show lower activity and poor selectivity to CH₄, and the stability of the catalysts is usually poor due to the serious carbon deposition on the surface of the catalysts.⁶ Co-based catalysts are not dominant for the selectivity of CH₄, even though they can be used under harsh reaction conditions.^7,8 Due to the relative low price, good activity and excellent selectivity to CH₄, Ni based catalysts become one of the most promising catalysts.^9,10

Great efforts have been made to resolve the problem of sintering and carbon deposition of the Ni catalysts. The sintering of Ni nanoparticles in the catalysts may be suppressed by loading Ni nanoparticles on carriers with large specific surface or limiting the size of Ni nanoparticles within ordered mesoporous structures, or by adding oxide promoters as the barrier of Ni nanoparticles, as reported in literatures.^11–14 The interaction between the carriers or oxide promoters and active components can confine metal and inhibit the migration of Ni nanoparticles. Carbon deposition can be reduced by adding promoters, such as La₂O₃, which can help eliminate carbon deposition on the surface of metal.^13,15 In addition, adding another metal to form alloy also can refrain the carbon deposition.¹⁶ Interestingly, Y. Li et al. reported that microstructured catalysts like open-cell metal foam can reduce the “hotspot” formation owing to its high thermal conductivity, and the sintering and carbon deposition were inhibited during the syngas methanation.^17–19

Perovskite-type oxides (PTO) are generally written as ABO₃, where A sites are usually occupied by rare-earth, alkaline-earth, or other large ions, while the B sites are filled with transition-metal cations.^15,20 PTOs are extensively used as catalysts extensively for oxidation/reduction reactions, such as methane catalytic combustion, carbon dioxide reforming of methane, partial oxidation of methane and so on.^21,22

LaNiO₃ with perovskite structure has been used as the precursors for preparing Ni nano catalysts. D. Lima et al.²³ found that highly dispersed Ni on La₂O₃ can be obtained by the reduction of LaNiO₃, and the resulted Ni/La₂O₃ catalyst is highly resistant to carbon deposition for SRE reaction. J. Gao et al.²⁰ found that after treatment under the reaction conditions of CO₂ methanation, LaNiO₃ perovskite would be transferred to highly dispersed Ni nanoparticles supported on La₂O₂CO₃, and the catalysts showed high catalytic activity for CO₂ methanation.

However, there is a defect for Ni/La₂O₃ catalyst by using LaNiO₃ as the precursor. Usually the surface area of LaNiO₃ is rather small, so after long time reaction, active species of nickel would be sintered inevitably. Meanwhile, the loading amount of Ni cannot be adjusted. An effective way for improving the surface of PTOs is to load PTOs on a support which has a high specific surface area. As for methanation reaction, ZrO₂ is a good choice for the support. Takenaka et al.²⁴ found that Ni/ZrO₂ has the best catalytic performance compared with that of Ni/MgO, Ni/Al₂O₃, Ni/SiO₂ and Ni/TiO₂ for CO methanation. L. G. Appel et al.²⁵ pointed out that the methanation of CO can occur on ZrO₂, which is related to the presence of acid sites on ZrO₂ surface. The acid sites on ZrO₂ can adsorb CO and then adsorbed CO reacts with H₂ spillover from Ni.²⁶

Based on the above considerations, a new scheme for the preparation of LaNiO₃/ZrO₂ is proposed in this work, and the LaNiO₃/ZrO₂ catalyst is firstly used for CO methanation. The design process has the following advantages: (1) ZrO₂ is a promising support for Ni, and ZrO₂ possesses a high specific surface area, thus the sintering of Ni nanoparticle on the support would be restricted; (2) after the reduction of LaNiO₃/ZrO₂, Ni would mix with La₂O₃ nanoparticles uniformly and highly disperse on ZrO₂ for that La³⁺ and Ni³⁺ are evenly distributed at atomic level in the precursor of LaNiO₃. The uniform mixing of Ni and La₂O₃ favors the interaction between Ni and La₂O₃; (3) La₂O₃ can act as a barrier between the neighbouring Ni nanoparticles and inhibits the sintering of Ni nanoparticles; (4) La₂O₃ can react with CO₂ to form La₂O₂CO₃, which help eliminate carbon deposition. Moreover the interaction between Ni and La₂O₃ will reinforce the ability of carbon elimination.

2. Experimental section

2.1. Catalysts preparation

The support of ZrO₂ was prepared according to the precipitation method as reported in the literature.²⁷ The aqueous solution of 0.3 M ZrOCl₂ was added to a 0.3 M aqueous NH₄OH under stirring at a drop pace of 1 mL min⁻¹ until the pH of the solution was 9–10. The hydrous zirconia, after staying in its mother solution for 24 h, was filtered and washed with deionized water until no Cl⁻ could be detected by AgNO₃. The obtained filter cake was dried at 110 °C overnight and finally calcined at 700 °C for 5 h.

LaNiO₃/ZrO₂ was prepared according to the incipient wetness impregnation method combined with citrate complexing method. Stoichiometric lanthanum and nickel nitrate at La/Ni molar ratio of 1 : 1 were dissolved in deionized water. Then, citric acid and ethylene glycol were added into the solution with the molar ratio of the total metal ion/citric acid/ethylene glycol = 1 : 1.2 : 0.48 under stirring. The obtained aqueous solution was impregnated on ZrO₂ support. After staying overnight, the resulting sample was dried at 80 °C for 6 h and 120 °C for 12 h. The dried samples were calcined at 350 °C and 700 °C for 2 and 5 h, respectively, at a heating rate of 2 °C min⁻¹. And xLaNiO₃/ZrO₂ (x = 15%, 20% and 25%) were obtained. The synthesized xLaNiO₃/ZrO₂ (x = 15%, 20% and 25%) were respectively named as xLN-Z (x = 15, 20, 25). Meanwhile, unsupported LaNiO₃ was prepared according to the citrate complexing method.

For comparison, NiO–La₂O₃/ZrO₂ was prepared by incipient-wetness impregnation method. Unlike the preparation method of xLaNiO₃/ZrO₂, no citric acid and ethylene glycol in the impregnation solution were used. In NiO–La₂O₃/ZrO₂, the content of Ni and La were same as that in 20% LaNiO₃/ZrO₂ catalyst. The obtained NiO–La₂O₃/ZrO₂ was labeled as 20L-N-Z-imp.

2.2. Characterization of catalysts

Nitrogen adsorption and desorption isotherms were tested on a Trwastar 3000 micromeritics apparatus at −196 °C. The samples were out gassed at 300 °C for 4 h under vacuum before starting N₂ adsorption. The specific surface areas were calculated according the BET method, the pore size distributions and pore volume were derived from the BJH method using the desorption branch of the isotherms.

X-ray diffraction (XRD) patterns of the catalysts were recorded on a Bruker D8-Focus X-ray diffractometer with Ni-filtered Cu Kα radiation (λ = 0.15406 nm). The scanning range (2θ) was between 20 to 55° at a scanning speed of 5° min⁻¹. The Scherrer equation was used to calculate the crystal size of the samples from the X-ray patterns.

Transmission electron microscopy (TEM) experiments were conducted on a FEI (PHILIPS) Tecnai-G2F20 instrument equipped with Field Electronic Emitter. After ultrasonic dispersion of the catalysts in absolute ethanol, the samples were deposited on copper grids with a holey carbon film support.

Temperature programmed reduction (TPR) tests were carried out on a Thermo-Finnigan instrument. In each run, 50 mg of the catalyst was loaded into a quartz tube reactor and pretreated in N₂ at 200 °C for 1 h to remove the air in the reactor. Then, the catalyst was heated from room temperature to 850 °C with a heating rate of 10 °C min⁻¹ in 5% H₂/Ar with a flow rate of 50 mL min⁻¹.

H₂ chemisorption test were conducted on the same instrument as H₂-TPR. 100 mg of the sample were in situ reduced at 600 °C for 1 h in H₂ (60 mL min⁻¹), and then flushed by Ar (60 mL min⁻¹) at 600 °C for 1 h. After the temperature was cooled to 30 °C, 50 mL of H₂ pulses were injected into the Ar stream until the effluent area of consecution pulses was constant. Ni dispersion and the average nickel crystallite diameter were calculated based on the assumption of spherical crystallites of uniform size.²⁸

Thermogravimetric analysis (TG) was conducted on a DTG-50/50H thermal analyser to obtain the amount of carbon deposited on the spent catalysts. During TG tests, 10 mg of the sample was used and heated from 20 to 800 °C with a ramping rate of 10 °C min⁻¹ in air atmosphere.

2.3. Catalytic performance

Catalytic performance tests were carried out on a continuous-flow fixed tubular quartz micro-reactor with an 8 mm inside diameter at atmospheric pressure. In each test, 200 mg of catalyst was pre-reduced at 600 °C for 2 h under H₂. The reaction feed stream is composed of 20% CO, 60% H₂ and 20% N₂ in volume, where N₂ was used as the internal standard gas to analyze the composition of the off-gases. The reaction temperature was monitored by a K-type thermocouple placed in the catalyst bed and controlled by a temperature controller. The effluent gases were analyzed using an on-line gas chromatograph (GC) of SP-3420 with a TCD detector. The catalytic performance of catalyst is evaluated by CO conversion (X_CO) and CH₄ selectivity (S_CH4), which are defined as the following formulas:here, [CO]_in is the moles of CO in the feed stream, [CO]_out is the moles of CO in the effluent gas; [CH₄]_out is the moles of CH₄ in the effluent gas.

3. Results and discussion

3.1. Catalyst characterizations

3.1.1. N₂ adsorption–desorption. The N₂ adsorption–desorption isotherms and the corresponding pore size distribution (the inset picture) of the catalysts are shown in Fig. 1. The N₂ adsorption–desorption isotherm of the samples exhibit typical IV adsorption stripping curve with a H₂-type hysteresis loop, suggesting the existence of mesopores. All of the pores in the samples were formed by particles packing, and not in parallel arranged structure.

Fig. 1 N₂ adsorption–desorption isotherms of (a) BJH pore size distributions and (b) of ZrO₂ and xLN-Z.

With the increasing loading amount of LaNiO₃, the onset of the hysteresis loop moved to the higher relative pressure, and the hysteresis loops become small, meaning that the loaded LaNiO₃ entered into the pores of ZrO₂.

The BET surface area of the ZrO₂ is 90.1 m² g⁻¹ (listed in Table 1). After loading LaNiO₃, the specific surface areas decrease from 90.1 m² g⁻¹ of ZrO₂ support to 59.5 m² g⁻¹ of 25LN-Z, which should be due to the blockage of the meso-pores of ZrO₂ by the perovskite-type oxide.

Table 1 Physical properties of ZrO₂ and xLN-Z

Sample	SBET (m^2 g^−1)	Pore size (nm)	VBJH (cm^3 g^−1)	DZrO2^a (nm)	Dperovskite^a (nm)
a Calculated from XRD results with Scherrer formula: D = kλ/(β cos θ).
ZrO2	90.1	29.32	0.257	16.3	—
15LN-Z	83.3	17.18	0.192	15.4	13.3
20LN-Z	62.7	17.06	0.157	16.3	13.9
25LN-Z	59.5	16.91	0.153	15.7	15.1

---

The most frequency pore sizes and the pore volumes were determined from the desorption isotherms, and the results are listed in Table 1. The pore volumes and the pore sizes decrease after loading the perovskite-type oxide of LaNiO₃ compared with that of ZrO₂, which indicate that subsequently impregnated LaNiO₃ entered into the pores of ZrO₂. Meanwhile, the sizes of LaNiO₃ calculated by XRD are about 15 nm, which is smaller than the pore size of ZrO₂ (29.32 nm). Thus, the LaNiO₃ would be dispersed on the surface of ZrO₂ support.

3.1.2. XRD. The XRD patterns of xLN-Z (x = 15, 20, 25), 20L-N-Z-imp and ZrO₂ are presented in Fig. 2. For ZrO₂, the main crystal phase in all the samples is monoclinic ZrO₂ (m-ZrO₂). For xLN-Z (x = 15, 20, 25) in Fig. 2a, a new diffraction peak at 2θ ≈ 33° appears, and the new diffraction peaks are attributed to the perovskite structure LaNiO₃. With the increasing of the loading amount of LaNiO₃, the diffraction intensity of LaNiO₃ increases. For 20L-N-Z-imp, another two diffraction peaks at 2θ ≈ 37.2 and 43.3° can be observed in Fig. 2a, and the two diffraction peaks are assigned to NiO. Namely, pure LaNiO₃ with perovskite structure was loaded on ZrO₂ in xLN-Z, which agrees with the catalyst designing scheme, while NiO was formed in the 20L-N-Z-imp.

Fig. 2 XRD patterns of xLN-Z, 20L-N-Z-imp and ZrO₂: (a) the fresh catalysts; (b) catalysts after reduction; (c) catalysts after reaction (△) m-ZrO₂; (●) Ni; (♠) LaNiO₃; (◆) NiO.

The XRD patterns of reduced catalysts are shown in Fig. 2b. After reduction, the diffraction peak for each sample attributed to perovskite structure disappear and a very faint diffraction peak corresponding to Ni at 2θ ≈ 44.48° can be observed in the enlarged XRD patterns, suggesting the high dispersion of Ni on ZrO₂. The atomic ratio of La/Ni in LaNiO₃ equals to 1, so the reduction of nickel ions from LaNiO₃ occurred concomitantly with the formation of La₂O₃. However, the diffraction peaks of La₂O₃ cannot be seen after reduction, which should be attributed to the highly dispersed La₂O₃ on ZrO₂. The similar situation was also reported by Sun et al., who found that diffraction peaks of La₂O₃ in the XRD patterns could be seen only when the loading amount of La₂O₃ on ZrO₂ was higher than 14 wt%.²⁹ The XRD results of the reduced catalysts suggest that Ni nanoparticles were highly dispersed on ZrO₂ and modified with La₂O₃.

Meanwhile, after the reduction of xLN-Z, La₂O₃ would mix with Ni nanoparticles uniformly for that La³⁺ and Ni³⁺ are evenly distributed at atomic level in the precursor of LaNiO₃/ZrO₂. In this way, La₂O₃ can act as barriers to prevent Ni nanoparticles contacting each other, and the sintering of Ni nanoparticles could be restricted. Besides, La₂O₃ can react with CO₂ to form La₂O₂CO₃, which help eliminate carbon deposition.²⁰ Therefore, the uniform distribution of Ni and La₂O₃ favors to improve the anti-coking and anti-sintering ability of the catalysts.

Fig. 2c shows the XRD patterns of the sample after reaction. The diffraction patterns of ZrO₂ have little change compared with the fresh samples, indicating that ZrO₂ is stable in the reaction process. After reaction, diffraction peaks corresponding to Ni at 2θ ≈ 44.48° are still much weak, meaning that Ni is stable under the reaction process. While for 20L-N-Z-imp, the diffraction peak of Ni seems a little sharper than 20LN-Z, meaning that the sintering of Ni occurred and the anti-sintering ability of xLN-Z is much better.

3.1.3. H₂ chemisorption. H₂ chemisorption of xLN-Z and 20L-N-Z-imp after reduction and after reaction are listed in Table 2. For xLN-Z (x = 15, 20, 25) after reduction, the H₂ uptake increases with the increasing of the loading amount of LaNiO₃. The highest dispersion of Ni nanoparticles can be seen in 20LN-Z, and for 25LN-Z the crystal size of LaNiO₃ is bigger (see Table 1) which would contribute to the lower dispersion of Ni. Compared with xLN-Z, the adsorption amount of hydrogen and dispersion of Ni in 20LN-Z-imp are obviously lower, indicating that nickel catalyst prepared from LaNiO₃/ZrO₂ is more conducive to high dispersion of nickel.

Table 2 H₂ chemisorption results of the catalysts of xLN-Z (x = 15, 20, 25) and 20L-N-Z-imp

Sample	After reduction		After reaction
	H2 uptake (μmol g^−1)	Ni dispersion^a (%)	H2 uptake (μmol g^−1)	Ni dispersion^a (%)
a Based on H2 uptake using the equation: D = Ni atoms on the surface/total Ni atoms.
15LN-Z	31.6	10.1	29.2	9.5
20LN-Z	45.9	11.0	41.8	10.3
25LN-Z	47.1	9.3	42.5	8.4
20L-N-Z-imp	27.4	6.5	16.1	3.8

---

After reaction, the uptake of hydrogen and dispersion of Ni on LN-Z only change a little, meaning that sintering of Ni nanoparticles does not occur and that the anti-sintering ability is excellent. While for 20L-N-Z-imp, the uptake of hydrogen and dispersion of Ni decrease significantly after reaction, indicating the severer sintering of Ni nanoparticles. The result of H₂ chemisorption is consistent with the XRD results of above.

3.1.4. TPR. The reducibility analysis of the catalyst precursor was carried out by using TPR technique, and the H₂-TPR profiles of the samples are presented in Fig. 3. No reduction peak of ZrO₂ can be seen, meaning that ZrO₂ is stable enough and all the reduction peaks showed are contributed to the reduction of nickel ions.

Fig. 3 TPR profiles of xLN-Z (x = 15, 20, 25), LaNiO₃, 20L-N-Z-imp and ZrO₂.

It's seen that H₂-TPR profiles of xLN-Z (x = 15, 20, 25) exhibit two main reduction peaks, which appear at around 380 and 585 °C. The reduction profiles of xLN-Z are similar with that of LaNiO₃, indicating that perovskite-type oxide of LaNiO₃ had been loaded on ZrO₂ successfully, which is consistent with the XRD results and supports the catalyst designing scheme.

Studies on the reduction of LaNiO₃ are many,^30–32 according to the studies, the first peak is contributed to the reduction of Ni³⁺ to Ni²⁺, forming La₂Ni₂O₅, and the second peak is corresponding to the reduction of Ni²⁺ to Ni⁰. The two reduction steps can be expressed in the following formulas, respectively:

2LaNiO₃ + H₂ → La₂Ni₂O₅ + H₂O (7)

La₂Ni₂O₅ + H₂ → La₂O₃ + 2Ni + H₂O (8)

H₂ consumptions are calculated and listed in Table 3. The H₂ consumptions of low temperature peaks derived from the TPR results are close to the total theoretic H₂ consumptions of Ni³⁺ → Ni²⁺, and the H₂ consumptions of high temperature peaks derived from the TPR results are close to the theoretic H₂ consumptions of Ni²⁺ → Ni⁰. This supports the attributions of the TPR peaks above.

Table 3 H₂ consumption (μmol mg⁻¹) for xLN-Z and 20L-N-Z-imp

Samples	Experimental values		Theoretical values
	TL	TH	Ni^3+ → Ni^2+	Ni^2+ → Ni^0
15LN-Z	12.6	25.9	15.5	31.0
20LN-Z	16.0	32.4	20.1	41.7
25LN-Z	22.5	46.3	25.3	50.7
20L-N-Z-imp	43.3	3.2	—	41.7

---

Compared with pure LaNiO₃, the two reduction peaks of xLN-Z (x = 15, 20, 25) shift to higher temperatures. The reason is that the LaNiO₃ particles are highly dispersed on ZrO₂ and the strong interaction between LaNiO₃ and ZrO₂ restrains the reduction of LaNiO₃. With the increasing content of LaNiO₃, the particles size of LaNiO₃ increases and the interaction between LaNiO₃ and ZrO₂ is weakened, leading to the lower reduction temperature.

For 20L-N-Z-imp, one main hydrogen consumption peak is located at around 440 °C, and another small hydrogen consumption peak is observed at around 600 °C. From the XRD result of 20L-N-Z-imp, both NiO and LaNiO₃ phases are existed. So the first reduction peak at 450 °C is attributed to the reduction of NiO to Ni⁰ and Ni³⁺ in LaNiO₃ to Ni²⁺. The small peak at around 600 °C is contributed to the reduction of Ni²⁺ in La₂Ni₂O₅ to Ni⁰, see eqn (8).

3.1.5. TEM. The TEM images of 20L-N-Z and 20L-N-Z-imp after reduction are shown in Fig. 4. The particle size of ZrO₂ is about 15 nm (see Fig. 4a and b), which is in according with the XRD results above. In order to obtain the feature of reduced catalyst of 20LN-Z, high resolution TEM micrograph (HRTEM) and Fast Fourier Transform (FFT) were investigated (Fig. 4a, a′ and a′′). In the Fig. 4a′, the lattice distance of 1.76 Å is assigned to Ni (100) and the size of the Ni on ZrO₂ is about 5 nm. Around the Ni nanoparticles, the lattice distance of 2.84 Å is the lattice fringes of support ZrO₂ (111). The analysis result suggests that Ni is highly dispersed on ZrO₂. Lattice fringes corresponding to La₂O₃ could not be observed, suggesting La₂O₃ is in amorphous state.

Fig. 4 TEM micrographs of the catalysts after H₂ reduction (a and a′) 20LN-Z; (b) 20L-N-Z-imp; (a′′) EDS line scanning of selected area for 20LN-Z.

Fig. 5 The variation of (a) CO conversion, (b) CH₄ selectivity, (c) CH₄ yield with the reaction temperature at GSHV = 15 000 mL g⁻¹ h⁻¹, in the gas mixture of H₂/CO/N₂ = 3/1/1.

The line scanning profiles (Fig. 4a′′) of two particles present the element distributions of O, Zr, La and Ni. The signal of La is weak, therefore the signal of La was enlarged and shown in Fig. 4a′′. Seen from the species dispersion on the abscissa in the insets of Fig. 4a′′, nickel dispersed mainly in the area between 25 to 35 nm, lanthanum mainly between 15 to 45 nm, zirconium and oxygen between 15 to 50 nm. This means that the nickel nanoparticle is supported on lanthanum oxide and the lanthanum oxide is spread on zirconia.

It is known that La₂O₃ tends to interact with ZrO₂ as stated by H. Sun et al.²⁹ and the interaction between La₂O₃ and ZrO₂ would pull La₂O₃ to spread on ZrO₂ surface. Meanwhile, both metal nickel nanoparticle and La₂O₃ came from a LaNiO₃ grain, hence metal nickel nanoparticle is on La₂O₃ and surrounded by La₂O₃. This situation is not observed in Fig. 4a′, for that La₂O₃ is in amorphous state, and no lattice corresponding to La₂O₃ exists.

3.1.6. Structure evolution of catalysts before and after reduction. From XRD, TPR and TEM analysis for the catalysts of LN-Z, the structure variation of the catalysts can be described as the Scheme 1a. In the preparation, citric acid and ethylene glycol were added to the impregnation liquid, and during the complexing at 80 °C, all the metal ions entered into the macromolecular complex, that is, all the metal ions were homogeneously dispersed, leading to the formation of LaNiO₃ perovskite after calcination (see Fig. 2). In the catalyst precursor of LaNiO₃/ZrO₂, Ni³⁺ and La³⁺ are uniformly dispersed at the atomic level in the crystalline of LaNiO₃. In the reduction process Ni would be separated out from the lattice of LaNiO₃ to form Ni nanoparticles, which should be well dispersed in La₂O₃ matrix and loaded on ZrO₂ (see Fig. 4). In this way, Ni nanoparticles contact with La₂O₃ closely and isolated by La₂O₃ spatially, which favors the interaction between Ni and La₂O₃. The interaction between Ni and La₂O₃ will inhibit the sintering of Ni nanoparticles and eliminate carbon deposition, contributing to the good stability for CO methanation.

Scheme 1 The structural evolution of xLN-Z (a) and 20L-N-Z-imp (b).

In contrast, the structure evolution of 20L-N-Z-imp can be described in Scheme 1b. Without the citric acid and ethylene glycol in the impregnation solution, Ni³⁺ and La³⁺ are not uniformly dispersed at the molecular level in the dried samples. After calcination, NiO and La₂O₃ were formed and unevenly dispersed on ZrO₂, besides a few grains of LaNiO₃ perovskite generated (see Fig. 2). During reduction, the Ni will release from NiO and couldn't be well contact with La₂O₃, which will not favor the interaction between Ni and La₂O₃. So Ni nanoparticles are easier to sintering during the reaction process.

From the analysis of structural evolution of the two catalysts, the catalyst LaNiO₃/ZrO₂ fabricated by citrate complexing method is more beneficial to form highly dispersed Ni nanoparticles than the catalysts prepared by traditional impregnation method.

3.2. Catalytic performance for CO methanation

The variation of CO conversion, CH₄ selectivity and CH₄ yield with temperature for LaNiO₃/ZrO₂ and 20L-N-Z-imp are shown in Fig. 5. Completely CO conversion can be obtained over 20LN-Z and 25LN-Z in the temperature range of 260 to 400 °C, suggesting that the catalysts are much active, see Fig. 5a. As the reaction temperature is higher than 400 °C, CO conversion decreases with the increase of the reaction temperature, for that CO methanation is a strong exothermic reaction, and in high temperature CO methanation reaction would be suppressed.

Fig. 6 Conversion of CO, selectivity to CH₄ and CO₂ over 2LN-Z (a); 20L-N-Z-imp (b) vs. reaction time on stream at 550 °C, GHSV of 15 000 mL g⁻¹ h⁻¹ and in H₂/CO/N₂ = 3/1/1.

From Fig. 5b and c the highest CH₄ selectivity and CH₄ yield are obtained at 350 °C. The decrease of the selectivity and the yield of methane is attributed to the occurrence of WGSR at high temperature, seen eqn (2), resulting in the lower methane selectivity, methane yield and more CO₂.

For the catalysts 20LN-Z and 20L-N-Z-imp, the contents of loaded active metal are identical (4.9 wt%). The temperature for the maximum conversion of CO over catalyst of 20L-N-Z-imp is 350 °C, where the CO conversion is 96% and CH₄ selectivity is 78%. While over 20LN-Z, 100% of CO conversion and 89% of CH₄ selectivity were obtained at 260 °C. So 20LN-Z is much better. The higher activity of 20LN-Z is attributed to the higher dispersion of Ni nanoparticles derived from perovskite structure. Seen from Table 2, the Ni dispersion (11%) in the reduced 20LN-Z is much higher than that in 20L-N-Z-imp (6.5%). The higher selectivity to methane over 20LN-Z suggests that 20LN-Z does not favor the WGSR reaction, which is of interest and further research is needed.

In addition, the effect of LaNiO₃ loading in LaNiO₃–ZrO₂ on the catalytic performance for CO methanation was investigated. 20LN-Z is the most active catalyst among the three catalysts with 100% CO conversion and 89.5% CH₄ selectivity at 260 °C, and the catalytic performance remains almost constant in the temperature range from 260 °C to 400 °C. While the performance of 15LN-Z is relatively poor with CO conversion of 100% at 270 °C and the corresponding CH₄ yield is 86.9%, which should be attributed to the lower loading amount of the active component nickel (3.6%). The activity of 25LN-Z is comparable with that of 20LN-Z, due to that the active component of nickel is dispersed more highly in 20LN-Z. Seen from Table 2, the Ni dispersion in 25LN-Z is about 9.3%, being lower than that of 20LN-Z (11%).

3.3. Stability test

The catalyst is very stable at low temperature. In order to accelerate the deactivation of the catalyst and shorten the test time, the stability tests of catalysts 20LN-Z and 20L-N-Z-imp were carried out at a high temperature of 550 °C and the results are summarized in Fig. 6. Seen from Fig. 6a, 20LN-Z catalyst maintains its good activity and selectivity during the 220 hours stability test, meaning that no obvious deactivation of 20LN-Z occurred. While, 20L-N-Z-imp catalyst only remain stable in the initial 120 hours and then deactivation occurs. Namely, the prepared 20LN-Z is more stable and showed excellent stability for CO methanation.

During the reaction of CO methanation, the stability of the catalyst is subject to carbon deposition on the active metallic nickel and sintering of nickel significantly. Carbon deposition will cover active sites on the surface of the catalyst, leading to the deactivation of catalysts. Sintering will lead to the formation of large metal particles, lowering the overall surface area and activity. The excellent stability of LaNiO₃/ZrO₂ is closely associated with the synergistic effect of Ni and La₂O₃. The synergistic effect between Ni and La₂O₃ will suppress the sintering of Ni nanoparticles. In addition, La₂O₃ can react with CO₂ to form La₂O₂CO₃ which help eliminate carbon deposition,¹⁵ and the synergistic effect between Ni and La₂O₃ will reinforce the ability of carbon elimination.

3.4. Characterization of the used catalysts

XRD analysis of 20LN-Z and 20L-N-Z-imp after long time test is shown in Fig. 7. For all the samples, only diffraction peak of m-ZrO₂ could be seen and the m-ZrO₂ is stable. It was reported that La₂O₃ can stabilize zirconia and can improve the anti-sintering ability of zirconia effectively.^33,34 No obvious change can be seen for 20LN-Z after reduction and after stability test, suggesting that Ni nanoparticles were still highly dispersed on the ZrO₂ after long time test at high temperature. However, for the used 20L-N-Z-imp, the intensity of Ni diffraction peak increases (the inset in Fig. 8), implying that Ni nanoparticles in 20L-N-Z-imp are sintered after long time reaction. The results of XRD suggest that the catalysts prepared by citric acid complexing method possess stronger ability for anti-sintering compared with the catalysts prepared by traditional impregnation method.

Fig. 7 XRD patterns of ZrO₂, the reduced and used catalysts. The catalysts were used for 220 hours at 550 °C with GHSV of 15 000 mL g⁻¹ h⁻¹ in H₂/CO/N₂ = 3/1/1.

Fig. 8 TG curves of the used catalysts: (a) 20LN-Z, (b) 20L-N-Z-imp. The catalysts were used under the same conditions as in Fig. 7.

To investigate the anti-coking ability of the prepared catalysts, TG analysis was taken out and the results are shown in Fig. 8. For used 20LN-Z, the initial weight loss before 300 °C is attributed to the removal of H₂O and some physically adsorbed CO₂ species or formate/carbonate species.^35,36 And a weight gain about 0.7% is observed at around 600 °C, which is due to the oxidation of Ni to NiO. Except that, no obvious weight loss occurs, indicating that coking deposition on the surface of the catalyst is much small. While for 20L-N-Z-imp, obvious weight loss can be seen in the temperature range between 300 and 600 °C, which was attributed to the combustion of carbon deposited.^11,37 Namely, the 20LN-Z possesses excellent anti-coking ability compared with 20L-N-Z-imp.

TEM images of the used catalysts 20LN-Z and 20L-N-Z-imp are shown in Fig. 9. For 20LN-Z, no carbon deposition can be observed even after 220 hours stability test (seen in Fig. 9a). In contrast, some amorphous carbons can be observed on the surface of the used 20L-N-Z-imp catalyst. The amorphous carbon deposited on the catalyst would wrap up the active sites of Ni, leading to deactivation of the catalyst. The TEM results are consistent with the TG results.

Fig. 9 TEM micrographs of the used catalysts: (a) 20LN-Z; (b) 20L-N-Z-imp. The catalysts were used under the same conditions as in Fig. 7.

Carbon deposition is generally formed via disproportionation of carbon monoxide and/or decomposition of methane. The reactant CO can be absorbed on Ni active site, carbon species would be formed and absorbed on the surface of Ni, see the eqn (9). In addition, when the reaction temperature is too high, the produced CH₄ adsorbed on Ni active sites will be dissociated to form carbon species and hydrogen, see the eqn (10). Those are the main reasons for carbon deposition on Ni active site in the CO methanation.

2CO + Ni → C–Ni + CO₂ (9)

CH₄ + Ni → C–Ni + 2H₂ (10)

The better ability of anti-coking of the 20LN-Z catalyst may be attributed to the synergistic effect between Ni and La₂O₃. It is reported that La₂O₃ species generated from the reduction of LaNiO₃ can reacted with CO₂ to form a carbonate layer of La₂O₂CO₃, see the eqn (11). And the carbonate species subsequently would react with surface carbon species on Ni sites, see the eqn (12), hence converting carbon species to CO.¹⁵

La₂O₃ + CO₂ → La₂O₂CO₃ (11)

C–Ni + La₂O₂CO₃ → 2CO + Ni + La₂O₃ (12)

In the catalyst precursor 20LaNiO₃/ZrO₂, Ni³⁺ and La³⁺ are uniformly dispersed at the atomic level in the crystalline of LaNiO₃, when treated with hydrogen Ni is released from LaNiO₃ along with La₂O₃ and highly dispersed on ZrO₂. In this way, lanthanum oxide has an intimate contact with Ni. In the reaction process, La₂O₃ converted to La₂O₂CO₃ as eqn (11) shown. When carbon species deposited on Ni surface, La₂O₂CO₃ around the Ni will quickly react with those carbon species, hence inhibiting the formation of carbon deposition, see the eqn (12). Therefore the synergistic effect between Ni and La₂O₃ contributed to the elimination of carbon deposition.

From XRD, TEM, and TGA analysis of the used catalysts, the good stability of 20LN-Z catalysts can be attributed to the much improved anti-coking and anti-sintering properties.

4. Conclusion

LaNiO₃ with the perovskite structure was loaded on the surface of ZrO₂ by using the citrate complexing method. After reduction, LaNiO₃/ZrO₂ favors the formation of the highly dispersed Ni nanoparticles supported on ZrO₂ and modified by La₂O₃. The Ni/ZrO₂–La₂O₃ catalyst reduced from LaNiO₃–ZrO₂ exhibited very good activity for CO methanation, due to the much high dispersion of Ni nanoparticles.

In the reduced catalysts of Ni/ZrO₂–La₂O₃, nanoparticles Ni are surrounded by or loaded on La₂O₃, for that both of them come from the nano grains of LaNiO₃. Compared with NiO–La₂O₃/ZrO₂ prepared by traditional impregnation method, the LaNiO₃/ZrO₂ catalysts exhibited higher stability, for the higher ability of anti-sintering and anti-carbon deposition. The interaction between Ni and La₂O₃ will suppress the sintering of Ni nanoparticles as well as help eliminate the carbon deposited.

Acknowledgements

The financial support of this work by NSFC (No. 21376170, 21576192) is gratefully acknowledged.

References

1. J. Sehested, S. Dahl, J. Jacobsen and J. R. Rostrup-Nielsen, J. Phys. Chem. B, 2005, 109, 2432–2438.
2. J. W. Snoeck, G. Froment and M. Fowles, J. Catal., 1997, 169, 250–262.
3. A. Beuls, C. Swalus, M. Jacquemin, G. Heyen, A. Karelovic and P. Ruiz, Appl. Catal., B, 2012, 113, 2–10.
4. J. N. Park and E. W. McFarland, J. Catal., 2009, 266, 92–97.
5. C. Galletti, S. Specchia and V. Specchia, Chem. Eng. J., 2011, 167, 616–621.
6. A. L. Kustov, A. M. Frey, K. E. Larsen, T. Johannessen, J. K. Nørskov and C. H. Christensen, Appl. Catal., A, 2007, 320, 98–104.
7. A. N. Akin, M. Ataman, A. E. Aksoylu and Z. I. Önsan, React. Kinet. Catal. Lett., 2002, 76, 265–270.
8. R. Razzaq, C. Li, M. Usman, K. Suzuki and S. Zhang, Chem. Eng. J., 2015, 262, 1090–1098.
9. H. Tian, S. Li, L. Zeng, H. Ma and J. Gong, Front. Mater. Sci. China, 2015, 58, 9–15.
10. J. Zarfl, D. Ferri, T. J. Schildhauer, J. Wambach and A. Wokaun, Appl. Catal., A, 2015, 495, 104–114.
11. Q. Liu, J. Gao, F. Gu, X. Lu, Y. Liu, H. Li, Z. Zhong, B. Liu, G. Xu and F. Su, J. Catal., 2015, 326, 127–138.
12. X. Lu, F. Gu, Q. Liu, J. Gao, Y. Liu, H. Li, L. Jia, G. Xu, Z. Zhong and F. Su, Fuel Process. Technol., 2015, 135, 34.
13. B. Valle, B. Aramburu, A. Remiro, J. Bilbao and A. G. Gayubo, Appl. Catal., B, 2014, 147, 402–410.
14. C. Guo, Y. Wu, H. Qin and J. Zhang, Fuel Process. Technol., 2014, 124, 61–69.
15. K. Sutthiumporn, T. Maneerung, Y. Kathiraser and S. Kawi, Int. J. Hydrogen Energy, 2012, 37, 11195–11207.
16. D. Harshini, Y. Kwon, J. Han, S. P. Yoon, S. W. Nam and T.-H. Lim, Korean J. Chem. Eng., 2010, 27, 480–486.
17. Y. Li, Q. Zhang, R. Chai, G. Zhao, F. Cao, Y. Liu and Y. Lu, Appl. Catal., A, 2016, 510, 216–226.
18. Y. Li, Q. Zhang, R. Chai, G. Zhao, Y. Liu and Y. Lu, ChemCatChem, 2015, 7, 1427–1431.
19. Y. Li, Q. Zhang, R. Chai, G. Zhao, Y. Liu, Y. Lu and F. Cao, AIChE J., 2015, 61, 4323–4331.
20. J. Gao, L. S. Jia, W. P. Fang, Q. B. Li and H. Song, J. Fuel Chem. Technol., 2009, 37, 573–577.
21. Y. J. Su, K. L. Pan and M.-B. Chang, Int. J. Hydrogen Energy, 2014, 39, 4917–4925.
22. M. Morales, F. Espiell and M. Segarra, Int. J. Hydrogen Energy, 2014, 29, 6454–6461.
23. S. M. de Lima, A. M. da Silva, L. O. O. da Costa, J. M. Assaf, G. Jacobs, B. H. Davis, L. V. Mattos and F. B. Noronha, Appl. Catal., A, 2010, 377, 181–190.
24. S. Takenaka, T. Shimizu and K. Otsuka, Int. J. Hydrogen Energy, 2004, 29, 1065–1073.
25. D. C. D. da Silva, S. Letichevsky, L. E. P. Borges and L. G. Appel, Int. J. Hydrogen Energy, 2012, 37, 8923–8928.
26. R. F. Wu, Y. Zhang, Y. Z. Wang, C. G. Gao and Y. X. Zhao, J. Fuel Chem. Technol., 2009, 37, 578–582.
27. G. Liu, Y. Geng, D. Pan, Y. Zhang, T. Niu and Y. Liu, Fuel Process. Technol., 2014, 128, 289–296.
28. J. Guo, Z. Hou, J. Gao and X. Zheng, Energy Fuels, 2008, 22, 1444–1448.
29. H. Sun, Y. Ding, J. Duan, Q. Zhang, Z. Wang, H. Lou and X. Zheng, Bioresour. Technol., 2010, 101, 953–958.
30. F. Liu, Y. Qu, Y. Yue, G. Liu and Y. Liu, RSC Adv., 2015, 5, 16837–16846.
31. C. Guo, J. Zhang and X. Zhang, React. Kinet. Catal. Lett., 2008, 95, 89–97.
32. G. R. Moradi, M. Rahmanzadeh and F. Khosravian, J. CO2 Util., 2014, 6, 7–11.
33. P. Thangadurai, A. C. Bose and S. Ramasamy, J. Mater. Sci., 2005, 40, 3963–3968.
34. K. R. Reddy, T. Bhaskar and K. V. Chary, Langmuir, 2003, 19, 10795–10802.
35. V. M. Shinde and G. Madras, AIChE J., 2014, 60, 1027–1035.
36. R. Razzaq, C. Li, M. Usman, K. Suzuki and S. Zhang, Chem. Eng. J., 2015, 262, 1090–1098.
37. J. Gao, C. Jia, J. Li, F. Gu, G. Xu, Z. Zhong and F. Su, Ind. Eng. Chem. Res., 2012, 51, 10345–10353.

---