Abstract

Macroporous nanocast perovskites, LaFe1−xNixO3 (x = 0.3, 0.5, and 0.7), were synthesized by using a nanocasting technique with SBA-15 as a template and applied to methane dry reforming (MDR). The prepared catalysts were characterized by X-ray diffraction, transmission electron microscopy, specific-surface-area analysis, hydrogen temperature-programmed reduction, and thermogravimetric analysis. LaFe1−xNixO3 revealed a large specific surface area, which could enhance its catalytic activity. The catalysts were reduced to Ni/LaFeO3-La2O3 in the MDR reaction. The alkaline additive, La2O3, and perovskite oxide, LaFeO3, strongly interacted with the active component to reduce the surface energy of metal particles and prevent aggregation of active Ni. The results showed that LaFe0.5Ni0.5O3 and LaFe0.3Ni0.7O3 perform better than LaFe0.7Ni0.3O3. More importantly, LaFe0.5Ni0.5O3 had a very long lifetime (>80 h) in the MDR reaction. The LaFe0.5Ni0.5O3 catalyst showed excellent stability in the MDR reaction and has potential use in industrial applications.

1. Introduction

The extensive use of traditional energy sources has led to the production of large amounts of CO2, leading to two main global problems: an energy crisis and environmental pollution. Scientific researchers are currently seeking methods to solve these two issues [14]. Ongoing breakthroughs in the technology of shale gas extraction imply that natural gas, which comprises mostly CH4, can be expected to replace coal as the second most abundant fossil fuel. CH4 dry reforming (MDR, CH4 + CO2 ⟶ 2H2 + 2CO) has recently received increased attention [3, 5]. The reaction simultaneously converts two greenhouse gases (CO2 and CH4) into syngas (H2 and CO), which reduces the amount of the former gases in the atmosphere. Syngas is mainly applied as a fuel or feedstock in the chemical industry. Therefore, the MDR reaction presents positive environmental benefits as well as efficient energy conversion and utilization [69].

Noble metals and Ni-based and Co-based catalysts are commonly used in the MDR reaction [4, 1015]. Although noble metal catalysts possess excellent catalytic activity, stability, and anti-C deposition capability, they are expensive and limited because they are difficult to use on a large scale. By comparison, Ni-based catalysts are inexpensive and can achieve catalytic effects comparable with those of noble metal catalysts by employing different carriers, additives, and preparation methods; hence, they are considered to be among the most promising industrial catalysts. However, Ni-based catalysts tend to become deactivated when subjected to long-term reactions due to C deposition and the sintering of Ni grains. Therefore, resistance to C deposition and prevention of Ni particle sintering must be studied [13, 1624].

Perovskite-type oxides (PTOs) have shown great potential as precursors for the catalytic reformation of CH4 and CO2 [3, 15, 2533]. PTO has the general formula ABO3, wherein A is usually a La-based ion or a second main-group metal ion and B is usually a transition metal ion. PTOs have good tunability and thermodynamic and chemical stabilities. Previous studies have reported the use of perovskite-type catalysts for MDR. Gallego et al., for example, prepared LaNiO3 as a catalyst precursor; the material featured small highly dispersed Ni particles supported on La2O3 during the MDR reaction and exhibited high catalytic activity [34]. Despite this achievement, however, two issues were uncovered: the small surface area (<10 m2·g−1) of the oxide affects its potential applications and the perovskite structure of LaNiO3 is completely decomposed during the MDR reaction [5, 26]. Such perovskite decomposition results in the weakening of the interactions between Ni and the carrier, leading to Ni sintering and C scattering. One method to solve the sintering problem is to scatter the PTOs over a material with a large specific surface area, thermal stability, and metal sintering resistance, such as ordered SBA-15 silica [3537]. The issue of C deposition has been considered in recent studies, which have shown that controlling the LaNiO3 precursor or partial substitution of Ni-containing perovskites largely suppresses Ni sintering and C deposition. Partial substitution of Ni by Fe in LaNiO3 stabilizes the perovskite structure by providing stronger metal-support interactions and maintaining a larger specific surface area [9, 28]; thus, it allows the catalyst to endure the MDR reaction.

In this study, we synthesized large specific surface area macroporous LaFe1−xNixO3 perovskite-type catalysts by using nanocatalyst technology and SBA-15 as a template. The prepared catalysts enabled the thorough dispersion of active Ni and showed high catalytic activity. Active Ni strongly interacts with the carrier to prevent the growth of Ni grains, thereby remarkably improving the stability of the prepared catalysts. The activity of LaFe1−xNixO3 during MDR was examined.

2. Materials and Methods

2.1. Catalyst Preparation
2.1.1. Preparation of Ordered SBA-15 Silica

SBA-15 silica was synthesized according to a previously reported method [35, 3739]. Briefly, 4.0 g of P123 (Aldrich, typical Mn = 5800) was dissolved in HCl (60 g, 4 M) and 90 ml of distilled water with stirring at 40°C for 4 h. Subsequently, 8.5 g of tetraethylorthosilicate was added to the solution, which was then kept at 40°C for 22 h. The solution was transferred to a Teflon bottle and crystallized at 100°C for 24 h. Finally, the product was washed to pH 7, filtered, dried, and calcined at 550°C for 7 h.

2.1.2. Synthesis of Macropore Perovskites

The synthesized SBA-15 was used as a template to synthesize macroporous LaFe1−xNixO3 [36, 39]. First, 1.5 g of SBA-15 was dissolved in 15 ml of distilled water for 30 min to obtain a suspension solution. Then, lanthanum nitrate, iron nitrate, nickel nitrate, and citric acid were added to citric acid dissolved in 20 mL of ethanol. The molar ratio of metal ions in the solution of La : Fe : Ni was 1 : 1 − x : x, where x = 0.3, 0.5, or 0.7. The prepared suspension solution was added to this mixture and stirred at 80°C until a gelatinous solid formed. This solid was dried at 110°C for 24 h and calcined at 750°C for 7 h. Finally, 2 M NaOH aqueous solution was added to the solids 3-4 times with stirring to remove the silica template. The obtained samples were washed to pH 7 with deionized water and ethanol and dried at 100°C for 10 h. The catalysts obtained via this nanocasting method were designated as LaFe1−xNixO3 (x = 0.3, 0.5, and 0.7). For comparison, LaFe1−xNixO3 was also synthesized by following the same protocol described above but without addition of the SBA-15 template. The prepared perovskite was designated as CA-LaFe1−xNixO3 (Scheme 1).

2.2. Characterization of Catalysts

X-ray diffraction (XRD) experiments were conducted on an X-ray diffractometer. Hydrogen temperature-programmed reduction (H2-TPR) of the catalysts was performed with an Auto Chem 2720 catalyst characterization system (Micromeritics Instrument Corporation, USA). Thermogravimetric/differential scanning calorimetry (TG/DSC) was performed on a TA SDT Q600 system (USA). Transmission electron microscopy (TEM) was conducted on a FEI Tecnai G2 F20 instrument (USA). The BET testing was performed on an ASAP 2460 instrument.

2.3. Catalytic Performance Tests

The MDR reactions were conducted in a fixed-bed reactor. The reactor was loosely filled with 75 mg of the LaFe1−xNix03 catalysts and fed a mixture of CH4, CO2, and N2 (CH4 : CO2 : N2 = 1.3 : 1.3 : 1, GHSV = 30000 mL·h−1·gcat). The catalytic tests were performed from 550°C to 850°C. The effluent product gases were analyzed by a GC-9790 gas chromatograph with a thermal conductivity detector. The stability tests were also conducted at 800°C for 80 h.

Conversions and H2/CO were calculated according to the following equations [40, 41]:where “in” represents the amount of a substance in the feed gas and “out” represents the amount of a substance in the gaseous effluent.

3. Results and Discussion

3.1. Physicochemical Properties of the Catalysts

The XRD patterns of the prepared LaFeO3 and LaFe1−xNixO3 catalysts are presented in Figure 1. Catalysts prepared by the nanocasting technique exhibited a single-phase crystalline perovskite structure. All peaks observed were consistent with those of the perovskite crystals of LaFeO3 (PDF#75-439) [42, 43]. The well-resolved and highly intense peaks reveal the excellent crystalline structures of the LaFe1−xNixO3 catalysts. As x increased in LaFe1−xNixO3, the catalysts exhibited broader, lower-intensity peaks that shifted toward higher 2θ values. This result can be explained by the fact that Ni3+ ions in LaFe1−x NixO3 are in the low-spin state and smaller than the high-spin state Fe3+ ions. Replacing Fe3+ in LaFeO3 with Ni3+ leads to a decrease in the crystal plane spacing of the PTO, which corresponds to the shift of the diffraction peaks toward higher 2θ values.

The TPR profiles of LaFe1−x NixO3 (x = 0.3, 0.5, 0.7) are shown in Figure 2. LaFe0.7Ni0.3O3 in the 375–550°C range shows two main reduction peaks. The first peak indicates that Ni3+ is reduced to Ni2+, while the second peak represents the reduction of Ni2+ to Ni0. Interactions between Ni and Fe complicate the reduction of perovskite and lead to expansion of the second peak, thus showing a strong interaction between the active Ni and carrier [2]. LaFe0.5Ni0.5O3 exhibits nearly the same trend as LaFe0.7Ni0.3O3, thus suggesting that both catalysts have similar structures. Interestingly, the Ni3+ and Ni2+ reduction peaks of LaFe0.3Ni0.7O3 move toward higher temperatures (415–750°C). This phenomenon is quite different from the findings of a previous report; thus, LaFe0.3Ni0.7O3 may have reduction intermediates different from those of LaFe0.7Ni0.3O3 and LaFe0.5Ni0.5O3 as the interaction between Ni and La/Fe species is likely to be altered as the Ni content increases [4244].

As shown in Table 1, the macroporous nanocast LaFe1−xNixO3 catalysts have significantly larger specific surface areas than the uncast CA-LaFe1−xNixO3. Among the catalysts prepared, LaFe0.5Ni0.5O3 showed the highest specific surface area. The improved textural characteristics of LaFe1−xNixO3 are attributed to the templating effect of SBA-15 during perovskite synthesis [39]. Because the prepared catalyst has a large specific surface area, dispersion of the active metal of the catalyst could be greatly improved. The porous structure of the catalyst also improves its activity and stability.

3.2. Catalytic Performance

The initial catalytic activity of the LaFe1−xNixO3 samples was tested to study the effect of Ni content on catalytic performance. In Figure 3(a) and 3(b), the conversion rates of CO2 and CH4 increased significantly at elevated reaction temperatures, thus displaying the endothermic feature of MDR [2, 7]. The CH4 and CO2 conversion rates of LaFe0.5Ni0.5O3 and LaFe0.3Ni0.7O3 indicated higher catalytic performance compared with that of LaFe0.7Ni0.3O3 at 550–850°C. This finding could be attributed to the Ni content of the catalysts. Further, the CO2 conversion was over 80%, which is higher than that of CH4, suggesting the RWGS reaction (CO2 + H2 ⟶ H2O + CO). LaFe0.5Ni0.5O3 and LaFe0.3Ni0.7O3 exhibit nearly the same catalytic performance at 550–850°C. The Ni content of LaFe0.3Ni0.7O3 is higher than that of LaFe0.5Ni0.5O3 (Figure 3), but LaFe0.3Ni0.7O3 does not display greater activity than LaFe0.5Ni0.5O3. This finding may be explained by the decomposition of the perovskite structure of LaFe0.3Ni0.7O3 during the reaction; this decomposition reduces the specific surface area of the catalyst and promotes agglomeration of Ni species. The high activity of LaFe0.5Ni0.5O3 may be due to its stable perovskite structure and high dispersion of active Ni.

3.3. XRD Patterns after Catalytic Performance

The XRD patterns of LaFe1−xNixO3 (x = 0.3, 0.5, 0.7) after reduction and the MDR reaction are shown in Figure 4. No significant characteristic diffraction peak of Ni (111) was observed at 44.7° (PDF#1-1260) in the XRD patterns of the LaFe1−xNixO3 catalysts in Figure 4(a) after reduction at 600°C for 2 h, which suggests that active Ni is highly dispersive with particles in small size. Among the three catalysts, LaFe0.7Ni0.3O3 and LaFe0.5Ni0.5O3 clearly maintain a good perovskite structure (Figures 4(a) and 4(b)), which reveals their high stability. After the reduction of LaFe1−xNixO3 (Table 2), LaFe0.5Ni0.5O3 shows the largest specific surface area, which means this catalyst has the best perovskite structure and Ni dispersion among the prepared catalysts. On the basis of the Scherrer formula (Figure 4(b)), the average particle sizes of the Ni phase of LaFe0.5Ni0.5O3 and LaFe0.3Ni0.7O3 were calculated to be 12.3 and 16.5 nm, respectively. The particle size of active Ni in LaFe0.5Ni0.5O3 is relatively small; thus, the catalyst could be expected to have a long lifetime [45]. No obvious diffraction peak of Ni was found in LaFe0.7Ni0.3O3, which performs very poorly in the activity test. Overall, Figure 4 reveals that LaFe0.5Ni0.5O3 maintains a stable perovskite structure and smaller Ni particle size compared with the other catalysts after reduction and MDR; thus, this catalyst possesses excellent catalytic activity and stability.

3.4. Stability of the LaFe0.5Ni0.5O3 and LaFe0.3Ni0.7O3 Catalysts

Based on the XRD images and BET data of the three LaFe1−xNixO3 catalysts, LaFe0.5Ni0.5O3 and LaFe0.3Ni0.7O3 were selected and their stabilities were tested at 800°C for 80 h (Figure 5). The results illustrate that LaFe0.5Ni0.5O3 exhibits long-term stability than LaFe0.3Ni0.7O3. The CH4 and CO2 conversion rates of LaFe0.5Ni0.5O3 remained stable, and no significant deactivation was detected throughout the 80 h stability test; by contrast, the activity of LaFe0.3Ni0.7O3 was significantly reduced during the stability test. Deactivation of Ni-based catalysts during MDR is mainly caused by C deposition and sintering of activated Ni grains. Ni grain size is a critical factor determining the performance of the catalyst, and smaller-sized Ni particles can effectively prevent C deposition and sintering, thereby facilitating the dry reformation of CH4. TEM and XRD analyses reveal that the size of Ni grains on the surface of LaFe0.5Ni0.5O3 remains small size after 80 h of testing. Figure 5 demonstrates that the H2/CO molar ratio LaFe0.5Ni0.5O3 of the produced syngas is close to 1, and the syngas is used as a raw material for the Fischer–Tropsch reaction. Therefore, we can conclude that the prepared mesoporous nanocast perovskite LaFe0.5Ni0.5O3 catalyst shows good resistance to carbon deposition and has promising prospect for future applications [45, 46].

3.5. Catalyst Characterization after Stability Test

Figure 6 reveals the XRD patterns of LaFe0.5Ni0.5O3 and LaFe0.3Ni0.7O3 after a long period of stability testing. The XRD pattern of LaFe0.5Ni0.5O3 (curve a) indicates a good perovskite structure, which means the prepared catalyst has good stability. By contrast, the perovskite structure of LaFe0.3Ni0.7O3 (curve b) is completely destroyed. This finding may be explained by the catalyst’s inability to maintain a large specific surface area, which clearly reduces its stability. In Table 3, LaFe0.5Ni0.5O3 has a larger specific surface area and smaller Ni particle size than LaFe0.3Ni0.7O3. These features indicate that the former can effectively limit the growth of Ni grains and thus has good sintering and carbon deposition resistance.

Figures 7(a) and 7(b) show the TEM images of LaFe0.5Ni0.5O3 in different stages. In Figure 7(a), small and relatively uniform Ni grains may be observed on the reduced catalyst [11, 47, 48]. This result demonstrates the good dispersion of Ni grains on the surface of LaFeO3. Figure 7(b) presents the catalyst after the 80 h MDR test; no significant carbon deposition could be found, which shows the good resistance of the catalyst to carbon deposition [1]. The Ni particle size is very small, consistent with the XRD results, especially after the stability test; thus, sintering of Ni grains does not occur during the MDR reaction. LaFe0.5Ni0.5O3 revealed good resistance to carbon deposition and sintering; thus, the catalyst possesses excellent stability.

The TG/DSC profiles of LaFe0.5Ni0.5O3 and LaFe0.3Ni0.7O3 are shown in Figures 8(a) and 8(b). During TG testing, at first, the small weight gains may due to the adsorption of oxygen on the oxygen vacancies of the PTO surface or from the oxidation of metallic Ni particles. Two distinct exothermic peaks could be observed in the DSC profile, which means two types of C are deposited in Ni active sites. The weight loss peak at low temperatures (250–350°C) can be attributed to the oxidation of active C (α-C), which is an active species in the MDR reaction [7, 27]. The mass loss of LaFe0.5Ni0.5O3 is 1.2%, while that of LaFe0.3Ni0.7O3 is very small. The mass loss at high temperatures (>550°C) can be ascribed to oxidation of inert C (γ-C), which is a major factor in catalyst deactivation. The mass loss of LaFe0.5Ni0.5O3 is approximately 13%, while that of LaFe0.3Ni0.7O3 is very large. LaFe0.5Ni0.5O3 remained highly active after the 80 h stability test, and the data indicated the remarkable inhibition of C deposition on the surface of the prepared catalysts.

4. Conclusion

In summary, a series of LaFe1−xNixO3 perovskite-type catalysts were prepared, among which LaFe0.5Ni0.5O3 exhibited the highest activity in MDR. The prepared catalysts had a large specific surface area, which could improve their catalytic activity. The catalysts showed excellent stability at 800°C during the MDR reaction with no significant deactivation over 80 h. The XRD, TEM, and TG-DSC data revealed no significant increase in the size of the Ni particles and no obvious carbon deposition on the catalyst after a long period of stability testing. These results demonstrate the promising application prospects of the LaFe0.5Ni0.5O3 catalyst.

Data Availability

The data used to support the findings of this study are available from the corresponding author upon request.

Conflicts of Interest

The authors declare that there are no conflicts of interest.

Authors’ Contributions

Z.W. and C.C. designed and administered the experiments. C.C. and Z.M. performed the experiments. All authors discussed the data and wrote the manuscript.

Acknowledgments

The authors are grateful to the inorganic functional material group of Shihezi University. The authors also acknowledge Shihezi University for the necessary support provided for the material characterization. This work was supported by the Natural Science Foundation of China (21566031 and 21766029) and the Scientific Research Program of Shihezi University (RCZX201411).