A Mini-Review on Lanthanum–Nickel-Based Perovskite-Derived Catalysts for Hydrogen Production via the Dry Reforming of Methane (DRM)

Abstract

Given that the attempts to head toward a hydrogen economy are gathering pace, the dry reforming of methane (DRM) to produce hydrogen-rich syngas is a reaction that is worthy of investigation. Nickel-based catalysts have been extensively examined as a cost-effective solution for DRM, though they suffer from fast deactivation caused by coke accumulation. However, a number of published studies report high catalytic performance in terms of both activity and stability for La–Ni-based perovskite-derived catalysts used in DRM in comparison to other corresponding materials. In the work presented herein, a thorough analysis regarding the application of La–Ni-based perovskite catalysts for DRM is carried out. LaNiO₃ is known for its anti-coking ability owing to the strong interaction between CO₂ and La₂O₃. A further modification to improve the catalytic performance can be achieved by the partial or complete substitution of A or/and B sites of the perovskite catalysts. The latest developments with respect to this topic are also discussed in this manuscript. Even though the low surface area of perovskite catalysts has always been an obstacle for their commercialization, new supported and porous perovskite materials have recently emerged to address, at least partly, the challenge. Finally, conclusions and future outlooks for developing novel perovskite catalysts that may potentially pioneer new technology are included.

1. Introduction

Energy holds a prominent role in modern societies, and the increase in world population and living standards will inevitably lead to further rises in the energy demand [1]. However, even though tremendous effort has been put toward developing renewable energy sources (RESs), fossil fuels are still responsible for nearly 80% of global energy consumption, with the remaining 20% also including nuclear and large-scale hydropower [2,3]. Because of the energy model followed for the past 200 years, the concentration of carbon dioxide (CO₂) has been steadily rising (it stood at around 419 ppm by June 2021 [4]), driving the continual increase in global surface and ocean temperatures [5,6]. Global warming, in turn, is responsible for climate change, which threatens human societies worldwide with catastrophic consequences [7,8].

A potential candidate to replace fossil fuels that is attracting increasing attention in the development of future energy systems is hydrogen [9,10,11,12]. Hydrogen has the major advantage of being a clean burning molecule, and its versatility means that it can facilitate the decarbonization of a range of sectors that have proved difficult to clean up in the past. This includes industries such as iron and steel, as well as transportation, especially long haul [13,14]. Hydrogen can also be employed to heat homes and store renewable electricity that would otherwise be wasted [15,16,17]. Another advantage is that hydrogen can also be produced from renewable feedstocks such as biomass [18,19,20].

Today, the annual worldwide hydrogen consumption is approximately 0.4–0.5 billion Nm³ [21]. Hydrogen production through the natural gas reforming process, which accounts for nearly 80–85% of the global hydrogen production, is advantageous because there is no need for investment in facilities or transport. In general, reforming processes include dry, steam, bi-, tri-, or partial oxidation, depending on the co-reactants (oxidizing agents). Among these methods, the dry reforming of methane (DRM) with carbon dioxide (Equation (1)) has the added advantage of consuming the main greenhouse gases (i.e., CH₄ and CO₂). The process produces syngas, which is a chemical building block for many petrochemical and refining processes. Syngas can be converted into valuable hydrocarbon products via Fischer–Tropsch (FT) synthesis or by CO separation into H₂ [22,23].

$${\mathrm{CH}}_4+{\mathrm{CO}}_2\to 2\mathrm{CO}+2{\mathrm{H}}_2 \Delta H^o{}_{298K}=247.3{\mathrm{kJmol}}^{-1}$$

(1)

Moreover, DRM offers a lower carbon footprint in comparison to other conventional methods for methane conversion to syngas (e.g., steam reforming) and, as result, can lower the net emissions of the CO₂ and CH₄ gases, especially if the energy used for it comes from non-hydrocarbon sources (e.g., solar energy) [24,25,26]. Despite its obvious attractions, DRM has yet to be fully commercialized owing to certain challenges and limitations that need to be addressed. For example, the reaction is carried out at very high temperatures because of its high endothermicity, which significantly increases the energy costs [27,28]. Moreover, even though numerous conventional metal/support catalytic systems have been shown to exhibit high catalytic activity, deactivation, owing to rapid carbon deposition on the catalyst, and the sintering of the metals that constitute the active phase remain obstacles yet to be surmounted [29,30].

Perovskites are one of the most common crystal structures found on the planet, making up 93% of the lower mantle’s mass, and 38% of the total mass of the planet. While it was originally discovered as a mineral, this crystal shape shows up everywhere and is a common subject for cutting-edge scientific research [31]. There are several different kinds of perovskites, namely CaTiO₃, BaTiO₃, CaSiO₃, and MgSiO₃, but they share a similar, dense crystalline structure. Nowadays, this class of materials has received considerable attention because such materials have been used in many areas, including heterogeneous catalysis [32,33,34].

In the last years, the use of perovskite-based catalysts for the dry reforming of methane has been growing rapidly, and a significant number of related scientific articles have been published. However, when searching through the available literature, one can find only a handful of review papers focusing on the use of perovskites in the DRM reaction [35,36]. This work focused exclusively on the use of lanthanum–nickel-based perovskite-derived catalysts for use in the DRM reaction, also presenting information with respect to kinetic and mechanistic aspects. It is hoped that, by highlighting the enhanced catalytic properties of these materials in the DRM, we will contribute to the discussion on the improvement of their structural characteristics and, resultantly, help them to achieve a higher catalytic performance.

2. Brief Summary of Ni-Based Catalysts for Methane Dry Reforming

A plethora of different metal-supported catalysts have been used for the DRM, with nickel-based systems showing comparative advantage over several noble-metal-based systems (e.g., Ru and Rh) due to the cost effectiveness of such formulations. As an example, Hou et al. [37] showed that Ni/Al₂O₃ exhibited increased initial activity compared to noble metal catalysts; however, its activity was compromised owing to carbon deposition. The same group also reported a beneficial synergetic effect between Ru and nickel that drastically reduced carbon deposition and enhanced catalytic activity.

As is well understood, different side reactions, such as the Reverse Water–Gas Shift (RWGS) (Equation (2)), CH₄ decomposition (Equation (3)), and CO disproportionation or Boudouard reaction (Equation (4)), synchronously take place during the DRM.

$${\mathrm{CO}}_2+{\mathrm{H}}_2\to \mathrm{CO}+{\mathrm{H}}_2\mathrm{O},\Delta H^o{}_{298K}=41{\mathrm{kJmol}}^{-1}$$

(2)

$${\mathrm{CH}}_4\to {\mathrm{C}}_{\mathrm{ads}}+2{\mathrm{H}}_2,\Delta H^o{}_{298K}=75{\mathrm{kJmol}}^{-1}$$

(3)

$$2\mathrm{CO}\to {\mathrm{C}}_{\mathrm{ads}}+{\mathrm{CO}}_2,\Delta H^o{}_{298K}=-172{\mathrm{kJmol}}^{-1}$$

(4)

A major issue for the DRM is the absence of oxygen, which does not allow for the gasification of the carbon formed by the Boudouard and methane-cracking reactions. Consequently, the catalysts used in the DRM often suffer from heavy carbon accumulation, which shortens their lifetime [29,30,38]. The Boudouard reaction is not favored thermodynamically at high temperatures (above 800 °C), as it is highly exothermic; hence, the carbon formed beyond this limit is mainly due to methane cracking [39,40,41]. However, as carbon species generated from methane cracking are more reactive in comparison to those produced by the Boudouard reaction, they can be oxidized by carbon dioxide; thus, coke accumulation beyond 800 °C is relatively insignificant [42,43,44]. Nevertheless, at a lower temperature range, the carbon species formed by the Boudouard reaction are not that active, meaning that carbon deposition is facilitated [45,46,47].

Along these lines, the development of a stable, highly carbon resistant nickel-based catalyst is a critical issue. Thus far, it is well known that the size of nickel metal particles plays a key role in carbon deposition. However, the nickel nanoparticles are prone to thermal sintering due to the low Tammann temperature of the metallic nickel [48,49]. Consequently, stabilization of the supported nickel particles to avoid their agglomeration is a matter of great importance. In addition, the surface structure of Ni particles also affects the carbon deposition. Specifically, Wang et al. [50] reported that CH₄ decomposition occurs faster on the surface of nickel (100) and nickel (110) in comparison to nickel (111) probably because the preferential catalytic pathway on different surfaces varies. Differences can also be found regarding the segregation behavior and diffusion of carbon atoms. For instance, it is easier for the diffusion of surficial carbon to take place in the bulk of nickel (110) compared to that of nickel (100) [51]. Thus, it is generally accepted that one of the most important factors for eliminating carbon deposition is associated with metal surface modifications. For example, it has been reported for phyllosilicate-derived materials that are used as catalysts for the DRM that a significant number of hydroxyl radicals on the support surface can promote H₂O activation, as well as gasify carbon species on the adjacent nickel surface [52,53,54].

Several techniques by different scientific groups have been developed to transfuse higher carbon resistance and reduce carbon deposition [55,56]. For this purpose, materials with different structures, such as phyllosilicates [30,52,54,57], spinel [58,59,60], and perovskites [61,62,63,64,65], have been widely used as catalyst precursors.

Perovskite-based catalysts for DRM have been thoroughly investigated in recent years and have been addressed in an increasing number of publications. The reason why they are so popular lies in fact that the exsolution taking place during the reduction of perovskite-based catalysts can provide a strong metal–support interaction and resistance to particle aggregation to alleviate their deactivation. In short, during exsolution, transition metal dopants are reduced to the elemental state, and metal nanoparticles emerge from the oxide structure, forming supported metal catalysts which are strongly anchored in the parental matrix. Those that are strongly socketed in the support nanoparticles are endowed with anti-coking properties, making them great candidates for long-term operations. In addition, numerous studies have shown that exsolution is cyclable, meaning that the metal nanoparticles are able to dissolve and return to being dopants in the oxide matrix and then regenerate on the material surface following a second reduction treatment [60,62,66,67,68,69]. The work presented herein attempts to make a constructive contribution to the existing literature by providing updated and clear information on the subject under consideration.

3. Perovskites for DRM

Generally speaking, a perovskite is a material that has a structural formula of ABO₃ (Figure 1), where A is a large cation (i.e., lanthanides, alkaline, or alkaline earth metals) that is 12-fold coordinated, B is a small cation (i.e., 6-fold coordinated) of the d-transition series, and O is an anion that bonds to both [70,71]. A₂BO₄ oxides (Figure 1) belong to the Ruddlesden–Popper (RP) phases (i.e., subclass of layered perovskite materials), and they are composed of alternating ABO₃ and AO layers; A₂BO₄ are also referred to as perovskite in many scientific works [71]. Owing to their unique structural characteristics, they can accommodate the majority of the metals at A and/or B sites without damaging the crystallinity of the structure [72,73]. In addition, perovskites feature high thermal stability, making them ideal choices for gas-phase reactions carried out at high temperatures, as well as increased oxygen mobility, which has a significant influence on eliminating coke formation during DRM [74,75,76].

DRM can be catalyzed by nickel; thereby, Ni-containing perovskite oxides can serve as precursors for DRM. Following hydrogen-reduction pretreatment, nickel cations are reduced to their active metal form (Ni⁰). However, the coexistence of additional metals at A or/and B sites can be challenging, as mixed oxides or/and bimetallic particles may be generated. In the following subsections, the perovskite-based catalysts are going to be grouped in accordance with their basic ABO₃ type, as well as their A- and B-site substitution. In addition, considering that an excellent review of the synthesis and characterization methods of perovskite has already been published, only a brief summary of the effect of the preparation method on catalytic performance for this type of materials is presented in this paper [71].

3.1. La–Ni Perovskite-Derived Catalysts for DRM

The most widespread nickel-based perovskite oxide used as a precursor for DRM is LaNiO₃. One of the earliest works reported by Verykios and co-workers [77] showed that a LaNiO₃ catalyst had a stable performance during time on stream, in contrast with conventional Ni-based catalysts, whose performance declined over time. Provendier et al. [78] also showed considerably high initial activity for a LaNiO₃ sample synthesized via a sol–gel method; however, carbon formation led to rapid deactivation. The authors also observed that the inclusion of small amounts of Fe to the perovskite structure significantly enhanced (to approximately 200 h) the stability of the sample without substantially compromising the activity. Over a decade later, Batiot-Dupeyrat et al. [79] reported that a LaNiO₃, which was synthesized via the auto-ignition method, was stable at 700 °C (90% conversion of reactants) during a time-on-stream test of 100 h (Figure 2). Another more recent study was carried out by Cao et al. [80], who showed that LaNiO₃ maintained a stable performance at different cycling stages.

In another noteworthy work carried out by Batiot-Dupeyrat et al. [81], the authors studied preactivated/reduced LaNiO₃ materials synthesized via the explosion method and found that the catalysts were stable during time-on-stream runs, showing low carbon deposition. The authors also argued that the formation of La₂NiO₄ (spinel structure) acted as an intermediate product participating in the DRM. In addition, the authors observed the formation of NiO and La₂NiO₄ following the oxidation of the spent LaNiO₃; however, the original perovskite structure was not reinstated. Using the self-combustion method, Gallego et al. [82] also synthesized a stable Ni/La₂O₃ catalyst, which was derived through the reduction of a LaNiO₃ precursor. This group also showed that La₂NiO₄ outperformed LaNiO₃ in terms of catalytic activity at 700 °C. In a similar work, Gallego et al. [83] prepared LaNiO₃ and La₂NiO₄ via the wet impregnation method and reported higher activity for the former perovskite in comparison to that of the latter perovskite, a fact that was assigned to the incomplete nickel reduction at lower temperatures (i.e., 500 °C). The authors denoted that the generation of slightly small nickel particles during hydrogen reduction increased the catalytic activity of the perovskites. Furthermore, the concurrent occurrence of RWGS resulted in lower CO₂ conversions compared to CH₄ conversions. On the other hand, the H₂/CO ratio was close to unity due to another simultaneous side reaction (i.e., CH₄ decomposition) that counterbalanced the hydrogen consumed from RWGS.

A noteworthy work was conducted by Pereñiguez et al. [84], who investigated the effect of different preparation methods, namely spray pyrolysis, spray pyrolysis–combustion, combustion, and hydrothermal synthesis, on the catalytic performance and physicochemical behavior of LaNiO₃ for DRM. It was shown that LaNiO₃ prepared through combustion performed poorly in comparison to that obtained through the other three methods, probably owing to the participation of increased levels of the NiO phase in the pristine calcined sample. Typically, larger Ni⁰ particle clusters obtained from the NiO phase during hydrogen pretreatment can weaken metal–support interactions and resultantly compromise catalytic performance. Furthermore, it was shown that the spray-pyrolysis–combustion method resulted in well-formed LaNiO₃ nanocrystals, which, in addition to producing catalysts with high porosity comprising small LaNiO₃ nanocrystals agglomerates, suppressed sintering phenomena during calcination. It is also worth noticing the considerable higher surface area values of the obtained perovskites in comparison to the ones reported in the literature. Another interesting study was conducted by Chawla et al. [85], who reported higher reactant conversions for LaNiO₃ perovskites prepared through the coprecipitation method when evaluated against sol–gel-prepared ones.

Typically, hydrogen temperature-programmed reduction (TPR) profiles of LaNiO₃ perovskites show two main reduction peaks for nickel at two different temperature ranges (i.e., 300–450 °C and 500–600 °C, respectively), as shown in Figure 3 [86].

One of the most widely accepted reduction pathways for LaNiO₃ was introduced by Kuras et al. [87], shown in Equations (5) and (6) below.

$$2{\mathrm{LaNiO}}_3+{\mathrm{H}}_2\to {\mathrm{La}}_2{\mathrm{Ni}}_2{\mathrm{O}}_5+{\mathrm{H}}_2\mathrm{O}\left(Low peak\right)$$

(5)

$${\mathrm{La}}_2{\mathrm{Ni}}_2{\mathrm{O}}_5+2{\mathrm{H}}_2\to 2{\mathrm{Ni}}^0+{\mathrm{La}}_2{\mathrm{O}}_3+2{\mathrm{H}}_2\mathrm{O}\left(High peak\right)$$

(6)

Valderrama et al. [88] corroborated the detection of a La₂Ni₂O₅ intermediate phase under a hydrogen atmosphere, using X-ray diffraction (XRD) analysis.

On the other hand, a three-step reduction pathway for LaNiO₃ was proposed by Batiot-Duperyrat et al. [81], as shown below.

$$4{\mathrm{LaNiO}}_3+2{\mathrm{H}}_2\to {\mathrm{La}}_4{\mathrm{Ni}}_3{\mathrm{O}}_{10}+{\mathrm{Ni}}^0+2{\mathrm{H}}_2\mathrm{O}$$

(7)

$${\mathrm{LaNiO}}_3+3{\mathrm{H}}_2\to {\mathrm{La}}_2{\mathrm{ONi}}_4+2{\mathrm{Ni}}^0+{\mathrm{La}}_2{\mathrm{O}}_3+3{\mathrm{H}}_2\mathrm{O}$$

(8)

$${\mathrm{La}}_2{\mathrm{ONi}}_4+{\mathrm{H}}_2\to {\mathrm{Ni}}^0+{\mathrm{La}}_2{\mathrm{O}}_3+{\mathrm{H}}_2\mathrm{O}$$

(9)

In a more recent study, Papargyriou et al. [89] proposed a more generalized reduction pathway which considers the phase change to various Ruddlesden–Popper phases (A_m+1B_mO_3m+1) under hydrogen atmosphere. The potential differences that occurred can be assigned to the calcination and preparation conditions, as well as the promoters used to synthesize the materials.

The evidently lower reducibility of nickel in LaNiO₃ compared to conventional supported catalysts suggests a more solid interaction in the perovskite framework between nickel atoms and lanthanum oxide, which, in turn, limits thermal agglomeration during hydrogen pretreatment, generating highly dispersed nickel particles (Ni⁰) on A-cation oxide.

The exceptional carbon-resistant properties of LaNiO₃ in DRM have been thoroughly discussed by numerous relevant studies, in which both preparation and pretreatment methods played a crucial role in the catalytic performance [69,88,90,91,92,93].

Another important determinant of catalytic behavior is the particle size of active metal. Chai et al. [94] prepared a nickel-doped La_0.46Sr_0.34Ti_0.9Ni_0.1O₃ catalyst with two different-sized nickel particles. The reduction of weakly interacting surface NiO at temperatures lower than 700 °C led to the formation of larger nickel particles (almost 14 nm), while the nickel incorporation into the perovskite structure at temperatures greater than 900 °C resulted in smaller nickel particles. DRM tests were carried out at 700 °C, using catalysts reduced at different temperatures. The authors noted increased catalytic activity and stability for the samples (active for 100 h) reduced at 950 °C, probably because of the presence of smaller nickel particles, which are known for their coke-resistant properties.

Pereñiguez et al. [95] managed to prepare LaNiO₃ samples, using the spray pyrolysis method, with nickel particles that were highly resistant to oxidation even after several cycles. A temperature-programmed oxidation (TPO) analysis of the reduced samples suggested similar oxidation profiles (OPs) for all the samples (Figure 4): (a) is the OP of pristine reduced LaNiO₃, (b) is the OP of pristine reduced LaNiO₃ after one TPR-TPO cycle, and (c) is the OP of pristine reduced LaNiO₃ after two TPR-TPO cycles.

The appearance of a change in the maximum peak at decreased temperatures (profile (a) to profile (c)) can be assigned to the higher dispersion due to a reduction in the nickel particle (Ni⁰) size. Furthermore, smaller metallic particles can be easier to oxidize in comparison to their larger counterparts.

The literature shows that La–Ni-containing perovskite oxides were the most popular catalysts for DRM reaction, whereas only a few other combinations with nickel in the structure were investigated recently. For example, Dama et al. [96] prepared MZr_0.8Ni_0.2O_3-δ (Μ = Ba, Sr, Ca) perovskites synthesized by a conventional sol–gel method for DRM at 800 °C. The authors noted that Ca-containing sample exhibited high activity and stability during a time-on-stream test of 500 h, suggesting potential for commercial exploitation. The catalyst’s stability was assigned to the redox property of the support which facilitated coke removal.

3.2. A₁A₂BO₃ Perovskite Catalysts

LaNiO₃ can be modified via the partial substitution of the A site with other metals to obtain the M_xL_1−xNiO₃ form. Research that focusses on finding the appropriate element and identifying the ideal substituted quantity is crucial to obtain a high catalytic performance.

Table 1. A₁A₂BO₃ perovskite catalysts for DRM.

Perovskite	Operating Conditions	CH4 Conversion Order	Carbon Deposition	Ref.
La0.95Ba0.05NiO3 La0.90Ba0.10NiO3 La0.85Ba0.15NiO3 La0.80Ba0.20NiO3 La0.75Ba0.25NiO3 La0.70Ba0.30NiO3	T = 650–750 °C WHSV = 60,000 mL g−1 h−1 CH4:CO2:He = 1:1:8	La0.90Ba0.10NiO3 > La0.85Ba0.15NiO3~La0.85Ba0.15NiO3 > La0.75Ba0.25NiO3 > La0.70Ba0.30NiO3 Ba-substituted catalysts outperformed the Sr and Mg ones.	-	[97]
LaNiO3 La0.95Ca0.05NiO3 La0.90Ca0.10NiO3 La0.70Ca0.30NiO3 La0.50Ca0.50NiO3 La0.20Ca0.80NiO3	T = 650–750 °C WHSV = 720,000 mL g−1 h−1 CH4:CO2 = 1:1	La0.50Ca0.50NiO3~La0.70Ca0.30NiO3 > La0.95Ca0.05NiO3 > La0.20Ca0.80NiO3 > La0.90Ca0.10NiO3 > LaNiO3	-	[98]
LaNiO3 La0.9Sr0.1NiO3 La0.8Sr0.2NiO3 La0.7Sr0.3NiO3 La0.6Sr0.4NiO3 Ni(5%)/La2O3	T = 700 °C CH4:CO2 = 1:1	LaNiO3 > La0.6Sr0.4NiO3 > Ni(5%)/La2O3 > La0.9Sr0.1NiO3 The order of activity depends on the content of strontium.	no coke no coke no coke no coke no coke no coke	[88]
La2Ni0.3Al0.7O3 La0.8Sr0.2Ni0.3Al0.7O2.9 La0.5Sr0.5Ni0.3Al0.7O2.75 La0.2Sr0.8 Ni0.3Al0.7O2.6	T = 750 °C WHSV = 15,000 mL g−1 h−1 CH4:CO2:N2 = 1:1:8	La0.2Sr0.8 Ni0.3Al0.7O2.6 > La0.5Sr0.5Ni0.3Al0.7O2.75 > La0.8Sr0.2Ni0.3Al0.7O2.9 Ca-substituted catalysts were more stable but less active.	8.71%—15 h 8.31%—15 h 3.21%—15 h -	[99]
LaNiO3 La0.98Pr0.02NiO3 La0.90Pr0.10NiO3 La0.60Pr0.40NiO3	T = 700 °C WHSV = 600,000 mL g−1 h−1 CH4:CO2 = 1:1	La0.50Pr0.10NiO3 > LaNiO3 > La0.98Pr0.02NiO3 > La0.60Pr0.40NiO3	63%—8 h 51%—8 h traces—8 h 52%—8 h	[100]
LaNiO3 La0.90Ce0.10NiO3 La0.70Ce0.30NiO3 La0.50Ce0.50NiO3	T = 600–800 °C GHSV = 10,000 h−1 CH4:CO2 = 1:1	LaNiO3 > La0.90Ce0.10NiO3 > La0.50Ce0.50NiO3 > La0.70Ce0.30NiO3	-	[101]

summarizes the related research.

Moradi et al. [97] compared the catalytic performance for the DRM of three different La_1−xM_xNiO₃-type (M = Ca, Sr, and Ba) perovskites. The results suggested that the Ba-containing sample outperformed the other tested catalysts in terms of catalytic activity. Next, the authors tried to evaluate the effect of Ba promotor (5% ≤ x ≤ 30%) on the catalytic performance and found that 20% of Ba offered the optimum promotion effect. De Lima et al. [98] investigated the substitution of calcium, drawing similar conclusions. The authors reported a promotion effect when a certain amount of Ca was added onto the substituted catalysts. On the other hand, beyond that calcium threshold, a high concentration of calcium can be accumulated on the surface, consequently compromising this promotion effect.

Lima et al. [102] investigated ceria-substituted perovskites in the DRM reaction. The authors reported that significant amounts of La₂NiO₄, CeO, and NiO were segregated due to the low solubility of ceria in the La_1−xCe_xNiO₃ matrix when x > 0.4. The optimum amount of ceria to produce the most active catalyst was found to be x = 0.05; meanwhile, increasing the value of x beyond that threshold led to lower activities. Moreover, the redox chemistry of the ceria oxide, which was produced after the reduction of the perovskite oxide, created oxygen vacancies that hindered the coke formation during DRM. Similarly, Gallego et al. [100] showed that the existence of highly dispersed and easily reducible nanosized nickel particles in ceria-substituted perovskites resulted in increased catalytic activity. The substitution of the bigger lanthanum by the smaller ceria affected the structural stability of the material, assisting in improved reducibility. In addition, the redox chemistry of ceria promoted the oxidation of carbon species. Another study that focused on the importance of oxygen vacancies resulting from the addition of ceria in a perovskite matrix was conducted by Wang et al. [103]. Interestingly, it was reported that the oxygen vacancies in the perovskite structure were created indirectly by the activation of B-site cations, since the valency state of ceria is the same as that of the A site. Firstly, the group employed a sol–gel self-combustion method to prepare a series of La_1−xCe_xNi_0.5Fe_0.5O₃ perovskites. Next, a TPR analysis was performed to evaluate the activity of the materials with respect to reducibility. Figure 5 shows the TPR profiles of the pristine sample with different x-values. Along these lines, peaks 1, 2, and 3 represent the reduction of nickel and iron species. It can be seen that these peaks shifted to lower temperatures as the x-values increased, suggesting that the addition of ceria at a site facilitated the reducibility process. On the other hand, peaks 4 and 5 correspond to reduced NiO and NiFe₂O₄ phases, as the perovskite structure was not formed when x > 0.6. Su et al. [101] also reported that lower ceria content resulted in higher catalytic activity. It was shown that La_0.9Ce_0.1NiO₃ was the optimized catalyst due to the increased number of oxygen atoms, as confirmed by the TPR results.

Lanthanides such as praseodymium (Pr) and samarium (Sm) can also be incorporated to substitute for lanthanum, either completely or partially, at the A site to produce a perovskite structure. These oxides have the ability to generate oxygen vacancies, enhancing the activity and durability of catalytic systems [104,105]. Gallego et al. [100] tested praseodymium (Pr)- and ceria (Ce)-substituted catalysts for DRM and showed that their addition facilitated the reduction process, as corroborated by the TPR results. It was also noted that the inclusion of either praseodymium or ceria could significantly reduce carbon deposition. La_0.9Pr_0.1NiO₃ had the best catalytic behavior, a fact that was assigned to the redox potential of Pr₂O₃ and the small (6 nm) nickel particles.

Sr-substituted perovskites were investigated even more thoroughly. Yang et al. [106] mentioned that the La₂O₂CO₃ surface was homogeneously covered by strontium species, resulting in a less strong metal–support interaction. Along these lines, the larger nickel particles obtained resulted in lower methane conversions. Nevertheless, the substitution of Sr improved the perovskite’s carbon resistance since CO₂ and CH₄ activation was promoted. These results were in line with Rynkowski’s relevant study [107]. Valderrama et al. [88] reported that the partial substitution of lanthanum by strontium in the perovskite matrix decreased the oxidation state of nickel, facilitating its reduction and producing spinel-structured solids owing to the increase in vacancies that promoted the oxygen mobility toward the surface of the catalyst. Sutthiumporn et al. [47] studied the effect of alkaline earth metal promoter over Ni/La₂O₃. X-ray photoelectron spectroscopy (XPS), temperature-programmed desorption (O-TPD), and H₂-TPR characterizations showed the existence of surface oxygen vacancies on substituted perovskites (M = Sr, Ca, and Mg), which consecutively promoted CO₂ adsorption/desorption in the samples. Firstly, the authors mentioned the importance of the La₂O₂CO₃ intermediate phase, which was found in La-based perovskites, in oxidizing surface carbon. Next, the group reported that the Sr-substituted catalyst had the highest CH₄ and CO₂ conversion, as well as the smallest amount of coke formation. The addition of strontium caused a lattice distortion and valence imbalance (Sr²⁺ and La³⁺), which led to an oxygen defect, thus increasing the lattice oxygen. The increased amount of lattice oxygen, in turn, can promote C–H activation, and it can react with carbon species, which improves the anti-coking properties of the perovskite.

Wang et al. [50] theoretically approached the DRM reaction mechanism by applying Density Functional Theory (DFT) calculations. The proposed reaction pathway, which is shown in Figure 6, suggested that the oxidation pathway of CH was more favorable in comparison to that of C.

The morphology dependence of the catalytic properties of Ni/CeO₂ for DRM was investigated by Du et al. [108], who concluded that nanorods displayed more excellent catalytic activity and enhanced anti-coking properties in comparison to nanocubes, owing to the oxygen vacancies and the mobility of lattice oxygen. According to previously published relevant works, the lattice oxygen can transfer to the adjacent nickel particles and then react with CH_x, generating carbon monoxide and hydrogen. Then, CO₂ dissociation, as well as oxygen mobility, can support the supplement of lattice oxygen. These last works underline the key part of lattice oxygen in C–H bond dissociation.

Khalesi et al. [99] conducted a comparative study with the aim of identifying the best alkaline earth metal substitution in La-Ni-Al perovskite oxide. The results showed that the strontium-substituted catalysts outperformed their calcium-substituted counterparts in terms of CH₄ conversion, selectivity, and yields, with Sr_0.8L_0.2Ni_0.3Al_0.7O_2.6 showing the higher value. On the other hand, calcium-substituted catalysts were proved to be more stable in comparison to strontium-substituted samples in the temperature range of 650–800 °C.

3.3. AB₁B₂O₃ Perovskite Catalysts

Partial substitution on B-site cations is more difficult in comparison to that of A-site cations. Initially, following catalyst reduction, the addition of reducible metallic elements at the B site would lead to the generation of metal alloy particles. Conversely, considering that the substituting element is inert, the reduction of the catalyst would result in the formation of nickel nanoparticles supported over mixed oxides.

Table 2. AB₁B₂O₃ perovskite catalysts for DRM.

Perovskite	Operating Conditions	CH4 Conversion Order	Carbon Deposition	Ref.
LaNiO3 LaNi0.8Zn0.2O3 LaNi0.6Zn0.4O3 LaNi0.4Zn0.6O3 LaNi0.2Zn0.8O3 LaZnO3	T = 750 °C WHSV = 180,000 mL g−1 h−1 CH4:CO2:He = 1:1:1	LaNi0.8Zn0.2O3 > LaNi0.6Zn0.4O3 > LaNiO3 > LaNi0.4Zn0.6O3 > LaNi0.2Zn0.8O3 > LaZnO3	0.7%—75 h 0.4%—75 h - - - -	[109]
LaNiO3 LaNi0.9Ru0.1O3 LaNi0.8Ru0.2O3 La3.5Ru4.0O3	T = 750 °C WHSV = 7200 mL mL g−1 h−1 CH4:CO2 = 1:1	LaNiO3 > LaNi0.9Ru0.1O3 > LaNi0.8Ru0.2O3~La3.5Ru4.0O3	65.7%—14 h 20.3%—14 h 6.7%—14 h 0.9%—14 h	[110]
LaNiO3 La2NiO4 La2Ni0.5Fe0.5O4 LaNi0.5Fe0.5O3	T = 750 °C WHSV = 120,000 mL g−1 h−1 CH4:CO2 = 1:1	LaNiO3~La2NiO4 > La2Ni0.5Fe0.5O4 > LaNi0.5Fe0.5O3	31.0%—4 h 18.0%—4 h 3%—4 h 9%—4 h	[111]
LaNiO3 LaNi0.8Cu0.2O3 LaNi0.6Cu0.4O3 LaNi0.4Cu0.6O3 LaNi0.2Cu0.8O3 LaCuO3	T = 750 °C WHSV = 180,000 mL g−1 h−1 CH4:CO2:He = 1:1:1	LaNiO3 > LaNi0.6Cu0.4O3 > LaNi0.4Cu0.6O3 > LaNi0.8Cu0.2O3 > LaNi0.2Cu0.8O3 > LaCuO3	- - - - - -	[112]
La2NiO4 La2Ni0.9Cu0.1O4 La2Ni0.8Cu0.2O4 La2Ni0.7Cu0.3O4 La2Ni0.6Cu0.4O4	T = 750 °C WHSV = 18,000 mL g−1 h−1 CH4:CO2 = 1:1	La2NiO4 > LaNi0.9Cu0.1O4 > LaNi0.8Cu0.2O4 > LaNi0.7Cu0.3O4 > LaNi0.6Cu0.4O4	0.4 gcg−1h−1—4 h 0.18 gcg−1h−1—5 h 0.01 gcg−1h−1—5 h 0.01 gcg−1h−1—5 h 0.01 gcg−1h−1—5 h	[113]
LaNiO3 LaNi0.8Mn0.2O3 LaNi0.6Mn0.4O3 LaNi0.4Mn0.6O3 LaNi0.2Mn0.8O3 LaMnO3	T = 750 °C GHSV = 15,000 mL g−1 h−1 CH4:CO2:N2 = 1:1:2	LaNi0.6Mn0.4O3 > LaNi0.4Mn0.6O3 > LaNi0.8Mn0.2O3 > LaMnO3 > LaNi0.2Mn0.8O3 > LaNiO3	- - - - - -	[114]

summarizes the related research.

The literature suggests that the addition of appropriate amounts of zirconium, zinc, and titanium has a promotional effect on the activity and stability (carbon resistance) [115,116]. For example, zinc can serve as a promoter, reinforcing the M-O-La bond, thus preventing the migration of active nickel nanoparticles [109,117]. Dama et al. [118] also used the Ruddlesden–Popper-type perovskite oxide Sr_n+1Ti_n−xNi_xO_3n+1 to investigate syngas production. The results indicated a direct relationship between the order of the Ruddlesden–Popper phase and activity, oxygen vacancies, and metal–support interaction, as the SrTi_1−xNi_xO_3−δ (n = ∞) sample had the highest catalytic activity.

The partial substitution of nickel by ruthenium on B sites to produce bimetallic catalysts has also been examined. Rivas et al. [119] reported that a low degree of ruthenium loading (LaNi_0.95Ru_0.05O₃) can modulate the surface of the catalyst and improve nickel reduction, as well as dispersion. The above results were in contrast to Araujo et al.’s work [110], which reported that Ru-O-La’s bond was reinforced by the addition of ruthenium on the B site. Even though the results with respect to catalytic activity were contradictory, they also confirmed the positive effect of the ruthenium addition on the stability of the LaNi_0.9Ru_0.1O₃ and LaNi_0.8Ru_0.2O₃ samples. Yasyerli et al. [120] tested nickel–ruthenium catalysts supported on MCM-41 and varied the amount of ruthenium (i.e., 0.5–3.0 wt%). It was found that the highest catalytic activity was achieved when x = 0.5 wt%. The positive effect of adding small amounts of ruthenium was also discussed by Zhou et al. [121]. The authors observed that increasing the ruthenium loading in Ru-Ni-Mg-O perovskites increased heterogeneity, as RuO₂ was easily segregated and remained undissolved in the NiO-MgO during calcination. That being said, the correlation between adding too much ruthenium and poor catalytic performance is indisputable. However, the contrastive results in terms of catalytic activity need to be further clarified. A possible explanation could be that an excess amount of ruthenium cannot be incorporated into B sites, as it would compromise the synergetic effect between the two metals. Moreover, optimizing the quantity of noble metals such as ruthenium or rhodium to ensure a high catalytic performance is needed owing to the cost limitations.

The addition of Fe (secondary metal to nickel) on B sites producing bimetallic catalyst has been shown to enhance both the catalytic activity and stability. For example, the beneficial effect on the catalytic performance of nickel–iron catalysts used in DRM was corroborated by Wang et al. [122,123], a fact that was ascribed to the generation of uniform nickel–iron alloy particles. Oxide precursors with different structures, such as perovskites, hydrotalcites, and spinels, have been employed to generate these alloy nickel–iron nanoparticles due to the increased oxygen affinity of iron, which endows the catalyst with anti-coking properties [124]. A noteworthy study was conducted by Jahangiri et al. [125], who revealed that LaNi_xFe_1−xO₃ can have different structures based on the value of x (i.e., x < 0.5 rhombohedral phase, x > 0.5 orthorhombic phase, x = 0.5 both phases). De Lima et al. [117] reported that low amounts of iron added as a secondary metal to nickel in LaNiO₃ perovskites improved the catalytic performance with respect to both the catalytic activity and stability. The H₂-TPR tests showed that an increase in iron content caused an increase to the second reduction peak at temperatures higher than 600 °C, corroborating iron’s stabilizing effect. The authors also noticed that both nickel and iron are reduced during the second reduction step and transformed into more stable nickel–iron alloys. Provendier et al. [78] also corroborated the alloy formation by carrying out several characterization techniques. The results of the transmission electron microscopy–energy-dispersive X-ray spectroscopy (TEM-EDS) measurements conducted at different areas of the sample (LaNi_0.3Fe_0.7O₃) demonstrated that the original structure of the fresh perovskite was destroyed under DRM conditions (time-on-stream tests at 800 °C). Figure 7 showcases the differences in lanthanum content between the initial area 1 and areas 2–9, where they were either lanthanum-rich or lanthanum-deficient.

In addition, the EDS analysis in the spent sample showed that the Ni/Fe ratio was almost constant (Figure 8), indicating the formation of a nickel–iron alloy. In addition, the Mossbauer spectroscopic analysis also confirmed the presence of a nickel–iron alloy. In particular, a shift in the hyperfine field was detected, which corresponds to the metallic iron peak in relation to the alpha-iron peak, and this shift varied according to the different iron loadings.

As previously discussed, Fe-substituted perovskites showed increased stability for DRM. Song et al. [111] carried out TPR measurements and showed that the addition of iron into B sites enhanced the stability of the structure and caused an upward change in the reduction temperature, but with a tradeoff on CH₄ conversion. This was consistent with Tomishige’s work [126], in which it was reported that nickel–iron alloys can reduce the number of the active nickel surface sites, which were responsible for the dissociation of C–C and C–H bonds, and resultantly compromise the activity. On the other hand, the addition of iron facilitated the adsorption of oxidative species, which were responsible for the suppression of carbon deposition. Theofanidis et al. [127] highlighted the importance of adding the appropriate amount of iron to the Ni-Fe/MgAlO₄ catalysts used in DRM. The authors reported improved activity for certain iron amounts added and decreased activity when too much iron was added. An in situ XRD analysis demonstrated that the DRM reaction pathway proceeded via the Mars–van Krevelen mechanism, as presented by the following equations.

$${\mathrm{CH}}_4\to {\mathrm{C}}_{\mathrm{Ni}}+2{\mathrm{H}}_2$$

(10)

$$\mathrm{Fe}+\mathrm{x}{\mathrm{CO}}_2\to {\mathrm{FeO}}_{\mathrm{x}}+\mathrm{xCO}$$

(11)

$${\mathrm{FeO}}_{\mathrm{x}}+{\mathrm{C}}_{\mathrm{Ni}}\to \mathrm{xCO}+\mathrm{Ni}+\mathrm{Fe}$$

(12)

$${\mathrm{FeO}}_{\mathrm{x}}+{\mathrm{H}}_2\to \mathrm{Fe}+{\mathrm{H}}_2\mathrm{O}$$

(13)

The positive influence of the redox cycle (Fe²⁺O/Fe⁰) on resisting carbon deposition was also discussed by Kim et al. [128]. Their results suggested the dynamic nature of the catalyst, which underwent dealloying via oxidation by carbon dioxide and re-alloying via its reduction by carbon species. The originality of this work lies with the fact that a significant phase separation in the catalysts under atmospheres with a high H₂/CO₂ ratio was avoided [128]. Tsoukalou et al. [129] tested the catalytic performance of a reduced LaNi_0.8Fe_0.2O₃ in DRM and reported the deallocation of the initially formed nickel–iron alloy particles and iron oxidation to FeO_x. The presence of FeO_x eventually led to the formation of LaFeO₃, which, in turn, encapsulated nickel nanoparticles and compromised the catalytic activity.

Nickel–cobalt bimetallic-supported catalysts (LaNi_xCo_1−xO₃) have been widely studied for reforming reactions and especially for the reforming of ethanol [130,131]. Valderrama et al. [132] investigated the catalytic behavior of this type of perovskite (LaNi_xCo_1−xO₃) in DRM. The results indicated increased activity for samples with x bigger than 0.2. It is noted that information regarding the anti-coking ability of the tested catalysts was not provided. Mousavi et al. [133] synthesized LaNi_0.5Co_0.5O₃ perovskites and examined the catalytic performance under the severe DRM conditions. Cobalt partial substitution brought improved carbon resistance; however, the catalytic activity was partly afflicted.

Copper can also be added as a secondary metal to nickel catalysts, generating bimetallic-supported catalysts. When searching through the available literature, one can see that copper-doped samples showed similar catalytic behavior to their cobalt-doped counterparts, meaning that carbon resistance was improved while the catalytic activity dropped [113]. For example, Touahra et al. [134] prepared LaCu_0.53Ni_0.47O₃ and LaCuO₃ perovskites via the sol–gel method and 5%NiO/LaCuO₃ by using impregnation. The copper-doped sample, as expected, showed increased resistance to coke formation but poor catalytic activity in comparison to LaCuO₃ and 5%NiO/LaCuO₃. Contrary to the consensus, Moradi et al. [112] synthesized LaNi_1−xCu_xO₃ (ternary perovskite oxides) by employing a sol–gel-related method in propionic acid and suggested higher activities and selectivities toward syngas products for the substituted samples. Increasing the amount of added copper generated CuO and La₂CuO₄ phases, in addition to the perovskite phase, which resulted in increasing the mean particle size of the particles involved in DRM.

Valderrama et al. [74] investigated the influence of manganese in LaNiO₃ perovskite and reported that low manganese loadings were able to enhance the catalytic performance of the catalyst in terms of both activity and stability. The addition of manganese led to high dispersion of the reduced nickel on La₂O₂CO₃-MnO_x, resulting in the hindrance of carbon deposition. Furthermore, the redox behavior of manganese on Mn-substituted perovskites was similar to that of iron’s, meaning that it can impart stability to the perovskite. Wei et al. [135] drew the same conclusions for nickel-substituted LaMnO₃ samples. Another interesting work was carried out by Kim et al. [136], who prepared a trimetallic lanthanum–nickel perovskite containing both manganese and cobalt at the B site. Interestingly, the catalyst showed exceptional catalytic behavior, as the addition of manganese induced high catalytic activity, while the addition of cobalt imparted high stability to coke formation.

A recent noteworthy work was carried out by Shahnazi et al. [114], who prepared manganese-substituted LaNiO₃ perovskite catalysts (LaNi_1−xMn_xO₃) via ultrasonic spray pyrolysis. The addition of manganese to the perovskite matrix increased the pore size and volume, as well as the surface area by 2.11 times. TPR analysis revealed a lower reduction tendency for LaNi_1−xMn_xO₃, indicating increased stability. O₂-TPD showed considerably higher oxygen mobility, which is attributed to the addition of manganese (Figure 9).

In addition, partial manganese substitution improved the catalytic behavior of the catalyst in terms of both activity and stability, with LaNi_0.6Mn_0.4O₃ achieving the better performance. It was also reported that the presence of manganese suppressed coke formation, and amorphous carbon was formed instead of whiskers. The XPS results indicated that the presence of amorphous carbon was assigned to the reversible transformation of Mn⁴⁺ and Mn³⁺, leading to increased oxygen mobility. Higher resistance to sintering for LaNi_1−xMn_xO₃ compared to bulk LaNiO₃ was also corroborated by the microstructural characterization of the used samples. The results showed that the catalytic mechanisms were significantly affected by the addition of manganese, where the formation of La₂O₂CO₃ intermediate aided LaNiO₃ in removing carbon. On the other hand, the stable LaNi_1−xMn_xO₃, which had a high oxygen capacity and mobility, removed coke through a cyclic redox mechanism.

3.4. A₁A₂B₁B₂O₃ Perovskite Catalysts

Numerous works have been carried out with a view to investigate the effect of the substitution of both A and B sites on catalytic activity and stability, with the majority of these studies trying to optimize the stoichiometry in order to achieve improved catalytic behavior. Valderrama et al. [137] focused on optimizing La_xSr_1−xNi_yCo_1−yO₃ perovskite. The authors reported improved activity and resistance toward coke for the sample under consideration. High carbon resistance was attributed to the SrCO₃ intermediary phase, which promoted the formation of La₂O₂CO₃ and SrO regeneration phases that, in turn, could serve as inhibitors against carbon deposition.

Sutthiumporn et al. [138] studied the DRM over La_0.8Sr_0.2Ni_0.8X_0.2O₃ (X = Fe, Cu, Cr, Co, and Bi) prepared via the sol–gel method and tried to shed light on the effect of lattice oxygen on both carbon resistance and C–H activation. The results showed higher methane conversion for the cobalt- and copper-doped samples at the initial stage. Nevertheless, the copper-doped (La_0.8Sr_0.2Ni_0.8Cu_0.2O₃) perovskite exhibited poor carbon resistance due to particle agglomeration. On the other hand, even though La_0.8Sr_0.2Ni_0.8Fe_0.2O₃ perovskite was not that active initially, it was considerably stable, a fact that was confirmed by a Thermogravimetric Analysis (TGA) analysis, with negligible amounts of carbon detected. The XPS results suggested higher oxygen lattice mobility over the copper-doped sample (La_0.8Sr_0.2Ni_0.8Cu_0.2O₃) in comparison to that of the iron-doped one (La_0.8Sr_0.2Ni_0.8Fe_0.2O₃). Moreover, the methane-TPR analysis showed that iron-doped (La_0.8Sr_0.2Ni_0.8Fe_0.2O₃) perovskite outperformed copper-doped (La_0.8Sr_0.2Ni_0.8Cu_0.2O₃) perovskite in terms of initial methane activation, presumably because of mobile lattice oxygen species, which activated the C–H bond. Furthermore, the addition of iron seemed to improve the metal–support interaction, which consequently enhanced carbon suppression.

In summary, A-site substitution is usually correlated with the oxygen vacancy formation, which is generally considered to be beneficial to enhance carbon resistance. On the other hand, B-site substitution with reducible metals often results in a synergetic effect, which can be either beneficial or detrimental. However, the mechanism of partial substitution of both the A and B sites and its subsequent effect on DRM have yet to be fully clarified, and further research would help address the issue.

3.5. Mesoporous/Supported Perovskite Catalysts

A significant limitation regarding the perovskite catalysts is the absence of micropores and surface area (≥10 m² g⁻¹), a problem that cannot be solved by partial substitution of the A or B site. A small surface area, along with weak mechanical strength, seems to prevent perovskite oxide catalysts from commercialization. Nowadays, different techniques that depend on morphology modulation have been adopted to address this issue.

The surface area could be increased by decreasing the particle size of the perovskite catalysts. A typical method to achieve this is by dispersing perovskite onto a support such as alumina, mesoporous silica, etc., which are known for their high surface area [139,140,141]. Therefore, this type of supported perovskite catalyst, which is usually prepared using the impregnation method (i.e., support is inserted into the mixture of precursor solution) to accurately control the perovskite loading, exhibits strong metal–support interaction and the ability to produce highly dispersed nanosized particles on the support [71].

Yadav et al. [141] prepared LaNi_xFe_1−xO₃ perovskites supported on different supports (i.e., SiO₂, Al₂O₃, and MgO) and tested their performance in DRM. Firstly, it was reported that SiO₂ support led to higher dispersion and improved conversions due to the increased number of active sites compared to Al₂O₃ and MgO support. Next, the authors observed that 40 wt% LaNi_0.75Fe_0.25O₃ supported on silica outperformed, in terms of catalytic activity and stability, the bulk perovskite-derived counterparts. In this regard, the XRD analysis showed that the peaks of the individual perovskite oxide phases were insignificant in the supported samples, indicating high dispersion. Nevertheless, the formation of silicate species was noticed at higher loadings (Figure 10). The FTIR analysis corroborated the existence of silicate species accommodated on the periphery of the samples, though without affecting the selectivity or conversions. This denotes the inability of silicate phase to encapsulate the nickel nanoparticles on the surface. Similarly, Sellam et al. [63] investigated LaNiO₃/SiO₂-supported catalysts and reported an improved catalytic performance owing to the higher surface area and nanosized nickel particles.

Zhang et al. [142] prepared LaNiO₃ perovskites supported on mesoporous silica and tried to capitalize the confinement effect derived from the use of this mesoporous material. It is known that the catalyst precursor can be confined with a layer of mesoporous barrier to produce more stable nickel particles and, thus, avoid sintering and carbon deposition. That being said, the authors used LaNiO₃ nanocubes embedded in SiO₂ for the DRM, as shown in Figure 11. The Ni/La₂O₃@SiO₂ obtained upon the hydrogen pretreatment was able to both support and confine nickel nanoparticles, resulting in enhanced anti-coking properties.

Wang et al. [143] synthesized LaNiO₃ perovskites supported on SiO₂, SBA-15, and MCM-41 carriers and investigated the influence of pore topology on their catalytic performance. As shown in Figure 12, isotherms of the supported samples are less steep and vertical in comparison to the those of their carriers. In addition, the slightly lower values of P/P₀ suggested the incorporation of perovskite particles into the mesopores of the carries. With respect to the narrow pore size distributions, SiO₂ showed higher values in comparison to those of two other carriers (i.e., SBA-15 and MCM-41). Moreover, LaNiO₃/MCM-41 showed the highest activity, followed by LaNiO₃/SBA-15 and LaNiO₃/SiO₂. However, time-on-stream tests at 700 °C for 60 h revealed that LaNiO₃/SBA-15 was more stable than LaNiO₃/MCM-41. From the TPR results (Figure 13), it can be seen that a higher reduction temperature peak was shifted to higher temperatures in LaNiO₃/MCM-41 and LaNiO₃/SBA-15 catalysts, suggesting strong metal–support interaction and, thus, improved stability. The authors also reported the co-occurrence of an RWGS reaction since the ratio of H₂/CO was less than unity.

Rivas et al. [144] prepared LaNiO₃, La_0.8Ca_0.2NiO₃, and La_0.8Ca_0.2Ni_0.6Co_0.4O₃ as catalyst precursors that were both bulk and supported on an SBA-15 carrier and tested them in DRM. The results showed better catalytic behavior for the supported catalysts. Moreover, the authors reported that, apart from the previously mentioned confinement effect, there was the dilution effect exerted by the mesoporous catalyst, which limited the heat diffusion issues associated with DRM’s high endothermicity.

A noteworthy work was conducted by Nair et al. [145], who synthesized mesoporous LaNiO₃ perovskite by using SBA-15 as a hard template for DRM. Following the removal of the SBA-15 template, ordered structures resembling nanowires were obtained. The Brunauer–Emmett–Teller (BET) analysis that was performed showed an increased surface area for the nanocast LaNiO₃ (150 m² g⁻¹) compared to that of the bulk LaNiO₃ (10 m² g⁻¹). The reducibility of the bulk and nanocast LaNiO₃ was monitored by carrying out H₂-TPR. It was reported that both samples had similar reduction profiles; however, no shoulder was apparent for the nanocast LaNiO₃, as result that can be attributed to the lack of grain boundaries in the nanocast samples (Figure 14). The peak position of the second reduction step indicated that the complete reduction of the nanocast LaNiO₃ occurred at a barely lower temperature compared to that used for the bulk LaNiO₃. In addition, the XRD pattern of the reduced perovskites (Figure 15) showed the total destruction of the perovskite structure at temperatures below 700 °C (black and blue curves) for both bulk and nanocast LaNiO₃, resulting in the formation of highly dispersed nickel in La₂O₃. The results from the catalytic tests revealed increased catalytic activity presumably due to the higher surface area and volume pore that provided more nickel active sites that were accessible to CO₂ and CH₄.

A similar technique to prepare mesoporous LaNiO₃/SBA-15 catalysts was adopted by Duan et al. [146] but for the reaction of the partial oxidation of methane.

In a recent study, Ruan et al. [147] prepared a LaAl_0.25Ni_0.75O₃ perovskite catalyst by using SBA-15 as a hard template for DRM. The results from different characterization techniques (i.e., H₂-TPR, CO₂-TPD, FT-IR, TEM-EDS, and XRD) revealed that the strong metal–support interaction, larger surface area, and increased number of strong basic sites of LaAl_0.25Ni_0.75O₃/SBA-15 were due to the unique structural characteristics of the catalyst, as well as the addition of proper amounts of silica. The authors also reported that the LaAl_0.25Ni_0.75O₃/SBA-15 catalyst outperformed, in terms of both activity and stability, the catalyst using commercial silica with a specific irregular channel structure and the bulk LaAl_0.25Ni_0.75O₃ catalyst. Following a 36-hour-long stability test, a conversion of more than 75% was maintained for the reactants (GSHX = 192,000 mL h⁻¹).

Rabelo-Neto et al. [10] used CeSiO₂ and Al₂O₃ basic oxides as carriers for LaNiO₃. The formation of Ce–O–Si bonds, which were produced via the introduction of Si to CeO₂, aided in the catalyst’s resistance to sintering and increase in surface area. The TPR results showed an increase in the reduction temperature for the supported LaNiO₃/CeSiO₂ catalyst compared to the bulk LaNiO₃, thus indicating a strong metal–support interaction. Along with the increased anti-coking ability brought in by the strong metal–support interaction, the redox behavior of ceria oxide (Ce⁺⁴/Ce³⁺) instigated high oxygen mobility, thus assisting in the oxidization of carbon residues. Massaoudi et al. [148] carried out a study on supported La_xNiO_y/MgAl₂O₄ and bulk LaNiO₃ perovskites in DRM. The H₂-TPR analysis showed that the supported samples were less reducible than the bulk ones, owing to the formation of NiAl₂O₄ during the calcination. Typically, the presence of NiAl₂O₄ improves the nickel dispersion over the support surface [148], which explains the enhanced catalytic performance of the supported La_xNiO_y/MgAl₂O₄ catalysts.

Yadav et al. [141] prepared and tested, in the DRM, perovskite catalysts (Ni 75wt% and Zr-Ce 25wt%, changing the ratio) supported on silica (SiO₂) and focused on the influence of alumina and magnesia on the support. The results showed that the introduction of a certain amount of ceria (x = 0.05) aided in increasing the surface area of the catalyst; however, the surface area was almost unchangeable after the SiO₂ was modified with magnesia and alumina. However, the modification of SiO₂ support with alumina and magnesia assisted in enhancing the basicity of the surface of SiO₂ and improving the reactant conversions. On the other hand, excess amounts of ceria and/or zirconia prevented the perovskite structure formation and led to the production of segregated phases of CeO₂, NiO, and La₂O₃ in the SiO₂-supported sample. The increased H₂/CO ratio that was observed (>1) can be attributed to the dominance of the methane-cracking reaction, which eventually led to coke formation in all catalysts. The authors postulated that the 40LaNi_0.75Ce_0.05Zr_0.20O₃/8MgO-SiO₂ sample exhibited the higher catalytic activity.

3.6. Three-Dimensionally Ordered (3DOM) Macroporous Perovskites

Furthermore, 3DOM perovskite materials have attracted global attention owing to their flexibility, tailorability, high stability, and unique macroporous structure that facilitates diffusion toward the active sites [149]. Typically, 3DOM perovskite-type oxides are prepared using colloidal crystal templates (e.g., polystyrene, poly (methyl-methacrylate), or silica), which are commercial products of high availability [150,151,152]. At the outset, the synthesis of macroporous perovskites is considered much more feasible compared to that of mesoporous perovskites since they are commercial products with low costs and are available in a variety of sizes, making them adaptable to demand. On the other hand, mesoporous silica has limited availability. Secondly, the synthesis of macroporous perovskites is more environmentally friendly compared to that of mesoporous silica, as harmful chemicals such as NaOH are required to dissolve the silica [146]. Thus far, this group of materials has been reported on works regarding the methane combustion; however, they have yet to be applied in DRM. Non-noble nickel-based catalysts with a 3DOM structure, such as La₂NiB′O₆, could be fabricated and tested for DRM in the near future.

4. Kinetic and Mechanistic Considerations

Several mechanisms have been proposed that rely on the type of the catalyst employed for this reaction [113,153]. Typically, CO₂ activation energy is lower than CH₄ activation energy due to the fact that methane has higher stability. A dual active-site mechanism has been proposed for LaNiO₃ perovskites used in the DRM. According to this mechanism, CH₄ is decomposed at site X₁, while CO₂ is activated at site X₂. La₂O₃, which is formed after the reduction of LaNiO₃ perovskite (LaNiO₃ → reduction → Ni/La₂O₃), can readily react with CO₂ to produce a stable phase of La₂O₂CO₃ (Equation (16)), as corroborated by FTIR and XRD analysis [82]. Generally, supports with higher basicity, such as La₂O₃, can promote the chemisorption of CO₂ to produce La₂O₂CO₃, which, in turn, has a positive effect on the activation of CH₄ on the active sites [154].

Tsipouriari et al. [155,156] confirmed the formation of La₂O₂CO₃, which was involved in CO formation, acting as an oxygen generator during the DRM reaction (Equation (17)).

This group proposed the following widely accepted kinetic mechanism for the DRM reaction over Ni/La₂O₃ catalysts, where RDS (Equations (15), (17), (24) and (26)) denotes the rate-determining step.

$${\mathrm{CH}}_4+\mathrm{Ni}\leftrightarrow \mathrm{Ni}-{\mathrm{CH}}_4,K_1\left(\mathrm{equilibrium}\right)$$

(14)

$$\mathrm{Ni}-{\mathrm{CH}}_4\to \mathrm{Ni}-\mathrm{C}+2{\mathrm{H}}_2,k_2\left(\mathrm{RDS}\right)$$

(15)

$${\mathrm{CO}}_2+{\mathrm{La}}_2{\mathrm{O}}_3\leftrightarrow {\mathrm{La}}_2{\mathrm{O}}_2{\mathrm{CO}}_3,K_3\left(\mathrm{equilibrium}\right)$$

(16)

$${\mathrm{La}}_2{\mathrm{O}}_2{\mathrm{CO}}_3+\mathrm{Ni}-\mathrm{C}\to {\mathrm{La}}_2{\mathrm{O}}_3+2\mathrm{CO}+\mathrm{Ni},k_4\left(\mathrm{RDS}\right)$$

(17)

$${\mathrm{H}}_2+2-\mathrm{Ni}\leftrightarrow 2\mathrm{H}-\mathrm{Ni},\left(\mathrm{equilibrium}\right)$$

(18)

The concurrent occurrence of RWGS includes the following reaction steps:

$${\mathrm{CO}}_2+\mathrm{Ni}\leftrightarrow \mathrm{Ni}-{\mathrm{CO}}_2,\left(\mathrm{equilibrium}\right)$$

(19)

$$\mathrm{Ni}-{\mathrm{CO}}_2+\mathrm{H}-\mathrm{Ni}\to \mathrm{Ni}-\mathrm{CO}+\mathrm{OH}-\mathrm{Ni},\mathrm{slow}$$

(20)

$$\mathrm{OH}-\mathrm{Ni}+\mathrm{H}-\mathrm{Ni}\leftrightarrow 2\mathrm{Ni}-{\mathrm{H}}_2\mathrm{O},\left(\mathrm{equilibrium}\right)$$

(21)

In this regard, the group proposed the following rate expression for the CH₄ consumption:

$$r_{C{H}_4}=\frac{K_1{k}_2{K}_3{k}_4{p}_{C{H}_4}{p}_{C{O}_2}}{K_1{K}_3{k}_4{p}_{C{H}_4}{p}_{C{O}_2}+K_1{k}_2{p}_{C{H}_4}+K_3{k}_4{p}_{C{O}_2}},\mathrm{Model}1$$

(22)

where K₁ represents the adsorption equilibrium constant of methane, k₂ is the rate constant of the surface reaction (methane cracking), K₃ corresponds to the adsorption equilibrium of the carbonatation reaction, and k₄ is the rate constant of the reaction between the carbon (carbon deposition) and oxycarbonate species.

Gallego et al. [82,83] proposed a similar equally widespread dual-site reaction mechanism for LaNiO₃ and LaNiO₄ perovskite catalysts that is shown below. Interestingly, the authors mentioned that the non-reduced perovskite oxide can improve the resistance to coke formation.

$${\mathrm{CH}}_4+{\mathrm{X}}_1\leftrightarrow {\mathrm{X}}_1-{\mathrm{CH}}_4,K_1 \left(\mathrm{equilibrium}\right)$$

(23)

$${\mathrm{X}}_1-{\mathrm{CH}}_4\to {\mathrm{X}}_1-\mathrm{C}+2{\mathrm{H}}_2,k_2\left(\mathrm{RDS}\right)$$

(24)

$${\mathrm{CO}}_2+{\mathrm{X}}_2\leftrightarrow {\mathrm{X}}_2-,K_3 \left(\mathrm{equilibrium}\right)$$

(25)

$${\mathrm{X}}_1-{\mathrm{CH}}_4+{\mathrm{X}}_2-{\mathrm{CO}}_2\to {\mathrm{X}}_2+2\mathrm{CO}+{\mathrm{X}}_1,k_4\left(\mathrm{RDS}\right)$$

(26)

where X₁ corresponds to the nickel active sites (Ni⁰) and X₂ represents the La₂O₃ supports as the active site. On this basis, the following rate expression for the CH₄ consumption was assumed:

$$r_{C{H}_4}=\frac{K_1{k}_2{K}_3{k}_4{p}_{C{H}_4}{p}_{C{O}_2}}{K_3{k}_4{p}_{C{O}_2}+K_1{K}_3{k}_4{p}_{C{H}_4}{p}_{C{O}_2}+K_1{k}_2{p}_{C{H}_4}+K_1{k}_2{K}_3{p}_{C{O}_2}},\mathrm{Model}2$$

(27)

Moradi et al. [157] carried out kinetic studies on the CO₂ reforming of CH₄ over La–Ni-based perovskite and found that the kinetic profile of LaNiO₃ resembled that of Ni/La₂O₃.

Table 3. Overview of kinetic rate parameters for DRM.

Kinetic Parameters	Temperature Range (°C)	Ref.
Ni/La2O3 K1 k2 = 2.61 × 10−3exp(−4300/T) [mol g−1 s−1] K3 = 5.17 × 10−5exp(8700/T) [kPa−1] k4 = 5.35 × 10−1exp(−7500/T) [mol g−1 s−1]	650–750	[155]
Ni/La2O3 derived from LaNiO3, at T = 700 °C K1 = 141 × 10−3 [kPa] k2 = 0.22326 × 10−3 [mol g−1 s−1] K3 = 15.98 × 10−3 [kPa] K4 = 13.22 × 10−3 [mol g−1 s−1]	500–700	[82]
LaNiO3 K1 = 279.55exp(−7502.5/T) [kPa−1] k2 = 12.27exp(−10,219.2/T) [mol g−1 s−1] K3 k4 = 0.034exp(−6968.2/T) [kPa−1 mol g−1 s−1]	650–750	[157]

summarizes and compares the calculated kinetic rate parameters found in the literature.

When searching through the available literature, it can be deduced that most of the reaction mechanisms proposed for DRM follow the Langmuir–Hinshelwood–Hougen–Watson (LHHW) formalism.

Batiot-Dupeyrat et al. [79] corroborated the presence of La₂O₂CO₃ hexagonal phase by XRD analysis on the spent samples and indicated that the concurrent realization of nickel reduction with oxycarbonate formation is crucial to limit the effect of RWGS and keep the ratio of H₂/CO close to unity. It was interesting that the catalytic activity improved when the reactants were added at ambient temperature compared to the case in which they were introduced at higher temperatures. This enhanced activity can be attributed to the production of oxycarbonate species, which promoted the CH₄ activation on nickel active sites. Slagtern et al. [154] also denoted the suppressive effect of lanthanum carbonate on coke accumulation, which eventually results in catalyst deactivation. In a recent study, Singh et al. [91] studied the relationship between the solid-phase crystallization and the shape of LaNiO₃. The results from 100-hour-long stability tests suggested decreased stability due to coke accumulation for the catalysts developed from LaNiO₃ cubes compared to the ones that were derived from rods and spheres whose catalytic stability was unaffected. The remarkable stability of the rod-derived LaNiO₃ catalyst can be attributed to both the smaller size and number of nickel nanoparticles in the structure. The results from a more recent study carried out by the same group suggested that solid-phase crystallization defects may be considered a possible cause for the high stability observed [158]. In general, many groups associated the anti-coking ability of the catalyst with the production of carbonate species during the DRM. Nevertheless, Dama et al. [96] proposed a mechanism in which hydroxyl species were involved in the mechanism. Considering these proposed mechanisms, Bhattar et al. [36], in their excellent review, presented the following sequence of reactions for CH₄ and CO₂:

CH₄ activation:

$${\mathrm{CH}}_4\left(\mathrm{g}\right)+\left(5-\mathrm{x}\right)\mathrm{X}\to {\mathrm{CH}}_{\mathrm{x}}-\mathrm{X}+\left(4-\mathrm{x}\right)\left[\mathrm{H}-\mathrm{X}\right]$$

(28)

$$2\left[\mathrm{H}-\mathrm{X}\right]\leftrightarrow {\mathrm{H}}_2+2\mathrm{X}$$

(29)

$$\mathrm{OH}-\mathrm{S}\leftrightarrow \mathrm{OH}-\mathrm{X}$$

(30)

$$\mathrm{OH}-\mathrm{X}+\mathrm{X}\leftrightarrow \mathrm{O}-\mathrm{X}+\mathrm{H}-\mathrm{X}$$

(31)

$$\mathrm{H}-\mathrm{X}+\mathrm{OH}-\mathrm{X}\leftrightarrow {\mathrm{H}}_2\mathrm{O}\left(\mathrm{g}\right)+2\mathrm{X}$$

(32)

$${\mathrm{CH}}_{\mathrm{x}}-\mathrm{X}+\mathrm{O}-\mathrm{X}\leftrightarrow \mathrm{CO}-\mathrm{X}+\mathrm{x}\left[\mathrm{H}-\mathrm{X}\right]$$

(33)

$$\mathrm{CO}-\mathrm{X}\leftrightarrow \mathrm{CO}\left(\mathrm{g}\right)+\mathrm{X}$$

(34)

$$2\mathrm{CO}-\mathrm{X}\leftrightarrow \mathrm{C}\left(\mathrm{s}\right)+{\mathrm{CO}}_2+2\mathrm{X}$$

(35)

CO₂ activation:

$${\mathrm{CO}}_2\left(\mathrm{g}\right)+\mathrm{S}\leftrightarrow {\mathrm{CO}}_2-\mathrm{S}$$

(36)

$${\mathrm{CO}}_2-\mathrm{S}+{\mathrm{O}}^{-2}\leftrightarrow {\mathrm{CO}}_3^{-2}-\mathrm{S}$$

(37)

$${\mathrm{CO}}_2-\mathrm{S}+{\mathrm{OH}}^{-}-\mathrm{S}\leftrightarrow {\mathrm{HCO}}_3^{-}-\mathrm{S}$$

(38)

$${\mathrm{CO}}_3^{-2}-\mathrm{S}+2\left(\mathrm{H}-\mathrm{S}\right)\leftrightarrow {\mathrm{HCO}}_2^{-}-\mathrm{S}+{\mathrm{OH}}^{-}-\mathrm{S}$$

(39)

$${\mathrm{HCO}}_3^{-}-\mathrm{S}+2\left(\mathrm{H}-\mathrm{S}\right)\leftrightarrow {\mathrm{HCO}}_2^{-}-\mathrm{S}+{\mathrm{H}}_2\mathrm{O}-\mathrm{S}$$

(40)

$${\mathrm{HCO}}_2^{-}-\mathrm{S}\leftrightarrow \mathrm{CO}-\mathrm{S}+\mathrm{OH}-\mathrm{S}$$

(41)

$$\mathrm{CO}-\mathrm{S}\leftrightarrow \mathrm{CO}\left(\mathrm{g}\right)+\mathrm{S}$$

(42)

$$2\left[\mathrm{H}-\mathrm{X}\right]\leftrightarrow 2\left[\mathrm{H}-\mathrm{S}\right]$$

(43)

Equations (28) and (35) contribute to the coke formation on the catalyst’s surface, while Equation (43) refers to the hydrogen atoms spilled over to the support, following methane decomposition [96]. Through this process, hydroxyl groups are produced in strong basic sites. In the previous equations, X represents the metal active site, while S corresponds to the active site of the support. However, specific criteria need to be met so that the above equations can be meaningful and valid, such as the following [157,159,160,161]:

(1)

Perovskite undergoes reduction (active metal should be reduced),

(2)

Perovskite is supported on basic oxide,

(3)

Rate of carbon dioxide dissociation is inconsiderable in comparison to that of methane,

(4)

Negligible surface coverage of hydrogen and carbon monoxide,

(5)

A part of active metal is carbon-free under DRM conditions.

The determination of mechanisms for LaNiO₃ perovskites used in the DRM has recently gained added impetus via in situ microscopy and spectroscopy studies that can provide evidence for activity–structure correlations. Thus, these techniques can contribute inter alia to a more comprehensive understanding of how oxygen defects influence catalytic behavior during DRM and illuminate their significance as active sites, as well as elucidate their influence on catalytic performance [162,163,164].

Nezhad et al. [162] carried out in situ synchrotron-based PXRD tests in both DRM mixtures, as well as under pure H₂. The results showed the formation of the perovskite-related counterpart of LaNiO₃ (i.e., La₂NiO₄) during the reaction, but this was not the case under pure H₂. The authors also reported the formation of oxygen-deficient structures and transient-related perovskite structures when subjected to heat in the DRM mixture.

Bonmassar et al. [61] employed quantitative in situ X-ray diffraction to uncover the dynamic structural changes taking place in LaNiO₃ perovskites during the DRM reaction. The results suggested structure–activity correlations, revealing monoclinic LaNiO_2.5 and transient oxygen-deficient triclinic LaNiO_2.7 phases, as well as transformations from the rhombohedral to cubic LaNiO₃ structure. The authors also discussed the association of these transitions with the increased DMR activity, which is likely attributed to surface-near nickel exsolution (not detectable by PXRD initially). The formation of the La₂NiO₄ phase was also reported [162] alongside the decomposition of monoclinic LaNiO_2.5, also affecting the catalytic activity. In addition, a substantial enhancement of DRM activity was observed once the lattice oxygen release from phase transformations toward Ni⁰/La₂O₂CO₃ was completed, probably owing to the full conversion of the Ni surface to its metallic state that was facilitated by monoclinic La₂O₂CO₃. Further transformations resulted in hexagonal La₂O₃, as well as hexagonal La₂O₂CO₃, formation for temperatures above 750 °C. This direct activity–structure correlation displayed that the active phase in DRM comprises a mixture of metallic Ni in contact with monoclinic La₂O₂CO₃ that acts as a stabilizer for the metallic Ni particles (similar behavior to the CO₂-activated spaces). Notably, the temperatures at which La₂O₂CO₃ formation and decomposition occurred were in line with those found in earlier studies of La₂O₃ with pure CO₂. A crystalline La₂O₃ phase at temperatures below 750 °C was not detected, whilst a sequence of reaction steps could not be ruled out [163].

5. Conclusions

The work presented herein provides information regarding the recent advances in the field of La–Ni-based perovskite-derived catalysts used in DRM. The main conclusions are summarized below:

-

LaNiO₃ gained considerable attention due to the high affinity between CO₂ and La₂O₃, which results in La₂O₂CO₃ formation, as well as the strong metal–support interaction provided. The presence of La₂O₂CO₃ plays a key role in the catalyst’s stability, as it can actively react with coke and act as an inhibitor against carbon accumulation on the nickel’s surface.

-

A-site partial substitutions with alkaline earth or rare earth metals into the perovskite matrix can both induce oxygen vacancies and modify the basicity of the surface of the catalyst, thereby improving coke resistance. The increased amount of lattice oxygen can also promote C–H activation, increase nickel dispersion, and increase the reducibility of the catalyst. Furthermore, redox chemistry of rare earth metals such as Sm, Pr, and Ce has a positive influence on the stability of the catalyst due to the fact that more oxygen for carbon removal is supplied.

-

B-site substitution can be reflected in two aspects. Non-reducible metals such as zirconium, and titanium can modify the structure and enhance the metal–support interaction, thus suppressing carbon deposition. On the other hand, reducible metals such as Fe, Co, Ru, and Cu can provide a synergetic effect which can be either beneficial or detrimental. To illustrate this point, the addition of iron generates a nickel-containing alloy which improves the stability of the perovskite. Moreover, the redox chemistry of the iron induces oxygen for coke removal due to the dynamic profile of the catalyst, which undergoes dealloying via oxidation by carbon dioxide and re-alloying via its reduction by carbon species. A smaller nickel particle size and decreased reducibility are related to the addition of iron or cobalt. In addition, cobalt can serve as a promoter owing to its high oxygen affinity, which aids in carbon gasification. Certain amounts of copper and manganese can also improve the catalytic behavior of the perovskite oxides in terms of both activity and stability.

-

The preparation of perovskites requires considerably high calcination temperatures to produce materials with considerably small surface areas, thus affecting their catalytic behavior. One method to overcome this problem is to disperse perovskite on certain supports, such as SBA-15 and MCM-41, leading to smaller perovskite particles. Another method is to synthesize porous perovskite catalysts by using templating methods.

However, there are many issues regarding this group of materials that still need to be addressed in order to consider them to be appropriate and effective for industrial DRM applications. Further research is necessary to come up with a better understanding of the thermal stability and coke formation of perovskites. For example, even though perovskites are known for their thermal stability, their structure can irreversibly decompose under extreme DRM conditions. However, the regeneration of the perovskite via multiple TPR-TPO cycles has been reported by a few studies, also resulting in increasing nickel dispersion. From an industrial perspective, the regeneration of the catalyst is a matter of great concern, as is the improvement of the catalytic activity at the industrially desired high WHSV/GHSV.

As was previously mentioned, the addition of reducible metal on B-sites can be either beneficial or detrimental. It has not been established yet whether a core–shell-like structure or uniform alloy is formed between the two metals that is clearly associated with their catalytic performance.

DFT calculations could be a powerful tool for designing perovskite catalysts, as it allows us to predict crucial properties, such as lattice oxygen mobility, which could aid in finding the best stoichiometry without educated guesses.

Finally, it would be interesting to test macroporous perovskites such as La₂NiB′O₆ in DRM, as, to the best of our knowledge, this has not yet been carried out.