Enhanced catalytic performance of Ni/Al₂O₃ for the carbon dioxide reforming reaction by controlling the pore structure and Ni addition in catalyst design

Abstract

Carbon dioxide reforming (CDR) of methane is a promising technology for the production of hydrogen and the removal of greenhouse gases. However, deactivation of catalysts via carbon deposition and sintering remains a limitation for the commercialization of CDR reactions. In this study, Ni/Al₂O₃ catalysts were prepared using different methods to investigate the effects of pore properties, as well as the interaction between Ni and supports, on the catalytic performance. Mesoporous and bimodal porous Ni/Al₂O₃ catalysts were prepared to confirm the relationship between the pore properties and catalytic performance. The addition of Ni was controlled to determine the effect of the interaction between Ni and Al₂O₃ on the catalytic performance. Various techniques were applied to understand the pore and interaction properties of Ni/Al₂O₃ catalysts prepared by different methods. The performance of Ni/Al₂O₃ catalysts prepared by different preparation methods and their pore properties were analyzed at a CH₄/CO₂ ratio of 1.0, gas hourly space velocity of 900 000 mL g⁻¹ h⁻¹, and reaction temperature of 700 °C. The highest catalytic performance was demonstrated by the bimodal porous Ni/Al₂O₃ catalyst, in which Ni was added after the calcination of the Al₂O₃ support. This result was owing to the high mass transfer due to the bimodal pore structure and proper interaction between Ni and Al₂O₃. This study will contribute to informing the importance of the pore structure and the addition of Ni for the development of catalysts that are both highly active and stable.

---

1. Introduction

Climate change threatens the safety of human life and maintenance of ecosystems on Earth. One of the main contributors to climate change is the use of fossil fuels, which emit greenhouse gases.^1,2 Therefore, international efforts, such as the Paris Agreement and Glasgow Climate Pact, have been made to cope with the global climate crisis.^3,4 In particular, the high dependence on fossil fuels is a significant barrier to overcoming the problems generated by climate change.^5,6 The establishment of a hydrogen society can be a solution to decrease the dependence on fossil fuel because hydrogen is a clean energy carrier with a high energy density per unit mass, emitting only water during combustion.⁷ Hydrogen can be produced by reforming hydrocarbons, water electrolysis, and nuclear power.⁸ The dominant method is hydrocarbon reforming, which can be classified into steam reforming (SR), carbon dioxide reforming (CDR), combined steam and carbon dioxide reforming (CSCDR), and autothermal reforming (ATR) depending on the type of oxidizing agent added.^9,10 Among the various reforming reactions, CDR has the advantage of producing hydrogen from carbon dioxide and methane, which are greenhouse gases.^11–13 In addition, it is easy to produce high-value fuels, which consist of long-chain hydrocarbons, from the syngas generated by the CDR reaction, owing to its low H₂/CO ratio.^14,15

A catalyst is an important factor for the energy-effective performance of CDR because it decreases energy consumption by reducing the activation energy. Precious-metal (Ru, Rh, and Pt)-based catalysts were initially used because of their high activity and stability.¹⁶ Recently, Ni-based catalysts have been actively studied by developing design technologies for cost-effective catalysts using nonprecious metals. In particular, the catalytic performance of Ni-based catalysts is affected by the metal loading. M. R. B. Pirshahid et al. reported the effect of nickel loading (5, 10, 15, and 20 wt%) of Ni-based catalysts in dry methane reforming.^17–20 It was reported that the catalyst with 15 wt% showed the highest activity and stability. This is because the appropriate level of Ni content increased exposure to the active site to increase accessibility to the reactants, and excessive Ni loading lowered the catalyst performance. W.-S. Dong et al. investigated the effect of Ni content on methane reforming in NiCe–ZrO₂ catalysts and reported that a 15 wt% Ni loading catalyst exhibited the best activity, selectivity, and stability.²¹ This was interpreted as due to an appropriate balance of active sites for methane and oxidants, and therefore, a 15 wt% Ni loading catalyst was used in this study. Nevertheless, Ni-based catalysts still suffer from deactivation due to carbon deposition and sintering at high temperatures. In addition, it is required to design strategies to inhibit the deactivation of Ni-based catalysts by carbon deposition on the catalyst surface and sintering Ni particles. Many studies have reported improvements in resistance to sintering and carbon deposition. In particular, the use of Al₂O₃ as a support to overcome the limitations of Ni-based catalysts in the CDR reaction has garnered significant attraction.^22–24

Al₂O₃ supports are widely used because of their excellent pore size dispersion, high specific surface area, high mechanical strength, strong chemical inertness, and low cost.²² Wang et al. studied the effect of active metal dispersion on the specific surface area characteristics and pore structure of an Al₂O₃ support and confirmed that a high specific surface area and large pore volume are helpful for the deposition of NiO particles inside mesopores, thereby resulting in stronger metal–support interactions.²³ Bychkov et al. compared the activity and carbon accumulation of Ni catalysts supported by α-Al₂O₃, γ-Al₂O₃, and θ-Al₂O₃ support.²⁴ The activity of the Ni/γ-Al₂O₃ catalyst was the highest, with low carbon accumulation. Zhong, et al. reported that when the NiAl₂O₄ spinel phase formed after firing at a high temperature, it was advantageous for maintaining high stability owing to the formation of small Ni particles and strong metal–support interactions.²²

The selection of a preparation method is important for preparing highly active and stable Ni/Al₂O₃ catalysts because the properties of the catalyst, which are directly related to catalytic performance, are strongly influenced by the preparation method.²⁵ Among the various properties, the pore properties, which are controlled by the preparation method, contribute to the improvement of the mass transfer ability, which is related to the contact between the reactant and active sites on the surface of the catalysts.^26,27 Recently, bimodal porous materials, including mesopores and macropores, have been considered promising supports for catalytic reactions because the bimodal porous structure compensates for the weaknesses of single-porous materials.^28–31 In addition, physicochemical properties were affected by preparation methods such as interaction between Ni and Al₂O₃, Ni dispersion, and Ni reducibility.

However, this is not sufficient to understand the impacts of the bimodal pore on the performance of catalysts in the CDR reaction. Therefore, it is necessary to investigate the preparation methods of Ni/Al₂O₃ catalysts that affect the pore and physicochemical properties to design highly active and stable catalysts. In this study, Ni/Al₂O₃ catalysts with bimodal porous and mesoporous structures were synthesized for understanding the effects of the bimodal structure on the performance of catalysts. In addition, Ni, an active metal, was loaded onto the supports using different methods to optimize the catalytic performance. Various techniques were used to investigate the properties of Ni/Al₂O₃ catalysts synthesized using different preparation methods. This study is expected to provide a guide for the selection of preparation methods and pore properties to design uniform porous catalysts for chemical reactions.

2. Experimental

2.1. Materials and methods

2.1.1. Materials. Aluminum isopropoxide (≥98%), styrene (≥99%), poly (vinyl pyrrolidone) (PVP, M_W ≈ 55 kDa, ≥ 99%), sodium 4-vinylbenzenesulfonate (NaSS, ≥ 90%), ethanol (≥99.5%), and nickel nitrate hexahydrate (≥97%) were obtained from Sigma-Aldrich. Azobisisobutyronitrile (AIBN, 98%) was purchased from Junsei Chemical.

2.1.2. Synthesis of polystyrene (PS) beads for the microporous template. Polystyrene (PS) beads with a diameter of 1.5 μm were synthesized via a dispersion polymerization method. A mixture of styrene (7.3 g), AIBN (0.14 g), PVP (0.795 g), and NaSS (0.074 g) was added to a three-necked round-bottomed flask containing ethanol (36 g) and deionized (DI) water (4 g). The solution was stirred under a nitrogen atmosphere for 30 minutes and then heated to 70 °C. After 24 h, the resulting micrometer-sized PS particles were collected and washed several times with DI water to remove residual components.

2.1.3. Synthesis of mesoporous alumina (MA) and bimodal porous alumina (BPA) as catalytic supports. Bimodally macro–mesoporous γ-alumina was synthesized using a solvent-deficient hydrolysis method, following a previously reported procedure.²⁶ Mesoporosity and macroporosity were controlled independently. Macropores were created using PS as a sacrificial template, while mesopores were formed by adjusting the hydrolysis reaction between aluminum alkoxides and DI water (H₂O : Al³⁺ ratio = 5 : 1). Aluminum isopropoxide (3 g) and the synthesized PS particles (1.5 μm, 1.5 g) were mixed with water using a mortar and pestle until a gel-like intermediate was formed. Subsequently, the material was calcined at 700 °C (ramping rate: 0.5 °C min⁻¹) for 10 h to remove the PS template and crystallize γ-alumina. For comparison, mesoporous γ-alumina was prepared by the same procedure but without the PS template.

2.1.4. Preparation of Ni/Al₂O₃ catalysts with a bimodal porous or mesoporous structure. The Ni/Al₂O₃ catalysts were prepared in two ways. In the first method which is Ni added during support preparation, nickel precursors were introduced during the synthesis of the alumina support, allowing added together and synthesized in the same manner as making the alumina support to the catalyst to form simultaneously with the support. In the second method which is Ni added after the calcination of the support, nickel nitrate hexahydrate was mixed with the alumina (Ni : Al₂O₃ = 15 : 85 wt%) in 100 mL of DI water and stirred at 500 rpm for 1 h. This solution was placed in the oven at 120 °C for 24 h to evaporate the DI water. Finally, to form NiO or NiAl₂O₄, the materials prepared from two routes were calcined at 700 °C (ramping rate: 10 °C min⁻¹) for 5 h. The difference in the catalyst preparation methods is summarized in Scheme 2.

The Ni/Al₂O₃ catalysts prepared by the first method are denoted as MNA with a mesoporous structure and BPNA with a bimodal porous structure. The Ni/Al₂O₃ catalysts prepared by the second method were denoted as Ni/MA with a mesoporous structure and Ni/BPA with a bimodal porous structure.

2.2. Characterization of catalysts

Various analytical techniques were applied to understand the characteristics of the prepared samples. The various techniques were Brunauer–Emmett–Teller (BET) measurements, field-emission scanning electron microscopy (FE-SEM), X-ray diffraction (XRD), H₂-temperature programmed reduction (H₂-TPR), H₂-chemisorption, X-ray photoelectron spectroscopy (XPS), field-emission transmission electron microscopy (FE-TEM), thermogravimetric analysis (TGA), and Raman spectroscopy. The reduction of catalysts was performed to obtain the information of metallic Ni as an active species prior to FE-SEM, FE-TEM, XRD and XPS analyses. N₂ adsorption/desorption was performed at −196 °C using an ASAP 2020 (Micromeritics, USA) instrument, and the BET surface area was calculated based on the amount of nitrogen adsorbed by the sample obtained. The microstructures and morphologies of the catalyst surfaces were investigated using FE-SEM (JSM-7900F, JEOL, Japan). The XRD patterns of the catalysts were obtained using an X'Pert PRO MPD diffractometer (PANalytical, Netherlands; Ni filtered Cu-Kα radiation, 40 kV, 30 mA). The patterns were obtained in a diffraction angle range (2θ) of 20° to 80°. The crystallite size was calculated using the Debye–Scherrer equation and XRD patterns. H₂-TPR was performed using a chemisorption AutoChem II 2920 (Micromeritics, USA) instrument. The catalyst was heated in a 10% H₂/Ar atmosphere from 25 °C to 1000 °C at a rate of 10 °C min⁻¹. The dispersion of Ni in the catalyst was measured using an AutoChem II 2920 instrument (Micromeritics, USA) via H₂-chemisorption. Prior to H₂-chemisorption, reduction was performed at 700 °C for 3 h in 10 vol% H₂/Ar. Thereafter, hydrogen was added at a catalyst bed temperature of 50 °C until the catalyst was saturated. The XPS analysis was conducted using an X-ray microprobe (ESCALAB 250XI, Thermo Fisher Scientific, USA). The results were plotted after calibration, based on the binding energy peak C 1s at 284.8 eV. An FE-TEM analysis was performed at an operating voltage of 200 kV using a JEM-F200 instrument (JEOL, Japan). All samples were sonicated, dispersed in ethanol, dropped onto a copper grid, and deposited with a carbon film. Subsequently, FE-TEM measurements were performed. The amount of carbon accumulated in the catalyst was measured by TGA (SDT 650, TA Instruments, USA). The catalyst was heat-treated in air from 30 °C to 800 °C at a heating rate of 10 °C min⁻¹. Raman spectroscopy was performed to determine the binding state of carbon deposited in the catalyst recovered after the reaction. The analysis was performed using NRS-3000 Series Raman spectrometers (JASCO, Japan) with a laser excitation wavelength of 532 nm at wavenumber 57 cm⁻¹.

2.3. Carbon dioxide reforming reaction

CDR of methane was conducted in a fixed-bed reactor with a diameter of 4 mm at 1 atm to evaluate its ability for hydrogen production. A quartz cylindrical reactor was fixed to the heat jacket, and 10 mg of catalyst was loaded over the quartz wool in the center of it. The reaction temperature was monitored using a K-type thermocouple (Omega Eng., Stamford, CT, USA) installed in the catalyst layer. The temperature of the catalyst layer was controlled using a thermal controller (Hanyoung Nux Co., Daegu, Republic of Korea). The catalyst before the catalytic activity test was reduced to 5% H₂/N₂ gas for 3 h at 700 °C. The gas hourly space velocity (GHSV) was fixed at 900 000 mL g⁻¹ h⁻¹. The feed gases were CH₄ and CO₂, supplied at a ratio of 1 : 1. The flow rate of the feed gas was 50 NmL min⁻¹ of CH₄, CO₂, and N₂. N₂ was used as a balanced (or internal standard) gas to calculate the conversion rate of CH₄ and CO₂ and yield of H₂. The outflow gases were measured using an online micro-gas chromatograph (Micro-GC 490, Agilent Technologies, CA, USA), which was equipped with a thermal conductivity detector (Micro-GC 490, Agilent Technologies, CA, USA). The CH₄ and CO₂ conversion rates, along with the H₂ and CO yields, were calculated as follows:Here, [A]_in and [A]_out represent the concentrations of gas A at the inlet and outlet, respectively.

2.4. Kinetic measurements

Kinetic experiments were conducted under various reaction temperatures and feed gas compositions to evaluate the reaction rate characteristics of the CDR reaction. This experiment was carried out in the same manner as the CDR reaction until the catalytic reduction step, and then the reaction was performed by varying the reaction temperature and the flow rate of N₂ and CH₄ or CO₂. It was carried out under reaction temperatures ranging from 650 to 800 °C, and the reaction was carried out by varying the flow rate of N₂ and CH₄ or CO₂ in the range of 10 to 50 mL min⁻¹ at a fixed flow rate of CH₄ or CO₂. As shown in Table S1, the reaction gas was composed of N₂, CH₄, and CO₂, and the total flow rate was fixed at 70 sccm. Accordingly, GHSV was designed to maintain 420 000 mL g⁻¹ h⁻¹. The generated gas was analyzed using gas chromatography. The reaction rate according to the reactants and products was calculated as follows:

The Mears criterion related to the external mass transfer was calculated by eqn (10).³²

where γ_A is the reaction rate calculated from the experimental [mol kg⁻¹ s⁻¹], ρ_b is the bulk density of the catalyst [kg m⁻³], R is the radius of the catalyst particle [m], K_g is the mass transfer coefficient [m s⁻¹], and C_A,b is the bulk concentration [mol m⁻³].

The Weisz–Perater number assigned to the internal mass transfer was calculated by eqn (11).³²

where γ_A is the reaction rate calculated from the experimental [mol kg⁻¹ s⁻¹], ρ_b is the bulk density of the catalyst [kg m⁻³], R_p is the radius of the catalyst particle [m], C_A,s is the bulk concentration [mol m⁻³], and D_eff is the effective diffusion coefficient [m² s⁻¹].

3. Results and discussion

3.1. Characterization of Ni/Al₂O₃ catalysts

3.1.1. Pore properties. Fig. 1 displays the isotherms of nitrogen adsorption/desorption, respectively, of the Ni/Al₂O₃ catalysts. For all prepared catalysts, a type-IV isotherm (IUPAC) was observed when the materials had a mesoporous structure, as shown in Fig. 1.^33–35 The pore properties, such as pore volume, pore size, micropore area, and surface area of the prepared catalysts, are listed in Table 1. The pore properties of the prepared catalysts were affected by the preparation method used. Regardless of the pore structure, the MNA and BPNA catalysts showed H3 type curves, while the Ni/MA and Ni/BPA catalysts exhibited H2 type curves. This result indicates that the pore structures of MNA and BPNA were slit- or wedge-shaped, whereas those of Ni/MA and Ni/BPA were ink-bottle-shaped.^36,37 In the H2 type with bottleneck pores, some of the pore inlets were blocked by nickel during the impregnation process.³⁸ In addition, the MNA and BPNA catalysts produced hysteresis loops at higher relative pressures (P/P₀). This indicates the presence of larger mesopores, as shown in Fig. S1.^39,40 This may have formed by integrating small mesopores and destroying some pore walls through reactions with nickel nitrate hexahydrate.²³

Fig. 1 N₂ adsorption/desorption isotherms of the Ni/Al₂O₃ catalysts: (a) MNA, (b) Ni/MA, (c) BPNA, and (d) Ni/BPA.

Table 1 Pore properties of the Ni/Al₂O₃ catalysts

Catalyst	Pore volume^a (cm^3 g^−1)	Pore size^a (nm)	Micropore area^a (m^2 g^−1)	BET surface area^a (m^2 g^−1)	Mears criterion^b	Weisz–Prater number^b
a Estimated from N2 adsorption at −196 °C. b Estimated from kinetic experiments.
MNA	0.67	7.3	29	202	316 × 10^5	43 116 × 10^5
Ni/MA	0.35	2.6	21	207	520 × 10^5	69 899 × 10^5
BPNA	0.69	9.4	24	186	379 × 10^5	50 531 × 10^5
Ni/BPA	0.29	2.6	19	190	569 × 10^5	84 702 × 10^5

---

Fig. S1 shows the pore diameter distribution curves of Ni/Al₂O₃ catalysts prepared by different preparation methods. The Ni/MA and Ni/BPA catalysts exhibited more uniform pore structures than the MNA and BPNA catalysts. The characteristics of the MNA and BPNA catalysts were consistent with those of the H3 type, which had primary irregular pores with large diameters.²³ This means that the Ni species added when manufacturing the support was inserted into the Al₂O₃ bulk and thus interfered with uniform pore generation.^23,41 Therefore, it was confirmed that the pore structure was deformed according to the preparation method. Because uniform pores suppress carbon deposition and improve mass transfer, the Ni/MA and Ni/BPA catalysts are expected to exhibit higher activities.^42,43

Table 1 lists the mesoporous properties of the Ni/Al₂O₃ catalysts prepared by different methods, as obtained from the N₂ adsorption at −196 °C. The surface areas of the Ni/Al₂O₃ catalysts were similar regardless of the preparation method. The mesoporous properties of the Ni/BPA catalyst were similar to those of the Ni/MA catalyst because the Ni/BPA catalyst included mesopores and macropores. Interestingly, the mesoporous properties were significantly affected by the addition of Ni. The pore size values presented in Table 1 correspond to the x-axis (pore diameter) of Fig. S1. The MNA and BPNA catalysts exhibited a broad pore diameter distribution in the range of 5–20 nm, whereas the Ni/MA and Ni/BPA catalysts showed a narrow and uniform pore diameter of approximately 2.6 nm. This observation is consistent with the pore diameter distribution results shown in the corrected Fig. S1. In addition, it should be noted that the N₂ adsorption desorption analysis employed in this study only characterizes pores which have a diameter smaller than 50 nm. The reduced pore size and narrowed distribution in the Ni/MA and Ni/BPA samples are attributed to the destruction of mesopores during the Ni loading process on porous Al₂O₃. According to previous research, the higher the BET surface area, the faster the reaction rate increases by promoting mass transfer and the adsorption of reactants.²⁶ Therefore, Ni/MA and Ni/BPA catalysts are expected to exhibit high catalytic activities.

The Mears criterion values are summarized in Table 1 when the average particle size diameter of catalysts was 300 μm. The external mass transfer effects can be minimized if the Mears criterion is less than 0.15.⁴⁴ In addition, the Weisz–Prater numbers are also summarized in Table 1.

Although the presence of macropores in the bimodal catalysts (Ni/BPA and BPNA) was expected to facilitate mass transfer, the Weisz–Prater analysis showed that Ni/BPA exhibited the largest N_WP. This is primarily because the intrinsic rate (γ_A) of Ni/BPA was substantially higher than that of the other catalysts, while its overall pore volume (0.29 cm³ g⁻¹) was the smallest among the series, leading to a lower effective diffusivity (D_eff). As a result, the ratio N_WP ∝ γ_A/D_eff increased, even though macropores were present. In contrast, MNA, which exhibited a lower intrinsic activity and larger pore volume, showed the smallest N_WP. It should also be noted that our N₂ sorption only quantifies pores <50 nm, so the contribution of large macropores (>50 nm) is underestimated. Therefore, the bimodal structure still enhances transport qualitatively, but under the current kinetic regime the very high activity of Ni/BPA accentuates the intraparticle concentration gradients in the Weisz–Prater criterion. In addition, the Weisz–Prater criterion effect is neglected when it is lower than 1.⁴⁴

Fig. 2 shows the FE-SEM images of reduced Ni/Al₂O₃ catalysts at 700 °C prepared using different methods. Mesopores were detected in all the prepared catalysts, whereas macropores were detected in the catalysts with added PS. The presence of mesopores and macropores negates the disadvantages of each type of pore structure. Mesoporous structures can suppress metal agglomeration because uniformly sized pores are regularly arranged. This phenomenon leads to the formation of well-dispersed Ni in the support, which improves its thermal stability.^45,46 However, this mesoporous structure can be easily blocked by the deposited carbon produced during the reforming reaction. Macropores show strong resistance to carbon deposition because of their large pore diameter, enhancing the movement of reactants and products and causing low Ni dispersion, which determines the number of active sites.^47,48 Therefore, we investigated the preparation of Al₂O₃ supports with bimodal porous structures composed of mesopores and macropores.^26,47,49,50 The MNA and BPNA catalysts had irregular pore structures, consistent with the characteristics of the H3 type, similar to the nitrogen adsorption/desorption isotherms.

Fig. 2 FE-SEM images of the Ni/Al₂O₃ catalysts: (a) MNA, (b) Ni/MA, (c) BPNA, and (d) Ni/BPA.

3.1.2. Properties of Ni active species. The XRD patterns of Ni/Al₂O₃ catalysts which were reduced at 700 °C and 900 °C are shown in Fig. 3(a and b), respectively, and the crystallinity of the prepared catalysts was investigated. Fig. 3(b) was included because the peak size of the metallic Ni phase in (a) was too small, making it difficult to calculate the grain size. In all the prepared catalysts shown in Fig. 3(a and b), it was observed that diffraction peaks of 2θ = 44.5°, 51.7°, and 76.4° were related to the metallic Ni nanoparticles as reported in the literature.^22,51,52 Diffraction peaks related to NiO were not detected, as confirmed by the H₂-TPR patterns. In all the prepared catalysts, a diffraction peak at 45.9° appeared, which was ascribed to the spinel phase of NiAl₂O₄.⁵³

Fig. 3 XRD patterns of reduced Ni/Al₂O₃ catalysts at (a) 700 °C and (b) 900 °C.

Table 2 shows the characteristics of Ni/Al₂O₃ catalysts prepared using different methods. It was difficult to calculate the crystallite size of Ni using the Debye–Scherrer equation because the diffraction peak of Ni was small and broad. Therefore, for the comparison of the crystallite sizes of the Ni/Al₂O₃ catalysts prepared via different methods, XRD analyses were conducted for the Ni/Al₂O₃ catalysts reduced at 900 °C. The diffraction peak of Ni/Al₂O₃ catalysts reduced at 900 °C was sharp compared to that of the Ni/Al₂O₃ catalysts reduced at 700 °C because the higher calcination temperature led to an increase in the crystallite size.²² The Ni crystallite size for reduced Ni/Al₂O₃ catalysts at 900 °C was as follows: Ni/BPA (3.1 nm) < Ni/MA (3.8 nm) < BPNA (5.0 nm) < MNA (5.2 nm). This result indicates that the Ni/MA and Ni/BPA catalysts have smaller Ni crystal sizes unlike the prepared MNA and BPNA catalysts. Generally, Ni dispersion is inversely proportional to the Ni crystallite size. The trend in Ni dispersion in the Ni/Al₂O₃ catalysts prepared using different methods, as confirmed by H₂-chemisorption, was as follows: MNA (0.42%) < Ni/BPA (0.63%) < Ni/MA (0.70%) < BPNA (1.08%). When considering the reduction degree (RD), the trend of Ni dispersion (RD) was observed to be: MNA (1.43%) < BPNA (1.90%) < Ni/BPA (1.94%) < Ni/MA (2.68%). The BPNA catalyst showed the highest Ni dispersion, but the Ni dispersion considering the reduction degree (RD) was higher in the Ni/MA and Ni/BPA catalysts. It is important to consider the reduction degree of catalysts because metallic Ni is an active site. Especially, NiO in Ni/Al₂O₃ which is not fully reduced when it is exposed in reaction condition as confirmed by TPR analysis. In addition, the amount of metal Ni formed during the reaction process varies depending on the reduction characteristics of the catalyst. In particular, in catalysts where Ni is introduced sequentially, the strong interaction between Ni and Al₂O₃ makes reduction relatively difficult, which may lead to a lower H₂-chemisorption based Ni dispersion as confirmed by XPS and TPR. Therefore, when analyzing the dispersion, the reduction characteristics must be considered together. This is consistent with the crystal size trend measured by XRD. Dispersion is a key property determining the catalytic activity of Ni-based catalysts for hydrocarbon reforming.⁵⁴ In addition, highly dispersed Ni enhances carbon deposition resistance and catalytic activity.^38,47,55 Therefore, Ni/MA and Ni/BPA catalysts are expected to exhibit excellent catalytic performance.

Table 2 Ni crystallite size and Ni dispersion of Ni/Al₂O₃ catalysts

Catalyst	Ni crystallite size^a (nm)	Ni dispersion^b (%)	Ni dispersion (RD)^c (%)
a Calculated from the XRD diffraction peak at 2θ = 52.0° of reduced Ni/Al2O3 catalysts at 900 °C. b Calculated from H2-chemisorption. c Calculated from H2-chemisorption considering the reduction degree.
MNA	5.2	0.42	1.43
Ni/MA	3.8	0.70	2.68
BPNA	5.0	1.08	1.90
Ni/BPA	3.1	0.63	1.94

---

Fig. 4 shows the H₂-TPR patterns of the prepared catalysts, which are related to the redox properties, including the reducibility and interaction between Ni and Al₂O₃. This interaction originates from the existing Ni–O–Al bridges.^56,57 Four reduction peaks were observed for the Ni/MA and Ni/BPA catalysts, whereas three peaks were detected for the MNA and BPNA catalysts. According to the literature, the first reduction peak is observed below 500 °C due to the reduction of free NiO, which does not interact with Al₂O₃.^58,59 The second reduction peak appearing around 550 °C is due to the reduction of NiO, which weakly interacts with Al₂O₃. The third reduction peak appearing around 700 °C was assigned to the reduction of NiO, which strongly interacts with Al₂O₃. The last reduction peak appearing around 800 °C was related to the reduction of spinel NiAl₂O₄.^58,59 In addition, a reduction peak related to NiO, which strongly interacted with Al₂O₃ of Ni/MA and Ni/BPA catalysts, was detected at higher temperatures than those for the MNA and BPNA catalysts. This indicates that the addition of Ni is important for determining the interactions between Ni and Al₂O₃. The intimate interaction between Ni and Al₂O₃ improves stability by enhancing the resistance of the Ni particles to carbon deposition and sintering.^60,61 Therefore, the Ni/MA and Ni/BPA catalysts are expected to be more stable and durable.

Fig. 4 H₂ temperature programmed reduction of the Ni/Al₂O₃ catalysts.

Fig. S2 shows the Ni 2p_3/2 spectra of the prepared catalysts, which were collected via XPS to explain the chemical properties of the catalyst surfaces. The Ni/Al₂O₃ catalyst was reduced at 700 °C. Three peaks related to various Ni species and one peak assigned to the satellite peak were observed for all the prepared catalysts. These three peaks can be divided into those of metallic Ni (851.9–852.6 eV), NiO (853–855 eV), and NiAl₂O₄ (855.6–856.2 eV).^52,62–66 In particular, all Ni 2p_3/2 peaks were detected at higher binding energies when the Ni/MA and Ni/BPA catalysts were compared with MNA and BPNA. This result indicates that a strong interaction between the active metal and support occurs in the Ni/MA and Ni/BPA catalysts, regardless of the pore structure.^46,47,67 This metal–support interaction is known to be an essential factor for enhancing catalytic stability and activity.^68,69 In addition, the strong interactions between Ni and Al₂O₃ in the Ni/MA and Ni/BPA catalysts matched well with the H₂-TPR analysis. Table 3 summarizes the amount of each Ni species on the catalyst surface, which was calculated using the peak area of the Ni 2p_3/2 spectra. The amount of Ni and NiO species was low when Ni was added to the porous Al₂O₃ supports, regardless of the pore structure, because the Ni species were inserted into the Al₂O₃ bulk during the preparation procedure.⁶⁰ In particular, the proportion of Ni⁰ species was higher in Ni/MA and Ni/BPA catalysts than in other catalysts. Chen et al. reported that when the proportion of Ni²⁺ species is excessively high, an excessively strong metal–support interaction (MSI) can encapsulate metal nanoparticles on the support, reducing the exposure of active metal sites. An appropriate level of MSI can promote the generation of Ni⁰ species during the reduction process, whereas an excessively strong MSI can suppress the reduction of NiO, thereby reducing the amount of metallic Ni.⁷⁰ Therefore, the high content of Ni⁰ species and high binding energy observed in the Ni/MA and Ni/BPA catalysts indicate that an appropriate level of MSI was formed that balanced structural stability and reducibility, which is considered to have contributed to the improvement of catalyst performance.

Table 3 Absolute and relative Ni 2p peak area of the Ni/Al₂O₃ catalysts

Catalyst	Relative peak area^a (%)
	Ni	NiO	NiAl2O4
a Calculated from XPS Ni 2p3/2 spectra.
MNA	17.1	21.6	61.3
Ni/MA	25.9	32.9	41.3
BPNA	18.1	35.0	46.8
Ni/BPA	19.2	38.6	42.2

---

Fig. 5 shows the morphologies of Ni/Al₂O₃ catalysts of the Ni/Al₂O₃ catalyst reduced at 700 °C at 200 kV. The EDS elemental mapping results of the Ni/Al₂O₃ catalysts show the distribution of elements in the catalyst. The analysis revealed that each element was uniformly distributed in all catalysts. However, the Ni distributions of the MNA and BPNA catalysts were weaker than those of the Ni/MA and Ni/BPA catalysts. This was due to the insertion of the Ni species into the Al₂O₃ bulk during catalyst preparation, as confirmed by the XPS analysis. Among the prepared Ni/Al₂O₃ catalysts, only rod-like crystallites were observed in the Ni/MA catalyst. The rod-like crystallites of Al₂O₃ originated from the contact between Al³⁺ and H₂O. For MNA and BPNA, the Ni precursor disrupted the contact between Al³⁺ and H₂O, whereas for Ni/BPA, this contact was inhibited by the PS template. It appears that the addition of the PS template influences the hydrolysis and condensation reactions between aluminum alkoxide and water, leading to the formation of mesopores.²⁶ Therefore, the morphology of the porous Ni/Al₂O₃ catalyst depends on the amount of H₂O added.

Fig. 5 FE-TEM images and energy-dispersive X-ray spectroscopy (EDS) elemental mapping results of the Ni/Al₂O₃ catalysts: (a) MNA, (b) Ni/MA, (c) BPNA, and (d) Ni/BPA.

3.2. Catalytic performance

Ni/Al₂O₃ catalysts have been chosen as the general commercial catalysts for the steam reforming (SR) reaction due to their high activity and cost efficiency. In addition, most Ni/Al₂O₃ catalysts exhibited a high activity in the CDR reaction as shown in Table S2. However, it is performed under severe conditions for the catalysts compared to the SR reaction. A higher reaction temperature is needed due to the dissociation of stable CO₂ leads to sintering of the catalyst. In addition, the condition without steam promotes the carbon formation phenomenon. Therefore, it is important to control pore properties which can increase the structural stability and can improve mass transfer to enhance the resistance to sintering and carbon formation. Accordingly, this study is of great significance in that it experimentally verified the effects of high GHSV conditions and double pore structures and preparation methods on CH₄ and CO₂ conversion, which were not covered in previous studies.

To investigate the catalytic performance, the CDR reaction was carried out in a time-on-stream manner, as shown in Fig. 6. The results show higher CO₂ conversion compared to CH₄ conversion resulting from the reverse water gas shift reaction (CO₂ + H₂ ⇌ CO + H₂O), which is a side reaction to produce CO and H₂O by consuming CO₂ and H₂.^71,72 All of the prepared catalysts exhibited stable performance over 20 hours without significant deactivation. As shown in Fig. 6a, the CH₄ conversion decreased by approximately 11%, 10%, 10%, and 8% for MNA, Ni/MA, BPNA, and Ni/BPA, respectively. Although MNA appeared relatively stable, it actually exhibited the highest loss in CH₄ conversion over time. On the other hand, Ni/BPA showed the smallest decrease and the highest stability. In addition, compared to the deactivation relative rate of catalysts in the literature as summarized in Table S2, the Ni/BPA showed a high ranking. This result indicates that the developed catalyst can be a promising candidate as commercial catalysts for the CDR reaction.

Fig. 6 (a) CH₄ conversion, (b) CO₂ conversion, and (c) H₂/CO ratio of the Ni/Al₂O₃ catalysts at 700 °C (CDR reaction).

This excellent stability was mainly due to the strong interaction between Ni and support based on the spinel NiAl₂O₄ structure. Ni/BPA shows the best activity among the prepared catalysts. The initial CH₄ and CO₂ conversions over the catalyst were 82% and 85%, respectively. This excellent performance resulted from the highly dispersed Ni active sites owing to the strong interaction between Ni and the support, which enhanced the catalytic activity and resistance to carbon deposition and sintering, as confirmed by the various characterization techniques mentioned above. However, the CH₄ and CO₂ conversions over this catalyst decreased slightly by 75% and 78%, respectively. Therefore, further studies are required to improve the catalytic stability. H₂/CO ratio was proportional to CH₄ and CO₂ conversion. In addition, the CO₂ conversion rates of the BPNA and MNA catalysts showed relatively large fluctuations. This is because the CO₂ conversion rate can fluctuate more easily than the CH₄ conversion rate due to the influence of various side reactions such as reverse water gas shift and methanation reactions.^73–75 On the other hand, since the H₂/CO ratio is mainly determined by the main CDR reaction, it tends to show a similar tendency to the CH₄ conversion rate. The CO₂ conversion rate can be varied due to side reactions such as the RWGS reaction, but the overall H₂/CO ratio remains relatively stable. This result might be due to the reverse water gas shift reaction and methanation when CH₄ conversion was low.^76–78

To evaluate the long-term stability of the Ni/BPA catalyst, which exhibited the highest initial activity, the time-on-stream experiment was extended to 100 h, as shown in Fig. S3. The CH₄ and CO₂ conversion rates decreased by approximately 16% and 14%, from initial conversions of 82% and 85% to 66% and 71% after 100 h, respectively. In the long-term stability test, the activity of the Ni/BPA catalyst tended to decrease slightly with the time on stream despite the high GHSV condition of 900 000 mL g⁻¹ h⁻¹. This indicates that the catalyst has excellent stability even under long-term operating conditions of an industrial scale reforming process.

Fig. S4 shows the XRD patterns of spent Ni/Al₂O₃ catalysts which were recovered after the CDR reaction for 20 h. The diffraction peaks assigned to the metallic Ni, Al₂O₃, and NiAl₂O₄ species are observed in the same position as shown in Fig. 3. The Ni crystallite size of spent catalysts was calculated by the Debye–Scherrer equation. The crystallite size trend of spent catalysts was as follows: Ni/BPA (3.1 nm) < Ni/MA (3.5 nm) < MNA (5.4 nm) < BPNA (6.4 nm). As above mentioned, the Ni crystallite size which was impossible to calculate due to the small size was detected in all the prepared catalysts. Therefore, the sintering phenomenon is observed in all the catalysts. This increasing trend of crystallite size results from the interaction between Ni and Al₂O₃, as verified by H₂-TPR analysis. Interestingly, characteristic peak at 2θ = 26° was detected for all catalysts, which was attributed to graphite carbon species deposited from the methane in the CDR reaction.^79,80 In addition, it was confirmed that the presence of graphite carbon species peaks in all catalysts contributed to the decrease in CH₄ and CO₂ conversion rates from the CDR reaction by covering the active sites in the surface of catalysts. The peak relating to the carbon species of Ni/MA and Ni/BPA catalysts was relatively smaller than that of MNA and BPNA catalysts. The smaller peak of carbon species is due to the small amount of deposited carbon. Among the prepared catalysts, the Ni/BPA catalyst with the smallest graphite carbon peak showed great resistance to carbon deposition due to its bimodal porous structure, improving catalyst stability and activity for the CDR reaction.

TGA analysis was performed to quantify the amount of deposited carbon on the surface of used Ni/Al₂O₃ catalysts after the CDR reaction and the results are shown in Fig. S5. According to the literature, the weight loss observed at about 220 °C is mainly due to moisture removal.^22,72,81,82 The slight increase from 220 °C to about 400 °C is due to the generation of NiO by oxidation of the reduced metallic Ni. The reduction in weight loss above 600 °C is due to the oxidation of graphite carbon. The weight loss decreased in the following order: MNA (25.5%) > BPNA (19.1%) > Ni/MA (12.8%) > Ni/BPA (5.6%). The low weight loss of Ni/BPA and Ni/MA catalysts is because the Ni active point is highly dispersed by the strong interaction between Ni and Al₂O₃, improving carbon deposition resistance.²⁸ As a result of comparing the catalysts prepared by the same preparation method, the catalyst with the bimodal structure showed lower weight loss than the catalyst with the mesoporous structure. This is because the adsorption of –OH₂ or –OH molecules occurred smoothly and interfered with the adsorption of carbon species, and it is confirmed that relatively little carbon deposition occurred.⁸³ As a result, the bimodal structure and the addition of Ni are important to design the Ni/Al₂O₃ catalyst with high resistance to carbon deposition. Furthermore, the resistance to carbon deposition is strongly influenced by the Ni crystallite size and the strength of the metal–support interaction. It is confirmed that the Ni introduction played a crucial role in determining these characteristics using XRD, H₂-TPR, and H₂ chemisorption analyses. A smaller Ni crystal size leads to higher dispersion of active sites, while a stronger Ni–Al₂O₃ interaction suppresses the sintering of Ni particles and contributes to improved catalytic stability in the CDR reaction. Therefore, by optimizing the Ni incorporation strategy, these critical structural parameters can be controlled, thereby effectively minimizing carbon deposition even under severe reaction conditions.

Fig. S6 shows Raman spectra used to analyze the type of carbon species deposited on the surface of the catalyst after the CDR reaction. The graphite carbon causes more serious deactivation problems in the catalysts, compared to disordered carbon. The D band and G band are observed at 1300–1350 cm⁻¹ and 1500–1600 cm⁻¹, respectively.²⁹ The D band is related to disordered carbon, and the G band is related to the stretching vibration of sp² of ordered carbon and graphite carbon species.^84,85 Therefore, the I_D/I_G representing the intensity ratio of the D band and the G band represents the degree of disorder and graphitization, and the higher the I_D/I_G, the higher the degree of disorder and the lower the graphitization degree.⁴⁷ The I_D/I_G ratios for all catalysts were in the following order: Ni/BPA (1.54) > Ni/MA (1.43) > BPNA (1.38) > MNA (1.37). It was confirmed that the carbon species deposited on the Ni/BPA catalyst after the reaction had a relatively low graphitization degree compared to other catalysts. Lyu et al. reported that the effect of the bimodal porous structure improves mass transfer and quickly removes carbon deposition, reducing the conversion of amorphous carbon to graphite carbon.⁴⁶ In addition, the low peak intensity of the carbon species of the Ni/BPA catalyst after the CDR reaction is lower than that of other catalysts, which means that the content of carbon deposition is small.⁴⁷ Therefore, it was confirmed that the bimodal porous structure and Ni addition control are influencing factors in the design of Ni/Al₂O₃ catalysts with high resistance to carbon deposition.

The FE-TEM images of the spent catalysts after TOS studies were obtained and are shown in Fig. S7. In Fig. S7(a–c), numerous deposited carbons were observed in the spent catalysts.²² In particular, the encapsulated carbon which covered with Ni particles was detected in the MNA and BPNA catalysts, which causes the deactivation of the catalyst. In addition, it can be clearly confirmed that the size of the Ni particles has increased compared to before the reaction, and it is considered that sintering has occurred. On the other hand, no carbonaceous deposits were observed in the Ni/BPA catalyst. This result is not fully consistent with the TGA and Raman results. This might be due to the small amount of deposited carbon which is difficult to detect with TEM analysis.

When carbon atoms diffuse to the interface adjacent to the support through Ni particles, a filamentous carbon structure is formed.⁸⁶ The proper growth of the filament can prevent encapsulation and prolong catalytic activity.⁸⁶ However, encapsulation blocks the active sites, resulting in final catalytic deactivation.⁸⁶ This is related to the metal–support interaction, and it is considered that the Ni/BPA catalyst blocked the growth of carbon due to the strong interaction between Ni and alumina support.⁸⁷ Also, most deposited carbon was filamentous carbon as confirmed by TEM images. It was reported that the effect of this carbon is not bigger than that of encapsulated carbon or sintering phenomenum.^86,88,89 When large amount of it was generated, the activity can be dramatically decreased by blocking gas flow in the reactor. In our study, the main reason for the deactivation might be sintering of active sites.

Fig. S8 shows the reaction rates at different CH₄ and CO₂ partial pressures at various reaction temperatures. It was clearly found that the higher the reaction temperature, the higher the CH₄ and CO₂ consumption rates. This result is mainly because the CDR reaction is thermodynamically preferred at a higher reaction temperature. The higher CO₂ consumption rate compared to CH₄ might be partly attributed to the contribution of the RWGS reaction and the relatively easier activation of CO₂. Furthermore, the difference between CO₂ and CH₄ consumption rates became more pronounced at higher temperatures, possibly due to the enhanced RWGS activity and faster CO₂ activation kinetics under such conditions.⁹⁰ The consumption rates of both CH₄ and CO₂ increased with an increase in either CH₄ or CO₂ partial pressure, indicating that higher reactant concentrations promoted the CDR reaction. The CO₂ consumption rate tended to increase more rapidly than that of CH₄ with increasing CH₄ or CO₂ partial pressure, which can be attributed to the relatively easier activation of CO₂ compared to the dissociation of CH₄.⁹¹

3.3. Proposed effect of the porous structure and addition of Ni on the catalytic performance

3.3.1. Bimodal porous structure. In this study, the effect of the bimodal porous structure on the catalysis in the CDR reaction was investigated using N₂ adsorption/desorption, FE-SEM, XRD, TGA and Raman spectroscopy analysis. The Ni/Al₂O₃ catalysts with a bimodal porous structure were suppressed for the deposition of carbon on the surface of the catalyst. In addition, higher methane conversion was observed in bimodal porous Ni/Al₂O₃ catalysts. These positive effects of the bimodal porous structure are based on the excellent ability of mass transfer. Scheme 1 shows the flow of reactants and products in the CDR reaction with mesoporous and bimodal porous structures. The reactants and products are easily transferred in the interface of the catalyst when a bimodal porous structure including macropores is used. This effect can contribute to achieving a higher conversion of reactants by promoting the supply of reactants into the interface and emission of products to the outside. In addition, the intermediates of deposited carbon will be reacted with the reactant, which is quickly supplied, and it will be moved to the outside of the interface due to the higher mass transfer of the bimodal porous structure. As a result, enhanced catalytic activity and resistance to carbon deposition are attributed to the bimodal porous structure.

Scheme 1 The effect of the bimodal porous structure on the catalysis in the CDR reaction.

3.3.2. The effect of the addition of Ni on the preparation of Ni/Al₂O₃. In this study, the Ni precursor was added when Al₂O₃ support was prepared, or it was added on the Al₂O₃ support which was already prepared for the synthesis of bimodal porous and mesoporous Ni/Al₂O₃ catalysts. Scheme 2 shows the effect of addition of Ni on the key properties which affect the catalytic performance. Based on the results of H₂-TPR, FE-TEM, XPS, TGA and Raman spectroscopy analyses, it was confirmed that several Ni particles were inserted into the bulk Al₂O₃ for Ni/Al₂O₃ catalysts when the Ni precursor was added in the preparation process for the Al₂O₃ support. This insertion decreased the CH₄ conversion by decreasing the Ni dispersion on the surface of Al₂O₃. In addition, it causes the interaction between Ni and Al₂O₃ to weaken, which leads to lower stability by decreasing resistance to sintering and carbon deposition. Therefore, it should be noted that Ni must be added after bimodal porous or mesoporous Al₂O₃ supports are finally prepared.

Scheme 2 The effect of the addition of Ni in the preparation of Ni/Al₂O₃.

4. Conclusions

In this study, different preparation methods were used to determine the pore structure and interactions between Ni and Al₂O₃. Mesoporous and bimodal porous Ni/Al₂O₃ catalysts were prepared to compare their pore properties, and the addition of the Ni precursor was controlled to compare the interactions between Ni and Al₂O₃. The Ni/Al₂O₃ catalysts with Ni added after the calcination of the support (Ni/MA, Ni/BPA) exhibited higher catalytic performance than those with Ni added during support preparation (MNA, BPNA). Furthermore, the bimodal porous catalyst (Ni/BPA) outperformed its mesoporous counterpart (Ni/MA). In addition, the bimodal porous Ni/Al₂O₃ catalyst (Ni/BPA) showed better catalytic performance than the mesoporous Ni/Al₂O₃ catalyst (Ni/MA). This result indicates that the macropores in bimodal porous Ni/Al₂O₃ contribute to enhanced catalytic performance by improving the mass transfer. This was because the amount of Ni inserted into bulk Al₂O₃ was higher when the precursors of Ni and Al₂O₃ were mixed during the preparation procedure. The Ni/BPA catalyst exhibited the highest catalytic performance for the CDR reaction. The remarkable performance of this catalyst was mainly due to the high mass transfer and proper interaction between Ni and Al₂O₃. However, the catalytic activity decreased slightly with time-on-stream. Therefore, further studies are required to control the pore properties of the bimodal porous Ni/Al₂O₃ catalysts. In addition, the Ni loading effect should be investigated because Ni loading affects the pore properties such as surface area, pore size, and pore volume and reduction properties including interaction between Ni and Al₂O₃.

Author contributions

Ji-Won Son: writing – original draft, investigation, data curation. Hak-Min Kim: writing – original draft, writing – review & editing, investigation, validation. Jae-Min Song: methodology, visualization. Beom-Su Cheon: methodology, visualization. Dae-Woon Jeong: conceptualization, supervision, resources, funding acquisition.

Conflicts of interest

There are no conflicts to declare.

Data availability

The data supporting this article have been included as part of the supplementary information (SI). Supplementary information: additional tables and figures presenting the kinetic experiment design, catalyst characterization results (BET, XPS, XRD, TGA, Raman, and TEM), and stability test data supporting the findings of this study. See DOI: https://doi.org/10.1039/d5ta07037f.

Acknowledgements

This work was supported by the Korea Institute of Energy Technology Evaluation and Planning (KETEP) grant funded by the Korea government (MOTIE) (20214000000090, Fostering human resources training in advanced hydrogen energy industry). This work was supported by the National Research Foundation of Korea (NRF) and the Commercialization Promotion Agency for R&D Outcomes (COMPA) grant funded by the Korea government (Ministry of Science and ICT) (RS-2024-00432910). This work was supported by the National Research Foundation of Korea (NRF) grant funded by the Korea government (MSIT) (RS-2023-00214068).

Notes and references

1. S. Sun, C. Zhang, Y. Wang, X. Zhao, H. Sun and C. Wu, Chem. Eng. J., 2023, 468, 143712.
2. J. Wang and W. Azam, Geosci. Front., 2024, 15, 101757.
3. L. Aleluia Reis and M. Tavoni, iScience, 2023, 26, 105933.
4. D. J. van de Ven, S. Mittal, A. Gambhir, R. D. Lamboll, H. Doukas, S. Giarola, A. Hawkes, K. Koasidis, A. C. Köberle, H. McJeon, S. Perdana, G. P. Peters, J. Rogelj, I. Sognnaes, M. Vielle and A. Nikas, Nat. Clim. Change, 2023, 13, 570–578.
5. E. Hille, Energy Econ., 2023, 127, 107061.
6. D. Eze Ekechukwu, O. Ogundipe, A. Esiri, A. Heavens Ikevuje, J. Mensah Kwakye, O. Benjamin Ogundipe and A. Emuobosa Esiri, Artic. Open Access Res. J. Multidiscip. Stud., 2024, 2024, 94–104.
7. T. Capurso, M. Stefanizzi, M. Torresi and S. M. Camporeale, Energy Convers. Manag., 2022, 251, 114898.
8. M. Younas, S. Shafique, A. Hafeez, F. Javed and F. Rehman, Fuel, 2022, 316, 123317.
9. M. J. Park, H. M. Kim, Y. J. Gu and D. W. Jeong, Energy, 2023, 265, 126273.
10. M. J. Park, H. M. Kim, Y. H. Lee, K. W. Jeon and D. W. Jeong, Waste Manag., 2022, 144, 272–284.
11. H. Wang, X. Zhu, A. Adogwa, Y. Shen, M. Yang and T. B. Lu, Chem. Eng. J., 2024, 493, 152501.
12. D. H. Le Phuong, M. Alsaiari, C. Q. Pham, N. H. Hieu, T. P. T. Pham, N. Rajamohan, D. D. Pham, D. V. N. Vo, T. H. Trinh, H. D. Setiabudi, D. L. T. Nguyen and T. M. Nguyen, J. Taiwan Inst. Chem. Eng., 2024, 155, 105253.
13. R. Gao, C. Zhang, Y. J. Lee, G. Kwak, K. W. Jun, S. K. Kim, H. G. Park and G. Guan, J. Clean. Prod., 2020, 272, 122552.
14. P. K. Gupta, V. Kumar and S. Maity, J. Chem. Technol. Biotechnol., 2020, 96, 853–868.
15. J. J. Torrez-Herrera, S. A. Korili and A. Gil, Chem. Eng. J. Adv., 2021, 5, 100080.
16. S. Yuan, Y. Ge, P. Su, H. Cheng, Z. Wang and F. Yu, J. Environ. Chem. Eng., 2024, 12, 112308.
17. M. R. B. Pirshahid, S. M. Alavi, M. Rezaei, E. Akbari, M. Varbar and K. Khosravi, J. Energy Inst., 2023, 110, 101361.
18. A. Choya, B. de Rivas, M. L. No, J. I. Gutiérrez-Ortiz and R. López-Fonseca, Fuel, 2024, 358, 130166.
19. M. Dan, M. Mihet, G. Borodi and M. D. Lazar, Catal. Today, 2021, 366, 87–96.
20. C. A. Schwengber, F. A. Da Silva, R. A. Schaffner, N. R. C. Fernandes-Machado, R. J. Ferracin, V. R. Bach and H. J. Alves, J. Environ. Chem. Eng., 2016, 4, 3688–3695.
21. W. S. Dong, H. S. Roh, K. W. Jun, S. E. Park and Y. S. Oh, Appl. Catal., A, 2002, 226, 63–72.
22. Z. Bian, W. Zhong, Y. Yu, Z. Wang, B. Jiang and S. Kawi, Int. J. Hydrogen Energy, 2021, 46, 31041–31053.
23. H. Yi, Q. Xue, S. Lu, J. Wu, Y. Wang and G. Luo, Fuel, 2022, 315, 123262.
24. V. Y. Bychkov, Y. P. Tyulenin, A. A. Firsova, E. A. Shafranovsky, A. Y. Gorenberg and V. N. Korchak, Appl. Catal., A, 2013, 453, 71–79.
25. S. Hu, M. Xue, H. Chen and J. Shen, Chem. Eng. J., 2010, 162, 371–379.
26. E. Im, H. J. Seo, D. I. Kim, D. C. Hyun and G. D. Moon, Chem. Eng. J., 2021, 416, 129147.
27. Z. Y. Lim, J. Tu, F. Zhou, B. Chen and K. L. Choy, Appl. Catal., B, 2024, 350, 123891.
28. K. Tao, Y. Zhang, S. Terao, Y. Yoneyama, T. Kawabata, K. Matsuda, S. Ikeno and N. Tsubaki, Chem. Eng. J., 2011, 170, 258–263.
29. L. Lyu, Y. Han, Q. Ma, S. Makpal, J. Sun, X. Gao, J. Zhang, H. Fan and T.-S. Zhao, Catalysts, 2020, 10, 1220.
30. A. H. K. Owgi, A. A. Jalil, I. Hussain, N. S. Hassan, H. U. Hambali, T. J. Siang and D. V. N. Vo, Environ. Chem. Lett., 2021, 19, 2157–2183.
31. F. Wen, C. Xu, N. Huang, T. Wang, X. Sun, H. Li, R. Zhang and G. Xia, Int. J. Hydrogen Energy, 2024, 69, 1481–1491.
32. I. C. Sophiana, S. Steven, A. Revieri, A. Permatasari, R. Andika, N. Nishiyama and B. H. Susanto, Case Stud. Chem. Environ. Eng., 2025, 11, 101078.
33. M. S. Mel’gunov, Adsorption, 2023, 29, 199–208.
34. D. W. Jeong, A. Jha, W. J. Jang, W. B. Han and H. S. Roh, Chem. Eng. J., 2015, 265, 100–109.
35. J. Wang, X. Wang and X. Zhang, J. Mater. Chem. A, 2017, 5, 4308–4313.
36. L. Xu, J. Zhang, J. Ding, T. Liu, G. Shi, X. Li, W. Dang, Y. Cheng and R. Guo, Miner, 2020, 10, 72.
37. S. Jayaprakash, N. Dewangan, A. Jangam, S. Das and S. Kawi, Int. J. Hydrogen Energy, 2021, 46, 18338–18352.
38. K. Tao, L. Shi, Q. Ma, D. wang, C. Zeng, C. Kong, M. Wu, L. Chen, S. Zhou, Y. Hu and N. Tsubaki, Chem. Eng. J., 2013, 221, 25–31.
39. R. Daroughegi, F. Meshkani and M. Rezaei, Int. J. Hydrogen Energy, 2017, 42, 15115–15125.
40. C. Italiano, J. Llorca, L. Pino, M. Ferraro, V. Antonucci and A. Vita, Appl. Catal., B, 2020, 264, 118494.
41. N. D. Charisiou, L. Tzounis, V. Sebastian, S. J. Hinder, M. A. Baker, K. Polychronopoulou and M. A. Goula, Appl. Surf. Sci., 2019, 474, 42–56.
42. X. Dong, W. Wang, G. Yang and L. Wang, Ind. Eng. Chem. Res., 2022, 61, 4255–4263.
43. Z. Song, H. Chen, S. Bao, Z. Xie, Q. Kuang and L. Zheng, J. Mater. Chem. A, 2020, 8, 3754–3762.
44. R. Kumar, S. Shah, J. Bahadur, Y. B. Melnichenko, D. Sen, S. Mazumder, C. P. Vinod and B. Chowdhury, Microporous Mesoporous Mater., 2016, 234, 293–302.
45. W. Gao, Q. Yin, X. Meng, X. He and Z. Xin, Fuel, 2022, 310, 121813.
46. L. Lyu, M. Shengene, Q. Ma, J. Sun, X. Gao, H. Fan, J. Zhang and T. S. Zhao, Fuel, 2022, 310, 122375.
47. Q. Ma, Y. Han, Q. Wei, S. Makpal, X. Gao, J. Zhang and T. S. Zhao, J. CO₂ Util., 2020, 35, 288–297.
48. C. Ma, J. Gong, S. Zhao, X. Liu, X. Mu, Y. Wang, X. Chen and T. Tang, Green Energy Environ., 2022, 7, 818–828.
49. B. Li, X. Lin, Y. Luo, X. Yuan and X. Wang, Fuel Process. Technol., 2018, 176, 153–166.
50. Y. Chai, Y. Fu, H. Feng, W. Kong, C. Yuan, B. Pan, J. Zhang and Y. Sun, ChemCatChem, 2018, 10, 2078–2086.
51. Z. Bao, Y. Lu, J. Han, Y. Li and F. Yu, Appl. Catal., A, 2015, 491, 116–126.
52. N. Pham-Ngoc, A. Jamsaz, M. Wang and E. Woo Shin, Chem. Eng. J., 2024, 494, 153207.
53. J. Zhang, Z. Xiao, L. Wang, X. Zhang and G. Li, ChemistrySelect, 2022, 7, e202203339.
54. J. J. Torrez-Herrera, S. A. Korili and A. Gil, Catal. Rev., 2023, 65, 1300–1357.
55. F. Rahbar Shamskar, M. Rezaei and F. Meshkani, Int. J. Hydrogen Energy, 2017, 42, 4155–4164.
56. W. Han, P. Yuan, Y. Fan, G. Shi, H. Liu, D. Bai and X. Bao, J. Mater. Chem., 2012, 22, 25340–25353.
57. F. E. Massoth, J. Catal., 1975, 36, 164–184.
58. S. V. Moghaddam, M. Rezaei and F. Meshkani, Energy Technol., 2020, 8, 1900778.
59. Y. Bae and J. Hong, Chem. Eng. J., 2022, 446, 136978.
60. S. Xu, T. J. A. Slater, H. Huang, Y. Zhou, Y. Jiao, C. M. A. Parlett, S. Guan, S. Chansai, S. Xu, X. Wang, C. Hardacre and X. Fan, Chem. Eng. J., 2022, 446, 137439.
61. Z. Wang, Z. Mei, L. Wang, Q. Wu, C. Xia, S. Li, T. Wang and C. Liu, J. Mater. Chem. A, 2024, 12, 24802–24838.
62. P. Liu, J. Ran, B. Xia, S. Xi, D. Gao and J. Wang, Nano–Micro Lett., 2020, 12, 1–12.
63. H. J. Byeon, K. W. Jeon, H. M. Kim, Y. H. Lee, Y. S. Heo, M. J. Park and D. W. Jeong, Fuel Process. Technol., 2022, 231, 107229.
64. S. Ali, M. M. Khader, M. J. Almarri and A. G. Abdelmoneim, Catal. Today, 2020, 343, 26–37.
65. D. Gallego-García, U. Iriarte-Velasco, M. A. Gutiérrez-Ortiz and J. L. Ayastuy, Appl. Catal., B, 2024, 344, 123671.
66. S. Yuan, Y. Yang, Z. Xiong, P. Guo, S. Sun, Z. Li, J. Du and Y. Gao, Green Energy Environ., 2024, 9, 321–332.
67. Z. Liu, W. Han, D. Hu, S. Sun, A. Hu, Z. Wang, Y. Jia, X. Zhao and Q. Yang, J. Catal., 2020, 387, 62–72.
68. H. Wang, Z. Fang, Y. Wang, K. Meng and S. Sun, J. Alloys Compd., 2024, 986, 174006.
69. J. Chen, Y. Zhang, Z. Zhang, D. Hou, F. Bai, Y. Han, C. Zhang, Y. Zhang and J. Hu, J. Mater. Chem. A, 2023, 11, 8540–8572.
70. Y. Chen, S. Zhu, J. Zhou, Y. Wang, Z. Zhang, N. Liu, R. Meng and S. He, Int. J. Hydrogen Energy, 2024, 96, 309–320.
71. M. G. Jeong, S. Y. Kim, D. H. Kim, S. W. Han, I. H. Kim, M. Lee, Y. K. Hwang and Y. D. Kim, Appl. Catal., A, 2016, 515, 45–50.
72. M. J. Park, J. H. Kim, Y. H. Lee, H. M. Kim and D. W. Jeong, Int. J. Hydrogen Energy, 2020, 45, 30188–30200.
73. V. A. Panchenko, Y. V. Daus, A. A. Kovalev, I. V. Yudaev and Y. V. Litti, Int. J. Hydrogen Energy, 2023, 48, 4551–4571.
74. M. Tommasi, S. N. Degerli, G. Ramis and I. Rossetti, Chem. Eng. Res. Des., 2024, 201, 457–482.
75. M. Ayoub, C. C. Chong, A. Zamir, Y. W. Cheng, S. Farrukh, S. R. Naqvi, H. D. Setiabudi, N. Riaz and N. Ramzan, E3S Web Conf., 2021, 287, 1–8.
76. S. Tada, A. Yanagita, N. Shimoda, T. Honma, M. Takahashi, A. Nariyuki and S. Satokawa, J. Jpn. Pet. Inst., 2018, 61, 80–86.
77. S. Tada, D. Shoji, K. Urasaki, N. Shimoda and S. Satokawa, Catal. Sci. Technol., 2016, 6, 3713–3717.
78. N. Shimoda, D. Shoji, K. Tani, M. Fujiwara, K. Urasaki, R. Kikuchi and S. Satokawa, Appl. Catal., B, 2015, 174–175, 486–495.
79. S. S. Negi, H. M. Kim, B. S. Cheon, C. H. Jeong, H. S. Roh and D. W. Jeong, ACS Appl. Mater. Interfaces, 2023, 15, 51013–51024.
80. Y. Shi, S. Wang, Y. Li, F. Yang, H. Yu, Y. Chu, T. Li and H. Yin, Materials, 2022, 15, 3044.
81. J. Niu, S. E. Liland, J. Yang, K. R. Rout, J. Ran and D. Chen, Chem. Eng. J., 2019, 377, 119763.
82. Y. Gao, Q. Li, C. Wang, D. Yan, J. Chen and H. Jia, J. Mater. Chem. A, 2022, 10, 16016–16028.
83. Q. Zhang, Z. Jiang, Y. Zhang, X. Xu, Y. Yang, Y. Qin, L. Song, Y. Mei and Y. Zu, Appl. Catal., B, 2025, 362, 124722.
84. S. Sun, Y. Wang, Y. Xu, H. Sun, X. Zhao, Y. Zhang, X. Yang, X. Bie, M. Wu, C. Zhang, Y. Zhu, Y. Xu, H. Zhou and C. Wu, Appl. Catal., B, 2024, 348, 123838.
85. F. Rosenburg, E. Ionescu, N. Nicoloso and R. Riedel, Mater, 2018, 11, 93.
86. M. A. Abdel-Fatah and A. Amin, Int. J. Hydrogen Energy, 2025, 137, 236–246.
87. P. K. Chaudhary and G. Deo, Colloids Surf., A, 2022, 646, 128973.
88. A. Serrano-Lotina and L. Daza, Appl. Catal., A, 2014, 474, 107–113.
89. J. Xu, W. Zhou, J. Wang, Z. Li and J. Ma, Chin. J. Catal., 2009, 30, 1076–1084.
90. W. J. Jang, D. W. Jeong, J. O. Shim, H. M. Kim, H. S. Roh, I. H. Son and S. J. Lee, Appl. Energy, 2016, 173, 80–91.
91. Z. Bao, L. Yongwu and Y. Fei, AIChE J., 2017, 63, 2019–2029.

---

Footnote

† This author contributed equally as the first author.

---