Plasma-catalytic reforming of CH₄–CO₂ over porous Ni/N-doped carbon: efficient syngas production and mechanistic insights

Abstract

The integration of non-thermal plasma and catalysis offers a promising route for low-temperature reforming of methane (CH₄) and carbon dioxide (CO₂) into high-value syngas (H₂ and CO), enabling efficient activation of stable molecules and tunable product selectivity. However, limitations in energy efficiency, selectivity control, and catalyst durability remain. In this study, a water-cooled dielectric barrier discharge (DBD) reactor coupled with a porous nitrogen-doped carbon-supported nickel catalyst (Ni/N–C) was developed to enhance CH₄–CO₂ reforming performance. The results showed that increasing the CH₄/CO₂ molar ratio significantly enhanced both reactant conversion and syngas selectivity, while a higher gas flow rate adversely affected conversion efficiency. Under optimal conditions (60 mL min⁻¹ gas flow rate, 1 : 5 molar ratio of CH₄/CO₂), CH₄ and CO₂ conversions reached 44.1% and 20.0%, with CO and H₂ selectivities of 67.6% and 48.1%, respectively. The corresponding energy efficiency was 0.39 mmol kJ⁻¹. Mechanistic insights derived from catalyst characterization and performance analysis revealed that moderate acid sites promoted CH₄ activation and facilitated the selective formation of C₂ hydrocarbons, while abundant basic sites enhanced CO₂ adsorption and activation, thereby improving CO and H₂ selectivity. The synergistic effect of acid–base site modulation and plasma-driven activation played a key role in steering the reaction pathway and optimizing product distribution. This work highlighted the potential of tailored Ni-based catalysts for efficient and selective plasma-catalytic reforming of CH₄ and CO₂ into syngas.

---

1. Introduction

Currently, the world is facing dual challenges of global warming and escalating energy demands. The 2015 Paris Agreement established a goal to limit the global temperature rise to within 1.5 °C above pre-industrial levels, while according to the Intergovernmental Panel on Climate Change (IPCC) Sixth Assessment Report, the global average temperature has already risen by 1.1 °C.^1–3 The excessive emission of greenhouse gases has led to a range of environmental challenges, including climate change, air pollution, rising sea levels, and increased frequency of extreme weather events.⁴ CH₄ and CO₂ are recognized as two major greenhouse gases, with their mitigation strategies and emission reduction technologies drawing significant scientific and societal attention.⁵ Dry reforming of methane (DRM) offers a promising route for simultaneous conversion of CH₄ and CO₂ into CO and H₂ (eqn(1)), valuable intermediates for downstream processes such as Fischer–Tropsch synthesis to yield alcohols, acids, and aldehydes.⁶ This approach not only addresses carbon emission reduction and recycling but also contributes to chemical feedstock generation and partial fulfillment of energy needs.⁷ However, the DRM reaction is highly endothermic (ΔH_298K = +247.3 kJ mol⁻¹), and the activation of CH₄ (C–H, 439 kJ mol⁻¹) and CO₂ (C=O, 750 kJ mol⁻¹) requires high energy input. As a result, conventional catalytic DRM typically operates above 700 °C, imposing significant energy and operational costs.⁸ Therefore, developing efficient catalytic systems capable of activating CH₄ and CO₂ under milder conditions is critical to improving the energy efficiency and sustainability of DRM.

CH₄ + CO₂ → 2CO + 2H₂ ΔH_298K = +247.3 kJ mol⁻¹ (1)

Plasma technology has brought new development prospects for DRM. According to the degree of thermodynamic equilibrium among electrons, ions, and neutral species, plasma can be generally classified into two categories: thermal plasma and non-thermal plasma (NTP).⁹ Thermal plasma processes can achieve extremely high temperatures (typically above 4000 K), enabling near-complete conversion of reactants and effective treatment of solid waste, while simultaneously producing syngas with high CO and H₂ selectivities (>80%).^10,11 In contrast, NTP offers a promising strategy for enabling DRM under ambient or near-ambient conditions. With electron temperatures typically ranging from 1 to 10 eV, NTP provides sufficient energy to activate thermodynamically stable CH₄ (4.5 eV) and CO₂ (5.5 eV) molecules.¹² Through processes like excitation, collision, ionization, and dissociation, NTP creates high-energy electrons and reactive radicals from CH₄ and CO₂ under ambient conditions. This enables the technology to drive thermodynamically constrained reactions to proceed rapidly under low-temperature conditions, thereby significantly reducing the requirements for high temperatures and associated energy consumption.¹³ However, NTP systems still face challenges such as relatively low selectivity toward target products and the need for further improvement in energy efficiency.¹⁴ Consequently, researchers have introduced catalysts to lower activation energy and regulate reaction pathways, achieving efficient conversion of CH₄ and CO₂ with substantially improved product selectivity and energy efficiency.¹⁵ As the core component of DRM technology, catalysts have become the primary research focus, with most studies concentrating on catalyst development.¹⁶

Currently, noble metal catalysts loaded with rhodium (Rh), platinum (Pt), and palladium (Pd) exhibit excellent catalytic activity in DRM reactions. However, their extremely limited global reserves result in high costs, making industrial-scale applications challenging.¹⁷ In contrast, transition metal catalysts (e.g., Ni, Co, and Fe) combine high intrinsic activity with cost-effectiveness.¹⁸ Among them, Ni-based catalysts, typically supported on materials such as metal oxides,¹⁹ zeolite,²⁰ and composite oxides,²¹ have been widely employed in plasma-assisted DRM systems due to their cost-effectiveness and strong activity in C–C/C–H bond activation.¹⁸ For example, Abbas et al.²² fabricated a Ni–Co₃O₄/TiO₂ NR (nanorod) catalyst that, when integrated with NTP, achieved CO and H₂ selectivities of 49.0% and 50.1%, respectively, which were significantly higher than the 31.6% and 33% obtained in plasma-only operation. In addition, the hybrid system attained a maximum energy efficiency of 0.131 mmol kJ⁻¹, representing a 26% improvement compared to the plasma-only system. Besides catalytic activity, recent studies have highlighted the crucial role of support properties in determining active site dispersion, charge transfer, and plasma–catalyst synergy. For instance, Diao et al.²³ showed that Ni/NS-Al₂O₃ with finely dispersed Ni nanoparticles enhanced electron density and reduced CO₂ activation energy, while Xu et al.²⁴ found that embedding Ni into the mesopores of MCM-41 improved dispersion and accessibility of active sites. Despite these advances, conventional metal oxides and zeolites still face challenges related to limited activity or structural stability under plasma conditions. Hence, the development of novel support materials with tailored textural and electronic properties is essential for further enhancing DRM performance in plasma-catalytic systems.

Metal–organic framework (MOF)-derived carbon materials have emerged as highly promising supports due to their large specific surface area, well-developed porosity, and tunable pore structures. These properties facilitate the uniform dispersion of metal nanoparticles and enhance their stability under reaction conditions.²⁵ Among them, ZIF-8-derived carbon materials are particularly attractive due to their exceptional thermochemical stability and activation capacity after metal incorporation.²⁶ For example, Guo et al.²⁷ synthesized nitrogen-doped, hierarchically porous carbon polyhedra from a ZIF-8 framework. They found that the well-defined micro- and mesoporous structure significantly enhanced mass transport and exposed internal active sites, resulting in excellent CO₂ photoconversion performance under mild conditions. Similarly, Li et al.²⁸ developed M/ZIF-8-C catalysts (M = Ni, Fe, Co, and Cu) for CO₂ hydrogenation. Their study demonstrated that highly dispersed metal nanoparticles, along with pyridinic nitrogen and carbide species, effectively promoted CO₂ activation and provided abundant active sites. These findings collectively highlight the potential of ZIF-8-derived carbon materials as multifunctional supports for CO₂ conversion reactions. Moreover, nitrogen co-doping introduces abundant defect sites, which enhance charge transfer, improve metal anchoring, and contribute to higher mass activity and catalytic stability.²⁹ Despite these advantages, the application of MOF-derived carbon materials in DRM catalysis remains relatively underexplored. In particular, the underlying reaction mechanisms, those occurring under NTP conditions, are not yet fully understood. This knowledge gap presents both a challenge and an opportunity for future research.

In this study, a porous nitrogen-doped carbon material (N–C) was synthesized as a support for nickel and integrated into a water-cooled dielectric barrier discharge (DBD) reactor for low-temperature reforming of CH₄ and CO₂. The effects of key process parameters, including the gas flow rate and CH₄/CO₂ molar ratio, on conversion rates, product selectivity, and energy efficiency were thoroughly investigated. Based on the analysis of reaction products, identification of plasma-generated reactive species, and detailed catalyst characterization, a synergistic mechanism governing CH₄/CO₂ reforming over the Ni/N–C catalyst under NTP conditions was proposed.

2. Experimental

2.1. Chemicals and instruments

CH₄ (99.99%) and CO₂ (99.99%) were purchased from Xi'an Yuefeng Tianyi Gas Co., Ltd. (China). Zinc nitrate hexahydrate (Zn(NO₃)₂·6H₂O, analytical reagent (AR), 98%), 2-methylimidazole (C₄H₆N₂, AR, 98%), and methanol (CH₃OH, AR, ≥99.5%) were procured from Shanghai Macklin Biochemical Technology Co., Ltd. (China). Nickel nitrate hexahydrate (Ni(NO₃)₂·6H₂O, AR, 98%) and hydrochloric acid (HCl, AR, 36–38%) were obtained from Sinopharm Chemical Reagent Co., Ltd. (China). Deionized water was used throughout the experiments.

2.2. Catalyst preparation

2.2.1. Preparation of the N–C support. ZIF-8 was synthesized with reference to the conventional method reported in the literature.²⁹ Specifically, 5.578 g of Zn(NO₃)₂·6H₂O was dissolved in 150 mL of methanol, followed by ultrasonication at 40 °C for 10 minutes. Similarly, 6.160 g of 2-methylimidazole was dissolved in 150 mL of methanol to prepare the corresponding solution. First, a total of 4 mL of the metal salt solution was injected into the 2-methylimidazole solution and ultrasonicated for 10 minutes, followed by the addition of the remaining solution and another 10 minutes of ultrasonication. The mixture was then subjected to static growth at 35 °C for 12 hours. After centrifugation at 8000 rpm for 3 minutes, thorough washing with methanol, and drying at 80 °C for 12 hours, a white ZIF-8 powder was obtained. The dried ZIF-8 was carbonized in N₂ at 850 °C (heating rate 3 °C min⁻¹) for 3 hours. The resulting black sample was washed three times with 1 M HCl solution to remove residual inorganic components, ultimately yielding the N–C support.

2.2.2. Preparation of the Ni/N–C catalyst. The procedure involved adding 5 g of N–C support to an aqueous solution containing 2.750 g of Ni(NO₃)₂·6H₂O, which was stirred for 5 hours and then centrifuged at 6000 rpm for 5 minutes, followed by drying at 80 °C for 12 hours. Finally, the dried sample was calcined in N₂ at 550 °C (heating rate: 2 °C min⁻¹) for 4 hours. After cooling to room temperature, the 10 wt% Ni/N–C catalyst was obtained.

2.3. Experimental setup

Fig. 1 shows a schematic diagram of the experimental setup. High-purity CH₄ and CO₂ gases were regulated using mass flow controllers (D07, Sevenstar) and premixed before being introduced into a cylindrical DBD reactor. The reactor comprised a quartz tube (outer diameter: 12 mm; inner diameter: 8 mm) with a central stainless-steel high-voltage electrode (diameter: 4 mm) and a water-cooled outer electrode. The discharge gap was 2 mm, and the discharge length was 50 mm. The catalyst bed was prepared by uniformly mixing 200 mg of tablet-pressed and sieved catalyst with 400 mg of quartz sand. The mixture was loaded into the quartz tube, fixed in place using quartz wool plugs, and maintained at 10 °C using a water cooler equipped with a temperature detector sensor (PT100, Dongyang Sanxing Temperature Instrument Co., Ltd., China). An adjustable high-voltage alternating current (AC) power supply (CTP-2000 K, Nanjing Suman Plasma Technology Co., Ltd., China) delivering energy to the plasma system was employed for plasma generation. Electrical signals were recorded using a digital oscilloscope (DSOX2014A, Keysight Technologies, USA), with discharge power calculated based on Lissajous figure analysis. Liquid products (CH₃OH, C₂H₅OH, CH₃COOH, and CH₃COCH₃) were condensed in a cooling trap and collected in a liquid-phase bottle. The volumetric flow rate of the outlet gas was measured using a soap-film flowmeter. Gaseous reactants and products such as CO, H₂, and C_xH_y (C₂H₆, C₂H₄, C₂H₂, C₃H₈, and C₃H₆) were analyzed online via a gas chromatograph (9790 II, Fuli, China). Additionally, active species generated during plasma-driven CH₄–CO₂ reforming were detected using optical emission spectroscopy (OES) with a spectrometer (Maya-2000 Pro, Ocean Insight, USA).

Fig. 1 Schematic diagram of the experimental setup.

The conversion of reactants, selectivity to gaseous products and oxygenates, and energy efficiency were calculated using the following equations (eqn (2)–(13)):

S_oxygenates (%) = 100% − (S_CO + S_CxHy) − ca. 5% carbon deposition (7)

where X_CH4 and X_CO2 represent the conversions of CH₄ and CO₂, respectively; S_H2, S_CO and S_CxHy represent the selectivities to H₂, CO and C_xH_y, respectively; S_oxygenates, S_CH3COOH, S_CH3COCH3, S_CH3OH and S_CH3CH2OH represent the selectivities to oxygenates, CH₃COOH, CH₃COCH₃, CH₃OH and CH₃CH₂OH, respectively; E (mmol kJ⁻¹) represents the energy efficiency of individual components (CH₄, CO₂, CO, and H₂), E_total (mmol kJ⁻¹) represents the total energy efficiency, P (kW) represents the plasma discharge power, q (mol s⁻¹) represents the flow rate of each component, and subscripts ‘in’ and ‘out’ represent the inlet and outlet of the reactor, respectively.

3. Results and discussion

3.1. Discharge characteristics

Catalysts exert a pronounced influence on the discharge behavior of DBD plasma, while discharge power and energy density play vital roles in determining the type and quantity of active species generated. Fig. 2(a and b) present the discharge signal profiles for plasma-only and Ni/N–C catalyst-packed conditions at a discharge power of 25 W. In both cases, the voltage and charge waveforms exhibit quasi-sinusoidal characteristics.³⁰ In the plasma-only system, the charge peak is approximately 0.47 μC, accompanied by numerous sharp current spikes, indicating that the discharge mode is predominantly filamentary. After introducing the catalyst, both the number and amplitude of charge spikes decrease significantly, whereas the peak charge increases to ∼0.65 μC. This behavior suggests a transition toward a mixed discharge regime combining weak filamentary and surface discharges due to the presence of the catalyst in the discharge gap.³¹ The corresponding discharge images (Fig. S1) visually confirm this shift in the discharge pattern.

Fig. 2 Discharge signals: (a) plasma only; (b) packed with the Ni/N–C catalyst.

3.2. Characterization of materials

To investigate the structural evolution of the carbon support and the crystallinity of the active metal phase, X-ray diffraction (XRD) analysis was performed on the N–C support and Ni/N–C catalyst. As shown in Fig. 3a, the N–C support exhibits broad diffraction peaks at 2θ = 24.1° and 44.4°, corresponding to the (002) and (101) planes of graphitic carbon, respectively, which reflect its predominantly amorphous structure and low degree of graphitization.³² For the Ni/N–C catalyst, well-defined peaks are observed at 2θ = 44.5°, 51.8°, and 76.4°, indexed to the (111), (200), and (220) planes of metallic Ni (PDF#04-0850), confirming the successful incorporation of highly crystalline Ni nanoparticles.³³ It is noteworthy that no characteristic diffraction peaks of NiO were detected in the XRD patterns, which may be attributed to its small crystallite size, highly dispersed state, and low crystallinity.³⁴ The graphite-related peaks remained largely unchanged after Ni loading, indicating that the structural integrity of the N–C framework was preserved.

Fig. 3 (a) XRD patterns of the N–C support and Ni/N–C catalyst; (b) Raman spectrum of the Ni/N–C catalyst; (c) FTIR spectrum of the Ni/N–C catalyst; (d) N₂ adsorption–desorption isotherms of the N–C support and Ni/N–C catalyst; (e) pore size distribution profiles of the N–C support and Ni/N–C catalyst.

Raman spectroscopy (Fig. 3b) was employed to further probe structural defects in the carbon support. The spectrum of the Ni/N–C catalyst displays three characteristic bands: the D band (∼1368 cm⁻¹) arising from disordered carbon and structural defects, the D band at 1368 cm⁻¹ associated with sp³ defects, and the G band at 1585 cm⁻¹ corresponding to the stretching vibrations of all sp²-bonded pairs, with a 2D band at ∼2651 cm⁻¹, which may be related to the presence of layered or partially ordered carbon domains.³⁵ The intensity ratio of the D to G bands (I_D/I_G = 1.14) suggests a high density of surface defects and edge sites, likely introduced by nitrogen doping and carbon disorder. These features are expected to enhance metal dispersion, active site accessibility, and ultimately catalytic performance.³⁶

Fourier transform infrared spectroscopy (FTIR) spectra (Fig. 3c) were recorded to identify surface functional groups. The broad absorption band at 3410 cm⁻¹ is attributed to O–H stretching vibrations, including surface hydroxyl groups and adsorbed water.³⁷ The peak at 2922 cm⁻¹ corresponds to C–H stretching vibrations. A strong band at 1584 cm⁻¹ can be ascribed to overlapping C=N and C=C stretching vibrations in aromatic or nitrogen-doped structures. Additional peaks at 1500 cm⁻¹ and 1292 cm⁻¹ are assigned to C=N/C=C aromatic ring vibrations and C–N stretching modes, respectively.³⁸ These results confirm the successful formation of nitrogen-doped carbon structures in the Ni/N–C catalyst.

Fig. 3d displays the N₂ adsorption–desorption isotherms and pore size distribution profiles of the N–C support and Ni/N–C catalyst. Both samples exhibit type IV isotherms with H4-type hysteresis loops, characteristic of mesoporous materials.³⁹ Compared with the N–C support, the Ni/N–C catalyst shows higher nitrogen uptake, consistent with its increased specific surface area and pore volume, as summarized in Table S1. This enhancement may be attributed to structural modification of the carbon matrix during high-temperature calcination, as well as the formation of additional porosity induced by Ni nanoparticle dispersion.⁴⁰ Pore size distribution analysis (Fig. 3e) reveals that both materials contain dominant mesopores in the 2–5 nm range. Notably, the average pore diameter of the Ni/N–C catalyst is slightly smaller than that of the N–C support, suggesting that metal incorporation may lead to the generation of more finely distributed mesopores and micropores, likely due to restructuring or fragmentation of pore walls during Ni loading and thermal treatment.⁴¹ In addition, inductively coupled plasma-mass spectrometry (ICP-MS) analysis confirmed a nickel loading of 9.73 wt% for the Ni/N–C catalyst, which closely matches the theoretical value of 10.00 wt%. This result indicates that the metal incorporation process was highly efficient, with negligible Ni loss during impregnation and calcination.

Transmission electron microscopy (TEM) characterization was conducted to observe the morphological changes and Ni particle size of the Ni/N–C catalyst before and after the reaction. As shown in Fig. 4a and d, the N–C support derived from ZIF-8 templates exhibits a well-defined rhombic dodecahedral morphology with an average particle size of approximately 500 nm, consistent with previous reports.⁴² The surface appears smooth prior to the reaction, and well-dispersed Ni nanoparticles with an average diameter of 6.4 nm are clearly observed. This small particle size indicates high metal dispersion and increased density of accessible active sites, which are beneficial for initial catalytic performance.²³ The TEM images (Fig. 4e and h) of the fresh Ni/N–C catalyst show that the N–C framework retains its morphological integrity, with Ni nanoparticles remaining uniformly distributed across the surface. The Ni nanoparticles remain uniformly dispersed on the support with an average particle size of 7.9 nm, exhibiting a marginal increase compared to the fresh state, likely due to slight agglomeration. Notably, no significant carbon deposition was observed on the spent catalyst. High-resolution transmission electron microscopy (HRTEM) images revealed 0.203 nm lattice spacings corresponding to Ni (111) planes in both fresh and spent catalysts (Fig. 4c and g), consistent with the XRD data. More importantly, an interplanar spacing of 0.241 nm corresponding to the (111) plane of NiO (PDF#47-1049) can be clearly observed (Fig. 4b and f), indicating the coexistence of Ni and NiO particles on the Ni/N–C catalyst. These observations collectively indicate that the Ni/N–C catalyst maintains excellent structural stability and metal dispersion under reaction conditions.

Fig. 4 TEM, HRTEM, and particle size distribution histograms of the Ni/N–C catalyst: (a–d) fresh; (e–h) spent.

X-ray photoelectron spectroscopy (XPS) was employed to investigate the surface chemical states and elemental composition of the Ni/N–C catalyst (Fig. 5). In the Ni 2p spectrum (Fig. 5a), the peaks located at 852.4 and 854.8 eV correspond to Ni⁰ and Ni²⁺ species in the Ni 2p_3/2 region, respectively, and a prominent satellite peak at 860.4 eV further supports the assignment of Ni²⁺.⁴³ Additionally, peaks at 871.7 and 878.2 eV are attributed to Ni⁰ in the Ni 2p_1/2 region and its associated satellite feature, confirming the coexistence of both oxidized and metallic nickel species.⁴⁴ This observation is consistent with the HRTEM results shown in Fig. 4. Furthermore, the proportions of Ni⁰ and Ni²⁺ are 55.81% and 44.19% (Table 1), respectively, corresponding to a Ni⁰/Ni²⁺ ratio of 1.26. The relatively high content of metallic Ni⁰ suggests that the catalyst possesses good reducibility and abundant exposed metallic sites, while the remaining NiO species may provide additional basic sites that could influence the adsorption and activation of reactant molecules in subsequent reactions.^45,46 The O 1s spectrum (Fig. 5b) was deconvoluted into three distinct components: lattice oxygen (O_I, 528.8 eV), chemisorbed oxygen (O_II, 530.6 eV), and physisorbed oxygen (O_III, 531.7 eV).⁴⁷ The high proportion of chemisorbed oxygen indicates the presence of abundant surface-active oxygen species (Table 1), which are believed to play a key role in enhancing CH₄ and CO₂ adsorption and sustaining redox cycles during the reforming process.⁴⁶ In the N 1s spectrum (Fig. 5c), three types of nitrogen species were identified: pyridinic N (398.7 eV), pyrrolic N (400.2 eV), and graphitic N (401.6 eV).⁴⁸ Nitrogen functionalities can modulate the electronic environment and promote metal anchoring and dispersion, contributing to improved catalytic performance.²⁸ The C 1s spectrum (Fig. 5d) shows characteristic peaks at 285.1 eV (C–C), 286.1 eV (C–N), and 288.8 eV (C=O), confirming the presence of various oxygen- and nitrogen-containing surface functionalities.⁴⁸

Fig. 5 XPS spectra of the Ni/N–C catalyst: (a) Ni 2p, (b) O 1s, (c) N 1s, and (d) C 1s.

Table 1 Ni 2p and O 1s spectral results of the Ni/N–C catalyst

Catalyst	Ni^0 (%)	Ni^2+ (%)	Ni^0/Ni^2+	OI (%)	OII (%)	OIII (%)
Ni/N–C	55.81	44.19	1.26	7.52	49.46	43.02

---

To investigate the reducibility of the catalyst and the interaction between the metal and support, hydrogen-temperature programmed reduction (H₂-TPR) characterization was performed as shown in Fig. 6a. The N–C support shows a broad hydrogen consumption signal from 200 to 550 °C, with a main peak at 376 °C attributed to the reduction of nitrogen-containing surface groups.⁴⁹ In comparison, the Ni/N–C catalyst exhibits multiple distinct reduction peaks. The first peak at 303 °C is mainly assigned to the reduction of small NiO particles with weak interactions with the N-doped carbon support.⁵⁰ However, this peak overlaps with the reduction region of N–C, suggesting possible contributions from residual or Ni-perturbed nitrogen groups. The second peak at 458 °C likely originates from nitrogen functionalities that are partially covered or electronically influenced by nearby Ni/NiO species, leading to delayed reduction.⁵¹ The peak at 523 °C is attributed to NiO particles embedded within the mesoporous structure, where strong metal–support interactions with nitrogen dopants increase the reduction temperature by facilitating electron transfer from the support to NiO.³³ Finally, the broad high-temperature peak at 675 °C corresponds to highly dispersed NiO species strongly interacting with the support, possibly through Ni–N coordination. These strong interactions stabilize the Ni²⁺ state and hinder reduction.⁵² Overall, the reduction behavior of Ni/N–C reflects a complex interplay between Ni/NiO dispersion, metal–support interaction strength, and the influence of nitrogen-doped sites in the carbon matrix.⁵³

Fig. 6 (a) H₂-TPR, (b) NH₃-TPD, (c) CO₂-TPD, and (d) TGA profiles of the fresh and spent Ni/N–C catalyst.

Ammonia-temperature programmed desorption (NH₃-TPD) was used to probe the surface acidic characteristics of the N–C support and Ni/N–C catalyst (Fig. 6b), with semi-quantitative results based on peak areas summarized in Table S2. The desorption below 150 °C is attributed to physical adsorption.⁵⁴ A weak desorption peak between 150 and 200 °C indicates the presence of weak acid sites.⁵⁵ The N–C support exhibited a strong acid desorption peak at 485 °C, which may be attributed to Lewis acid sites associated with electronic vacancies in the mesoporous carbon. In contrast, the Ni/N–C catalyst showed a prominent medium-strength acid peak at 297 °C, likely originating from unsaturated Ni centers formed through interaction with nitrogen dopants.⁵⁶ Semi-quantitative analysis of the peak areas indicates that after Ni loading, strong acid sites were significantly reduced or eliminated, while the number of medium-strength acid sites increased notably. This redistribution of acid sites is considered beneficial, as excessively strong acidity may hinder CO₂ adsorption and promote carbon deposition.⁵⁷ The increase in medium-strength acid sites contributes to enhanced catalytic activity and improved resistance to carbon formation.

Characterization of the basic sites on the material surface was performed using carbon dioxide-temperature programmed desorption (CO₂-TPD) analysis (Fig. 6c), as shown in Table S3. The desorption peaks observed at 128 °C for the N–C support and 111 °C for the Ni/N–C catalyst are attributed to the physical adsorption of CO₂. In both samples, CO₂ desorption above 600 °C is primarily associated with the thermal degradation of the carbon framework.⁵⁸ Notably, the Ni/N–C catalyst exhibits a prominent desorption peak at 554 °C, which indicates the presence of abundant strong basic sites. This enhancement in basicity is attributed to the introduction of Ni, particularly the formation of NiO nanoparticles that contribute significantly to the generation of strong basic sites on the catalyst surface.⁴⁰ The increased density of basic sites improves the adsorption and activation of CO₂, thereby facilitating its conversion.⁵⁹

Thermogravimetric analysis (TGA) analysis of fresh and spent catalysts revealed that the fresh catalyst exhibited a 12.1% mass loss at 0–130 °C (Fig. 6d), attributed to desorption of guest methanol molecules from N–C crystal cavities and evaporation of moisture. Notably, the spent catalyst showed significantly reduced mass loss in this range, indicating effective removal of surface water and solvent during the reaction.⁶⁰ A substantial 51.7% mass loss occurred between 300 and 600 °C for the fresh catalyst, corresponding to oxygen-containing functional group decomposition in Ni/N–C.⁶¹ Remarkably, the total mass loss of the spent catalyst increased by 4.6% compared to the fresh catalyst, primarily due to carbon deposition during the catalytic process.⁶²

3.3. Catalytic performances

3.3.1. Synergistic interactions. The reforming reaction performance was investigated in distinct modes: plasma-alone and plasma–catalyst systems. As shown in Fig. 7a, the plasma-alone mode yielded CH₄ and CO₂ conversions of 11.1% and 8.7%, respectively, due to activation by high-energy electrons generated in the discharge.¹⁴ When plasma was coupled with the N–C support, the conversion rate of CH₄ increased, while the conversion rate of CO₂ remained almost unchanged. Notably, a significant enhancement was observed in the plasma–Ni/N–C system, where CH₄ and CO₂ conversions reached 16.2% and 13.3%, respectively—approximately 1.5 times higher than in the plasma-alone mode. The energy efficiency also improved from 0.15 to 0.25 mmol kJ⁻¹. Fig. 7b compares the product selectivities among the three systems. The plasma–N–C system increased the CO selectivity, but the H₂ selectivity remained almost unchanged. Meanwhile, the plasma–Ni/N–C system achieved markedly higher selectivities for both CO (59.2%) and H₂ (45.2%) compared to the plasma-alone system (44.5% and 41.1%, respectively). Moreover, the total selectivity for C₂–C₃ hydrocarbons also improved in the plasma–Ni/N–C system. The catalyst integration not only enhances conversion but also steers product distribution toward syngas and light hydrocarbons.⁶³

Fig. 7 (a) Conversion rates of CH₄ and CO₂; (b) selectivity to gaseous products; (c) component-specific selectivity to C₂–C₃ hydrocarbons (discharge power: 25 W; gas flow rate: 60 mL min⁻¹; CH₄/CO₂ molar ratio: 1 : 1).

As shown in Fig. 7c, C₂H₆ was the dominant C₂–C₃ hydrocarbon product across all modes, accounting for approximately 70% of total hydrocarbon selectivity. Other products followed the order C₃H₈ > C₂H₂ > C₂H₄ > C₃H₆; among them, C₃H₆ was present only in trace amounts. In the plasma–N–C system, although the total selectivity for C₂–C₃ hydrocarbons remained nearly unchanged upon incorporating the N–C support, the addition enhanced the selectivity for C₂H₆ while reducing that for C₃H₈. In the plasma–Ni/N–C system, the selectivity toward C₂H₆ (up to 17.2%) and C₂H₄ increased, while C₃H₈ selectivity slightly decreased. This shift suggests that the catalyst not only facilitates C–C coupling reactions but also helps regulate hydrogenation pathways. The presence of well-dispersed Ni nanoparticles and nitrogen-doped carbon may weaken the over-hydrogenation of intermediates, thereby promoting the formation of light olefins and alkynes. These results confirm the role of the Ni/N–C catalyst in tuning product distribution under plasma conditions. According to the NH₃-TPD and CO₂-TPD characterization results, the N–C support has almost no basic sites for chemisorption but a large number of acidic sites, indicating that the acidic sites may enhance both the CH₄ conversion rate and C₂H₆ selectivity. Meanwhile, the Ni/N–C catalyst possesses more basic sites and medium acid sites, suggesting that the synergetic effect between basic and medium acid sites can significantly improve CO₂ conversion as well as the selectivities towards CO and H₂.

3.3.2. Influence of the gas flow rate. The effect of the gas flow rate on the performance of the plasma-coupled N–C support and plasma–Ni/N–C catalyst systems was systematically evaluated. As shown in Fig. 8a and b, increasing the gas flow rate led to a general decline in CH₄ and CO₂ conversion rates for both systems. For the Ni/N–C catalyst, the highest conversions of CH₄ (23.1%) and CO₂ (15.1%) were obtained at 40 mL min⁻¹. However, when the flow rate was increased to 80 mL min⁻¹, the conversions decreased to 13.0% and 10.3%, respectively. This trend is attributed to shortened residence time and reduced molecular collision frequency at higher flow rates, which limit reactant activation and conversion efficiency.⁶⁴ Interestingly, although the conversions decreased, the energy efficiency of CH₄ and CO₂ reforming over the Ni/N–C catalyst increased with rising flow rate, owing to the greater absolute quantity of reactants processed within the same discharge period. This indicates that higher throughput compensates for lower per-molecule conversion, resulting in a net gain in energy efficiency.⁶⁵ In contrast, the N–C support system exhibited minimal changes in energy efficiency, highlighting the superior activation capacity of the Ni-loaded catalyst.

Fig. 8 Effects of the gas flow rate on CH₄–CO₂ reforming performance: (a and b) CH₄ and CO₂ conversion rates and energy efficiency; (c and d) CO and H₂ selectivity and energy efficiency; (e) selectivity to gaseous hydrocarbon products (discharge power: 25 W; CH₄/CO₂ molar ratio: 1 : 1).

As shown in Fig. 8c and d, the selectivity for CO and H₂ decreased with increasing flow rate, dropping from 67.9% to 58.5% and from 44.6% to 40.5%, respectively. Meanwhile, the total C₂–C₃ hydrocarbon selectivity increased from 19.9% to 26.3%. Notably, C₂H₆ selectivity showed a marked rise from 14.6% to 20.0% (Fig. 8e), which can be ascribed to enhanced recombination of radicals under shorter residence times, suppressing their decomposition into CO and H₂.⁶⁴ This shift reflects the kinetic competition between hydrogenation and reforming pathways. Furthermore, regardless of the gas flow rate variations, the Ni/N–C catalyst consistently outperformed the N–C support in terms of conversion, product selectivity, and energy efficiency. These improvements are primarily due to the high density of active Ni sites and the synergistic effect between the metal and nitrogen-doped carbon matrix, which together promote efficient plasma–catalyst interactions under dynamic flow conditions.^66,67

3.3.3. Influence of the CH₄/CO₂ molar ratio. The effect of the CH₄/CO₂ molar ratio on the reforming performance of the plasma–Ni/N–C system was investigated, as shown in Fig. 9. When the CH₄/CO₂ ratio decreased from 5 : 1 to 1 : 5, both CH₄ and CO₂ conversion rates increased significantly, with CH₄ conversion rising from 7.7% to 44.1% and CO₂ conversion from 9.7% to 20.0% (Fig. 9a and b). This improvement is attributed to the higher CO₂ content, which generates more reactive oxygen species via electron impact dissociation, enhancing C–H bond cleavage in CH₄ (eqn (14)).⁶⁵ Meanwhile, the increased CH₄ conversion produces more H* radicals; H* undergoes self-recombination to form H₂, which promotes the reverse water–gas shift (RWGS) reaction (eqn (15)) and further accelerates CO₂ consumption.

Fig. 9 Effects of the CH₄/CO₂ molar ratio on reforming performance: (a and b) CH₄ and CO₂ conversion rates and energy efficiency; (c and d) CO and H₂ selectivity and energy efficiency; (e) selectivity to gaseous hydrocarbon products (discharge power: 25 W; CH₄/CO₂ molar ratio: 1 : 1).

As shown in Fig. 9c and d, CO selectivity increased markedly with decreasing CH₄/CO₂ ratio, reaching 67.6% at 1 : 5, due to the excess CO₂ available for CO formation.⁶⁵ H₂ selectivity also increased, peaking at 48.0% under CO₂-rich conditions, benefiting from enhanced hydrogen radical generation. Notably, once the CH₄/CO₂ ratio dropped below 2.5 : 1, the energy efficiencies for CH₄, CO₂, CO, and H₂ over the Ni/N-C catalyst consistently surpassed those of the N–C support, reflecting the superior reforming capability of the metal-loaded system under oxidizing conditions. In contrast, the selectivity to C₂–C₃ hydrocarbons decreased with decreasing CH₄/CO₂ ratio (Fig. 9e). At a ratio of 5 : 1, the highest selectivities were observed for C₂H₆ (25.5%), C₃H₈ (13.8%), C₂H₂ (15.8%), and C₂H₄ (10.0%), with trace amounts of C₃H₆ (2.4%). These findings indicate that hydrocarbon chain growth and radical recombination are favored under CH₄-rich conditions, whereas high CO₂ concentrations suppress hydrocarbon formation and steer product selectivity toward syngas components (CO and H₂).^68,69

CO₂ + 4H₂ → CH₄ + 2H₂O (15)

3.3.4. Distribution of liquid products. As shown in Fig. 10a, the concentrations of liquid products were analyzed after 6 hours of reaction under three conditions: plasma-only, plasma–N–C and plasma–Ni/N–C systems. Evidently, under all three systems, the concentration of CH₃OH was the highest, followed by CH₃CH₂OH, while only trace amounts of other liquid products were detected. In the plasma-alone system, the concentration of CH₃OH was 7.8 mg mL⁻¹, and that of CH₃CH₂OH was 2.0 mg mL⁻¹. After coupling with the N–C support, the concentration of alcohols further increased. In the system combining plasma with the Ni/N–C catalyst, the concentration of alcohols, particularly CH₃OH, significantly increased, with the CH₃OH concentration reaching 15.4 mg mL⁻¹—nearly twice that in the plasma-alone system. This highlights the critical role of Ni/N–C catalysis in promoting CH₃OH production. This enhancement is attributed to the activation of CH₄ by Ni⁰ species after Ni loading on the N–C support, which facilitates CH₄ dissociation to produce radicals. These radicals can then react with OH* species derived from O* and H* to form CH₃OH (), where acts as a key intermediate.^11,70 Moreover, previous studies reported that Ni²⁺ (NiO) exhibited a weak adsorption capacity for CO*, which suppressed the CO* + OH* → COOH* and pathways, thereby favoring the direct formation of CH₃OH through the coupling of and OH*.⁷¹ As shown in Fig. 10b, the selectivity for CH₃OH remains the highest among the three systems, followed by CH₃CH₂OH. However, after coupling the plasma with either the N–C support or the Ni/N–C catalyst, the overall selectivity toward liquid products decreased. This may be attributed to the enhanced conversion of CH₄ and CO₂ toward gaseous products, such as CO and C₂–C₃ hydrocarbons. Nevertheless, the total concentration still shows a significant increase.

Fig. 10 (a) Liquid product concentration distribution and (b) selectivity for plasma-only, the N–C support, and the Ni/N–C catalyst (discharge power: 25 W; gas flow rate: 60 mL min⁻¹; CH₄/CO₂ molar ratio: 1 : 1).

3.4. Stability test of the Ni/N–C catalyst

Fig. 11 shows the 48 hour stability test of the CH₄–CO₂ reforming reaction using the plasma–Ni/N–C catalyst under optimal reaction conditions. As observed in Fig. 11a, the conversion rates of CH₄ and CO₂ remained largely constant throughout the test period, with only a slight decline toward the end. Specifically, the CH₄ conversion decreased from 42.3% to 37.4% (a reduction of 4.9%), while the CO₂ conversion dropped from 19.8% to 17.1% (a reduction of 2.7%). This minor decline may be attributed to the partial coverage of active sites by trace carbon deposits formed during the reaction. Fig. 11b reveals a modest increase in the selectivity to CO and H₂ during the test. The CO selectivity increased from 66.3% to 71.5% (an increase of 5.2%), and the H₂ selectivity increased from 47.2% to 49.5% (an increase of 2.3%). Fig. 11c and d further present the stable distribution of gaseous hydrocarbons and liquid products. The selectivity to gaseous hydrocarbons remained relatively stable at below 5%, while that to liquid products was maintained within 25%, with minor fluctuations in the selectivity to individual products. Methanol was consistently the dominant liquid product, exhibiting a selectivity of approximately 16% (Fig. 11d). Overall, the results demonstrate that the Ni/N–C catalyst maintains excellent catalytic stability during extended plasma-assisted CH₄–CO₂ reforming.

Fig. 11 Stability performance of the plasma–Ni/N–C system: (a) CH₄ and CO₂ conversion rates; (b) CO and H₂ selectivity; (c) gaseous hydrocarbon selectivity; (d) liquid product selectivity (discharge power: 25 W; gas flow rate: 60 mL min⁻¹; CH₄/CO₂ molar ratio: 1 : 5).

As shown in Table 2, which compares the catalytic performance of various plasma-coupled Ni-based catalysts, the Ni/N–C catalyst developed in this study achieved CO and H₂ selectivities of 67% and 48%, respectively, in CH₄–CO₂ reforming under relatively mild conditions (10 °C). Its energy efficiency reached 0.39 mmol kJ⁻¹, ranking among the highest levels in the compared systems.^71–75 Although some catalysts achieved higher absolute conversion rates, they required higher discharge powers (50–100 W) or lower gas flow rates, limiting their energy efficiency and scalability. The Ni/N–C catalyst, while maintaining competitive syngas selectivity, exhibited superior energy efficiency, underscoring the synergistic role of N-doped carbon in promoting active site dispersion, enhancing plasma–catalyst coupling, and improving overall energy utilization efficiency.

Table 2 Comparison of syngas selectivity and energy efficiency in CH₄–CO₂ reforming using plasma-coupled Ni-based catalysts

Catalysts	Discharge power (W)	Gas flow rate (mL min^−1)	CH4/CO2 molar ratio	Conversion (%)		Selectivity (%)		Energy efficiency (mmol kJ^−1)	Ref.
				CH4	CO2	CO	H2
Ni/N–C	25	60	1 : 5	44	20	67	48	0.39	This work
NiGa/NF	25	30	1 : 1	16	9	41	37	0.12	71
Ni/SiO2	50	20	1 : 1	55	44	61	48	0.27	72
Ni/La2O3–MgAl2O4	100	20	1 : 1	86	84	49	50	0.13	73
Ni/γ-Al2O3	50	50	1 : 1	56	30	52	31	0.32	74
Co/Ni-MOF	22.5	30	1 : 1	52	34	53	47	0.13	75

---

3.5. Reaction mechanism

3.5.1. Analysis of plasma active species. The active species generated during the plasma reforming of CH₄ and CO₂ were characterized using OES across the 200–800 nm spectral range. As shown in Fig. 12a, various excited radicals were observed, including CO₂⁺, CO, CH, OH, CO₂, C₂, H_α and O. The CO₂⁺ (A²Π–X²Π) bands were detected at 287.6, 337.7, 351.3, and 369.6 nm, while the CO emission bands corresponding to the b³Σ–a³Π and B¹Σ–A¹Π transitions were observed at 287.9 nm and within the 450–570 nm range, confirming CO formation as a major pathway.¹² The C₂ (d³Π–a³Π) Swan bands appeared between 460 and 520 nm, likely arising from CO decomposition and subsequent carbon–carbon bond formation.⁷⁶ CH radicals were identified via C²Σ⁺–X²Π, B²Σ–X²Π, and A²Δ–X²Π transitions at 315.4, 391.2, and 431.2 nm, indicating hydrocarbon fragmentation and recombination.^12,19 OH radicals, attributed to the A²Σ⁺–X²Π and ³Π–³Σ transitions, were observed at 311.3 and 356.1 nm, and likely originated from trace moisture or CO₂ dissociation within the DBD reactor.^12,19 The CO₂ (¹B₂–X¹Σ⁺) bands were also present at 375.2, 426.6, and 433.7 nm, confirming incomplete CO₂ decomposition under discharge conditions.^77,78 Additionally, relatively weak intensity peaks corresponding to the H_α (3d²D–2p²P⁰) transition at 656.3 nm and the O (3s⁵S⁰–3p⁵P) transition at 777.5 nm were detected.¹² Although and cannot be directly detected by OES, the concurrent presence of CH and H_α radicals suggests the stepwise dehydrogenation of CH₄ to (x = 1, 2, 3) intermediates, as outlined in (eqn (16)–(19)).

CH* + e⁻ → C* + H* + e⁻ (19)

Fig. 12 (a) OES spectrum (discharge power: 25 W; gas flow rate: 60 mL min⁻¹; CH₄/CO₂ molar ratio: 1 : 1) and (b) non-thermal plasma-coupled Ni/N–C catalytic CH₄–CO₂ reforming reaction mechanism diagram.

3.5.2. Mechanism of the CH₄–CO₂ reforming reaction. Based on the physicochemical properties of the catalysts, the nature of active species, and the observed product distribution, a reaction mechanism for CH₄–CO₂ reforming conditions is proposed (Fig. 12b).

In the plasma-alone system, energetic electrons collide with CH₄ and CO₂ molecules, generating reactive radicals such as (x = 1–3), H*, O*, and CO*. radicals may recombine to form C₂H₆, H* forms H₂, and O* contributes to carbon oxidation into CO.⁷⁹ However, due to the absence of catalytic control, radical utilization is inefficient, resulting in low selectivity for syngas and hydrocarbons, and significant carbon accumulation on surfaces.

The introduction of the N–C support significantly alters the plasma–catalyst interface. The abundant strong acid sites facilitate CH₄ activation and dehydrogenation, increasing CH₄ conversion and promoting C–C coupling reactions. Additionally, the mesoporous architecture of N–C enriches radicals within confined channels, enhancing their self-coupling into C₂H₆ while suppressing undesired routes such as – coupling that leads to C₃H₈. However, despite containing both strong acid and some basic sites, the CO₂ conversion with N–C is only marginally higher than that of the plasma-alone system. This suggests that the basic sites in N–C are either insufficiently exposed or lack adequate strength to effectively adsorb and activate CO₂ under plasma conditions. Some O* radicals generated from CO₂ dissociation may still react with surface carbon to form CO, partially enhancing CO selectivity and mitigating coking.

The incorporation of Ni into N–C substantially enhances both CH₄ and CO₂ activation pathways. The surface of Ni/N–C is dominated by metallic Ni species, which efficiently dissociate CH₄ into and H*, accelerating conversion. At the same time, unsaturated Ni centers generate moderate acid sites that stabilize intermediates, reducing over-dehydrogenation and suppressing carbon formation. These species are more likely to recombine into C₂–C₃ hydrocarbons, with self-coupling accounting for the high selectivity to C₂H₆ over C₃H₈. Some surface C₂H₆ may further dehydrogenate to yield C₂H₄. Importantly, the addition of Ni also introduces stronger and more abundant basic sites through the formation of finely dispersed oxidized Ni species. These basic sites significantly improve CO₂ adsorption and activation, promoting the generation of reactive oxygen species (O*). These O* radicals contribute to multiple reaction steps: oxidizing carbon deposits to form CO, reacting with CH₄ to produce OH*, and facilitating the formation of oxygenates such as CH₃OH. Furthermore, lattice oxygen (O²⁻) associated with these oxygen-containing species can participate in redox cycles near metallic Ni centers, continuously converting and CO₂ into CO and H₂ while regenerating O²⁻. This dynamic oxygen cycle supports sustained syngas production and improves catalyst durability.

In summary, the cooperative interaction between metallic Ni and oxygen-containing species, together with a well-balanced distribution of acid–base functionalities, plays a pivotal role in enhancing CH₄ and CO₂ conversion, improving selectivity toward syngas and light hydrocarbons, and effectively inhibiting catalyst deactivation due to carbon deposition.

4. Conclusion

This study demonstrated the efficient reforming of CH₄ and CO₂ into syngas using a NTP-catalytic system with a Ni/N–C catalyst. Under the optimal reaction conditions, including a discharge power of 25 W, a total gas flow rate of 60 mL min⁻¹, and a CH₄/CO₂ molar ratio of 1 : 5, the CH₄ and CO₂ conversions reached 44.1% and 20.0%, respectively. The corresponding selectivities for CO and H₂ were 67.6% and 48.1%, and the energy efficiency reached 0.39 mmol kJ⁻¹. Furthermore, the Ni/N–C catalyst maintained excellent stability during the prolonged 48 hour catalytic performance test. Catalyst characterization and performance results confirm that the mesoporous N–C support possesses a high specific surface area and a well-connected pore network, providing abundant sites for CH₄/CO₂ adsorption and activation while facilitating efficient mass transfer of reactive species to active Ni centers. Pyridinic nitrogen groups serve as robust anchoring sites that strengthen Ni–support interactions, effectively preventing nanoparticle agglomeration and maintaining uniform Ni dispersion. Metallic Ni species act as the key active centers for CH₄ dissociation, while oxygen-containing Ni species introduce basic sites that promote CO₂ adsorption and activation. Moreover, moderate acidity at Ni–N/C interfacial regions helps stabilize intermediates, suppressing over-dehydrogenation and favoring selective C–C coupling. The synergy between hierarchical porosity, N-doping, acid–base functionality, and plasma-generated reactive species enables efficient and selective CH₄–CO₂ conversion, providing valuable guidance for the rational design of next-generation plasma-catalytic systems.

Author contributions

Tian Chang: methodology, conceptualization, supervision, funding acquisition, project administration, writing – original draft, writing – review & editing. Zhao Yang: investigation, data curation, manuscript writing. Zuotong Zhao: investigation. Xuanchen Chang: writing – review & editing. Chuanyi Wang: writing – review & editing.

Conflicts of interest

There are no conflicts to declare.

Data availability

All data generated in this study are included in the article and its supplementary information (SI).

Supplementary information: characterization methods, discharge images, N₂ adsorption–desorption and ICP-MS results, NH₃-TPD and CO₂-TPD analysis data. See DOI: https://doi.org/10.1039/d5cy00936g.

Acknowledgements

This work was financially supported by the National Natural Science Foundation of China (22208204), the China Postdoctoral Science Foundation (2022M722015), the State Key Laboratory of Electrical Insulation and Power Equipment (EIPE23213), the Foundation of State Key Laboratory of High-efficiency Utilization of Coal and Green Chemical Engineering (2022-K45), the Innovation Capability Support Program of Shaanxi (2024WZ-YBXM-18), the Postdoctoral Research Project of Shaanxi Province, and the Graduate Scientific Research Foundation of School of Environmental Science & Engineering, SUST.

References

1. D. Welsby, J. Price, S. Pye and P. Ekins, Nature, 2021, 597, 230–234.
2. H. Adun, J. D. Ampah, O. Bamisile and Y. Hu, Sustain. Prod. Consum., 2024, 45, 386–407.
3. R. P. Madruga, Science, 2021, 374, 511.
4. K. Arshad, N. Hussain, M. H. Ashraf and M. Z. Saleem, Sci. Total Environ., 2024, 928, 172370.
5. H. Wang, G. Cui, H. Lu, Z. Li, L. Wang, H. Meng, J. Li, H. Yan, Y. Yang and M. Wei, Nat. Commun., 2024, 15, 3765.
6. S. Zhang, M. Ying, J. Yu, W. Zhan, L. Wang, Y. Guo and Y. Guo, Appl. Catal., B, 2021, 291, 120074.
7. E. Akbari, S. M. Alavi and M. Rezaei, J. CO2 Util., 2018, 24, 128–138.
8. B. Abdullah, N. A. Abd Ghani and D.-V. N. Vo, J. Cleaner Prod., 2017, 162, 170–185.
9. T. Gupta and H. Ayan, Appl. Sci., 2019, 9, 3548.
10. Y. Zhou, R. Chu, L. Fan, J. Zhao, W. Li, X. Jiang, X. Meng, Y. Li, S. Yu and Y. Wan, Sci. Total Environ., 2023, 866, 161453.
11. Y. Zhou, L. Zhu, B. Yang, L. Fan, X. Meng, R. Chu, X. Jiang, P. Li, W. Li and H. Chen, J. Hazard. Mater., 2024, 461, 132698.
12. Y. Wang, Y. Chen, J. Harding, H. He, A. Bogaerts and X. Tu, Chem. Eng. J., 2022, 450, 137860.
13. L. Wang, Y. Yi, C. Wu, H. Guo and X. Tu, Angew. Chem., Int. Ed., 2017, 56, 13679–13683.
14. S. Baig and B. Sajjadi, J. Energy Chem., 2024, 97, 265–301.
15. I. Yentekakis, P. Panagiotopoulou and G. Artemakis, Appl. Catal., B, 2021, 296, 120210.
16. D. He, S. Wu, X. Cao, D. Chen, L. Zhang, Y. Zhang and Y. Luo, Appl. Catal., B, 2024, 346, 123728.
17. O. V. Golubev, P. S. Il'chuk and A. L. Maximov, Pet. Chem., 2024, 64, 435–449.
18. 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.
19. D. Ray, P. Reddy and C. Subrahmanyam, Catal. Today, 2018, 309, 212–218.
20. J. Liu, J. Meng, C. Zou, X. Wang, C. Bu, J. Zhang, C. Liu, X. Cao, L. Liu and H. Xie, J. Energy Inst., 2025, 120, 102050.
21. A. Khoja, M. Tahir and N. Amin, Fuel Process. Technol., 2018, 178, 166–179.
22. T. Abbas, H. S. M. Yahya and N. A. S. Amin, Fuel Process. Technol., 2023, 248, 107836.
23. Y. Diao, H. Wang, B. Chen, X. Zhang and C. Shi, Appl. Catal., B, 2023, 330, 122573.
24. J. Xu, B. Yang, K. Chen, Q. Zhang, P. Xia, L. Jiang and F. Guo, Microporous Mesoporous Mater., 2024, 364, 112847.
25. W. Xu, M. Dong, L. Di and X. Zhang, Nanomaterials, 2019, 9, 1432.
26. B. An, Y. Ma, X. Han, M. Schröder and S. Yang, Acc. Mater. Res., 2024, 6, 77–88.
27. Y. Guo, L. Feng, C. Wu, X. Wang and X. Zhang, J. Catal., 2020, 390, 213–223.
28. Y. Li, X. Cai, S. Chen, H. Zhang, K. H. L. Zhang, J. Hong, B. Chen, D. H. Kuo and W. Wang, ChemSusChem, 2018, 11, 1040–1047.
29. Q. Liang, W. Li, L. Xie, Y. He, B. Qiu, H. Zeng, S. Zhou, J. Zeng, T. Liu, M. Yan, K. Liang, O. Terasaki, L. Jiang and B. Kong, Nano Lett., 2022, 22, 2889–2897.
30. D. Mei, M. Sun, S. Liu, P. Zhang, Z. Fang and X. Tu, J. CO2 Util., 2023, 67, 102307.
31. D. Mei, G. Duan, J. Fu, S. Liu, R. Zhou, R. Zhou, Z. Fang, P. Cullen and K. Ostrikov, J. CO2 Util., 2021, 53, 101703.
32. Z. Tai, M. Shi, S. Chong, Y. Chen, C. Shu, X. Dai, Q. Tan and Y. Liu, J. Alloys Compd., 2019, 800, 1–7.
33. Y. Diao, H. Wang, B. Chen, L. Wang, X. Zhang and C. Shi, Fuel Process. Technol., 2023, 243, 107676.
34. R. Zhou, M. Mohamedali, Y. Ren, Q. Lu and N. Mahinpey, Appl. Catal., B, 2022, 316, 121696.
35. B. Singh, S. Samal, S. Nayak, S. Majhi, L. Besra and S. Bhattacharjee, Surf. Coat. Technol., 2011, 206, 1319–1326.
36. S. Dai, Z. Liu, B. Zhao, J. Zeng, H. Hu, Q. Zhang, D. Chen, C. Qu, D. Dang and M. Liu, J. Power Sources, 2018, 387, 43–48.
37. G. Varadwaj, O. Oyetade, S. Rana, B. Martincigh, S. Jonnalagadda and V. Nyamori, ACS Appl. Mater. Interfaces, 2017, 9, 17291–17306.
38. S. Jafari, F. Ghorbani-Shahna, A. Bahrami and H. Kazemian, Appl. Sci., 2018, 8, 310.
39. Z. ALOthman, Materials, 2012, 5, 2874–2902.
40. X. Cao, J. Xia, X. Meng, J. Xu, Q. Liu and Z. Wang, Adv. Funct. Mater., 2019, 29, 1902237.
41. M. Hossain, M. Ahasan and R. Wang, Chem. Eng. J., 2024, 496, 154193.
42. B. Hatamluyi, S. Sadeghzadeh, M. Rezayi and S. Sany, Sens. Actuators, B, 2023, 381, 133424.
43. J. Diao, T. Zhang, Z. Xu and G. Guo, Chem. Eng. J., 2023, 467, 143271.
44. O. Daoura, N. El Hassan, M. Boutros, S. Casale, P. Massiani and F. Launay, ACS Appl. Nano Mater., 2022, 5, 18048–18059.
45. Y. Diao, X. Zhang, Y. Liu, B. Chen, G. Wu and C. Shi, Appl. Catal., B, 2022, 301, 120779.
46. Y. Wang, L. Li, G. Li, Q. Zhao, X. s. Wu, Y. Wang, Y. Sun and C. Hu, ACS Catal., 2023, 13, 6486–6496.
47. N. Abdullah, N. Ainirazali, C. Chong, H. Razak, H. Setiabudi, A. Jalil and D. Vo, Int. J. Hydrogen Energy, 2020, 45, 18426–18439.
48. M. Yi, C. Zhang, C. Cao, C. Xu, B. Sa, D. Cai and H. Zhan, Inorg. Chem., 2019, 58, 3916–3924.
49. S. Liu, H. Yang, X. Huang, L. Liu, W. Cai, J. Gao, X. Li, T. Zhang, Y. Huang and B. Liu, Adv. Funct. Mater., 2018, 28, 1800499.
50. T. Chang, T. Zhang, Y. Wang, A. Labidi, K. Leus, N. De Geyter, R. Morent and C. Wang, Chem. Eng. J., 2025, 520, 166113.
51. Y. Sun, G. Zhang, Y. Xu and R. Zhang, Int. J. Hydrogen Energy, 2019, 44, 16424–16435.
52. M. W. Dlamini, T. N. Phaahlamohlaka, D. O. Kumi, R. Forbes, L. L. Jewell and N. J. Coville, Catal. Today, 2020, 342, 99–110.
53. F. Ju, M. Wang, H. Luan, P. Du, Z. Tang and H. Ling, RSC Adv., 2018, 8, 33354–33360.
54. Y. Chen, J. Li, W. Teng, W. Liu, S. Ren, J. Yang and Q. Liu, J. Environ. Chem. Eng., 2023, 11, 110787.
55. S. Zhang, Q. Cao, Y. Meng and D. Xia, Appl. Surf. Sci., 2025, 686, 162216.
56. L. Wang, L. Fan, Y. Wang, Q. Chen, Y. Zhu and Y. Yi, Appl. Catal., B, 2024, 350, 123938.
57. F. Al-Mubaddel, R. Kumar, M. Sofiu, F. Frusteri, A. Ibrahim, V. Srivastava, S. Kasim, A. Fakeeha, A. Abasaeed, A. Osman and A. Al-Fatesh, Int. J. Hydrogen Energy, 2021, 46, 14225–14235.
58. S. Zhang, X. Zhang, K. Wang, C. Li, Q. Ma, T. Zhao, X. Gao and J. Zhang, J. Environ. Chem. Eng., 2025, 13, 116334.
59. M. Zhang, Y. Gao, Y. Mao, W. Wang, J. Sun, Z. Song, J. Sun and X. Zhao, Chem. Eng. J., 2023, 451, 138616.
60. J. T. Li and S. Xia, Catal. Surv. Asia, 2024, 28, 269–282.
61. M. Jiang, X. Cao, D. Zhu, Y. Duan and J. Zhang, Electrochim. Acta, 2016, 196, 699–707.
62. J. Lu, J. Wu, Y. Jiang, P. Tan, L. Zhang, Y. Lei, X. Liu and L. Sun, Angew. Chem., Int. Ed., 2020, 59, 6428–6434.
63. D. Mei, X. Shen, S. Liu, R. Zhou, X. Yuan, Z. Rao, Y. Sun, Z. Fang, X. Du, Y. Zhou and X. Tu, Chem. Eng. J., 2023, 462, 142044.
64. B. Wanten, S. Maerivoet, C. Vantomme, J. Slaets, G. Trenchev and A. Bogaerts, J. CO2 Util., 2022, 56, 101869.
65. D. Mei, P. Zhang, G. Duan, S. Liu, Y. Zhou, Z. Fang and X. Tu, J. CO2 Util., 2022, 62, 102073.
66. J. Wang, Y. Fu, W. Kong, F. Jin, J. Bai, J. Zhang and Y. Sun, Appl. Catal., B, 2021, 282, 119546.
67. M. Hu, W. Deng, M. Zhu, Y. Su, L. Wang and G. Chen, Chem. Eng. J., 2024, 499, 156437.
68. S. Li, J. Sun, Y. Gorbanev, K. Van't Veer, B. Loenders, Y. Yi, T. Kenis, Q. Chen and A. Bogaerts, ACS Sustainable Chem. Eng., 2023, 11, 15373–15384.
69. C. Qi, Y. Bi, H. Yu, H. Zhang, Q. Zhang, X. Wang, M. Fan, T. Xing, M. Wang, M. Wu and W. Wu, Ind. Eng. Chem. Res., 2024, 63, 9576–9583.
70. P. Chawdhury, Y. Wang, D. Ray, S. Mathieu, N. Wang, J. Harding, F. Bin, X. Tu and C. Subrahmanyam, Appl. Catal., B, 2021, 284, 119735.
71. J. Li, L. Dou, Y. Gao, X. Hei, F. Yu and T. Shao, J. CO2 Util., 2021, 52, 101675.
72. J. Wang, K. Zhang, M. Mertens, A. Bogaerts and V. Meynen, Appl. Catal., B, 2023, 337, 122977.
73. A. Khoja, M. Tahir and N. Amin, Energy Fuels, 2019, 33, 11630–11647.
74. Y. Zeng, X. Zhu, D. Mei, B. Ashford and X. Tu, Catal. Today, 2015, 256, 80–87.
75. K. Zheng, X. Gao, Y. Xie, Z. He, Y. Ma, S. Hou, D. Su and X. Ma, J. Colloid Interface Sci., 2025, 683, 564–573.
76. Q. Lu, W. Lei, W. Yue, W. Huang, Y. Dong, W. Yan, Y. Liu, Y. Chen and Y. Zhao, Fuel, 2023, 344, 128041.
77. X. Wang, S. Xu, W. Yang, X. Fan, Q. Pan and H. Chen, Carbon Capture Sci. Technol., 2022, 5, 100067.
78. P. Reyes, A. Gómez, J. Vergara, H. Martínez and C. Torres, Rev. Mex. Fis., 2017, 63, 363–371.
79. T. Chang, X. Chang, Z. Zhao, C. Ma, X. Zhao, Y. Du, A. Patrocinio, A. Nikiforov and C. Wang, J. Environ. Chem. Eng., 2025, 13, 117074.

---