Sustainable Biogas-to-Syngas Catalytic Dry Reforming of Methane (DRM) Using a Novel Fleece Reactor

Abstract

Driven by the growing global energy demand and the pursuit of carbon utilization goals, dry reforming of methane (DRM) has attracted considerable attention for its ability to convert CO₂ and CH₄ into syngas. Biogas, an eco-friendly product of processes such as anaerobic digestion, is primarily composed of CO₂ and CH₄ and ideally meets the feedstock requirements for DRM. In practice, biogas is generated via anaerobic digestion of livestock manure and other organic waste, providing a stable and sustainable source for the DRM reaction and thus enabling waste valorization. Supported Ni⁰ catalysts have become a research focus in this field due to their high catalytic activity and moderate cost. Conventional particulate Ni⁰ catalysts, however, are prone to carbon coking in fixed-bed applications and are difficult to effectively recover and regenerate after the reaction; thus, they are often being discarded, leading to resource waste and environmental burden. To address these issues, this study has designed a novel metal-sintered fleece catalyst support and developed a corresponding reactor. The effects of the catalyst preparation method, activation conditions, and the support structure on DRM performance have been systematically investigated. The spent Ni-based catalyst could be regenerated via calcination to restore catalytic activity and enable multiple cycles of use, significantly extending the catalyst’s lifespan and offering both economic and environmental benefits. Experimental results have demonstrated that the reactor achieved a conversion rate exceeding 80% with near-complete product selectivity.

1. Introduction

1.1. Feedstock Sources

In the operating steel mills, mineral processing plants, and coal-fired power plants, carbon dioxide obtained through carbon capture and storage technology needs to be effectively utilized [1,2,3]. At the same time, the supply of “green methane”, such as methane from biomass [4,5] or from renewable feedstock sources [6,7]. has become increasingly important. The combination of captured carbon dioxide and green methane can produce syngas (CO + H₂) through a process of dry reforming of methane (DRM), which is a key precursor for the synthesis of important oxygenated chemicals, such as methanol and dimethyl ether [8,9,10], or for fuel-alternative compounds by Fischer–Tropsch synthesis [11]. In addition, anaerobic digestion processes in the fields of sewage treatment, livestock and poultry manure, and landfills, among others, will produce a large amount of biogas, which is itself a mixed gas of carbon dioxide and methane. Although the specific composition of biogas varies with its different digested resources, it provides an off-the-shelf renewable pathway for the simultaneous provision of CO₂ and CH₄, thereby coupling waste treatment with the production of high-value-added chemicals and further promoting the development of a circular carbon economy.

1.2. Dry Reforming of Methane (DRM)

DRM, as an environmentally friendly catalytic process, can convert two major greenhouse gases, CH₄ and CO₂, into syngas, a mixture of H₂ and CO, which is an attractive and valuable gaseous feedstock for subsequent organic synthesis reactions [12,13,14]. It can synthesize oxygenated hydrocarbons, such as methanol, ethylene glycol, and dimethyl ether, as well as fuel-range hydrocarbons by the Fischer–Tropsch synthesis. DRM is a strongly endothermic reaction, so the highest conversion value can be achieved at high temperatures [15,16,17]. The endothermic and catalytic reaction is shown in Equation (1).

$${\mathrm{C}\mathrm{H}}_4+{\mathrm{C}\mathrm{O}}_2\leftrightarrow 2\mathrm{C}\mathrm{O}+2{\mathrm{H}}_2∆{\mathrm{H}}_{298\mathrm{K}}^0=247 \mathrm{k}\mathrm{J}/\mathrm{m}\mathrm{o}\mathrm{l}$$

(1)

$∆{\mathrm{H}}_{298\mathrm{K}}^0$ is the reaction enthalpy at 298 K.

Despite its environmental advantages, the catalysts used in DRM tend to deactivate due to coke formation caused by three side-reactions. Firstly, it is caused by methane decomposition, as shown in Equation (2), which occurs at a high temperature. Secondly, the CO disproportionation reaction is based on the exothermic Duja Equation (3), which occurs at lower temperatures only. Finally, the reverse water–gas shift Equation (4) proceeds simultaneously at moderate temperatures, resulting in a syngas ratio lower than the theoretical unity.

$${\mathrm{C}\mathrm{H}}_4\to \mathrm{C}+2{\mathrm{H}}_2∆{\mathrm{H}}_{298\mathrm{K}}^0=75.0 \mathrm{k}\mathrm{J}/\mathrm{m}\mathrm{o}\mathrm{l}$$

(2)

$$2\mathrm{C}\mathrm{O}\to \mathrm{C}+{\mathrm{C}\mathrm{O}}_2∆{\mathrm{H}}_{298\mathrm{K}}^0=-172.0 \mathrm{k}\mathrm{J}/\mathrm{m}\mathrm{o}\mathrm{l}$$

(3)

$${\mathrm{C}\mathrm{O}}_2+{\mathrm{H}}_2\to \mathrm{C}\mathrm{O}+{\mathrm{H}}_2\mathrm{O}∆{\mathrm{H}}_{298\mathrm{K}}^0=41.2 \mathrm{k}\mathrm{J}/\mathrm{m}\mathrm{o}\mathrm{l}$$

(4)

1.3. Objectives

The present research investigates the catalytic DRM concept and introduces a novel catalyst carrier. The objectives are fivefold:

• It examines the coating of Ni⁰-catalysts on a metal fleece candle to counteract fixed DRM problems;
• It assesses the reaction yields and efficiencies in long duration experiments;
• It demonstrates the quality of the produced syngas;
• It assesses the operational stability of the catalyst/fleece system and its easy regeneration (if coking occurs);
• It describes the possiblescalability and energy consumption.

2. Materials and Methods

2.1. The Sintered Metal Fleece Reactor

The stainless-steel sintered fleece candles were supplied by Boyue Filtration Equipment Co., Ltd., Henan, China. Details of the candles’ geometry are given in Supplementary Materials Figure S1. The Ni⁰-loaded fleece candle was centrally mounted in the reactor housing (Shanxi, China), and the required auxiliaries (gas feed (Beijing, China), T and P probes (Shanghai, China), electrical heating (Shanxi, China) and analytical equipment (Beijing, China)) were installed. Figure 1 illustrates the experimental layout. Figure 2 shows the physical diagrams of the experimental setup and the sintered fleece.

Figure 1. Layout of the experimental DRM fleece reactor. (1) CH₄ and CO₂ cylinders (Beijing, China) with relief valve for pressure, P, and mass flow meter, F; (2) N₂ cylinder (Beijing, China) with identical fittings; (3) Feed gas preheater (Shanxi, China); (4) temperature T controlled reactor (Shanxi, China) with (5) sintered metal fleece candle; (6) cooler (Shanxi, China); (7) vent (Beijing, China); and (8) GC (Beijing, China).

Figure 2. Photographically illustrates of (a) the overall setup; and (b) the fleece candle.

The fleece was fabricated from AISI 304 sintered stainless-steel mesh, with a candle of 50 mm outer diameter and 295 mm active length, fixed within a concentric housing of 76 mm inner diameter. The feed gas can be distributed uniformly to the outer tube and penetrate through the fleece pores. Table 1 lists the relevant properties of the selected fleece. The 1.7 mm thick metal fleece comprised several fiber layers stacked in a cell geometry, as illustrated in Figure 3. The metal fibers had a density of 8320 kg/m³. Other physical properties are outlined in Table 1.

Table 1. Fleece characteristics.

Key Fleece Material Properties	Value
$\mathrm{Thickness}\mathrm{of}\mathrm{the}\mathrm{fleece},\mathsf{\mu }\mathrm{m}$	1700
$\mathrm{Fiber}\mathrm{diameter},\mathsf{\mu }\mathrm{m}$	25
Fleece porosity, (-)	0.79
Amount of fibers in a horizontal fleece section, Nh	68
Amount of fibers along the candle height, Nl	11,800
Total fiber number, NT	802,400
$\mathrm{Layer}\mathrm{spacing}\mathrm{of}\mathrm{two}\mathrm{fibers},\mathsf{\mu }\mathrm{m}$	10
Fleece candle weight, g	400
Melting range, °C	1400–1450
Tensile strength, MPa	500–600
Elastic modulus, Gpa	193
Thermal conductivity, W/m	16.3 (at 100 °C)
Hardness, Brinell	≤215
Cr (wt%)	19.2
Ni (wt%)	9.5
C (wt%)	0.075
Mn (wt%)	≤2.0
Si (wt%)	≤1.0
Oxidative resistance	Excellent till 925 °C
Weight of Ni loading, g	2–5
Wash-coat thickness of Ni coating, $\mathsf{\mu }\mathrm{m}$	0.0052–0.013
BET (non-porous fleece), m2/g	0.024
BET (after etching), m2/g	0.52
Ni application method	Spray

Figure 3. Cell geometry of the fleece.

The wash-coat thickness on the fleece fibers is calculated from the catalyst weight, the specific surface area of the fleece, and the loaded weight and density of Ni.

At set flow rates of N₂ (inert carrier), CH₄ and CO₂ are mixed during electrical preheating, and the reaction temperature is set and regulated by K-type thermocouple (Shanghai, China) monitoring. All safety measures were taken to avoid gas leakage, and leak testing was performed by a pressure-hold measurement. Several thermocouples were fixed within and onto the fleece reactor. The flow rate of CH₄ and CO₂ was 25 mL/min each, although lower rates were also tested for the kinetic study. N₂ was only used as a gas at the start-up and stop procedures. The temperature setpoints were 650, 700, and 750 °C. The DRM was started once a stable reaction temperature was achieved. Samples were collected at regular intervals and analyzed according to the tools and methods in Supplementary Materials Table S1. The catalyst was reduced to Ni⁰ before the DRM reaction by H₂ for 3 h at 25 mL/min and 700 °C. The CH₄ and CO₂ conversions can be defined by their respective inlet (feed) and outlet concentrations:

$${\mathrm{C}\mathrm{H}}_4\mathrm{c}\mathrm{o}\mathrm{n}\mathrm{v}\mathrm{e}\mathrm{r}\mathrm{s}\mathrm{i}\mathrm{o}\mathrm{n}={\displaystyle \frac{{\mathrm{C}\mathrm{H}}_{4\mathrm{i}\mathrm{n}}-{\mathrm{C}\mathrm{H}}_{4\mathrm{o}\mathrm{u}\mathrm{t}}}{{\mathrm{C}\mathrm{H}}_{4\mathrm{i}\mathrm{n}}}}\times 100\%$$

(5)

$${\mathrm{C}\mathrm{O}}_2\mathrm{c}\mathrm{o}\mathrm{n}\mathrm{v}\mathrm{e}\mathrm{r}\mathrm{s}\mathrm{i}\mathrm{o}\mathrm{n}={\displaystyle \frac{{\mathrm{C}\mathrm{O}}_{2\mathrm{i}\mathrm{n}}-{\mathrm{C}\mathrm{O}}_{2\mathrm{o}\mathrm{u}\mathrm{t}}}{{\mathrm{C}\mathrm{O}}_{2\mathrm{i}\mathrm{n}}}}\times 100\%$$

(6)

and the H₂ and CO yields can be given by:

$${\mathrm{H}}_2\mathrm{y}\mathrm{i}\mathrm{e}\mathrm{l}\mathrm{d}={\displaystyle \frac{{\mathrm{H}}_{2\mathrm{o}\mathrm{u}\mathrm{t}}}{{2\times \mathrm{C}\mathrm{H}}_{4\mathrm{i}\mathrm{n}}}}\times 100\%$$

(7)

$$\mathrm{C}\mathrm{O}\mathrm{y}\mathrm{i}\mathrm{e}\mathrm{l}\mathrm{d}={\displaystyle \frac{{\mathrm{C}\mathrm{O}}_{\mathrm{o}\mathrm{u}\mathrm{t}}}{{\mathrm{C}\mathrm{H}}_{4\mathrm{i}\mathrm{n}}+{\mathrm{C}\mathrm{O}}_{2\mathrm{i}\mathrm{n}}}}\times 100\%$$

(8)

2.2. Coating of the Fleece Reactor with Catalyst

In previous investigations, wet-spray impregnation or sol–gel deposition of the metal catalyst active species onto inert supports of Al₂O₃, SiO₂, MgO, and others were mainly applied in the fixed-bed reactor. The results using Ni/ZSM-5 [18], Ni-Al₂O₃ [19,20], and other Ni-based catalysts [21,22,23], as well as the addition of Cerium [24,25,26] and other support-Ni combinations [27,28,29,30,31], were reported. The conversion of CH₄ and CO₂ ranged from 12 to 94% and 30 to 93%, respectively. With over 125 publications since 2017, the literature has been important, but not all process data have been supplied. Despite being an important operational parameter, the possible catalyst deactivation due to carbon deposit has rarely been reported with quantitative data. Ni has been the preferred active species, while MgO or CaO have been widely used as fixed-bed supports—although the use of CNT-support has been promoted since 2023. The tested temperatures have varied from 600 to 900 °C. The applied CH₄/CO₂ molar ratio was mostly 1/1. The present research has relied on the sole spray impregnation of the fleece by the Ni active species precursor.

The fleece was first immersed in 0.5 mol/L HCl (Beijing, China) solution for acid etching for 24 h. The fleece was subsequently rinsed with deionized water at a neutral pH. The saturated nickel nitrate solution was prepared by adding Ni(NO₃)₂ (Shanghai, China) to ultra-pure water with 18.75 g/L of nonionic detergent, AEO-3 (Shanghai, China) or AEO-9 (Shanghai, China). AEO-3 and AEO-9 are both RO(CH₂CH₂O)_nH, R=C_12–16 alkyl, where AEO-3 n = 3 and AEO-9 n = 9. The molecular weight is about 315 Da for AEO-3, and about 590 Da for AEO-9. Using a pressurized spray, the saturated solution was evenly deposited onto the fleece at a set weight. After drying (Shanghai, China) at 80 °C, the coated fleece was calcined (Tianjin, China) at 700 °C for 3 h in air. H₂ was used for reduction at 700 °C for 3 h. The catalyst and fleece were then ready for testing.

2.3. Catalyst Characterization Methods

2.3.1. SEM

The surface morphology of the Ni/fleece catalyst sample was observed using scanning electron microscopy (SEM) (Shenzhen, China). The sample was directly adhered to conductive tape (Beijing, China) and sputter-coated with gold for 45 s prior to imaging.

2.3.2. XRD

X-ray diffraction (XRD) was determined by a Rigaku Miniflex instrument (Tokyo, Japan) at 40 kV and 40 mA. The data were recorded at every 0.02° at a 2θ angle ranging between 5 and 85, which were assessed toward the phases of the fresh catalysts. The Scherrer formula in Equation (9) was applied for calculating the mean crystallite size of the catalyst:

$$\mathrm{D}={\displaystyle \frac{0.89\mathsf{\lambda }}{\mathsf{\beta }\cos \mathsf{\theta }}}$$

(9)

with λ is the radiation wavelength (0.15406 nm), while θ and β are the Bragg diffraction angle (in radians) and the line broadening at half-maximum intensity, respectively.

2.3.3. BET

The specific surface area of the Ni/fleece catalyst was measured by a specific surface area porosity adsorber (Shanghai, China). Before the test, the sample was vacuum degassed at 300 °C for 8 h.

2.4. The Reaction Rate Equation in the Fleece Reactor

With respect to kinetic rate expressions for the DRM reaction, the literature is limited, although it has accepted that CH₄ dissociation can be a rate-determining step (RDS) [32]. Zhang and Verykios [33] came to that conclusion by applying the Langmuir–Hinshelwood model to a Ni catalyst. The CH₄ RDS approach was confirmed by Nandini et al. [34] at moderate temperatures, while the RDS shifted to the reaction of CO₂ and CH₄ at high temperatures. Full details about the Langmuir, Hinschelwood, Hougen, and Watson (LHHW) model approach, including its advantages and shortcomings, are given in Hayes and Kolaczkowski [35]. Other studies also confirmed the CH₄-driven reaction model, as determined from experiments with a Ni/SiO₂ catalyst [36], a Ni-Co/Al-Mg-O bimetallic catalyst [37], or with other Ni catalysts in general [38].

From the mechanisms with the Langmuir model for the chemisorption and desorption steps, as well as the surface CH₄ reaction as a RDS, the rate equation can be based upon the catalyst volume and substance concentration. Apart from the chemical reaction rate, the external gas diffusion within the bulk gas phase and the internal pore diffusion should also be considered.

The resulting equation, with CH₄ as a RDS, can be given:

$${\displaystyle \frac{{\mathrm{d}\mathrm{C}}_{{\mathrm{C}\mathrm{H}}_4}}{\mathrm{d}\mathrm{t}}}=-\varnothing {\mathrm{k}}_{\mathrm{r}}{\mathrm{C}}_{{\mathrm{s},\mathrm{C}\mathrm{H}}_4}=-{\mathrm{k}}_0{\mathrm{C}}_{{\mathrm{C}\mathrm{H}}_4}$$

(10)

where k₀ is the apparent rate constant, s⁻¹; ${\mathrm{C}}_{{\mathrm{C}\mathrm{H}}_4}$ is the bulk concentration of CH₄, mol/m³; ${\mathrm{C}}_{{\mathrm{s},\mathrm{C}\mathrm{H}}_4}$ is the CH₄ concentration at the catalyst surface, mol/m³; $\mathrm{\varnothing }$ is the effectiveness factor, (-); and k_r is the reaction rate constant, s⁻¹.

Since there was no substantial internal diffusion resistance in the thin wash coat of the fleece, $\mathrm{\varnothing }$ ≈ 1, as previously demonstrated [14,16], the equation can then be written as:

$${\displaystyle \frac{{\mathrm{d}\mathrm{C}}_{{\mathrm{C}\mathrm{H}}_4}}{\mathrm{d}\mathrm{t}}}=-{\displaystyle \frac{{\mathrm{k}}_{\mathrm{r}}{\mathrm{C}}_{{\mathrm{C}\mathrm{H}}_4}}{1+\frac{{\mathrm{k}}_{\mathrm{r}}\mathrm{V}}{{\mathrm{k}}_{\mathrm{m}}\mathrm{A}}}}$$

(11)

where V/A is the ratio between the catalyst volume and the geometric catalyst surface area, i.e., the wash-coat thickness, $\mathsf{\delta }$, m, while k_m is the mass transfer coefficient of CH₄, m/s.

Previous investigations [39] demonstrated that k_m was in of the order of 5 × 10⁻³ m/s, with a wash-coat thickness, $\mathsf{\delta }$, on the fleece support in the order of 10⁻⁸ m. The factor $\mathsf{\delta }$/k_m was in of the order of milliseconds, whereas k_r was approximately 10⁻¹ s. The denominator was hence reduced to ≈1, and the rate equation was simplified to:

$${\displaystyle \frac{{\mathrm{d}\mathrm{C}}_{{\mathrm{C}\mathrm{H}}_4}}{\mathrm{d}\mathrm{t}}}=-{\mathrm{k}}_{\mathrm{r}}{\mathrm{C}}_{{\mathrm{C}\mathrm{H}}_4}$$

(12)

or

$${\displaystyle \frac{{\mathrm{C}}_{{\mathrm{C}\mathrm{H}}_4}}{{\mathrm{C}}_{{\mathrm{C}\mathrm{H}}_4,0}}}=\mathrm{e}\mathrm{x}\mathrm{p}(-{\mathrm{k}}_{\mathrm{r}}\mathrm{t})$$

(13)

$$\ln \left(1-\mathsf{\eta }\right)=-{\mathrm{k}}_{\mathrm{r}}\mathrm{t}$$

(14)

where $\mathsf{\eta }$ is the CH₄ conversion ratio as ${\displaystyle \frac{{\mathrm{C}}_{{\mathrm{C}\mathrm{H}}_4,0}-{\mathrm{C}}_{{\mathrm{C}\mathrm{H}}_4}}{{\mathrm{C}}_{{\mathrm{C}\mathrm{H}}_4,0}}}$

This simplified rate equation has been applied in Section 3.4 of the paper.

3. Results and Discussion

3.1. Preparation Results

3.1.1. The Beneficial Effect of Adding Detergents During the Catalyst Impregnation

In the preparation of supported Ni⁰ catalysts via the impregnation method, it was observed that due to the large contact angle between the aqueous Ni(NO₃)₂ precursor solution and the carrier surface, the interfacial wettability between both was poor. This resulted in the slow spreading of the solution on the carrier surface and inadequate penetration into the internal pore channels during the impregnation process, making it difficult for Ni(NO₃)₂ to effectively enter the pore structure of the carrier. This issue directly led to an uneven distribution of the active Ni component on the carrier, which subsequently affected the catalyst’s activity and stability.

To address the aforementioned issue, the detergent AEO-9 was introduced into the Ni(NO₃)₂ precursor solution. By reducing the surface tension of the solution and improving the solid–liquid interfacial wettability, the detergent facilitated the spontaneous, rapid, and uniform penetration of the precursor solution into the internal pore channels of the carrier via capillary action. This approach aimed to achieve uniform dispersion of the active Ni species within the three-dimensional space of the carrier, thereby facilitating the preparation of high-performance supported catalysts.

To investigate the effects of the detergent on the wettability and dispersion effectiveness of the precursor solution, contact angle measurements and EDS characterization on Ni(NO₃)₂ solutions, both before and after the addition of the detergent AEO-9, were performed.

As shown in Figure 4, the contact angle measurements indicated that when no detergent was added (Figure 4a), the Ni(NO₃)₂ solution exhibited a large contact angle on the surface of the metal-sintered carrier, indicating poor wettability. In contrast, after the addition of the detergent AEO-9 (Figure 4b), the contact angle was significantly reduced, demonstrating a marked improvement in wettability. These results confirmed that the introduction of detergent effectively enhanced the solid–liquid interfacial interaction, enabling the solution to spread rapidly and uniformly over the carrier surface.

Figure 4. Contact angle of fleece and precursor solution: (a) Ni(NO₃)₂; (b) Ni(NO₃)₂ with detergent AEO-9.

EDS mapping analysis of the impregnated carriers (Figure 5) revealed considerable differences in the Ni distribution. In the sample prepared without the detergent (Figure 5a), the Ni elemental distribution was localized and non-uniform. Conversely, with the addition of the detergent AEO-9 (Figure 5b), Ni(NO₃)₂ was significantly dispersed more uniformly on the carrier surface. Furthermore, the Ni signal intensity exhibited a gradient decrease from the surface toward the interior of the carrier cross-section, indicating the effective penetration of the precursor solution into the internal pore channels. This improvement in distribution characteristics facilitated the formation of more abundant active sites within the carrier, creating favorable conditions for enhancing the overall performance of the catalyst. According to the reported literature, the mechanism of action of the nonionic surfactant AEO-9 included: (i) reducing the surface tension of the solution, thereby improving the wettability of the support pores [40]; and (ii) during the drying stage, the surfactant molecules inhibited the migration and agglomeration of Ni²⁺ through steric hindrance effects, thus promoting the formation of smaller and more stable NiO particles. Smaller particle sizes could imply a higher specific surface area of the active phase and a greater number of surface active sites, which is key to the enhancement of catalytic performance [41].

Figure 5. EDS spectrum of (a) Ni(NO₃)₂ and (b) Ni(NO₃)₂ with detergent AEO-9.

As shown in Figure 6, the performances of the three types of Ni/fleece catalyst in the DRM were systematically compared over time under stable reaction conditions, including calcination at 700 °C, a catalyst loading of 5 g, and a gas hourly space velocity of 1200 NmL/g/h. The catalysts were evaluated without detergent, with detergent AEO-3, and with detergent AEO-9.

Figure 6. (a) CH₄ and (b) CO₂ conversion; (c) H₂/CO ratio with or without detergent versus time on stream: 700 °C calcination, 700 °C reaction, 1.25 wt% Ni, 1200 NmL/g/h gas hourly space velocity.

By comparing the effects of two detergents with different molecular weights, it was found that both detergent AEO-3 and detergent AEO-9 effectively enhanced the reaction performance compared to the catalyst without detergent. Specifically, for the catalyst with detergent AEO-3, the CH₄ conversion stabilized in the range of 62–68%, the CO₂ conversion was 36–39%, and the H₂/CO ratio remained around 1.8. In contrast, the catalyst with detergent AEO-9 exhibited superior activity and stability, achieving a CH₄ conversion of 69–73%, a CO₂ conversion of 40–42%, and an equally stable H₂/CO ratio of approximately 1.8.

A comparison of the long-term stability of the two detergent revealed that after 1000 min of reaction, the activity of the catalyst with detergent AEO-3 gradually declined to a level comparable to that of the catalyst without detergent, whereas the catalyst with detergent AEO-9 maintained higher catalytic efficiency throughout the entire testing period.

Based on these experimental findings, detergent AEO-9 was selected as the optimal additive for the precursor solution in the subsequent efficient preparation of Ni/fleece catalysts. Since the addition of the detergent to the Ni-based precursor solution was clear and easy, it could be recommended in scale-up applications. An additional simple soaking step could be introduced prior to the loading of catalyst, as the impact on the whole process in terms of extra time and operation would be limited. The increased CO₂ and CH₄ conversion by several % should merit its extra consideration, as suggested by You et al. [42].

3.1.2. Effect of Calcination Time

During the preparation of Ni/fleece catalysts, the sintered metamesh carrier was first impregnated with the Ni(NO₃)₂ solution. The samples were then placed in a tube furnace and calcined in air at 700 °C for 3 and 6 h. After a reduction treatment with H₂, their catalytic performances were evaluated under identical reaction conditions by introducing the reaction gas.

The comparative catalytic performance of the catalysts prepared with different calcination durations is presented in Figure 7, with error bars representing the standard deviation of three parallel measurements. In terms of CH₄ conversion (Figure 7a), the catalyst calcined for 3 h maintained 71.2 ± 1.1% throughout the reaction. When the calcination time was prolonged to 6 h, the methane conversion rate decreased to 69.4 ± 1.8%, with larger error bars indicating reduced stability. Regarding CO₂ conversion (Figure 7b), the sample calcined for 3 h showed stable performance at 41.5 ± 0.8%, while the 6 h calcined sample exhibited lower and more fluctuating conversions (36.2 ± 2.3%). The smaller error bars for the 3 h calcined catalyst suggested better reproducibility and stability. Although the H₂/CO ratio (Figure 7c) increased slightly to approximately 2.0 after 6 h of calcination, the catalyst calcined for 3 h demonstrated superior overall reaction efficiency with consistently smaller error bars across all performance indicators.

Figure 7. (a) CH₄ and (b) CO₂ conversion; (c) H₂/CO ratio with calcination times of 3 and 6 h versus time on stream: 700 °C reaction, 1.25 wt% Ni, 1200 NmL/g/h gas hourly space velocity.

In summary, calcination at 700 °C in air for 3 h was identified as a more suitable pretreatment condition, under which the prepared Ni/fleece catalyst exhibited higher catalytic activity.

SEM and EDS characterization analyses were performed on the Ni/fleece catalysts calcined at 700 °C under an air atmosphere for 3 and 6 h, with the results presented in Figure 8. The SEM image (Figure 8a) revealed that the catalyst calcined for 3 h exhibited small, densely packed, and uniformly distributed NiO particles on its surface. The corresponding EDS mapping result (Figure 8b) further confirmed a highly uniform dispersion of Ni across the carrier. In contrast, after extending the calcination time to 6 h, the SEM image (Figure 8c) showed significant coarsening of NiO crystallites and the formation of “island-like” aggregates due to sintering. The corresponding EDS analysis (Figure 8d) indicated a clear decline in the uniformity of Ni distribution.

Figure 8. SEM images and EDS of Ni distribution after calcination: (a,b) 3 h; (c,d) 6 h.

These structural changes aligned well with the previously observed reaction performance, as the reduced activity of the 6 h calcined catalyst can be attributed to a decrease in active surface area and diminished accessibility of active sites, both resulting from the sintering and agglomeration of NiO particles.

3.1.3. Effect of Active Species Loading

The effect of the metal loading on catalyst performance was studied for Ni loadings (0.5, 0.75, and 1.25 wt%). Their efficiencies for DRM were systematically evaluated and shown in Figure 9.

Figure 9. (a) CH₄ and (b) CO₂ conversion; (c) H₂/CO ratio with active species loading for 0.5, 0.75, and 1.25 wt% versus time on stream: 700 °C reaction, 1200 NmL/g/h gas hourly space velocity.

The catalyst with a nickel loading of 1.25 wt% exhibited the highest activity, with stable CH₄ conversion in the range of 71–73%, CO₂ conversion around 42%, and the H₂/CO ratio maintained at approximately 1.8. When the loading was reduced to 0.75 wt%, a slight decline in activity was observed: CH₄ conversion ranged from 66 to 69%, CO₂ conversion was about 44%, and the H₂/CO ratio was around 1.55. Conversely, with a further reduction in loading to 0.5 wt%, the catalyst not only showed lower initial activity (CH₄ conversion 47%, CO₂ conversion 32%), but also demonstrated a pronounced deactivation trend during the 1000 min reaction: CH₄ conversion gradually decreased to 22%, CO₂ conversion dropped to 20%, and the H₂/CO ratio fluctuated between 1.58 and 1.74.

These results indicated that the Ni loading had a significant impact on the catalysts’ stability. For the sample with 0.5 wt% Ni, the initial activity was relatively low and continued to decline over 1000 min of operation, exhibiting typical deactivation behavior. According to the reported literature [43,44], the deactivation of low-loading catalysts under DRM reaction conditions was generally attributed to the following two potential mechanisms: (i) insufficient density of active sites, where the number of Ni active sites per unit area was too low, leading to even trace amounts of carbonaceous species generated during the reaction rapidly covering the entire Ni surface and physically blocking the active sites [45]; (ii) sintering of metal particles, where at extremely low loadings, Ni species may exist as highly dispersed small clusters with high surface energy; however, these clusters are thermodynamically unstable under high-temperature reaction conditions and prone to migration and agglomeration, resulting in a reduction in the number of active sites [46].

In contrast, the catalysts with 0.75 wt% and 1.25 wt% Ni maintained relatively stable activity throughout the reaction, indicating that higher loadings facilitated the formation of a sufficient number of active centers with enhanced resistance to deactivation. Among these, the 1.25 wt% sample exhibited the best performance, which may be attributed to the moderate Ni particle size achieved at this loading, while providing an adequate number of active sites while suppressing sintering and carbon deposition to a certain extent.

3.1.4. Effect of Reduction Conditions

Effect of Reducing Gas

During the activation process of the Ni/fleece catalysts, the influences of different reduction atmospheres (CH₄ and H₂) on their final catalytic performance were systematically investigated.

Under the reaction condition of 700 °C, the influences of different reduction atmospheres (H₂ and CH₄) on the performance of Ni/fleece catalysts are shown in Figure 10. The catalyst reduced in H₂ exhibited excellent catalytic stability: CH₄ conversion remained stable at 66–67% (Figure 10a), CO₂ conversion was maintained between 35 and 40% (Figure 10b), and the H₂/CO ratio consistently stayed around 1.8 (Figure 10c). In contrast, the catalyst reduced in CH₄ exhibited a strikingly different and more complex behavior: while CH₄ conversion remained relatively stable at 39–40%, CO₂ conversion fluctuated dramatically within a wide range of 28–42%, and the H₂/CO ratio varied drastically between 0.9 and 1.9 (Figure 10). These results suggested that using CH₄ as the reducing gas may induce carbon deposition during the reduction process or incomplete reduction on the catalyst surface, which could cover the active sites and alter the reaction pathway, thereby leading to decreased catalyst activity.

Figure 10. (a) CH₄ and (b) CO₂ conversion; (c) H₂/CO ratio with reducing gas for H₂ and CH₄ versus time on stream: 700 °C reaction, 1.25 wt% Ni, 1200 NmL/g/h gas hourly space velocity.

To further investigate the influence of the reducing atmosphere on the surface structure of the catalysts, Raman spectroscopy was performed on the samples activated under different atmospheres. The results are presented in Figure 11. The Raman spectra were fitted with dashed and solid lines represent the experimental data and fitted curves, respectively. For the catalyst reduced in CH₄ (Figure 11a), the spectrum exhibits distinct D and G bands at approximately 1350 and 1590 cm⁻¹, respectively, along with several additional weak bands. The I_D/I_G ratio is determined to be 1.53, indicating the formation of defect-rich amorphous carbon species. Additionally, a peak at 670 cm⁻¹ corresponding to NiO was observed, suggesting an incomplete reduction and the coexistence of Ni⁰, NiO, and carbon species. In contrast, the H₂-reduced catalyst (Figure 11b) showed no characteristic carbon peaks and only a minor NiO feature at 550 cm⁻¹, indicating a predominantly clean metallic surface.

Figure 11. Raman spectra of Ni/fleece catalysts after different reduction atmosphere treatment: (a) CH₄ reduction; (b) H₂ reduction.

When correlated with the catalytic performance, these Raman features can point toward distinct dynamic pathways recently identified in Ni-catalyzed DRM [43,47,48,49], thereby providing a mechanistic explanation for the observed differences. For the CH₄-reduced catalyst, the coexistence of defect-rich amorphous carbon and residual NiO created a system prone to dynamic instability. The defective carbon actively participated in the reaction via the reverse Boudouard reaction (C + CO₂ → 2CO), generating carbon vacancies that continuously captured and released CH₄ fragments, while the residual NiO and metallic Ni established a reservoir of interconverting metastable oxygen species (surface atomic oxygen, subsurface oxygen, and bulk NiOₓ) with distinct catalytic activities. Under the DRM feed, this catalyst became trapped in a non-equilibrium oscillatory state, where the sequential transformation of these oxygen species, coupled with the dynamic carbon cycle, led to self-sustained fluctuations in surface chemistry. In this oscillatory regime, CH₄ conversion remained relatively stable because a subset of Ni⁰ sites—likely those regenerated by H₂ from CH₄ cracking—persistently activated methane. However, CO₂ conversion fluctuated dramatically because CO₂ activation was exquisitely sensitive to both the dynamic oxygen speciation and the oscillating carbon coverage, with CO₂ serving dual roles as a reactant and carbon-removal agent. The corresponding fluctuations in the H₂/CO ratio reflected the shifting balance between CH₄ cracking and CO₂ activation as the surface oscillated between different metastable configurations. For the H₂-reduced catalyst, the clean metallic surface with only minor residual NiO allowed the DRM reaction to establish a stable, self-regulating cycle from the outset, converging to a stable attractor state rather than becoming trapped in oscillation.

Based on the integrated results, the choice of reducing gas determined not only the initial surface cleanliness but, more critically, the dynamic trajectory the catalyst followed under reaction conditions. H₂ reduction generated a clean metallic surface that could establish a stable reaction pathway, while CH₄ reduction created a complex surface architecture that trapped the catalyst in a non-equilibrium oscillatory state. Therefore, H₂ was selected as the reducing gas for the activation stage of the Ni/fleece catalyst.

Effect of Reduction Time

To further optimize the reduction process of the Ni/fleece catalyst, the effects of different H₂ activation durations on its structure and catalytic performance were systematically investigated.

The catalytic performance tests, as presented in Figure 12, showed that under the condition of H₂ reduction for 1 h, the CH₄ conversion stabilized at 66–67%, the CO₂ conversion ranged from 35 to 40%, and the H₂/CO ratio was approximately 1.8. When the reduction time was extended to 3 h, the catalyst activity was further enhanced, achieving a CH₄ conversion of 70–73% and a CO₂ conversion of 41–42%, while the H₂/CO ratio remained stable at around 1.8. These results indicated that a longer reduction time was conducive to obtaining higher catalytic activity. This improvement could likely be attributed to a more complete reduction process, which converted a greater amount of NiO into active metallic Ni, thereby providing more abundant active sites.

Figure 12. (a) CH₄ and (b) CO₂ conversion; (c) H₂/CO ratio with reduction time of 1 and 3 h versus time on stream: 700 °C reaction, 1.25 wt% Ni, 1200 NmL/g/h gas hourly space velocity.

3.1.5. Fleece Before and After Reduction/Reaction

As shown in Figure 13, the values of 2θ = 44.5°, 51.8°, and 76.4° corresponding to Ni (111), (200), and (220), with no impurities appearing, indicated the inert carrier of the catalyst.

Figure 13. XRD patterns of the Ni/fleece catalyst after reduction and reaction.

To thoroughly investigate the evolution of the catalyst’s textural properties during the preparation and reaction, nitrogen physisorption was systematically conducted on the samples after calcination (NiO/fleece), after reduction (Ni/fleece), and after reaction. The relevant textural parameters are summarized in Table 2. The isotherms are given in Figure 14.

Table 2. Physical structure parameters of catalyst loading.

	Surface Area BET (m2/g)	Pore Volume (cm3/g)	Average Pore Size (nm)	t-Plot External Surface Area (m2/g)	BJH Adsorption Cumulative Volume of Pores (cm3/g)
After Calcination	5.29	0.024	17.85	1.40	0.022
After Reduction	18.22	0.017	3.64	25.99	0.015
After Reaction	15.93	0.017	4.19	24.56	0.016

Figure 14. (a) N₂ sorption isotherms and (b) pore size distributions of the catalyst at different stages after calcination, after reduction, and after reaction.

The BET of the bare and etched fleece were 0.024 and 0.52 m²/g, respectively. These low values proved that the fleece fibers were non-porous. With the Ni loading, the Np size, and the wash coat, the increased BET values proved that the wash-coat layer was responsible. Since the fleece was non-porous, pore blocking cannot occur. This finding confirmed earlier findings by Zhang et al. [50].

The data in Table 2 revealed that the reduction treatment was a critical step for optimizing the catalyst’s textural properties. The calcined sample (NiO/fleece) exhibited a relatively low BET specific surface area (5.29 m²/g) and a large average pore diameter (17.85 nm), characteristic of the initial macroporous-dominated structure of the carrier. After H₂ reduction, the sample (Ni/fleece) underwent significant textural changes: its specific surface area increased substantially to 18.21 m²/g, while the average pore diameter decreased sharply to 3.64 nm. This transformation indicated that the reduction process not only converted NiO into metallic Ni but may have also involved partial reconstruction of the oxide layer on the carrier surface, thereby exposing a fresher surface and generating a more abundant mesoporous structure. The post-reaction catalyst maintained a similar specific surface area (15.93 m²/g) and pore structure to the reduced sample, demonstrating that the catalyst skeleton remained stable during the prolonged DRM reaction without severe pore blockages or sintering.

The pore size distribution curves (Figure 14b) showed a distinct concentration of mesopores in the range of 20–60 nm for both the reduced and post-reaction samples, which was beneficial for the mass transfer of gaseous molecules in the reaction system. In contrast, the pore size distribution of the calcined sample was relatively broad and dispersed.

In summary, although the metal fleece carrier itself possessed a low specific surface area, its texture was significantly optimized through reduction activation, resulting in a pore system dominated by mesopores that was more conducive to catalysis. The combination of a high specific surface area and a favorable pore structure ensured sufficient exposure of active sites and efficient diffusion of reactants, serving as an important structural foundation for the catalyst’s high activity and stability.

3.2. Effect of the Fleece Support Characteristics

3.2.1. Effect of Pore Size of Filter Layer of Metal Sintering Mesh Carrier

To investigate the influence of the pore size of the metal-sintered mesh carrier filtration layer on the DRM, a series of Ni/fleece catalysts were prepared using carriers with pore sizes of 15, 10, 5, and 2 μm, and another without an intermediate filtration layer (denoted as “0 μm”). Their catalytic performances were evaluated and the results are given in Figure 15.

Figure 15. (a) CH₄ and (b) CO₂ conversion; (c) H₂/CO ratio with pore sizes of 15, 10, 5, 2, and 0 μm versus time on stream: 700 °C reaction, 1.25 wt% Ni, 1200 NmL/g/h gas hourly space velocity.

The performance test results (Figure 15) showed that the catalyst with a pore size of 15 μm exhibited the best overall performance, maintaining stable CH₄ conversion in the range of 73–77%, CO₂ conversion of 39–42%, and an H₂/CO ratio of approximately 1.8, demonstrating high activity and good stability. When the pore size was reduced to 10 μm, a slight decrease in activity was observed, with CH₄ conversion ranging from 70 to 73%, CO₂ conversion of 41–42%, and the H₂/CO ratio was still around 1.8.

Further reductions in the pore size to 5 μm and below led to significant performance deterioration and accelerated deactivation. For the 5 μm sample, CH₄ conversion gradually declined from an initial 69 to 58%, with CO₂ conversion between 34 and 37% and an H₂/CO ratio around 1.9. The 2 μm sample showed a drop in CH₄ conversion from 66 to 55%, CO₂ conversion of 29–31%, and an H₂/CO ratio of about 1.9. The carrier without a distinct intermediate filtration layer (indicated as “0 μm”) exhibited a decrease in CH₄ conversion from 65 to 53%, CO₂ conversion of 27–29%, and an H₂/CO ratio around 2.0.

Integrating structural and performance analyses, excessively small pore sizes (≤5 μm) severely restricted the mass transfer during the reaction, reducing the accessibility of active sites and exacerbating carbon deposition or sintering, thereby causing activity decline and rapid deactivation. Although the 15 μm carrier demonstrated optimal performance, its fabrication required finer metal wires and a greater amount of wire per unit area, resulting in significantly higher preparation costs compared to carriers with the 10 µm (and smaller) pore sizes.

Based on a comprehensive trade-off between performance and cost, the 10 μm carrier was selected for further in-depth investigation, which showed favorable activity and stability at a more advantageous cost.

3.2.2. Effect of the Number of Fiber Layers of the Sintered Metal Carrier

The influence of the number of metal-sintered mesh carrier filtration layers on the performance in DRM was further investigated, with the results presented in Figure 16.

Figure 16. (a) CH₄ and (b) CO₂ conversion; (c) H₂/CO ratio with filtration layers of 0, 1, 2, and 4 versus time on stream: 700 °C reaction, 1.25 wt% Ni, 1200 NmL/g/h gas hourly space velocity.

When no intermediate filtration layer was present in the carrier, the catalyst exhibited low activity and rapid deactivation: CH₄ conversion gradually declined from 65 to 53%, CO₂ conversion was only 27–29%, and the H₂/CO ratio was approximately 2.0. With the addition of a single filtration layer, performance improved significantly, stabilizing at a CH₄ conversion of 70–73%, a CO₂ conversion around 41%, and an H₂/CO ratio maintained near 1.8. When a double-layer configuration was employed, activity was further enhanced, achieving a CH₄ conversion of 76–77% while CO₂ conversion remained around 41% and the H₂/CO ratio stayed at about 1.8. In contrast, increasing to four layers showed no clear improvement over the double-layer structure, with CH₄ conversion (76–77%), CO₂ conversion (~40%), and H₂/CO ratio (~1.8) remaining comparable.

These results indicated that the introduction of filtration layers effectively improved the reactant distribution and mass transfer, leading to markedly enhanced catalytic activity and stability. However, given the similar performance observed between the double-layer and four-layer configurations, and considering that additional layers substantially increase material costs and bed pressure drop, the double-layer structure was selected for subsequent experiments, which offered excellent overall performance at a more reasonable cost.

3.3. Stability Test of Catalyst

To evaluate the long-term operational stability of the Ni/fleece catalyst, a 90 h lifetime test was conducted under the corresponding reaction conditions. As shown in Figure 17, no significant deactivation was observed throughout the entire test period, with all performance metrics remaining stable: CH₄ conversion consistently stayed around 77%, CO₂ conversion was stable near 37%, H₂ yield was approximately 67%, CO yield about 37%, and the H₂/CO ratio maintained around 1.8.

Figure 17. Conversion and yield versus time on stream: 700 °C reaction, 1.25 wt% Ni, 1200 NmL/g/h gas hourly space velocity.

These results demonstrated (i) that the prepared Ni⁰/fleece catalyst exhibited good structural stability and catalytic durability in DRM, indicating promising potential for practical application, and (ii) that both CH₄ and CO₂ conversions exceeded the values obtained in a fixed-bed reactor. Moreover, the fixed-bed operations were prone to higher coking, as reviewed by Deng et al. [51]—carbon-supported Ni has been reported to eliminate coking to a large extent [52].

3.4. Conversion and Yield

3.4.1. Experimental Results

Having applied the findings from the literature to determine the rate equation, it is important to investigate the influence of temperature, T, and catalyst contact time, t, on the conversions and yields—experiments were again performed in triplicate. The deviation in the repeated results and the averages were below 2%. Only the average values were hence applied to the further treatment. The essential data are reported in Table 3 below for both CH₄ and CO₂ conversions, $\mathsf{\eta }$.

Table 3. Conversion versus temperature (T) and reaction time (t) [CH₄:CO₂ = 1:1].

T (°C)	t (s)	${\mathit{\eta }}_{{\mathit{C}\mathit{H}}_4}$	${\mathit{\eta }}_{{\mathit{C}\mathit{O}}_2}$
650	11.04	0.488	0.324
700	10.47	0.715	0.422
750	9.963	0.814	0.458
650	22.08	0.566	0.331
700	20.95	0.796	0.515
750	19.93	0.843	0.555
650	36.81	0.668	0.343
700	34.92	0.853	0.583
750	33.21	0.886	0.607

Considering the exponential nature of gas–catalytic first-order reactions, the data were fitted, as illustrated in Figure 18.

Figure 18. Conversion (a) CH₄ and (b) CO₂ in function $\mathsf{\eta }$ T and t.

The gas velocity through the fleece (with porosity 0.79) was calculated from the total feed flow rate, corrected for temperature T, and divided by the free (open) fleece area:

$${\mathrm{N}}_{\mathrm{f}}=\sum \mathrm{m}\mathrm{L}/\min ({\mathrm{C}\mathrm{O}}_2,{\mathrm{C}\mathrm{H}}_4,{\mathrm{N}}_2)\times {\displaystyle \frac{\mathrm{T}+273}{273}}\times {\displaystyle \frac{{10}^{-6}}{60}}/(\mathsf{\pi }\times 0.05\times 0.295\times 0.79)\mathrm{m}/\mathrm{s}$$

(15)

The contact time was calculated as the ratio of the fleece volume and the gas flow rate at the operating temperature. The free volume of the fleece was determined from its 1.7 mm thickness as 6.22 × 10⁻⁵ m³. Since CH₄ was recognized as the RDS, the primary reaction rate can be tentatively presented as a first-order CH₄ reaction. The CO₂ conversion was stoichiometrically linked with the CH₄ conversion, but of a lesser rate-determining step. Both rate reactions allowed the determination of k_r1 and k_r2 for CH₄ and CO_2, respectively, as follows:

$${\mathrm{k}}_{\mathrm{r}\mathrm{i}}=\mathrm{l}\mathrm{n}(1-{\mathsf{\eta }}_{\mathrm{i}})/\mathrm{t}$$

(16)

and varied with temperature, as expected from the Arrhenius expression:

$${\mathrm{k}}_{\mathrm{r}\mathrm{i}}={\mathrm{A}}_{\mathrm{i}}\exp \left(-{\displaystyle \frac{{\mathrm{E}}_{\mathrm{A}\mathrm{i}}}{\mathrm{R}\mathrm{T}}}\right)$$

(17)

Transforming the results according to Equation (17) can provide the values for the activation energy E_A, J/mol, and the pre-exponential factor, A_i, s⁻¹, as reported in Table 4.

Table 4. Activation energy, E_A, and pre-exponential factor, A.

	EA (J/mol)	A (s−1)
CH4	102,500	32,000
CO2	95,600	4640

Values of E_A and A had a maximum deviation of ±3% on the averages. The Arrhenius expression defined k_r, as well as the expected conversion for any set or operating conditions (t, T, ${\mathrm{C}}_{{\mathrm{C}\mathrm{H}}_4}$, ${\mathrm{C}}_{{\mathrm{C}\mathrm{O}}_2}$). Table 4 clearly demonstrates that the CH₄ reaction was dominant (high A), whereas the CO₂ reaction had a similar E_A but a far lower A.

3.4.2. Effect of Space Velocity

The velocity and flow rate data can also be expressed as the ratio of the catalyst weight (see Table 1) and the defined the gas hourly space velocity, GHSV. For all experiments, these can be expressed in NmL/gcat/h and are presented in Table 5.

Table 5. Reaction condition of GHSV.

Flow Rate (NmL/min)	GHSV (NmL/g/h)
100	1200
50	600
30	360

At temperature T, the GHSV (in actual mL/g/h) must be multiplied by (T + 273)/273.

The experimental results are presented in Figure 19.

Figure 19. Conversion (CH₄, CO₂) and yield (H₂, CO) versus GHSV: 700 °C reaction, 1.25 wt% Ni.

As space velocity decreased, the conversion of reactants progressively increased. At the highest space velocity of 1200 NmL/g/h, CH₄ conversion ranged from 70 to 73%, CO₂ conversion was 40–42%, H₂ and CO yields were 64–67% and 37–39%, respectively, and the H₂/CO ratio remained stable around 1.8. When the space velocity was reduced to 600 NmL/g/h, CH₄ conversion increased to 74–79% and CO₂ conversion reached 46–51%; however, H₂ yield declined to 58–62%, CO yield decreased to 27–29%, and the H₂/CO ratio correspondingly rose to about 2.0. Upon further decreasing the space velocity to 360 NmL/g/h, although the initial CH₄ and CO₂ conversions were as high as 86 and 59%, respectively, they exhibited a gradual downward trend during the test period, eventually falling to 81 and 53%. Concurrently, H₂ yield was 52–54%, CO yield was only 18–21%, and the H₂/CO ratio decreased from an initial value of 3.0 to approximately 2.6.

These results indicated that lower space velocities favored higher reactant conversions but led to decreased yields of H₂ and CO, as well as an elevated H₂/CO ratio. This trend may be attributed to the occurrence of secondary reactions, such as the reverse water–gas shift reaction or methane cracking, which could lower the selectivity toward the target products or result in further consumption of products due to prolonged residence time in the bed. Therefore, in practical applications, a balanced optimization of space velocity is required, taking into comprehensive consideration conversion efficiency, product distribution, and catalyst stability.

3.4.3. Effect of Temperature

As the reaction temperature increased from 650 to 750 °C, both CH₄ and CO₂ conversions significantly improved. The results are illustrated Table 3.

At 700 °C, CH₄ conversion maintained at 70–85%, CO₂ conversion at 42–58%, and the H₂/CO ratio stable around 1.8. Upon further increasing the temperature to 750 °C, the initial activity was even higher, reaching a CH₄ conversion of 89% and a CO₂ conversion of 61%; however, both values gradually decreased during the reaction to 85% and 43%, respectively, while the H₂/CO ratio rose to about 2.0.

From a kinetic perspective, raising the temperature facilitated a higher activity for reactant molecules, promoting the rates of elementary reactions such as methane and CO₂ reforming, which explained the significant improvement in CH₄ and CO₂ conversions at elevated temperatures. Concurrently, higher temperatures favored methane cracking to produce H₂, leading to an increase in the H₂/CO ratio with temperature—a trend consistent with thermodynamic equilibrium. However, the gradual activity decline observed at 750 °C could likely be attributed to two factors: first, accelerated sintering of Ni grain at high temperature; second, although high temperatures helped suppress carbon deposition from CO disproportionation, it may have simultaneously enhanced methane cracking-induced coking (reaction 2). The combined effect of these two deactivation mechanisms could account for the observed slow performance degradation.

3.5. Selectivity

In the elemental selectivity analysis of the outlet gas, a quantitative breakdown of the sources for each element was performed in Figure 20. For carbon, its origin was distributed as follows: 11.5% from the feedstock CH₄, 29.54% from CO₂, 37.09% from CO, and the remaining 21.88% from other carbon-containing species or deposited carbon. For hydrogen, 23% from CH₄, 67.31% from H₂, and 9.69% from H₂O. For oxygen, 59.08% from CO₂, 37.09% from CO, and the remaining 3.83% from H₂O.

Figure 20. Elemental composition of output gas.

4. Scale-Up Recommendations

The concept of using a fleece candle as a catalyst carrier could facilitate the upscaling to industrial capacities. Not only can several candles be installed in a single reactor vessel and operated in parallel, but candles of diameter to 0.15 m and length to 3 m can be supplied. If a long contact time is required, the fleece thickness can be increased to 30 mm, thus allowing higher active species loadings, and again increasing the gas/catalyst contact time at a given surface gas flow rate. Experiments are currently ongoing with fleece candles of decreasing diameter mounted in the outer fleece of the longer diameter.

With a higher catalyst loading, the GHSV value obtained on laboratory scale could define the reaction yield. During the lab-scale experiments, some coking occurred (as monitored by the fleece ΔP). The DRM reaction was then stopped, and air was fed at 700 °C to burn the coke layer. When the ΔP was restored, the DRM was restarted. This was deemed necessary after about 90 h on stream.

The challenges for future scale-up will include the design and manufacturing of a multi-orifice distributor that can uniformly feed all parallel candles. Although the ΔP of the lab-scale candle was 6 Pa only, thicker or series fleeces will increase the working pressure, although this increase would be proportional with the fleece thickness since the flow mode is laminar. From a sustainability perspective, the final step could involve separating non-reacted CH₄ and CO₂ from the syngas product stream using techniques such as adsorption/desorption, membrane separation, or other equivalent methods. Crucially, this separated residual gas stream would not be discarded but rather recycled directly back to the reactor inlet. This closed-loop recycling strategy could significantly improve the overall raw material utilization rate, reduce feedstock consumption, and minimize greenhouse gas emissions. By converting unreacted gases back into valuable products, the process embodies the principles of a circular economy and could contribute to the environmental sustainability of the DRM technology.

The proposed scale-up fleece reactor is depicted in Figure 21. With these merits of the proposed technology, and the constant long-duration conversion results obtained, the objectives of Section 1.3 were deemed to be met.

Figure 21. Proposal setup for fleece–DRM. ① Fleece reaction with tubular feedstock distributor; ② electrical heating; ③ feedstock preheater; ④ exhaust fan; ⑤ adsorption/desorption of residual CO₂ + CH₄ with recycle to feedstock, and F, mass flow.

5. Conclusions

Biogas, typically composed of approximately 50 vol% CO₂ and 50 vol% CH₄, serves as an ideal and sustainable feedstock for the dry reforming of methane (DRM), transforming waste-derived resources into valuable syngas. Ni⁰ has been proven to be a good catalyst at low (0.5–1.25 wt%) loading of the carriers. In contrast to conventional fixed-bed reactors, a novel through-flow fleece reactor was developed and proven to offer both high effectiveness and operational efficiency, achieving very high conversions of CO₂ and CH₄. The optimal loading of active species on the sintered metal fleece was achieved (i) by adding detergent to the Ni-solution and (ii) through precise control of calcination and reduction conditions. Although a limited amount of coking was observed, the catalyst could be readily regenerated via offline combustion, restoring its activity and extending its service life—an important feature for process sustainability and economic viability. Owing to its simple configuration, the catalyst fleece reactor could be easily scalable for industrial applications. To further mitigate the slight coking observed during prolonged DRM operation, there is ongoing research investigating the addition of small amounts of zeolites or SAPO-11 to the catalyst precursor to enhance the Brønsted acidity of the active sites, thereby improving long-term stability and resource efficiency.