Fluidized Catalytic Oxidation of Ultralow Concentration Methane: A Pilot-Scale Study

Abstract

Ventilation air methane from coal mining is characterized by large flux and ultralow concentration, which is the largest source of methane emissions in coal mines. This study proposes a fluidized catalytic oxidation process to treat methane using cost-effective natural ore catalysts. Experiments were conducted in a pilot-scale fluidized bed reactor. The results indicate that the methane conversion rate reached 90% at 543 °C, while further temperature increment showed a minor effect on the conversion rate. Within the tested range, the effects of the bed inventory and catalyst particle size on efficiency were insignificant. In the stability test, methane conversion decreased from 97.4 to 90% in the initial 22 h. It was found that ∼35.6 wt % of the catalysts was blown out of the reactor during the reaction due to the low separation efficiency of the used cyclone, causing the conversion rate to decrease. After the reaction, the active component MnO₂ in the natural ore was partially converted to Mn₂O₃ with less surface oxygen vacancies, which also contributed to the decreased activity since the methane catalytic activity was much lower on Mn₂O₃ than that on MnO₂ (T₉₀ = 795 vs 505 °C).

1. Introduction

Methane, as a greenhouse gas, presents a global warming potential of 29.8 times greater than that of CO₂ for a 100-year horizon, which has contributed to ∼30% of global temperature rise since the industrial revolution.¹ The energy sector is one of the largest CH₄ emission sources, second only to the agriculture sector, among which the coal mining process is the largest contribution in China. The coal mining process is often accompanied by the extraction of coalbed methane. Gas with a high CH₄ concentration (usually >8 vol %) can be directly utilized in industrial applications, such as combustion in industrial boilers or gas engines for power and heat generation.² However, in coal mine ventilation air, the CH₄ concentration is typically less than 0.5 vol %, making resource utilization challenging and direct combustion unfeasible. Consequently, most mine ventilation air is vented directly into the atmosphere, leading to substantial greenhouse gas emissions. Ventilation air methane (VAM) consists of over 60% of the total CH₄ emissions in coal mine.³The treatment and utilization of VAM are, therefore, of great importance to reduce greenhouse gas emissions, offering both energy-saving and environmental benefits.

Catalytic oxidation enables the complete conversion of low-concentration methane at relatively low temperatures by using catalysts,⁴ making it an effective approach for VAM reduction.⁵ Regenerative catalytic oxidation (RCO) is the most extensively studied catalytic oxidation technology, with engineering applications in the treatment of low-concentration combustible gases such as volatile organic compounds (VOCs)⁶ and CO.⁷ VAM is characterized by its large flux, extremely low concentration, and fluctuations. When applying RCO in a fixed-bed process, insufficient radial diffusion may lead to localized hotspots, which would reduce the life of catalysts.⁸ On the other hand, scaling up the fixed-bed process to handle millions of cubic meters of VAM per hour is challenging, and the cost of catalysts is high, especially when using noble metal catalysts although with relatively higher activity.⁹ Therefore, there is an urgent need to develop new reaction systems and the corresponding catalytic materials for large-scale treatment of VAM in coal mines.

Fluidized bed reactors have advantages such as intense heat and mass transfer, ease of scaling up, and strong adaptability, making them a focus of attention for the catalytic conversion process.^10,11 For example, Iamarino et al.¹² used Cu/Al₂O₃ as a catalyst in a fluidized bed for methane combustion. The complete methane conversion could be achieved at 700 °C. Foka et al.¹³ studied the catalytic oxidation of premixed methane and air in a fluidized bed, and their research indicated that the turbulent fluidized bed was more suitable for catalytic oxidation. Zukowski¹⁴ investigated methane combustion in a fluidized bed using Mn-based catalysts as bed material and analyzed the fluctuations in bed temperature and methane conversion rate under different operating conditions. They found that effective heterogeneous reactions occurred on the catalyst surface at bed temperatures above 750 °C. Yang et al.¹¹ studied the catalytic oxidation of low-concentration methane in a bubbling fluidized bed and found that the temperature for 90% methane conversion (T₉₀) was 650 °C. As the inlet methane concentration and gas flow rate increased, the conversion rate decreased. They also investigated the catalytic oxidation of methane using a noble metal catalyst (0.5 wt % Pd/Al₂O₃), in which the complete methane conversion was achieved at 650 °C.¹⁵ Zhang et al.¹⁶ used Cu/γ-Al₂O₃ as bed material and studied the effects of fluidizing velocities and inlet concentrations on methane catalytic combustion efficiency. A methane conversion rate of 95% was reached at a bed temperature of 650 °C.

However, most studies on the catalytic oxidation of methane using a fluidized bed were conducted on lab-scale devices, making it difficult to provide guidance for practical engineering applications. The catalysts used generally exhibited lower catalytic activity (higher T₉₀) compared to those tested in fixed-bed reactors. Additionally, most of the reported catalysts are synthetically produced, even including noble metals, which are costly for large-scale applications. This highlights the need for more economical and efficient catalytic materials. Based on our preliminary experiments, this study selected natural ore-based catalysts with the ability to activate methane at low temperatures. To investigate their catalytic performance in a fluidized bed under industrially relevant conditions, experiments were conducted on a pilot-scale circulating fluidized bed reactor for the first time. Before and after the catalytic reaction, the natural ore catalyst was characterized to analyze the possible variations in its physicochemical properties. By varying conditions such as methane concentration, catalyst particle size, and fluidizing air velocity, the feasibility of using natural ore-based catalysts was explored, and the potential operational issues in fluidized bed reactors were identified, which is supposed to provide technical support for further scaling up of the process.

2. Experimental Section

The catalyst used in this study was natural ore originating from China at a cost of ∼200 USD/t. Its main compositions, i.e., Si, Al, Fe, and Mn, are listed in Table 1 as given by XRF. The obtained natural ore was crushed, dried, and sieved to produce catalyst particles of different sizes for further use.

Table 1. Main Compositions in Fresh Catalyst (Oxygen Excluded, wt %).

composition	Mn	Fe	Si	Al	others*
wt %	44.43	20.05	27.73	4.02	3.77

^*Containing K, Ca, Ti, Mg, V, etc. each with a concentration of less than 1 wt %.

The pilot reactor employed in this experiment is shown in Figure 1. The reactor consists primarily of an electric heating unit, fluidized bed reactor, cyclone separator, water-cooling system, feeding system, slag discharge unit, and a bag filter for dust removal. The basic parameters of the experimental setup are presented in Table 2. The reactor uses an electric heating unit with 31 segments of 15 kW electric heaters arranged around the reactor. The locations of the temperature and pressure sensors in the reactor are shown in Figure 1 and Table 3. The temperature and pressure changes were continuously monitored by using a data acquisition module.

Figure 1.

Layout of the pilot-scale circulating fluidized bed reactor.

Table 2. Basic Parameters of 100 kW Fluidized Bed.

parameters	unit	value
reactor height	m	18.8
inner diameter	mm	200
designed flow rate, Q	Nm3/h	100–200
preheating temperature	°C	360

Table 3. Temperature and Pressure Sensor Location.

items	label	height from the distributor (m)
pressure	P1	–0.1
	P2	0.1
	P3	9
	P4	18
temperature	T1	0.2
	T2	3
	T3	9
	T4	18

Before the experiment began, residual materials in the reactor were removed, and 15 kg of catalyst was added to the downcomer to accelerate the establishment of the circulation process. The primary air was adjusted to the target flow rate. Once the target temperature was reached, a predetermined mass of the catalyst was introduced into the main bed. After stable operation was achieved, a methane–air mixture at a set concentration was introduced. The methane concentration at different locations within the reactor was continuously measured online using a Fourier transform infrared (FTIR) gas analyzer (model AFS-B2T, Protea). Once the methane concentration at each sampling point stabilized, data recording began, with each test lasting 5–10 min. After all tests were completed, catalyst materials from the main bed, downcomer, and bag filter were collected for further analysis.

In the catalytic oxidation of low-concentration methane under fluidized conditions, factors including the reaction temperature, gas velocity, methane concentration, and catalyst particle size may significantly influence the catalytic reaction. In this study, a pilot-scale experiment was conducted to investigate the effects of these factors, as outlined in Table 4. Temperature measurement point T1, located in the dense phase zone, was designated as the reaction temperature for subsequent discussions, unless otherwise specified. In a typical operation case, the temperature was maintained at 550 °C; the particle size ranged from 380 to 830 μm with a median diameter of 500 μm; and the gas flow rate was set at 100 N m³/h. Under these operation conditions, the gas velocity was about 2.67 m/s at 550 °C, which was about 10.8 times the minimum fluidized velocity (V_mf) of the 500 μm particles.

Table 4. Experimental Conditions.

case	conditions	T (°C)	concentration (vol %)	loading amount (kg)	particle size (μm)
1	temperature	450–650	0.2	60	380–830
2	catalyst loading	550	0.2	40–80	380–830
3	CH4 concentration	550	0.1–0.5	50	380–830
4	particle size	550	0.2	50	250–3000
5	stability test	550	0.2	50–105	380–830

3. Results and Discussion

3.1. Blank Test

Blank tests were conducted to check the system errors and get background values. Figure 2 presents the pressure variations at different measurement points during the hot blank test (550 °C, without catalyst). The heating process took approximately 3 h to reach 550 °C. As shown in the figure, the pressure drops at various positions in the bed stabilized after 3 h. The pressure drop between P1 and P2 was about 27 kPa, which is primarily caused by the air distributor. The pressure drop between P2 and P3 was 2.7–2.9 kPa, corresponding to a bed height of 0.1–9 m, where the majority of the bed material presented to participate in the catalytic reaction. This pressure drop was mainly caused by residual materials deposited in the reactor that cannot be discharged completely. The pressure drop between P3 and P4 was 0.35–0.36 kPa, corresponding to a bed height of 9–20 m.

Figure 2.

Changes in pressure drop (a–c) and methane concentration (d) under hot state empty bed experiment (Q: 200 N m³/h,T: 550 ± 10 °C).

After the temperature stabilized, methane with a concentration of ∼2000 ppm was introduced. The methane concentrations at different heights are listed in Figure 3d. It can be observed that methane concentration gradually decreased with increasing measurement height, with the outlet concentration about 200 ppm lower than the inlet concentration. This decrease may be due to the residual material in the furnace, which could have promoted methane decomposition at 550 °C, or possibly due to air leakage along the reactor, leading to the reduction in outlet methane concentration. Overall, the blank experiments indicate that the system is relatively stable and suitable for subsequent experiments. The measured methane concentrations at different heights during the blank test were used as the baseline values for the calculation of methane conversion rates at the corresponding height.

Figure 3.

(a) Methane concentration and temperature change over time in a fluidized bed reactor (catalyst loading amount: 60 kg, Q: 100 N m³/h, particle size: 380–830 μm, and CH₄ concentration: 2000 ppm); (b) comparison of the T₉₀ with the Mn and Fe catalysts reported in previous studies (the references labeled in the figure are listed in the Supporting Information; operation conditions used in fixed-bed reactor were catalyst amount: 2g, weight hourly space velocity: 7500 mL/(g·h), particle size: 180–250 μm, and CH₄ concentration: 5000 ppm).

3.2. Effect of Temperature

Figure 3a shows the variation of the methane outlet concentration and temperature with time in the fluidized bed reactor. When the furnace temperature (T1) reached 450 °C, methane at a concentration of 2000 ppm was introduced into the reactor for reaction, while the furnace temperature was gradually increased at a heating rate of ∼20 °C/h during the reaction process. At the temperature of 450 °C, the methane outlet concentration was 700 ppm, resulting in a conversion rate of only 65%. As the temperature increased, the methane concentration gradually decreased. The methane conversion rate reached 90% at 543 °C. At approximately 600 °C, the methane concentration reached its lowest point, with a conversion rate of 93.7%. Beyond this temperature, the methane conversion rate remained relatively stable. The T₉₀ obtained in this study is compared with the previously reported Mn- or Fe-based catalyst as shown in Figure 3b. It can be found that the activity of the natural ore without any further purification or modification was even better than those of some synthetic catalysts. Furthermore, the cost of the natural ore is about 200 USD per ton, which is at least 1 order of magnitude lower than the synthetic catalysts, showing the potential of its application in the treatment of VAM with large flux. Based on these findings, a reaction temperature of 550 °C was selected for subsequent experiments.

3.3. Effect of Catalyst Loading Amount (Bed Inventory)

The reaction was maintained at a gas flow rate of 100 N m³/h. The amount of catalyst added to the main bed was varied to adjust the gas–solid contact time. Once the reactor temperature reached 550 °C, an initial amount of 50 kg of catalyst was added to the main bed, followed by subsequent additions of 10 kg at intervals until a total amount of 80 kg was reached. The corresponding mass space velocity ranged from 1250 to 2000 N m³/t-h at static state. Figure 4a–e shows the pressure drop variations at different measurement points during the reaction. After the addition of 50 kg of catalyst, there was an increase in the pressure drop between P1 and P2 (Figure 4a), which then remained relatively stable. Since the majority of the added material was located below the P3 measurement point, the pressure drop between P2 and P3 increased sharply after each addition of material, followed by a gradual decrease as the reaction progressed, as shown in Figure 4b. This decrease in the pressure drop indicates that some catalysts were blown out of the main bed during the reaction. The pressure in the returning chamber (Figure 4d) also gradually decreased after the addition of 50 kg of material, suggesting that the material in the downcomer was gradually depleted over time. The reduction in the pressure drop between P2 and P3 further indicated that the material was continuously being lost from the reaction system during the process. As more catalysts were added, both the main bed and returning chamber pressures gradually increased, allowing more catalysts to participate in the circulation and reaction within the main bed.

Figure 4.

Changes in pressure drops (a–e) and methane conversion rate (f) under different catalyst loading amounts (Q: 100 N m³/h, T: 550 ± 10 °C, particle size: 380–830 μm, and CH₄ concentraion: 2000 ppm).

Figure 4f shows the methane conversion rates under different catalyst loading amounts. As observed, the methane conversion rate generally increased with increasing amount of catalyst, but the change was not significant. At a loading of 50 kg, the conversion rate was 80%, and it reached a maximum of 91% when the loading was increased to 60 kg. Further addition of the catalyst resulted in only slight changes in methane concentration, with the conversion rate remaining around 90%. These results indicate that a catalyst loading of 50–60 kg is sufficient to achieve a 90% methane conversion rate. It is important to note that the cyclone separator used in this experimental setup has a relatively low separation efficiency, leading to the continuous loss of fine particles from the reactor during operation (as shown in Figure 4b). Therefore, the actual amount of catalyst participating in the reaction within the main bed was lower than the initial loading amount, which is also supported by the material balance discussed in the following section. In practical applications, improving the separation efficiency of the cyclone separator would allow more material to participate in the catalytic reaction, potentially achieving a methane conversion rate of over 90% with less catalyst and a higher space velocity.

3.4. Effect of CH₄ Concentration

Figure 5 illustrates the variation in the methane conversion rate with methane concentration. During the reaction, the inlet methane concentration was increased every hour. The methane conversion rate showed a decreasing trend as the methane concentration increased. When the methane concentration was 0.1 vol %, the conversion rate reached 90%, whereas at a methane concentration of 0.45 vol %, the conversion rate drops to 78%. In practical reactions, when methane concentration fluctuates, the methane conversion rate can be improved by increasing the catalyst circulation rate, thereby enhancing the catalyst bed inventory to improve the treatment capacity.

Figure 5.

Methane conversion rate with concentration variation (Q: 100 N m³/h, T: 550 ± 10 °C, particle size: 380–830 μm, and catalyst loading: 60 kg).

3.5. Effect of Catalyst Particle Size

Experiments were conducted to examine the effect of different catalyst particle sizes on the methane conversion rate under the same operating conditions. At an airflow rate of 100 N m³/h, preliminary tests revealed that a significant portion of particles smaller than 380 μm were blown out of the furnace and could not be effectively captured by the cyclone separator. This resulted in a noticeable decline in the catalyst amount participating in the reaction within the main bed over time, leading to a significant reduction in the methane conversion rate (as shown in Figure 6a). Consequently, it was difficult to compare the catalytic performance of these small particles with that of larger ones. Therefore, in this study, particle sizes larger than 380 μm were selected for analysis to evaluate their effect on methane conversion rate, i.e., 380–830, 830–2000, and >2000 μm.

Figure 6.

(a) Trend of methane conversion rate changing with flow rate, the catalyst loading amount was 60 kg with a particle size of smaller than 380 μm, the gas flow rate Q was increased from 50 to 100 N m³/h gradually while the operation temperature was maintained at 550 ± 10 °C; (b) methane conversion rate under different particle sizes, Q: 100 N m³/h, T: 550 ± 10 °C, and catalyst loading: 60 kg.

As seen in Figure 6b, the methane conversion rate for all three particle sizes exceeded 90%, with only minor differences. The conversion rate remained relatively stable, indicating that particle sizes larger than 380 μm have little influence on the reaction performance.

To further verify the effect of particle size, additional experiments were conducted in a laboratory-scale fluidized bed reactor with an inner diameter of 13 mm. The tested particle size range was 75–850 μm. The CH₄ concentration was maintained at 0.2 vol % for all of the tests. It is worth noting that different particle sizes have different V_mf; thus, to maintain a similar flow regime for different particles, the same U/V_mf was used, where U is the superficial gas velocity in the reactor. To further maintain the same gas–solid contact time at the static state, different catalyst amounts were used for particles with different sizes. The loading amount of the catalyst and the superficial gas velocity are listed in Table 5. The results are shown in Figure 7. The findings indicate that under the same residence time (static state), the size of the catalyst has little effect on methane conversion efficiency, with T₉₀ remaining essentially unchanged. This is consistent with the results observed in the pilot-scale tests. In gas–solid heterogeneous catalysis, smaller particle sizes typically provide greater surface area and more active sites, which generally promote catalytic reactions. However, in this study, the catalytic activity did not vary significantly within the tested particle size range, suggesting that even the largest particles have sufficient active sites for methane reactions. Thus, the catalytic performance is not sensitive to particle size variations in this case.

Table 5. Operation Conditions in Lab-Scale Fluidized Bed Reactor with Different Particle Sizes.

particle size (μm)	Vmf (m/s)	Q (mL/min)	catalyst amount (g)	static residence time (s)	T90 °C
425–850	0.1719	812.0	3.00	∼0.18	520
250–425	0.0622	882.2	3.17		530
180–250	0.0253	359.1	1.29		519
150–180	0.0149	211.5	0.76		520
125–150	0.0104	146.9	0.52		536
100–125	0.0069	98.5	0.35		524
75–100	0.0042	178.4	0.64		524

Figure 7.

Influence of particle size on reaction in lab-scale fluidized bed.

3.6. Stability Tests

In practical applications, the stability of the catalyst significantly affects the technoeconomic feasibility of the process. To investigate the catalytic stability of the catalyst in a circulating fluidized bed reactor, catalysts in the size range of 830–2000 μm were used for the stability test. At the initial stage of the reaction, 60 kg of catalyst was added, with an additional 5 kg of catalyst added each time to maintain a stable methane conversion rate.

Figure 8 shows the variation in methane concentration over a 78 h period. The initial methane conversion rate was 97.4%. But as the reaction progressed, the conversion rate gradually decreased. After 22 h, the methane conversion rate dropped below 90% for the first time. After adding more catalyst, the conversion rate increased significantly. However, when the cumulative catalyst amount reached 100 kg, further catalyst additions showed little effect on increasing the methane conversion rate, though it remained above 85%. In the early stages of the reaction, the gradual decline in methane conversion was primarily due to the loss of catalyst in the main bed. The fluidized bed reactor subjected the catalyst to significant collisions and wear, causing the formation of fine particles that were not effectively captured and involved in the reaction, which resulted in a decrease in efficiency. As the reaction continued, it can be found that even with additional catalyst supplementation, it became difficult to restore the conversion rate to above 90%. This is mainly due to the catalyst deactivation as will be discussed in the following characterization section.

Figure 8.

Catalyst stability test.

The loss of catalyst during operation can be evidenced by the variation in pressure drop over the 78 h test period shown in Figure 9. As the reaction progressed, the pressure drop between P2 and P3 in the main bed gradually decreased, and the pressure drops in the fluidization chamber and returning chamber of the downcomer also declined (data not shown). This indicates that the amount of material in the dense phase zone was gradually reduced, and the downcomer failed to maintain a stable inventory of material. Fine particles left the system with the flue gas, reducing the amount of material returning to the main bed for participation in the reaction. After continuous material replenishment, the amount of material in the main bed increased, and the pressure drop rose accordingly. By the end of the experiment, the pressure drop in the main bed was approximately the same as when the initial 60 kg of material was added, suggesting that the remaining material in the furnace was close to the initially added amount.

Figure 9.

Variation in pressure drop of (a) P2–P3 and (b) P3–P4 during the 78 h test.

It is noteworthy that the pressure difference in the upper section of the furnace (P3–P4) initially remained stable at around 0.3 kPa but later decreased to 0.15 kPa, which is comparable to the empty bed pressure drop under hot conditions (see Figure 2). Despite the furnace height is approximately 19 m in this pilot-scale reactor, the reaction mainly occurred within the dense phase region, which is located within 3 m from the air distributor, while the upper portion of the furnace did not form an effective particle concentration distribution. This led to an underutilization of the vertical space. In the design and operation of actual reactors, improving the efficiency of the cyclone separator at the furnace outlet is essential to ensure that more fine particles are effectively recovered and returned to the main bed to participate in the reaction. Establishing a certain concentration of material in the upper section of the furnace would allow for better utilization of the furnace height, which can increase gas–solid contact time and further enhance the methane conversion efficiency.

A mass balance analysis was performed using the gravimetric method after 78 h of operation, with the results shown in Table 6 and Figure 10. After the test, the material collected from the main bed of the reactor, the bag filter, and the fly ash deposit weighed 66.6, 17.5, and 0.6 kg, respectively. The amount of material remaining in the reactor was close to the initial amount added, which is consistent with the pressure drop results, indicating that only 62.8% of the catalyst added was actively involved in the reaction. Fly ash accounted for 21.37% of the total collected material, suggesting that a significant portion of the catalyst was blown out of the reactor. In total, 84.7 kg of material was collected, while the total catalyst input was 105 kg, resulting in a material loss rate of 19.33%. The likely causes of this loss include the fact that some of the catalyst particles were too fine to be captured by the cyclone separator and the subsequent bag filter, allowing them to escape from the system. Additionally, there may have been dead zones within the experimental system, where reaction products could not be fully recovered. From the particle size distribution of the catalyst shown in Figure 11, one can find that the particles larger than 830 μm accounted for 70.3% of the total weight after 78 h of operation, while particles smaller than 830 μm made up 29.7%, indicating that significant wear occurred during extended operation. In future catalyst design and preparation, it will be essential to further improve the wear resistance of the catalyst to reduce losses during operation.

Table 6. Recovery of Catalysts after Reaction.

total catalyst amount (kg)	initial size (μm)	Q (m3/h)	catalysts in the reactor (kg)	catalysts in the filter (kg)	catalysts in the fly ash deposit (kg)	recovered amount (kg)	lossa (%)	catalysts out of the reactorb (%)
105	830–2000	100	66.6	17.5	0.6	84.7	19.33	36.57

^a(Total catalyst amount – Recovered amount)/Total catalyst amount *100%.

^b(Total catalyst amount – Catalysts in the reactor)/Total catalyst amount *100%.

Figure 10.

Particle size distribution of the catalyst in the main bed after reaction.

Figure 11.

(a) X-ray diffraction (XRD) of the fresh catalyst and the catalyst after 78 h stability test and (b) catalytic activity of pure Mn₂O₃ and MnO₂.

3.7. Characterizations of the Catalyst before and after the Stability Test

As can be seen from Figure 11, for the fresh catalyst, the major component observed was SiO₂ with distinct diffraction peaks at 20.8, 26.6, 36.5, and 50.5°. Yet the diffraction peaks for active components of Mn and Fe as listed in Table 1 were not observed, which suggests that Mn and Fe were likely well dispersed in the natural ore or presented mainly as amorphous phase. After the 78 h reaction, the diffraction peaks of Mn₂O₃ and Fe₂O₃ appeared, indicating that the phase transformation occurred for portion of the manganese during the reaction, as will also be supported by the following XPS analysis. As illustrated in Figure 11b, the activity of Mn₂O₃ was much lower than that of MnO₂ (T₉₀ = 795 vs 505 °C). The appearance of Mn₂O₃ after the reaction is most likely the main reason for the gradual decreased catalytic activity during the reaction.

Figure 12 presents the XPS spectra of the polarized peaks of O 1s and Mn 2p before and after the stability test. The deconvoluted two peaks for Mn 2p_1/2 at around 641.9 and 642.6 eV can be ascribed to Mn(III) and Mn(IV), respectively.^17,18 For the fresh catalyst, Mn(IV) accounted for ∼70.6%, which indicates that Mn was mainly presented as MnO₂. After the reaction, however, Mn(III) amount increased from 29.4 to 63.1%, suggesting the transformation of MnO₂ to Mn₂O₃, which is consistent with the XRD result. The O 1s spectra shown in Figure 12b can be deconvoluted to three characteristic peaks, corresponding to lattice oxygen (O_I), adsorbed oxygen (O_II), and adsorbed water or carbonates (O_III).^17,19 The ratio of adsorbed oxygen to lattice, O_II/O_I, can be used to indicate the surface oxygen vacancies, which are considered active during the catalytic oxidation of CH₄. On fresh catalyst, the O_II/O_I was 2.31, higher than that of 1.16 for the catalyst after the 78 h reaction. This indicates the decrease in lattice oxygen activity, which also contributed to the activity decline during the long-term operation.

Figure 12.

XPS spectra of the catalyst before and after reaction: (a) Mn 2p and (b) O 1s.

The morphology of the catalyst before and after the reaction is shown in Figure 13. For the fresh catalyst, flake-shaped particles were found to be distributed on the surface, which leads to a relatively high Brunauer–Emmett–Teller (BET) surface area of 56.6 m²/g as shown in Figure 14. The isothermal adsorption–desorption plot presented a hysteresis loop in the P/P₀ range of 0.4–1.0, indicative of a typical IV–H3 adsorption curve according to the IUPAC classification. This indicates the irregular distribution of the pore structures, which are presented mainly in slit, crack, and/or wedge structures. This is inline with the SEM observation where flake-shaped structures were observed on a fresh catalyst surface. After the reaction, slight agglomeration was observed with the disappearance of some flakes when comparing Figure 13a,c. The specific surface area decreased from 56.6 to 37.9 m²/g, indicating the collapse of some pores, which may cause the loss of active sites for CH₄ activation.

Figure 13.

Surface morphology of the catalyst (a, b) before and (c, d) after reaction.

Figure 14.

BET-specific surface area analysis for catalyst before and after reaction.

4. Conclusions

In this study, to provide practical guidance for tackling the ventilation air methane (VAM) emission in coal mines, a cost-effective natural ore-based catalyst was employed to perform catalytic oxidation of ultralow concentration methane on a pilot-scale circulating fluidized bed reactor for the first time. The experiments explored the effects of the reaction temperature, catalyst bed inventory, catalyst particle size, and methane concentration on the methane conversion efficiency. Additionally, a 78 h stability test was conducted to provide a basis for the development of cost-effective catalyst materials and the application of fluidized bed technology in coal mine VAM treatment. The main conclusions of the study are as follows:

• (1)The catalytic efficiency of methane increased with rising reaction temperature, with T₉₀ being ∼543 °C. Beyond 630 °C, the conversion rate showed no significant change.
• (2)An increase in methane concentration (0.1–0.45 vol %) led to a gradual decrease in conversion efficiency from 90 to 78%. In practical operations, increasing the catalyst circulation rate can improve the conversion rate to accommodate concentration fluctuations. Catalyst particle size had limited effect on methane conversion rate when gas–solid contact time was kept constant.
• (3)The 78 h stability test indicates that the methane conversion rate gradually decreased over time, maintaining above 90% during the first 22 h and above 85% when fresh material was supplemented online. Significant catalyst particle attrition occurred during the reaction, with 36.6 wt % of the material being blown out of the furnace, preventing it from participating in the methane catalytic reaction. Future studies should focus on improving the wear resistance of the catalyst particles and enhancing the recycling efficiency of fine particles.
• (4)After reaction at 550 °C, the active MnO₂ component in fresh ore was found to be partially converted to less active Mn₂O₃. Furthermore, the surface lattice oxygen decreased and slight agglomeration was observed with decreased surface area. These combined factors caused the activity decrease in the long-term operation. The stability of the natural ore as catalyst needs to be improved in further studies.

Supporting Information Available

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acsomega.4c10532.

• Comparison of the T₉₀ with the Mn and Fe catalysts reported in previous studies (Figure S1) and references used for comparison in Figures 3b and S1 (PDF)

The authors declare no competing financial interest.