Study on the Spontaneous Combustion Characteristics and CO-CH₄ Formation Law of Low-Rank Bituminous Coal Based on In Situ Infrared

Abstract

Low-rank bituminous coal exhibits strong chemical reactivity and a relatively high spontaneous combustion tendency due to its high volatile matter content. The relationship between chemical structure evolution and gas generation during the oxidation of low-rank bituminous coal was studied by temperature-programmed and in situ infrared spectroscopy. The results show that the reaction between aliphatic side-chain −CH₃/–CH₂– groups in the coal and O₂ molecules is the main reason for CO generation. The aliphatic side-chain group decreased by 80.57% in the stage of coal oxidation at 180–380 °C. During the high-temperature oxidation stage, the oxygen-containing functional groups CO/C–O–C in the coal are thermally decomposed to generate CO. The CH₄ formation in the process of coal oxidation is not affected by O₂ concentration, and it is more sensitive to temperature. The adsorbed gas in the residual coal is gradually released by heating during the low temperature oxidation stage. When the coal is oxidized to 260 °C, it thermally decomposes to produce CH₄. As the temperature continues to rise to 330 °C, the CH₄ formation rate increases exponentially. The thermal decomposition of the aromatic ring branched-chain structure Ar–C–H– is the main reason for the CH₄ generation during the spontaneous combustion of residual coal. The apparent activation energy E _a of coal with different O₂ concentrations during the slow oxidation stage at 60–100 °C shows that an 8% O₂ concentration can be used as a criterion for determining the critical O₂ volume fraction of low-rank bituminous coal spontaneous combustion in the Shendong Mining Area . Monitoring of spontaneous combustion of residual coal in goaf should pay attention to O₂, CO, and CH₄ at the same time, especially when the CH₄ concentration rises abnormally, which indicates that the spontaneous combustion of residual coal in goaf has entered the stage of high-temperature oxidation. Continuous air leakage into the goaf can lead to residual coal combustion, thereby endangering mine safety.

1. Introduction

Western China possesses abundant coal resources with favorable mining conditions. However, the region faces a relatively high risk of coal spontaneous combustion. The coal seams are characterized by shallow burial depths and form close-distance seam groups. In addition, the high-intensity mining makes the coal spontaneous combustion disaster seriously threaten the health and safety of workers during the mining process. During low-temperature oxidation, the heat generated by coal accumulates, gradually raising the temperature to the ignition point and leading to combustion. In the whole process of coal spontaneous combustion, the original chemical structure of coal changes from each other to generate index gas. In the process of beam-tube gas monitoring in the goaf of an underground coal mine (the goaf refers to the cavity left underground after coal is mined and transported); if not handled properly, it is easy to cause ground collapse and mine safety accidents). It has been found that CH₄ is detected in areas with high CO concentrations, and the produced CH₄ may also lead to CH₄ accumulation and induce coal spontaneous combustion and gas explosion symbiosis disaster.

Previous studies have analyzed the coal–oxygen reaction process from the perspective of oxidation kinetics and typically divide coal self-heating into three stages: low-temperature oxygen chemisorption, rapid temperature rise, and combustion. However, this stage division is to a large extent an empirical simplification, which is highly dependent on experimental conditions (such as heating rate, oxygen concentration, coal type, and particle size). The study found that the high-temperature oxidation stage of coal spontaneous combustion involves changes in free radicals, but this study did not clarify the specific types of free radicals, their concentration evolution trends, and their quantitative correlation with temperature. The previous work has preliminarily discussed the relationship between highly active carbon-containing free radicals in coal and spontaneous combustion reactivity, but there are some debatable conclusions. First of all, the free radical parameters measured in the low temperature oxidation stage and the changes in various free radical concentrations are classified as weakly correlated with the spontaneous combustion tendency. However, this judgment lacks statistical significance testing and effect size reporting, and may be limited by specific coal types, heating rates, or oxygen partial pressure conditions. In studies of the temperature change law of coal oxidation process by domestic and foreign scholars, it is found that the coal oxidation process at different stages shows obvious temperature rise differences. To precisely obtain indicator information during coal oxidation, research has focused on gas product formation patterns, with findings further enriching understanding in this field. Under specific experimental conditions, low-temperature oxidation experiments of coal in a methane-containing atmosphere were carried out by predecessors, and the changes in the concentrations of indicator gases with the oxidation process were investigated. The cumulative production of indicator gases showed a decreasing trend with increasing methane concentration. It is inferred that methane may have a competitive adsorption or dilution effect on the coal–oxygen composite reaction. During early coal spontaneous combustion, oxygen-containing functional groups such as carbonyl and carboxyl in coal are more active under ambient conditions and can decompose to produce CO and CO₂ gases, and this process is highly temperature-dependent. Temperature elevation accelerates complex decomposition to generate CO, while O₂ consumption remains low at this stage, resulting in a slow coal–oxygen reaction. Through tests of the evolution of key groups during coal oxidation under a low-gas atmosphere, it was found that low-temperature oxidation promotes the conversion of C–O– and CO into COO–, which is further oxidized to CO and CO₂. The gas release characteristics of coal samples with different metamorphic degrees in the process of spontaneous combustion and heating were investigated by temperature-programmed experiments. It was found that there were significant differences in the temperature of CH₄ detected for the first time in each coal sample, but the temperature threshold was affected by experimental conditions such as heating rate, oxygen concentration, and coal particle size, which could not be extrapolated to the actual working conditions. The formation of CH₄ was attributed to the electrophilic attack of O₂ on the alkyl side chain of the benzene ring. However, this one-step reaction description ignores the cascade process of free radical chain reactions in the coal macromolecular structure, and does not consider the influence of side-chain substituent type (such as −CH₃, −CH₂−) and the degree of aromatic ring condensation on the reactivity. Infrared analysis indicates that hydrocarbon gas generation during coal oxidation results from O₂ molecule attack on coal side chains, a process involving the formation of unstable peroxides and their decomposition into aliphatic hydrocarbon free radicals. Methane formation is attributed to methyl radicals capturing hydrogen from other groups. However, the feasibility of this hydrogen abstraction step at low temperatures and whether there is a competitive path from other precursors still need to be further clarified by thermodynamic calculations and isotope tracing experiments.

Nevertheless, the mechanisms underlying self-structural evolution and CO/CH₄ cogeneration during low-rank bituminous coal spontaneous combustionespecially the interactions among O₂/CO/CH₄ gases influenced by spontaneous combustion temperature and their impact on coal spontaneous combustionrequire further investigation. Therefore, by studying the chemical structure evolution during low-rank bituminous coal oxidation, this paper clarifies the formation characteristics of CO/CH₄ under the guidance of key groups in the spontaneous combustion process and explores the mechanisms by which spontaneous combustion temperature and O₂ concentration of residual coal in the goaf affect CO–CH₄ gas composition.

2. Coal Sample Selection and Experiments

2.1. Coal Sample Selection

The experimental coal sample was an original block sample collected from a western mining area. Prior to the experiment, the coal sample was sealed in a vacuum bag, transported back to the laboratory, and stored under a nitrogen atmosphere. The outer shell of the coal block was removed, and the coal core was taken and weighed. The initial weight was recorded, and the sample was placed in a vacuum drying oven at 40 °C. The coal sample was taken out and weighed every 12 h until the weight change was less than 1%, after which it was unsealed and used. Vitrinite reflectance and proximate analysis results are presented in Table .

1. Vitrinite Reflectance and Proximate Analysis Results of Coal Sample .

		Proximate analysis/%
Coal sample properties	Vitrinite reflectivity/%	FCad	Vad	Aad	Mad
Low-rank bituminous coal	0.54	55.26	33.90	5.28	5.56

^aFC_ad is the air-dried base-fixed carbon; V_ad is the air-dried base volatiles; A_ad is the air-dried base ash; and M_ad is the air-dried base moisture.

2.2. In Situ Infrared Spectroscopy Experiment

A Fourier transform infrared spectrometer requires the sample particle size to be below 200 mesh (0.075 mm). Under nitrogen protection, coal samples were crushed and screened to obtain pulverized coal with a particle size of 0.048–0.075 mm for in situ infrared spectroscopy testing. The test apparatus is shown in Figure . The resolution of this infrared spectrometer is better than 0.4 cm^–1, and the wavenumber accuracy is better than 0.0005 cm^–1 (standard deviation of 10 repeated measurements). Therefore, the accuracy of the experimental measurements is ensured.

• (1)Pulverized coal powder and KBr powder were separately spread in two sample cups and smoothed with a scraper.
• (2)Connecting the in situ reaction chamber, gas circuit, and water-cooling system. The infrared scanning range was set to 600–4000 cm^–1, with the scanning resolution of 4 cm^–1, and spectra were acquired at 60 s intervals. The spectra were processed using the Kubelka −Munk.
• (3)The sample cup containing KBr powder was first placed into the reaction chamber, and a background reference spectrum was collected under ambient-temperature conditions and saved. When the sample is loaded, the sample should be set at 1 mm lower than the sample surface by using a sample press stick.
• (4)The KBr sample cup was taken out, and the gas flow rate into the reaction chamber was adjusted with the mass flow controller. In order to prevent the sample from being blown away, the flow rate was set to 20 mL/min. The raw material gas purity specifications were 99.999%.
• (5)The programmed temperature-control system was activated, with an experimental temperature range of 30–430 °C and a heating rate of 1 °C/min. Infrared spectra were continuously acquired during coal sample heating.

1.

In situ FTIR testing system.

The Kubelka–Munk is a theory used to describe the optical behavior of systems containing tiny particles that scatter and absorb incident light. Based on the Beer–Lambert principle, the value of KM was proportional to the concentration of the targeted functional groups. Thus, all spectra were transformed using the Kubelka–Munk function eq :

$$KM=\frac{{(1-D_r)}^2}{2D_r}=\frac{2.303\epsilon }{s}·c$$ 1

where D _r is the diffuse reflectance, D _r = 1/10 ^A , A is the absorbance, ε is the molar extinction coefficient, s is the scattering coefficient, and c represents the concentration of the targeted functional group.

2.3. Temperature-Programmed Experiment

By measuring the amount of gaseous products generated during coal heating, combined with in situ reaction test results, the structural changes and the product formation law of the coal–oxygen composite reaction process were analyzed. First, 50 g of coal samples with a particle size of 0.18–0.38 mm were weighed and placed into the coal sample tank inside the temperature-programmed control tube. The experimental gas was controlled at a stable flow rate of 50 mL/min, and temperature-programmed experiments were carried out under different atmospheres (O₂ concentration was 0%, 4%, 8%, 12%, 16%, and 21%, respectively). Due to the temperature-resistance limit of the thermally conductive copper coal sample tank and the limitations on gas concentration in gas chromatographic analysis, the temperature-programmed experiment was conducted from 30 to 430 °C at a heating rate of 1 °C/min, with the tail gas composition detected by gas chromatography. The ZRJ-1 temperature-programmed furnace and GC-4000A gas chromatograph were used in the experiment, and the experimental instruments are shown in Figure . Before each experiment, a standard gas with a known concentration was selected as the test sample to calibrate the gas chromatograph. The relative standard deviation of the gas chromatograph is less than 0.5%.

2.

Temperature-programmed test system.

3. Results

3.1. Infrared Spectroscopy

The test results for the surface chemical structure evolution of the coal sample during heating are presented in Figure . Figure a shows that when the temperature of the coal sample rises to 300 °C under dry-air conditions, the infrared absorption peaks for wavenumbers below 2000 cm^–1 rises obviously. In contrast, Figure b shows that the chemical structure of the coal remains almost unchanged up to 400 °C in a pure nitrogen environment, and the infrared absorption peaks show an upward trend after the temperature increases by 400 °C, indicating that the coal sample itself begins to undergo pyrolysis. Under the condition of pure nitrogen, the volume of coal samples decreases and agglomerates, and brown tar-like substances are found on the walls of the reaction chamber, indicating that anaerobic pyrolysis reaction released macromolecular organic matter that was carried out of the reaction chamber by the airflow without participating in chemical reactions. In both dry-air and pure-nitrogen environments, the absorption peaks of the chemical functional groups on the surface of the coal sample did not change significantly before 130 °C, indicating that the coal-oxygen composite reaction and the pyrolysis reaction of the coal body itself were not strong at this stage.

3.

3D infrared spectra of the coal sample during heating process under different atmospheric conditions: (a) dry-air ; (b) N₂.

Infrared absorption spectra of coal samples are typically divided into aromatic ring branch structures (700–900 cm^–1), oxygen-containing group structures (1000–1800 cm^–1), aliphatic structures (2800–3000 cm^–1), and hydroxyl structures (3000–3600 cm^–1). The absorption peak positions of different groups for the experimental coal sample surface are fixed, as shown in Table . Due to the superposition of absorption peaks on the surface of coal samples, a single infrared absorption peak with high strength or large width is often the result of the superposition of absorption peaks from multiple groups. Therefore, it is necessary to perform peak fitting on the spectrum obtained from the test. Using PeakFit software, the Fourier self-deconvolution method was used to improve the resolution of absorption peaks. First, the spectrum of the selected area is baselined by using a straight line connection at the left and right points of the interval. Second, the second-derivative method was used to derive the spectrum, and the position of each absorption peak and the number of fitting peaks were preliminarily determined. The Peak Type is selected as “Spectroscopy”. Finally, the fitting peak type is selected as the Gauss-Lorentz combination, which allows the shape and width of the fitted absorption peak for each band to differ from the initially predicted absorption peak. Then, the mode of “Autoscan” begins to scan the absorption peaks. The absorption peaks of the original spectrum are segmented, and the fitted curve formed by the superposition of the final multiple absorption peaks is obtained by multiple fittings, and the peak-fitting processing of the infrared spectrum of the experimental test is completed. The whole spectrum peak-fitting process is completely processed by the software, and the fitting parameters are all consistent, which avoids interference from human factors during data processing. The infrared spectrum fitting results of coal samples are shown in Figure .

2. Absorption Peak Positions of Different Functional Groups.

Wave number/cm–1	Types of functional groups
3600	Free hydroxyl (−OH)
3375	Associate hydrogen bond (−OH)
2955, 2925	Aliphatic stretching vibration (−CH3/–CH2−)
1680, 1730	CO vibration of aldehyde group, carboxyl group and ketone carbonyl group
1600	Aromatic nucleus (CC)
1275	Aromatic structure C–O–C

4.

Curve fitting of the infrared spectrograms for the coal sample: (a) Aromatic structures; (b) oxygen-containing functional groups; (c) aliphatic structures; (d) hydroxyl groups.

3.2. Evolution Law of Chemical Structure

The fitted peak areas of the infrared spectra were summed to obtain semiquantitative analysis results for each group, as shown in Figures –.

5.

Evolution law of hydroxyl groups in the coal sample: (a) Under dry-air atmospheric conditions ; (b) under N₂ atmosphere conditions; (c) semiquantitative analysis.

8.

Evolution law of aromatic groups in the coal sample: (a) Under dry-air atmosphere conditions; (b) under N₂ atmosphere conditions; (c) semiquantitative analysis.

3.2.1. Hydroxyl–OH

As shown in Figure , the number of hydroxyl groups in the coal sample gradually decreases with an increase in the temperature. According to the results of proximate analysis, the original moisture content of the coal sample is high and the water exists within the coal body in the form of free water and bound water. The coal sample is heated in dry air and in a pure nitrogen environment, and the internal water gradually evaporates, leading to a progressive decline in hydroxyl group abundance.

3.2.2. Aliphatic Side Chains −CH₃/–CH₂–

It can be seen from Figure that the aliphatic side-chain −CH₃/–CH₂– group has strong activity and is easy to react with O₂ or break by heat. In a dry air environment, the aliphatic side-chain groups decreased by (10 ± 1)%, (80 ± 5)%, and (5 ± 1)% in the temperature ranges of 30–180 °C, 180–380 °C, and 380–430 °C, respectively, indicating that the reaction of coal samples with O₂ began to accelerate at 180 °C. Until the temperature of 480 °C, the aliphatic side chains were almost exhausted, and the reduction of aliphatic side-chain groups under aerobic conditions was the result of the combined effects of coal oxidation reaction and self-thermal decomposition. In a pure nitrogen environment, the aliphatic side-chain groups decreased by (5 ± 1)%, (15 ± 2)%, and (14 ± 2)% at the heating stages of 30–180 °C, 180–380 °C and 380–430 °C, respectively. After 380 °C, the pyrolysis reaction of the coal samples began to accelerate.

6.

Evolution law of aliphatic side-chain groups in the coal sample: (a) Under dry-air atmospheric conditions ; (b) under N₂ atmosphere conditions; (c) semiquantitative analysis.

3.2.3. Active Oxygen-Containing Groups CO/Aromatic Structure C–O–C

As observed in Figure , in the dry-air environment, the reduction of aliphatic side-chain −CH₃/–CH₂– groups is accompanied by a marked enhancement of CO stretching vibration absorption peaks (belonging to aldehyde, carboxyl, and ketone carbonyl groups) in the 1750–1680 cm^–1 infrared spectral range. Additionally, the C–O–C absorption peak of the oxygen-containing aromatic structure near 1275 cm^–1 also increases significantly, indicating that a large number of active oxygen-containing groups are generated during heating, providing favorable conditions for CO indicator gas production. In the dry-air environment, the oxygen-containing groups increased by (28 ± 2)% and (65 ± 5)%, respectively, in the heating stages of 30–180 °C and 180–380 °C. In the heating stage of 380–430 °C, the increase in oxygen-containing groups decreased by (43 ± 2)%. In contrast, in the pure nitrogen environment, the oxygen-containing groups on the surface of the coal sample always rise slowly, showing a linear relationship with the temperature, which increases by (21 ± 2)% at the final temperature of 430 °C. This shows that in the process of coal spontaneous combustion, the carbon atom groups are broken and combine with O₂ atoms to form more active oxygen-containing groups. The transformation of various groups in coal is always in a dynamic process. After reaching a certain temperature condition, the active groups are rapidly consumed by chemical reactions.

7.

Evolution law of active oxygen-containing groups in the coal sample: (a) Under dry-air atmosphere conditions; (b) under N₂ atmosphere conditions; (c) semiquantitative analysis.

3.2.4. Aromatic Ring Branched-Chain Structure Ar–C–H–

Absorption peaks in the 700–900 cm^–1 infrared spectral range are primarily attributed to the out-of-plane bending vibration of the aromatic ring branched structure Ar–C–H–. The condensation degree of aromatic rings in low-rank bituminous coal is lower, the aromatic clusters are smaller, and the substituents are more. Therefore, the absorption peak in this region is more obvious than in medium and high-rank coal. It can be seen from Figure that in the dry-air environment, 230 °C is the key temperature for the transformation of coal structure. Before and after this temperature point, the aromatic ring branched structure Ar–C–H– decreased by (12 ± 2)% and (83 ± 5)%, respectively. At the final temperature of 430 °C, Ar–C–H– was almost completely consumed. In contrast, in the pure nitrogen environment, due to the lack of O₂ molecules, the coal sample only has a pyrolysis process during the heating process of 30–430 °C, and the number of Ar–C–H– branched chains in the aromatic ring is significantly lower than that in the dry air environment. At the final temperature of 430 °C, it only decreased by (28 ± 2)%, and the coal sample remained in its original state and no residue was generated after the experiment.

3.3. CO/CH₄ Gas-Generation Law

The formation law of CO and CH₄ with temperature are shown in Figure . During the low-temperature oxidation stage, the coal sample is in the process of heat storage and heating up. After 100 °C, the CO production under different O₂ concentration conditions begins to show obvious differences. The O₂ concentration is positively correlated with the CO production. When the coal body temperature rises to 230 °C, the CO production shows an exponential growth trend. In contrast, in the anaerobic environment, a higher temperature threshold (above 380 °C) is needed to generate a large amount of CO. However, since this process does not involve O₂ under anaerobic environment, therefore, in addition to the oxidation reaction, the pyrolysis of oxygen-containing functional groups in coal is also one of the reasons for the spontaneous combustion of residual coal to generate CO under high temperature environment.

9.

Gas products during the spontaneous combustion process of coal samples: (a) CO generation; (b) CH₄ generation.

In contrast, from Figure b, it can be seen that almost no CH₄ is formed during the early low-temperature oxidation heating and early high-temperature spontaneous combustion stages. However, when the temperature exceeds 260 °C, the CH₄ formation rate begins to increase significantly. After 330 °C, the CH₄ formation rate rises exponentially, and CH₄ formation is less affected by the O₂ concentration. The CH₄ formation trends under different O₂ concentration conditions are nearly identical.

Under different O₂ concentration conditions, the apparent activation energy E _a reflects the minimum energy required for the coal–oxygen composite reaction. A lower apparent activation energy indicates that less energy is needed for the coal–oxygen reaction, making the composite reaction more likely to occur. Based on the Arrhenius equation and the reaction rate equation, the quantitative relationship between O₂ consumption rate and apparent activation energy can be established as follows:

$$V_{{\mathrm{O}}_2}=A{C}_{{\mathrm{O}}_2}^n\exp (-\frac{E_{\mathrm{a}}}{RT})$$ 2

where $V_{{\mathrm{O}}_2}$ is real O₂ consumption rate, mol/(m³·s); A is the pre-exponential factor, s^–1; n is the order of reaction, for spontaneous combustion process of coal, the reaction order is 1; $C_{{\mathrm{O}}_2}$ is O₂ content in the reaction gas, mol/m³; E _a is the apparent activation energy, J/mol; R is the ideal gas constant, R = 8.314 J//(mol·K); T is the thermodynamic temperature, K.

During the test, the coal sample is loosely packed in the sample tank, achieving uniform heating and a small temperature gradient within the tank. Therefore, the temperature of the coal sample in the tank can be considered uniform. Integrating over the coal sample in the tank:

$${\int }_0^L\frac{S}{Q}·A\exp (-\frac{E_{\mathrm{a}}}{RT})dx={\int }_{C_{in}}^{C_{out}}(-\frac{1}{C_{{\mathrm{O}}_2}})dC$$ 3

where L is the length of the coal sample tank, m; S is the cross-sectional area of the coal sample tank, m²; Q is the gas flow, m³/s; C _in is the O₂ concentration at the inlet end of the coal sample tank, mol/m³; and C _out is the O₂ concentration at the outlet end of the coal sample tank, mol/m³.

After the indefinite integral of the above formula is obtained and simplified, the natural logarithm on both sides of the formula is taken again, and it can be obtained that

$$\ln \left[\ln \left(\frac{C_{in}}{C_{out}}\right)\right]=-\frac{E_{\mathrm{a}}}{R}·\frac{1}{T}+\ln \left(\frac{VA}{Q}\right)$$ 4

The O₂ concentration and temperature data measured in the experiment were expressed in the form of a scatter plot of −1/T versus In­[In­(C _in /C _out )], and linear fitting of the data points was performed to obtain the apparent activation energy for the slow oxidation stage (60–100 °C) and the accelerated oxidation heating stage (100–180 °C), the fitting results are shown in Table . As shown in Figure , the apparent activation energy E _a for coal samples under different O₂ concentration conditions in the 60–100 °C range is 83.51–118.77 kJ/mol, while in the 100–180 °C range it is 43.51–67.47 kJ/mol. The apparent activation energy in the slow oxidation stage is significantly higher than that in the accelerated oxidation heating stage.

3. Fitting Results for Apparent Activation Energy E _a of Coal Samples.

	Oxidation stage
	60–100 °C		100–180 °C
O2concentration	E a‑1 (kJ/mol)	R 2	E a‑2 (kJ/mol)	R 2
4%	118.77	0.9910	43.51	0.9255
8%	115.35	0.9856	61.22	0.9979
12%	104.41	0.9505	66.57	0.9995
16%	93.92	0.9910	67.47	0.9972
21%	83.51	0.9923	67.07	0.9955

10.

Apparent activation energy of coal samples under different O₂ concentrations: (a) 4% O₂ concentration; (b) 8% O₂ concentration; (c) 12% O₂ concentration; (d) 16% O₂ concentration; (e) 21% O₂ concentration; (f) relationship between O₂ concentration and activation energy.

In the slow oxidation stage, the coal sample surface temperature is low, and many original chemical functional groups have not yet been activated, requiring more energy to react with O₂. Additionally, the collision frequency between the functional groups involved in the reaction and O₂ molecules is higher, resulting in a larger apparent activation energy at this stage. As the temperature increases, the surface functional groups of the coal sample become fully activated and begin to release substantial heat. At this point, active functional groups decrease due to consumption, leading to a gradual reduction in the activation energy required for the coal–oxygen reaction. When the crossing-point temperature is reached, under favorable heat-storage and O₂ supply conditions, the coal sample can sustain the spontaneous combustion process through its own oxidation and heat release. Comparing the relationship between the apparent activation energy of coal samples and the change in O₂ concentration during the slow oxidation stage (60–100 °C), as shown in Figure (f), it is evident that the apparent activation energy in the early oxidation stage is influenced by O₂ supply conditions. Combined with the CO release pattern during the low-temperature oxidation stage, when the O₂ concentration exceeds 8%, the CO formation temperature decreases from 70 to 60 °C, and the apparent activation energy of the coal sample decreases with increasing O₂ concentration. This indicates that an O₂ concentration of 8% can serve as a reference value for the critical O₂ volume fraction for low-rank bituminous coal spontaneous combustion in the western region.

4. Discussion

The trend of CO generation during the coal sample heating process is consistent with the trend of aliphatic side-chain −CH₃/–CH₂– consumption. Moreover, under anaerobic conditions, CO generation is delayed, indicating that the cause of CO formation is not solely the coal–oxygen composite reaction but also the pyrolysis reaction of functional groups in the coal sample. CO is the first indicator gas produced during coal spontaneous combustion. Meanwhile, the aliphatic side chain is the most active group in coal and preferentially reacts with O₂. Therefore, the primary reason for CO production during coal sample heating is the reaction of the aliphatic side chain with O₂. Secondarily, the active oxygen-containing groupz CO/aromatic structure C–O–C undergo pyrolysis under high-temperature conditions, cleaving to release CO molecules. Thus, the CO formation pathway during coal spontaneous combustion is as follows: from the low-temperature to medium-temperature oxidation stage, the coal–oxygen composite reaction dominated by aliphatic side chains −CH₃/–CH₂– generates CO. Then, from the medium-temperature to high-temperature oxidation stage, the thermal decomposition reaction of oxygen-containing functional groups CO/C–O–C provides an additional CO formation process.

Predecessors have carried out relevant research on the law of CH₄ generation in the process of low-temperature oxidation of coal (temperature below 200 °C). It is generally believed that in the early stage of low-temperature oxidation, with the accumulation of energy in coal, the chemical reaction proceeds toward the process of gas product generation. The bond C–C in the −CH₂–CH₃ group becomes longer, which is conducive to the development of CH₄ generation. Finally, the bond C–C breaks, and −CH₃ is the main reason for the formation of CH₄ in coal, while the aromatic ring is very stable at low temperature and does not break. From Figure b, in dry air and pure N₂ environments, the formation pattern of CH₄ is nearly identical, and the presence of O₂ does not affect CH₄ formation, indicating that CH₄ generation results from the pyrolysis of functional group structures in the coal sample, with a weak dependence on O₂. As temperature increases, CH₄ production steadily rises and shows strong temperature sensitivity. During the early low-temperature oxidation stage of coal spontaneous combustion, nondesorbable gas adsorbed inside the coal body is continuously released as the temperature rises. Consequently, a small amount of CH₄ is detected early in the experimental process. The energy required for pyrolysis at this stage has not yet accumulated to a sufficient level, so the amount of CH₄ generated is very small. The CH₄ formation rate begins to increase significantly after 260 °C, and after 330 °C, it rises exponentially.

The dynamic changes of the oxygen-containing groups CO/C–O–C over the 30–430 °C temperature range do not parallel the CH₄ formation pattern in both dry air and pure N₂ environments, indicating that CH₄ formation is not related to these oxygen-containing groups. The variation trend of the aromatic ring-branched structure Ar–C–H– was found to match the CH₄ formation trend. Low-rank bituminous coal, with its low degree of metamorphism, contains more aromatic ring-branched structures Ar–C–H– due to its lower aromatic ring condensation degree, indicating that CH₄ is not synthesized by oxidation reactions but is released through the pyrolysis reaction of the coal sample.

The CH₄ formation pathway during coal spontaneous combustion is as follows: from the low-temperature to the medium-temperature oxidation stage, adsorbed CH₄ is desorbed after the coal body absorbs heat; then, from the medium-temperature to the high-temperature oxidation stage, thermal decomposition of the aromatic ring branched-chain structure Ar–C–H– generates large amounts of CH₄. From the above analysis, it can be seen that the specific void conditions in the goaf provide favorable heat storage conditions for residual coal spontaneous combustion. During spontaneous combustion, the coal body accumulates heat due to oxidation reactions, creating a high-temperature environment in localized areas of the goaf. During the oxidative heating process of low-rank bituminous coal, in addition to continuous CO gas generation, CH₄ gas is also produced. In the low-temperature oxidation stage, the nondesorbed atmospheric gas adsorbed in the residual coal is gradually released by heat. After entering the high-temperature oxidation stage, the coal body undergoes thermal decomposition to produce CH₄, resulting in a significant increase in CH₄ concentration in the coal spontaneous combustion area of the goaf.

In coal mining operations in western China, when monitoring residual-coal spontaneous combustion in goafs, it is necessary to track changes in the O₂ and CH₄ gas indicators in addition to the CO indicator. When the O₂ concentration in the goaf rises above 8%, special attention should be paid to changes in gas components in that region. The appearance of CO in the monitored gas indicates that a natural temperature rise of residual coal in the region has occurred. Due to the low gas content of coal seams in the western region, the amount of gas desorbed from residual coal in the goaf is small. If the CH₄ concentration rises in the monitored gases, it indicates that residual-coal spontaneous combustion has entered the accelerated oxidation stage after 180 °C. When the CH₄ concentration rises rapidly within a short period, it indicates that the residual coal has entered the high-temperature spontaneous combustion stage after 330 °C. If the air leakage intensity in the goaf is high, the residual-coal temperature will rise rapidly until combustion occurs, simultaneously posing a risk of CH₄ accumulation.

5. Conclusion

In this study, the relationship between changes in chemical structure and gas generation during the process of spontaneous combustion of low-rank bituminous coal is discussed from two aspects, the microstructure and the macroscopic products of coal oxidation and heating, and the following main conclusions are obtained:

• (1)In a dry-air environment, low-rank bituminous coal consumes 80.57% of its aliphatic side-chain groups and 83.58% of its aromatic groups during the high-temperature oxidation stage, releasing large amounts of CO gas. Concurrently, the coal sample’s pyrolysis reaction begins to accelerate after 380 °C. Carbon atom groups in the coal are cleaved and combine with O₂ atoms to form more active oxygen-containing groups, whose thermal decomposition generates CO gas.
• (2)CH₄ formation during coal oxidation heating is unaffected by O₂ concentration but is highly temperature sensitive. CH₄ begins to appear clearly after 260 °C, and its formation rate increases exponentially after 330 °C. The reasons for CH₄ production during coal sample heating are as follows: in the low-temperature to medium-temperature oxidation stage, adsorbed CH₄ is desorbed after the coal sample absorbs heat; after the high-temperature oxidation stage, thermal decomposition of the aromatic ring branched structure Ar–C–H– generates substantial amounts of CH₄.
• (3)The apparent activation energy of low-rank bituminous coal in the slow oxidation stage (60–100 °C) ranges from 83.51 to 118.77 kJ/mol. When the O₂ concentration exceeds 8%, the apparent activation energy decreases significantly with increasing O₂ concentration, indicating that an 8% O₂ concentration can serve as a criterion for determining the critical O₂ volume fraction for low-rank bituminous coal spontaneous combustion.

The authors declare no competing financial interest.