1. Introduction
Although significant progress has been made in safety management, coal mining worldwide still faces an enormous challenge from the spontaneous combustion of coal. In Queensland, Australia, there is an average of at least one spontaneous combustion-related accident per year, resulting in coal miners evacuating the mine and, in some cases, even the closure of the mine [
1]. From 1990 to 1999, about 17% of the 87 reported fires in underground coal mines in the United States were caused by CSC [
2]. In India, about 80% of mine fires are caused by CSC [
3]. In China, more than 50% of coal mines have experienced spontaneous combustion accidents, and it is estimated that a few key coal mines have experienced 360 spontaneous combustion fires each year [
4]. One-third of the 254 mine fires reported between 1970 and 1990 in South Africa were caused by the spontaneous combustion of coal [
5]. CSC causes massive resource and property losses, and adversely impacts society.
The safety management of spontaneous combustion in low-rank coal goaf is particularly difficult. On the one hand, the CSC risk of low-rank coal is generally high [
6], and mainly falls into classes Ⅰ and Ⅱ of spontaneous combustion. On the other hand, the early warning indicators, especially those from CO, are vulnerable to interference from multiple sources [
7,
8], such as the coal cutting effect, blasting, the ambient temperature oxidation (ATO) of coal, etc. The technical problem of how to ensure an early warning of CSC in low-rank coal goaf needs to be urgently solved by coal mine researchers.
CO plays an important role in CSC forecasting. It has a high sensitivity and low application cost, and the relationship between its concentration or change rate and temperature can provide an early warning of CSC. The main coal mining countries, including China, America, and Australia, regard CO as the most crucial gas for the early warning of CSC [
9]. Due to the complexity of goaf scenarios, CO production of CSC is easily effected by many factors. Through experimental research, on-site monitoring, density functional theory (DFT), and other methods, scholars have investigated the influence of an oxygen-poor atmosphere [
10,
11], moisture contents [
12,
13], and heating rates [
14] on CO output characteristics and their corresponding mechanisms, providing the basis for the scientific management and control of CSC.
In the massive process of mining low-rank coal seam, which utilizes a fully mechanized top-coal caving technology in China, the tailgate corner often presents a high CO concentration that exceeds the established limit (24 ppm in China), but without other warning signs of CSC, such as temperature rise, coal wall sweat, etc. For example, the daily average CO concentration of the 1305lignite-working face of Dananhu No.1 mine, Xinjiang (NW China), is over 100 ppm; that of the 12,206 long-flame coal working face of Shangwan mine, Inner Mongolia (NE China), is between 60 and 90 ppm; and that of the 52,214 non-stick coal working face of Yujialiang mine, Shaanxi (Central China), is above 70 ppm. Even if taking a significant number of CSC-fighting measures, including pressure equalizing and plugging, and the injection of nitrogen into the goaf, there is little effect. The fact that the CO concentration in the tailgate corner continuously exceeds the limit creates great difficulties for the early warning of CSC in goaf. Tang [
15] and Li [
16] took the lead in proving that ATO is the main CO source in low-rank coal mines by means of on-site monitoring and experimental research. Luo [
17] and Wu [
18] carried out an ATO experiment with lignite and long-flame coal, and proved that low-rank coal can oxidize and release CO at room temperature.
The ATO of coal usually refers to the spontaneous reaction between coal and oxygen to produce CO from 10 to 30 °C, and mainly occurs in low-rank coal, such as lignite, long-flame coal, and non-stick coal [
8]. In their research of coal’s ATO, Zhang [
18] investigated O
2 consumption and CO production characteristics in a confined space, and found that the two are proportional to each other. Ren [
8] studied the influence of oxygen concentrations, grain sizes, and oxidation times on coal’s ATO, and the corresponding mechanism was deduced. Liu [
19] preliminarily studied the possible elementary reaction path of the ATO of long-flame coal by means of a quantum chemistry calculation. At present, although much work has been conducted on the study of CO production in goaf, there are still large deficiencies. This is manifested in the following aspects: (1) In previous studies, the research on ATO and CSC was isolated, and the characteristics of specific oxidation scenarios have not been considered. Therefore, especially in the goaf of low-rank coal where two kinds of oxidation coexist, the relationship of the two has not been investigated. (2) The mechanism of CO production in low-rank coal goaf is still unclear, which leads to the existing situation where the prevention and control of high concentrations of CO in the tailgate corner are limited to CSC-fighting measures, but the effect is not significant.
In view of these problems, this study first analyzed the oxidation scenario of low-rank coal goaf, and the CO production characteristics of low-rank coal under ATO. The CO generation rules of the coal samples treated with different ATO processes in CSC were obtained through ATO and temperature-programmed experiments. Based on in situ FTIR analysis, the precursor group of the CO produced under ATO conditions was determined, and the evolution law of the CO precursor in the ATO and CSC processes was obtained. The CO production mechanism of low-rank coal from ATO to CSC was established by means of a DFT simulation. Finally, combined with field practice, the influence of the coal’s ATO on CSC was scientifically analyzed, and a corresponding theoretical strategy was proposed. The related research results have an important guiding significance for improving the safety management level of CSC in low-rank coal goaf.
2. Analysis of Oxidation Scenario in Goaf
The formulation of the research scheme must be based on the analysis of real scenarios. The typical oxidation scenario of residual coal in goaf is shown in
Figure 1. The goaf can be divided into three zones: dissipation zone (O
2 ≥ 15%), oxidation zone (5% ≤ O
2 < 15%), and suffocative zone (O
2 < 5%). According to the temperature rise and oxidation capacity of each zone, the dissipation zone is mainly oxidized at ambient temperature (usually 10 °C ≤ temperature < 30 °C), the oxidation zone is subject to CSC oxidation (usually temperature ≥ 30 °C), and the suffocative zone is generally free of oxidation. Therefore, the CO in the tailgate corner is jointly produced by the coal’s ATO in the dissipation zone and the CSC oxidation in the oxidation zone.
It should be noted that the ATO of coal may have an important influence on CSC. Before the residual coal enters the oxidation zone for CSC oxidation, it first experiences ATO in the dissipation zone. The ATO definitely changes the intrinsic chemical reaction properties of the coal, which then influences the CO output of CSC. In view of the coexistence of ATO and CSC in low-rank coal goaf, the study of its CO production mechanism must cover the two processes from ATO to CSC. As a result, the formulation of a follow-up research plan needs to focus on the two processes of coal’s ATO and the CSC of coal samples treated with different oxidation times at ambient temperature.
4. Results
4.1. Ambient Temperature Oxidation Results
The variation in O
2 and CO concentration (volume concentration) with time is shown in
Figure 4. Among the three low-rank coals, lignite has the highest O
2 consumption and CO production, with a final O
2 concentration of 7.79%, consumption concentration of 13.21%, and final CO concentration of 545 ppm. This is followed by long-flame coal, with a final O
2 concentration of 9.14%, consumption concentration of 11.86%, and final CO concentration of 311 ppm. The non-stick coal has the lowest O
2 consumption and CO production, with a final O
2 concentration of 14.24%, consumption concentration of 6.76%, and final CO concentration of 140 ppm.
This shows that there is a good correlation between the ATO characteristics and the metamorphic degree of coal. With a decrease in the metamorphic degree of coal, the ATO characteristics gradually increase. As lignite has the strongest oxidation property at ambient temperature, it is selected as the experimental coal sample for further research at the next stage.
4.2. Programmed Heating Oxidation Results
With lignite as the research object, changes in O
2 and CO concentration at the reactor outlet after different ATO time treatment is shown in
Figure 5.
The variation in O2 concentration in the different coal samples is basically similar, and the ATO has no obvious influence on the oxygen consumption characteristics. However, in terms of CO production, it can be seen through the amplification of local areas (20~80 °C) that oxidation at ambient temperature shows an inhibitory effect on the production of CO when the coal temperature is lower than 70 °C. The concentration of CO produced by the raw coal is the highest at the temperature of each measuring point, followed by coal-6 h, coal-12 h, coal-24 h, and coal-48 h, and the concentration of CO produced by coal-10 d is the lowest. At 70 °C, the difference in CO concentration between the most active raw coal and the least active coal-10 d is more than 30 ppm, a decrease of 67%.
When the coal temperature is greater than 70 °C, the ATO shows a completely opposite influence. That is, the ATO has a promoting effect on the production of CO, which is manifested in the highest CO production concentration of coal-10 d, followed by coal-48 h, coal-24 h, coal-12 h, and coal-6 h, and the lowest CO production concentration is raw coal. When the coal temperature is about 200 °C, the CO production concentration of raw coal and coal-10 d are 9435 ppm and 19,680 ppm respectively, and the former is 10,245 ppm higher than the latter, with an increase of 109%.
Furthermore, the change in CO output concentration with temperature in the two stages of temperature ≤ 70 °C and temperature > 70 °C is fitted, respectively, in the form of the exponential function of y = A × e
(R*x) and y = B × e
(W*x), and the relevant fitting parameters and effects are shown in
Table 2. The fitting effect of the correlation functions is good, and adj.R
2 is all above 0.98. Since the R and W values are fixed, respectively, in the fitting process, the effect of ATO on CO production during CSC is directly reflected in the influence on the fitting equation parameters A and B. The changes in A and B values with respect to ATO time (t) are shown in
Figure 6.
It can be seen that the value of A decreases gradually with an increase in ATO time. However, in the first 50 h, the A value decreases significantly, and thereafter decreases slowly. Based on this, it can be inferred that CO production in the process of CSC will no longer change with ATO time, when the temperature is less than or equal to 70 °C after a certain ATO time. The B value increases gradually with an increase in ATO time, while in the first 50 h, the B value increases rapidly, and thereafter the increasing rate decreases gradually. The function form of y = a × x
b is used to fit the change in A and B values with ATO time. The ideal effect is achieved, and adj. R
2 is not less than 0.93. As a result, the influence law of coal’s ATO on CO production during CSC can be presented by Equation (1):
where y represents the CO production concentration, unit: ppm;
t represents ATO time, unit: h;
x represents temperature, unit: °C.
4.3. In Situ FTIR Analysis
The study focuses on the groups related to CO production; consequently, it mainly analyzes the C=O absorption vibration range (1500~1850 cm
−1) in the spectrum. By fitting the peaks of the infrared spectra under ATO conditions, the results show that the aldehyde functional group (1717 cm
−1) changed significantly with an increase in time, which indicates that the aldehyde functional group is the precursor of CO under coal’s ATO. Its peak area and peak area change rate with time is shown in
Figure 7. It can be seen that the aldehyde functional groups experience a significant dynamic change in the process of ATO. At the initial stage, the content of the aldehyde functional groups and the absolute value of their change rates decrease rapidly. With an increase in time, the two gradually become slower. When a certain time is reached, the content of the aldehyde functional groups decreases to a minimum, and there is no more change after that.
At present, aldehyde functional groups are widely accepted as precursors for CO production during CSC [
15,
16,
17]. This means that the precursors of CO produced under ATO and CSC have good consistency, both of which are the aldehyde functional groups. The peak area of the aldehyde functional group and its change rate with temperature during CSC is shown in
Figure 8. When the temperature is lower than 60 °C, the content of the aldehyde functional group and its growth rate change relatively slowly and increase slightly with temperature. From 60 to 80 °C, the content of the aldehyde functional groups presents a transition-type change, and the content increases rapidly. The growth rate reaches its maximum at 70 °C, which is 2.56 a.u./°C. A temperature equal to 70 °C is the critical temperature for the experimental coal sample in the process of CSC. When the temperature is higher than 80 °C, the growth rate of the aldehyde functional group content decreases rapidly, and then it remains at a low state. When the aldehyde functional group content is at a high state, this ensures the continuous high output of CO in the later stage of CSC.
4.4. DFT Simulation Results
Based on an in situ FTIR analysis, it was determined that the aldehyde functional group is the precursor group of CO produced by coal’s ATO, and its change characteristics in the process of ATO and CSC were obtained.
As is well-known, coal is a kind of organic macromolecular substance, which is composed of aromatic nuclei, aliphatic rings, aliphatic side chains, oxygen-containing functional groups, and some heterocycles [
22]. Heterocycles are usually not considered because, due to their low content, they do not have much of an effect on the main chemical properties of coal due [
22,
23,
24,
25]. Aromatic nuclei are stable at low temperatures and do not participate in the low-temperature oxidation reaction of coal, but they are the main body connecting the other reaction structural units. Among the oxygen-containing functional groups, excepting the aldehyde group which is the precursor group of CO, the hydroxyl group also plays an important role in CO production [
23]. Although the stability of aliphatic rings and aliphatic side chains is very high at low temperatures, both of these can participate in the low-temperature oxidation process of coal through -CH
2· formed by mechanical damage [
26,
27]. Based on the above information, a chain chemical reaction model for CO production is constructed, with the typical molecular fragment C
6 H
5CH
2CH
2·of low-rank coal as the initial chemical reaction structure, as shown in
Figure 9.
After the structural optimization and frequency calculation of products and reactants in each reaction step, the transition state search was carried out. As reaction 1 is a complexation reaction between -CH
2· and O
2, which only have bond formation and no bond breakage processes, there is no transition state in the reaction. The transition state search method, the keywords, and the unique virtual frequency of the obtained transition state in reactions 2~4 are shown in
Table 3.
An IRC response path analysis was conducted around each transition state, and IRC curves of each reaction sequence were obtained, as shown in
Figure 10, which verified the accuracy of the chemical reaction model in
Figure 9. The thermodynamic parameters of the reactants, transition states, and products in each reaction sequence are shown in
Table 4. In step 1, the energy barrier (△E) is 0 kJ/mol as there is no transition state. The enthalpy change (△H) is as high as 293.56 k J/mol. In step 2, the energy barrier rises to 186.51 kJ/mol due to the simultaneous separation and bond breaking of the O atom and H atom from the main structure during the reaction. After the completion of this step, the reaction releases 104.76 kJ/mol of heat. The energy barrier in Step 3 is very low, only 10.79 kJ/mol. However, the reaction can release 88.08 kJ/mol of heat. Step 4 is directly related to CO production, and the energy barrier and heat released are 31.46 kJ/mol and 0.76 kJ/mol, respectively.
5. Discussion
5.1. CO Production Mechanism of Low-Rank Coal Goaf
Different from high metamorphic coal goaf, low-rank coal goaf not only has a CSC oxidation property, but also an ATO. The establishment of the CO production mechanism in low-rank coal goaf must cover the two processes of ATO and CSC. The elementary reaction with an energy barrier equal to 40 kJ/mol is usually attributed to the ATO reaction [
25]. When the energy barrier is greater than 40 kJ/moL, it means that the reaction needs to take place at a higher temperature. Combined with the results in
Section 4.4, the CO production mechanism of low-rank coal goaf is shown in
Figure 11.
Only reaction 1 and reactions 3–4 of the established four-step chain reaction model can occur at ambient temperature, and they constitute the reaction mechanism sequences of CO production under ATO. Among them, reactions 3–4 are directly related to CO production, which indicates that the CO produced at ambient temperature may only come from the primary aldehyde functional groups. When the coal temperature reaches a certain critical point, reaction 2 overcomes the energy barrier and is activated. Reactions 1~4 constitute the reaction mechanism sequences of CO production after the coal temperature exceeds the critical temperature.
The CO precursor before the critical temperature point is mainly the primary aldehyde functional group. As the content of the primary aldehyde functional groups is limited, CO output will inevitably decline with an increase in the reaction time at ambient temperature. The longer the reaction time of coal’s ATO is, the greater the consumption of the primary aldehyde functional groups will be, which will reduce CO output before the critical temperature point. On the other hand, coal will accumulate peroxide-free radicals at ambient temperature, and the content will gradually increase with time, which will lead to a greater CO output after the critical temperature point. The above analysis is consistent with the experimental research results in
Section 4.1,
Section 4.2 and
Section 4.3, which verifies the accuracy of the CO production mechanism in
Figure 11.
5.2. Influence of ATO on CSC
CSC oxidation and ATO coexist in low-rank coal goaf, and there is an interaction and connection between them, which is mainly reflected in the direct interference of ATO on CSC. It is very important to have a deeper understanding of the influence of ATO on CSC.
Low-rank coal can react at ambient temperature and release CO, which gives it a wide existing space. Not only the dissipation zone of goaf, but also the coal wall, floor, and roof of the working face, as well as the wall of the air inlet and return roadway, constitutes the source of the CO produced under ATO. Furthermore, the CO production of each source term is not uniform and will change with the freshness of the coal sample and the reaction time. The characteristics of the heterogeneous emission of CO from multiple ATO sources makes the variation law of CO concentration in the tailgate corner complex and difficult to accurately quantify, which directly interferes with the early warning of CSC.
The ATO also directly affects the CO output of CSC by changing the chemical reaction properties of coal. It will continuously consume the primary aldehyde functional groups and accumulate peroxide-free radicals. This means that the longer the ATO time is, the lower the level of CO production will be before the critical temperature, and the higher the CO output after that. Our study shows that, for the assessment of the severity of CSC in low-rank coal goaf, if the ATO phenomenon is not considered, a serious misjudgment of the CSC risk degree will occur. This in, turn, will lead to the misuse of disproportionate prevention measures, resulting in a great waste of CSC-fighting resources.
Although the ATO of coal can interfere with the early warning of CSC in goaf, it will improve the oxygen consumption capacity, which is conducive to shortening the length of the oxidation zone, and so inhibits or even avoids the threat of CSC. Take the 1306 lignite-working face of Dananhu No.1 coal mine, with a 3 m mining height and a 3.18 m top carving height in Xinjiang, as an example. Although there is a high CO concentration hazard in the tailgate corner (>52 ppm), under the influence of ATO, and even if no CSC prevention or extinguishing measures are taken in the goaf, the residual coal will quickly enter the suffocative zone when the length of the dissipation zone is not more than 20 m and the length of the oxidation zone is not more than 15 m. Under a mining speed of more than 10 m/d, there is absolutely no threat of CSC. Nevertheless, how to balance the relationship between ATO and CSC, and ensure that the CO concentration in the tailgate corner is within an acceptable range, needs further research.
5.3. Theoretical Strategy
The phenomenon of high CO concentration in the tailgate corner of low-rank coal goafs during normal mining operations not only challenges the early warning of CSC, but also poses a serious threat to the health of the underground workers. How to ensure the safety of CSC management and the health of the workers at the same time in low-rank coal mines is one of the problems that coal mine safety researchers need to urgently solve. The existence of these two hidden dangers is caused by coal’s ATO. Therefore, to control these risks, it is necessary to properly control the ATO of low-rank coal.
An inhibitor is a chemical substance that usually acts on the oxidation stage of CSC. It can prevent CSC by destroying or reducing the specific structure of coal reacting with O
2, and has been widely used in CSC treatment. Based on the CO production mechanism in
Section 5.1, future research will develop ATO inhibitors, and the relevant inhibition mechanism is shown in
Figure 12.
The ATO inhibitor is oriented to cut off the sequence reaction in reaction 4 (
Figure 12) and reaction 3, which can fundamentally inhibit the CO produced under ATO conditions. However, as CO plays an important role in the early warning of CSC, the ATO inhibitor cannot affect the CO output in the CSC stage. Therefore, the inhibitor must have an inhibiting effect at ambient temperature and decompose once the reaction enters into CSC oxidation. Through the development of this inhibitor, the emission of high concentrations of CO from low-rank coal goaf can be controlled as CO production at ambient temperature is restrained, the abnormal source of CO acting as an early warning of CSC can be removed, and the health environment of underground workers will be guaranteed, thus fundamentally controlling the potential safety hazards in the normal mining process of low-rank coal seam.