Skip to main content
NCBI home page
ACS Omega logoLink to ACS Omega
. 2021 Mar 12;6(11):7669–7679. doi: 10.1021/acsomega.0c06322

Oxidation Characteristics of Functional Groups in Relation to Coal Spontaneous Combustion

Yutao Zhang 1, Jing Zhang 1, Yaqing Li 1,*, Sheng Gao 1, Chaoping Yang 1, Xueqiang Shi 1
PMCID: PMC7992153  PMID: 33778277

Abstract

graphic file with name ao0c06322_0020.jpg

To investigate and better understand the mechanism of coal spontaneous combustion, the distributions, evolution, and oxidation characteristics of functional groups in different coal samples were characterized using in situ Fourier transform infrared (FTIR) and electron paramagnetic resonance (EPR) experiments. The macroscopic characteristics of coal spontaneous combustion in relation to functional groups were also analyzed using the thermogravimetric/differential scanning calorimetry–FTIR coupling technique. The experimental results indicated that −OH was the most active groups of coal spontaneous combustion. It not only could react with the absorbed oxygen spontaneously but also found to be the main product of the chemisorption. Consequently, −OH was believed to contribute most both for the loss and increase of coal mass during the process of spontaneous combustion. Aliphatic hydrocarbons were the main components to form −C–O–O• and could be further oxidized into C=O. However, reactions between aliphatic hydrocarbons and oxygen were nonspontaneous. EPR experiments suggested that the tendency of coal spontaneous combustion acutely depended on the stability and survival time of free radicals. The more the stable and longer survival time of free radicals are, the lower the tendency of coal spontaneous combustion is.

1. Introduction

Coal is the most abundant energy resource with significant reserves remaining in China. However, the spontaneous combustion continues to be a big challenge associated with the production, storage, and transportation of coal. It has been reported that more than 70% of the total mine fires in China result from spontaneous combustion. The spontaneous combustion may lead to coal ignition, a full-blown fire, and even explosion if it is not eradicated appropriately. This may not only result in interruption of operations and mine closures, but serious losses of miners’ lives.1,2 As the necessity for ensuring safe production, storage, and transportation of coal, the prevention and control of coal spontaneous combustion has become increasingly vital.

The mechanism of coal spontaneous combustion is the foundation to prevent and control the coal fires efficiently and effectively. As early as the 17th century, coal spontaneous combustion had attracted widespread attention from the industry. Soon after, the occurrence and development of coal spontaneous combustion has been attributed to be induced by pyrite, bacteria, phenolic, complex of coal and oxygen, and so forth. Among them, the phenomena of the coal spontaneous combustion explained by the hypothesis of the coal−oxygen complex agrees well with the macroscopic characteristics of coal spontaneous combustion. Consequently, the coal−oxygen complex hypothesis has been widely accepted.3 However, this hypothesis cannot clearly explain the microscopic processes such as species of functional groups reacting with oxygen, reacting conditions, and reacting characteristics so far.

Following the interpretation of coal–oxygen complex theory, heat accumulation is the direct cause of coal spontaneous combustion. Therefore, lots of studies toward a basic understanding of the mechanism of coal spontaneous combustion using thermal analysis technology have been presented. Garcia et al.4 calculated the kinetic parameters of different coal samples and analyzed the process of coal spontaneous combustion from the standpoint of activation energy. Pis et al.5 compared the oxidation characteristics of raw coal and oxidized coals. They found that characteristic temperatures of oxidized coals increased regardless of coal properties. Wang6 evaluated the effects of the heating rate on the characteristics of coal spontaneous combustion and found that the exothermic profiles of coal samples shifted to higher temperatures with the increase of heating rates. Yu et al.7 found that the heat released at the coal oxidation stage was much higher than that at the pyrolysis stage. Xiao et al. and Zhang et al.8−10 defined characteristic temperatures during coal spontaneous combustion according to thermogravimetric (TG) curves. Based on the experimental results, Deng et al.11 concluded that temperatures of thermal decomposition and the maximum rate of weight loss were critical to reflect the severity of coal spontaneous combustion. Zhang et al.12 proposed a differential scanning calorimetry (DSC) inflection method to determine autoignition temperatures of coals during the spontaneous combustion process. Actually, heat release is only the outward manifestation and an accompanying effect of reactions during the coal spontaneous combustion. The essential cause was the production and evolution of diverse functional groups. Consequently, identifying and characterizing the oxidation characteristics of functional groups during the spontaneous combustion have been of interest.

Later on, Fourier transform infrared (FTIR) spectroscopy technique has been widely used to study the functional groups in coals.13 Based on the FTIR spectra, substantial achievements have been made. Petersen14,15 speculated the structural parameters of coals using FTIR. Ibarra et al.16 identified the distributions of aromatic hydrocarbons and oxygen-containing species in coals. Zhang et al.17 addressed that the occurrence of coal spontaneous combustion was the result of reactions between active groups and oxygen molecules. Furthermore, active groups in coals were classified into oxygen-containing and oxygen-free groups by Yang.18 Encouraged by this finding, the free radical action hypothesis was proposed by Li,19 and Li tried to explain the mechanism of coal spontaneous combustion from the standpoint of chemistry. By analyzing the oxygen distribution during the process of coal spontaneous combustion, Retcofsky et al.20 validated the reasonability of the free radical action hypothesis. Afterward, quite a few studies have been conducted to investigate the mechanism of coal spontaneous combustion by employing the electron paramagnetic resonance (EPR) technique. Wood et al.21 observed that concentrations of free radicals in oxidized coal samples were higher than those of raw coals. Dai22 pointed out that the intensity of coal oxidation during the process of spontaneous combustion was determined by the amount of generated free radicals rather than that of the original. In addition, factors influencing the characteristics of free radicals in coals have been experimentally studied.23−25,37 However, both FTIR and EPR techniques could not quantify the functional group changes with the oxidation processes.

In this study, the in situ FTIR technology combined with the EPR technique were employed to characterize the evolution of multifarious functional groups as well as their oxidation characteristics. Meanwhile, the oxidation characteristics of the functional groups in relation to the macroscopic characteristics of coal spontaneous combustion were investigated on the basis of TG/DSC–FTIR experiments. This study would not only be beneficial for revealing the mechanism of coal spontaneous combustion but may also provide theoretical foundations for development of fire-extinguishing materials.

2. Results and Discussion

2.1. Distribution of Functional Groups

Quantitative analysis of the experimental FT-IR spectra was based on the Beer–Lambert law. Based on this law, curve fitting of coal infrared spectra can describe the peak area of every functional group in coal.26 The corresponding second-derivative and Fourier deconvolution spectra of different coal samples were then used as references and a Gaussian–Lorentzian function was used to fit the spectrum to obtain peak areas of the different functional groups shown in Table 1. The results are illustrated in Figure 1. To emphasize comparisons, corresponding regression curves are also illustrated in Figure 1. Based on the peak areas of the regression curves, relative contents of functional groups were obtained and summarized in Table 2.

Table 1. Categories and Motions of Functional Groups Based on Wavenumbers30 (Classification 2005).

wavenumber (cm–1) functional groups motion of functional groups
750 –OH bending vibration of −OH in alcohol or phenolic molecules
786 C–H bending vibration of C–H in 1,2,3 substituted benzene
912 C–H out-of-plane bending vibration of C–H in aldehydic molecules
1002–1174 C–O–C stretching vibration of –C–O–C–
1355 –CH3 shear vibration of −CH3
1396 –CH3, −CH2– in-plane bending vibration of C–H
1440 –CH2– shear vibration of −CH2–
1618 C=C stretching vibration of C=C on aromatic or condensed nucleuses
1637 C=O stretching vibration of C=O
2780–2350 –COO– stretching vibration of −COO–
2921 –CH3 stretching vibration of C–H in −CH3
2989 –CH2– stretching vibration of C–H in −CH2–
3100–3000 Ar-CH stretching vibration of C–C on aromatic rings
3415 –OH associating hydrogen bond of −OH among molecules
Open in a new tab

Figure 1.

Figure 1

Open in a new tab

FTIR spectra of different samples.

Table 2. Percentages of Main Functional Groups for Different Coal Samples.

  aliphatic hydrocarbons, %
 aromatic hydrocarbons, %
 oxygen-bearing groups, %
samples –CH2– –CH3 C=C Ar-CH subtotal –OH C=O –C–O–C– –COO– subtotal
C1 4.82 5.64 16.25 1.53 17.78 43.56 7.84 14.65 5.12 71.17
C2 11.40 4.06 20.37 2.16 22.53 36.21 2.87 20.63 1.94 61.65
C3 4.51 12.32 26.66 6.33 32.99 13.94 4.65 28.75 2.72 50.06
Open in a new tab

Data listed in Table 2 indicate that functional groups in coals, in general, can be classified into three types including oxygen-bearing groups, aliphatic hydrocarbons, and aromatic hydrocarbons.27,28 Of them, the content of the oxygen-bearing groups was the largest, accounting for more than 50% and that of the aliphatic hydrocarbons was the least. Furthermore, it was found that oxygen-bearing groups in the tested coal sample existed in the forms of −OH, C=O, −COO–, and −C–O–C–. This finding was consistent with the molecular models of coals constructed by Mathews and Chaffee.29 Connected to this finding was that aliphatic hydrocarbons in coals were present either as −CH3 or −CH2–. Aromatic hydrocarbons, as the backbone of the coal molecule, were characterized by stretching vibrations of Ar-CH and C=C on an aromatic or a condensed nucleus, located at 3026 and 1614 cm–1, respectively. Further analysis of the data listed in Table 1 suggests that the contents of aliphatic hydrocarbons and aromatic hydrocarbons were increased with the coalification. Nevertheless, an opposite change was observed for oxygen-bearing groups. It is a practical cognition that the lower the coalification is, the easier the coal tends to spontaneous combustion. Combing the experimental results with the practical cognition leads to the statement that oxygen-bearing groups were of importance for the tendency of coal spontaneous combustion.

Figure 2 illustrates the relative contents of oxygen-bearing groups in different coal samples. It was notable that −OH and −C–O–C– accounted for the majority of the oxygen-bearing groups regardless of coal properties when compared to C=O and −COO–. Furthermore, it was observed that the content of −C–O–C– increased with the coalification. On the contrary, a larger content of −OH was observed for the coal sample with a lower degree of coalification. Because of the high polarity, −OH was believed to be the main functional group in coal molecules to form hydrogen bonds. Different hydrogen bonds could be formed when −OH was combined with different acceptors. According to Fei et al.,31 −OH in coals existed in three forms including free −OH, intermolecular-associated −OH, and self-associated −OH whose infrared vibration peaks were located at 3674, 3535, and 3428 cm–1, respectively. Relative contents of these three forms of −OH for tested coal samples were obtained and displayed in Figure 3. Graphs in Figure 3 depict that the intermolecular-associated −OH was acutely dependent on the properties of the coal samples. With the increase of coalification, contents of the intermolecular-associated −OH decreased from 27.28 to 2.47%. The same trend was observed for self-associated −OH. On the contrary, contents of the free −OH increased from 2.79% for lignite to 9.32% for anthracite. This observation implies that the contribution of the hydrogen bond for the reactivity of coal diminished with the increase of the coalification.

Figure 2.

Figure 2

Open in a new tab

Relative contents of the oxygen-bearing groups in different coal samples.

Figure 3.

Figure 3

Open in a new tab

Different forms of −OH distributed in coal samples.

Figure 4 shows the distributions of −CH3 and −CH2– in various coal samples. The ratio of −CH2– to −CH3 usually indicates the length of branched chains and a high ratio represents a longer branched chain.32 Graphs in Figure 4 indicate that the ratio of −CH2– to −CH3 from high to low is in the order of C2, C1, and C3. This order indicates that the chain of the branched aliphatic hydrocarbons in C2 was the longest, followed by C1 and C3.

Figure 4.

Figure 4

Open in a new tab

Distributions of −CH2– and −CH3 in different coal samples.

2.2. Evolution of Oxygen-Bearing Groups with Temperatures

To ensure the accuracy of the fitting parameters, the peak areas of oxygen-bearing functional groups obtained by peak-fitting were corrected through PeakFit software. The criterion of the correction was deriving the data of peak separation when the interaction parameter of PeakFit software changes to 7. The sum of the corrected peak areas was regarded as the total peak area and the ratio of the corrected peak area of a certain functional group to the total peak area was expressed as the content of this functional group in the coal. Analysis of the ratios and sum of the corrected peak areas for oxygen-bearing functional groups as a function of the temperatures are shown in Figures 8–10.

Figure 8.

Figure 8

Open in a new tab

Evolution of the C–H bond in −CH2– and −CH3 as a function of the temperature for various coals (2918 cm–1).

Figure 10.

Figure 10

Open in a new tab

Evolution of aromatic hydrogen (Ar-CH) in different coal samples as a function of temperature (3026 cm–1).

2.2.1. Evolution of −OH with Temperatures

As mentioned above, −OH is of vital importance during the spontaneous combustion of coal. As a result, the investigation of the evolution of −OH was beneficial for a deep understanding of the microcosmic mechanism of coal oxidation. Figure 5 shows the response of −OH to temperatures. It can be seen from Figure 8 that the content of free −OH in coal decreased with the increase of the temperature regardless of coal properties. However, from further analysis of curves in Figure 8a, it was found that the change trend of the free −OH varies with the properties of coal. Curves in Figure 5a imply that the free −OH in C1 and C2 decreases sharply at 50–120 °C while the free −OH in C3 only exhibits a small decrease when the temperature was greater than 250 °C. On the basis of the TG and DSC curves, the process of coal spontaneous combustion could be classified into the dehydration and desorption stage, the oxidation stage, the combustion stage, and the burn-out stage.33 Consequently, the decrease of the free −OH in the temperature range of 50–120 °C indicates that the free −OH was mainly involved in the dehydration and desorption stage. In addition, a small decrease of free −OH at temperatures exceeding 250 °C in C3 coal imply that the free −OH in coals was also involved in the reactions during the oxidation stage.

Figure 5.

Figure 5

Open in a new tab

Evolution of −OH with the increase of the temperature. (a) Free −OH (3674 cm–1). (b) Intermolecular-associated-OH (3535 cm–1).

Figure 5b illustrates that the intermolecular-associated −OH of C1 decreases as the temperature increases from the beginning. Analysis of the curves in Figure 5b indicates that the content of the intermolecular-associated −OH in C2 exhibits a small decrease when the temperature increases from 25 to 170 °C. However, further increasing the temperature to 342 °C results in an increase of the intermolecular-associated −OH for C2. Afterward, a significant decrease was observed for C2 after 342 °C. For C3, the intermolecular-associated −OH was kept almost constant before 260 °C and increased as the temperature went up from 260 to 430 °C. The decrease for C1 and C2 indicates that the intermolecular-associated −OH had a high reactivity and participated in the reactions spontaneously. Nevertheless, the behaviors of −OH responding to the temperature for C2 and C3 suggest that new intermolecular-associated −OH was generated with the consumptions of the original ones during the reactions. According to organic chemistry theories and the quantum chemistry, −OH could be generated and consumed as follows

2.2.1. 1

Apparently, the generation of intermolecular-associated −OH during the process of coal spontaneous combustion strongly supported the rationality of these three speculations. Substituting these three speculations into the behaviors of the intermolecular-associated −OH in C2 and C3 would lead to the conclusion that energy had to be provided to induce the reactions illustrated in 1. Moreover, it was observed that the temperature at which the intermolecular-associated −OH started to increase for C2 was higher than that for C3. Combining this observation with the previous finding that the chain of the branched aliphatic hydrocarbons in C2 was the longest, it could be concluded that the shorter side chain may cause a higher reactivity in situ where other factors were the same.

2.2.2. Evolution of C=O with Temperatures

The stretching vibration of C=O was located at the wavelengths between 1780 and 1630 cm–1. Based on the experimental results, the strongest absorbance at 1776 cm–1 was selected as the representative to investigate the response of C=O to the temperature. Results are plotted in Figure 6.

Figure 6.

Figure 6

Open in a new tab

Evolution of C=O with the increase of the temperature (1776 cm–1).

Curves in Figure 6 illustrate that contents of C=O increased with temperatures before the temperature exceeded thresholds of 341 °C for C1 and 389 °C for C2, respectively. After exceeding these thresholds, the contents of C=O gradually decreased. This illustration suggests that C=O was an important transition functional group during the coal spontaneous combustion process. Before the thresholds, the formation of C=O was prominent, correlating with previous findings of other researchers.34−36 Based on the principles of chemical reactions, C=O could be produced as follows

2.2.2. 4

First, aliphatic hydrocarbons were attacked by oxygen, resulting in C•. Furthermore, C• was transformed into −C–O–O• by oxygen. Successively, −C–O–O• was converted into unstable •OH and −C–O•, and −C–O• was then decomposed into C=O and gases of CxHy. Following this explanation, aliphatic hydrocarbons were speculated to be the main components to form −C–O–O•. The conclusion was well supported by curves in Figure 6. Meanwhile, it was observed that contents of C=O in C2 and C3 almost kept constant until the temperatures exceeded thresholds. This observation further confirms the previous finding that aliphatic hydrocarbons and oxygen could not react spontaneously unless enough energy was provided.

2.2.3. Evolution of −COO– with Temperatures

Figure 7 shows the evolution of −COO– against temperatures. It is noteworthy that −COO– in C1 started to decrease at the beginning of coal spontaneous combustion. Nevertheless, contents of −COO– in C2 and C3 kept almost unchanged during the whole experimental process. This was because −COO– in C1 mainly existed in the form of −COOH; while −COO– in C2 and C3 were primarily present in the ester groups. Therefore, the response of −COO– in C1 to temperatures indicates that −COOH in coals exhibits a high reactivity.

Figure 7.

Figure 7

Open in a new tab

Evolution of −COO– as a function of the temperature (2733 cm–1).

2.3. Evolution of Aliphatic Hydrocarbons with Temperatures

According to Table 1, the main motions of branched aliphatic hydrocarbons include the stretching vibration of the C–H bond in −CH3 and −CH2– located between 2921 and 2989 cm–1. The strongest absorbance of the C–H bond at 2918 cm–1 was selected as the representative to investigate the behavior of branched aliphatic hydrocarbons during the process of coal spontaneous combustion. Then, evolution of the aliphatic hydrocarbons with the temperature was investigated by employing the method same as that discussed in Section 2.2 and the results are illustrated in Figure 8.

Curves in Figure 8 illustrate that the contents of aliphatic hydrocarbons almost kept constant before the temperatures reached certain values in spite of coal properties. However, they suddenly decreased when the temperature exceeded thresholds. The changes indicated that energy was required for −CH3 and −CH2– to fracture or to participate into chemical reactions. This suggestion emphasized the previous conclusion that reactions between oxygen and −CH3 or −CH2– in coals were nonspontaneous.

Based on curves in Figure 8, the temperatures when −CH2– and −CH3 started to decrease were 202 °C for C1, 215 °C for C2, and 178 °C for C3. These data were consistent with the previous statement that the shorter the side chains were, the stronger tendency of coal toward spontaneous combustion would be. This was mainly attributed to the high susceptibility of the side chains to heat and chemical reactions.

2.4. Evolution of Aromatic Hydrocarbons with Temperatures

Aromatic hydrocarbons in the coal samples could be classified into C=C and Ar-CH. According to Table 1, the stretching vibration of C=C on mononuclear aromatic hydrocarbons ranges from 1620 to 1430 cm–1 and that of Ar-CH on aromatic hydrocarbons was in the range between 3100 and 3000 cm–1. Consequently, the strongest absorbances located at 1614 and 3026 cm–1 were selected as representatives of C=C and Ar-CH, respectively. Using the same method described in Section 3.2, the changes in C=C and Ar-CH versus temperatures were obtained and illustrated in Figures 12 and 13.

Figure 12.

Figure 12

Open in a new tab

TG curves of different coal samples.

Figure 13.

Figure 13

Open in a new tab

Profiles of the water vapor released during the spontaneous combustion of coal (3735 cm–1).

As can be seen in Figure 9, C=C in C1, C2, and C3 changed little before the temperatures reached the thresholds of 310, 350, and 407 °C, respectively. These temperatures were close to the pyrolysis temperatures of the coal samples used in the TG experiment in Section 2.6. Before the temperature threshold, the variation range of the C=C curve was not large. After the temperature threshold, the C=C curve begins to decline greatly. The changes of C=C with temperatures demonstrate that the benzene rings of coal molecules were stable and only participated in reactions at the combustion stage.

Figure 9.

Figure 9

Open in a new tab

Evolution of the aromatic nucleus (C=C) in different coal samples as a function of temperature (1614 cm–1).

Curves in Figure 10 show that contents of Ar-CH in C1 and C2 decrease as the temperature increases. However, the decreasing rate of Ar-CH in C2 was lower than that in C1. Instead of decreasing, Ar-CH in C3 increased when the temperature increased from 25 to 106 °C. Then, the content of Ar-CH in C3 remained constant until the temperature increased to 220 °C. After 220 °C, Ar-CH in C3 decreases with the increase of the temperature. The behavior of Ar-CH in C1 responding to the temperature indicates that the aromatic nucleus in coal molecules possesses high reactivity and would have involved in reactions at the beginning of coal oxidation. Besides, the increase of Ar-CH in C3 with the increase of the temperature suggests that an aromatic nucleus was generated during the process of coal oxidation.

2.5. Characteristics of Free Radicals in the Tested Coal Samples

Figure 11 illustrates the EPR spectrograms of the tested coal samples. Based on the curves in Figure 11, the g factor, line width (ΔH), and concentrations of free radicals (Ng) were obtained and listed in Table 3.

Figure 11.

Figure 11

Open in a new tab

EPR spectra of the tested coal samples.

Table 3. Parameters of the EPR Spectra for the Tested Coal Samples.

samples g factor beam line width (ΔH/mT) concentration of free radicals (Ng/1017 g–1)
C1 2.00344 633.86 2.864
C2 2.00267 619.77 19.01
C3 2.00264 578.25 20.16
Open in a new tab

It is noticeable in Table 3 that g factors of the tested coal samples range from 2.00344 to 2.00264. This indicates that free radicals in the tested coal samples were organic free radicals regardless of coal properties. Connected with this is the observation that a larger g factor was obtained for a coal sample with a lower degree of coalification. Because the unpaired electrons of coal move along the highly nonlocalized molecular orbital, the orbital magnetic moment of coal was negligible. Consequently, the degree of g factor deviating from the free electron (ge) was determined by the quenching degree of the angular momentum. Generally, a greater quenching degree of the angular momentum results in a larger deviation of the g factor from ge. Combining the interpretation of the g factor with its responses to the coalification leads to the statement that: the lower the degree of coalification is, the more unstable the free radicals and the shorter their survival time is. This was speculated to be the main reason why the coal with lower coalification exhibited a higher tendency of spontaneous combustion.

By comparing the data listed in Table 3, it was found that the beam line width (ΔH) of the tested coal samples from high to low was in the order of C1, C2, and C3. It has been known that the beam line width strongly depends on spin–lattice relaxation and spin–spin relaxation. A narrower beam line width indicates a longer time of spin–spin relaxation. A longer time of spin–spin relaxation is a reflection of more violent electron exchange among free radicals. Combing this explanation with the analysis of the beam line width (ΔH) listed in Table 3, it can be stated that the intensity of electron exchange among free radicals increases with the increase of the coalification. Furthermore, it can said that chemical reactions in the coal sample with a higher degree of coalification were more violent.

As seen in Table 3, the larger concentration of free radicals was observed for the coal sample with a higher degree of coalification under the same conditions. This was mainly because the coal sample with a higher degree of coalification contains a larger amount of fused ring. Generally, benzene rings, which were adhered by a variety of branched aliphatic hydrocarbons and oxygen-bearing groups, were the backbone of coal molecules. As the coalification develops, those adhered side chains and functional groups separate from benzene rings because of thermolysis and polycondensation. Accordingly, the amount of fused ring increased, and free radicals tended to be in stable states. This strongly supports the finding of the g factor that free radicals in lower metamorphic coal were unstable and had a shorter survival time. These stable states of free radicals benefited polycondensation of the aromatic nucleus. Consequently, even larger and more stable macromolecular free radicals were generated. Simultaneously, the conductivity of electrons was enhanced. Consequently, a higher concentration of free radicals was observed in the coal sample with a higher degree of coalification. This finding is in agreement with the analysis of the evolution of aliphatic hydrocarbons as a function of temperature.

Additionally, it is noticeable in Table 3 that concentrations of free radicals in C1 were much lower than those in C2 and C3. This suggests that branched aliphatic hydrocarbons and oxygen-bearing groups of C1 adhere to aromatic rings by hydrogen bonds and/or van der Waals forces. This suggestion further confirms the previous findings that the contribution of hydrogen bonds to the reactivity of coal reduced with the increase of the coalification.

2.6. Macroscopic Characteristics of Coal Spontaneous Combustion in Relation to Functional Groups

TG/DSC–FTIR experiments were conducted to investigate the macroscopic parameters of coal spontaneous combustion in relation to functional groups. Figure 12 shows the TG curves of different coal samples during the process of coal spontaneous combustion. As can be seen, an obvious reduction of coal mass could be observed for C1 at the dehydration and desorption stage (25–120 °C). However, C2 and C3 exhibited inconspicuous changes in coal mass at this stage. As explained in the section regarding the evolution of −OH with the temperature, −OH could react with oxygen molecules spontaneously. According to the principle of chemical reactions, the main oxidation product of −OH in coal was water. Therefore, there should be a generation of water vapor for C1 at this stage. Figure 13 illustrates the profiles of water vapor productions during the process of coal spontaneous combustion. As can be seen from Figure 16, a high generation of water vapor was observed for C1 in the temperature range of 25 to 120 °C. This observation further validates previous explanation that −OH could react with oxygen molecules spontaneously. Consequently, it can be said that the mass loss of coal samples at the dehydration and desorption stages was mainly attributed to the oxidation of −OH.

Figure 16.

Figure 16

Open in a new tab

Schematic of the TG/DSC–FTIR experimental system used for coal spontaneous combustion tests.

Additionally, the absorbed oxygen molecules would react with coals following the procedures in 1 at the oxidation stage and thus increase the coal mass. This interpretation conform well to the observations illustrated in Figure 13. Consequently, it is suffice to say that the generation of intermolecular-associated −OH was the main cause leading to the mass increase during the coal spontaneous combustion. Upon further analysis of TG curves in Figure 13, it was found that the increment of mass for C3 was higher than that for C2 at the oxidation stage. This finding was the evidence of the statement drawn from the analysis of the beam line width (ΔH) that chemical reactions in the coal sample with a higher degree of coalification were more violent.

3. Conclusions

To investigate the mechanism of coal spontaneous combustion, three coals with different properties were subjected to in situ FTIR experiments and EPR experiments. Besides, the macroscopic characteristic parameters of coal spontaneous combustion in relation to functional groups were correlated via TG/DSC–FTIR experiments.

From the analysis of the FTIR spectra, it was found that the contents of functional groups in coals from low to high was in an order of aliphatic hydrocarbons, aromatic hydrocarbons, and oxygen-bearing groups regardless of coal properties. With the increase of the coalification, both aliphatic hydrocarbons and aromatic hydrocarbons increased gradually, while the content of oxygen-bearing groups decreased. Along the same line, the contribution of the hydrogen bond for the reactivity of coal diminished. −OH was validated to be the most active group during the spontaneous combustion. It could not only react with oxygen spontaneously, but also be generated by the chemisorption of oxygen. Besides, −OH was validated to contribute most to the mass decrease as well as the increase during the process of coal spontaneous combustion by TG/DSC–FTIR experiments. Different from −OH, energy was indispensable in inducing the reactions between aliphatic hydrocarbons and oxygen. As the oxidation product of aliphatic hydrocarbons, C=O was an important transition group of coal spontaneous combustion. Responses of C=C and Ar-CH to the temperature indicated that benzene rings of the coal sample were quite stable and could only involve in the reactions after ignition. Nevertheless, aromatic rings exhibited relatively high reactivity and would participate in reactions at the beginning of coal oxidation. Also, it was found that aromatic rings were generated during the process of coal spontaneous combustion.

From the analysis of the EPR spectra, it was found that free radicals in coal were limited into organic elements. With the increase of the coalification, the content of macromolecular free radicals increased. Simultaneously, both the stability and survive time of free radicals enhanced. In addition to this, a higher concentration of free radicals was found to be in the coal sample with a higher degree of coalification because of the increased fused rings. Accordingly, more violent chemical reactions were observed in the coal sample with higher coalification.

4. Experimental Section

4.1. Coal Samples

Three coals with different properties from Xinjiang and Anhui provinces in China were used in this study. The proximate analysis of each coal is listed in Table 4. According to the ISO 11760 classification of coal, C1, C2, and C3 could be categorized into lignite, bituminous coal, and anthracite, respectively.37

Table 4. Codes and Properties of the Coal Samples.

  proximate analysis, wt %
code of samples Mad Aad Vad FCad
C1 25.37 7.46 52.25 14.92
C2 1.81 14.62 33.36 50.21
C3 1.89 13.03 9.50 75.58
Open in a new tab

A two-step crushing process was employed to prepare coal samples. First, raw coals with different properties were crushed to less than 5 mm with a jaw crusher. Subsequently, the crushed coals were further ground to less than 74 μm in a hammer mill. To avoid oxidation, the crushing and grinding processes were all carried out under the protection of N2. Furthermore, the ground coal samples were put into a vacuum drying oven and allowed to stand still at room temperature until the mass change of coal samples became less than 0.1% to remove air-dried moisture. Then, the prepared coal samples were put into well-sealed flasks for carrying out tests.

4.2. Experimental Setup

4.2.1. In Situ FTIR Experiments

To characterize functional groups in different coal samples as well as, an in situ FTIR experimental system was set up to carry out the experiments their evolution during the spontaneous combustion. Figure 14 illustrates the schematic of the system. The system consists of a VERTEX 70v series Fourier transform infrared spectrometer manufactured by Bruker Incorporation and a diffuse reflection in situ cell. For each series experiments, three parallel experiments were conducted. Then, the typical experimental results were selected for analyzing. During each experimental run, a 15 mg coal sample was oxidized from 25 to 430 °C in the cell at an increasing rate of 1 °C/min. The wavenumber of FTIR was set to be between 400 and 4000 cm–1 with a resolution of 4 cm–1 and the data collecting frequency was 32 Hz. The functional groups were categorized on the basis of Table 1.

Figure 14.

Figure 14

Open in a new tab

Schematic of the in situ FTIR system.

4.2.2. EPR Experiments

Free radicals allow one to essentially understand the oxidation process of coal. The (EPR technique, which can directly detect unpaired electrons of molecules, plays an important role in the determination of free radicals in coals. To investigate in depth the oxidation characteristics of functional groups during the coal oxidation process, an EMSPlus-10-12 electron paramagnetic resonance system manufactured by Bruker Incorporation was employed to characterize free radicals of different coal samples. The schematic and set up of the apparatus are shown in Figure 15. Basically, the apparatus was composed of a microwave bridge, a resonant cavity, a magnet, a control box, a main engine, and a water cooling system.

Figure 15.

Figure 15

Open in a new tab

Schematic of the EPR system.

For each experimental run, a 15 mg prepared coal sample was put into the sample bin with a diameter of 4 mm for carrying out the test. For the purpose of experiments, 3360 G magnetic flux was loaded and the radiation frequency of microwave was set to be 935 MHz. The scanning time was 40 s with a magnification of 20.

4.2.3. TG/DSC–FTIR Experiments

An ASTA449F3 series simultaneous TG/DSC thermal analyzer of Netzsch Incorporation coupled with a VERTEX 70v serious FTIR manufactured by Bruker Incorporation system was set up to conduct relative tests, as illustrated in Figure 16. In this experimental system, the coal sample was oxidized using the TG/DSC thermal analyzer and the gaseous products were delivered to the Fourier transform infrared spectrometer via a specially designed interface in real time for identification.

For each experimental run, a 15 mg coal sample was put into the sample bin of the TG/DSC analyzer, nitrogen was selected as the protection gas and the total flow rate of the protection and carrier gases was constantly kept at 100 mL/min. For the purpose of this study, the oxygen concentration in the carrier gas was adjusted to be 21% and the coal sample was pretreated under the carrying gas around 10 min till its quality became stable. Then, the sample was heated from 25 to 430 °C at a rate of 5 °C/min. The wavenumber of the FTIR was set to be between 400 and 4000 cm–1 with a resolution of 4 cm–1. To avoid condensation of the gaseous products, temperatures of the interface and the transmission line were both kept constant at 220 °C.

Acknowledgments

This research was finically supported by National Key R&D Program of China (2018YFC0807900) and National Natural Science Foundation of China (grant nos. 51774233 and 51974235).

The authors declare no competing financial interest.

References

  1. Zhang Y.; Zhang Y.; Li Y.; Li Q.; Zhang J.; Yang C. Study on the characteristics of coal spontaneous combustion during the development and decaying processes. Process Saf. Environ. Prot. 2020, 138, 9–17. 10.1016/j.psep.2020.02.038. [DOI] [Google Scholar]
  2. Xu Y.-l.; Wang L.-y.; Tian N.; Zhang J.-p.; Yu M.-g.; Delichatsios M. A. Spontaneous combustion coal parameters for the Crossing-Point Temperature (CPT) method in a Temperature-Programmed System (TPS). Fire Saf. J. 2017, 91, 147–154. 10.1016/j.firesaf.2017.03.084. [DOI] [Google Scholar]
  3. Wang D. M.Science of Mine Fire. Xuzhou; China University of Mining and Technology Press: Xuzhou, 2008; pp 47–53.
  4. Garcia P.; Hall P. J.; Mondragon F. The use of differential scanning calorimetry to identify coals susceptible to spontaneous combustion. Thermochim. Acta 1999, 336, 41–46. 10.1016/s0040-6031(99)00183-5. [DOI] [Google Scholar]
  5. Pis J.; de la Puente G.; Fuente E.; Morán A.; Rubiera F. A study of the self-heating of fresh and oxidized coals by differential thermal analysis. Thermochim. Acta 1996, 279, 93–101. 10.1016/s0040-6031(96)90066-0. [DOI] [Google Scholar]
  6. Wang K.Study on the oxidation and spontaneous combustion characteristics of Jurassic coal in Northern Shaanxi. Master Thesis, Xi’an University of Science and Technology, 2013. [Google Scholar]
  7. Yu M.-g.; Zheng Y.-m.; Lu C.; Jia H.-l. Thermal analysis experiment on low-temperature oxidation and pyrolysis of coal. China Saf. Sci. J. 2009, 09, 83–86. [Google Scholar]
  8. Xiao Y.; Ma L.; Wang Z.; Deng J.; Wang W.; Xiang X. Research on characteristic temperature in coal spontaneous combustion with thermal gravity analysis method. Coal Sci. Technol. 2007, 35, 73–76. [Google Scholar]
  9. Zhang Y.; Shi X.; Li Y.; Liu Y. Characteristics of carbon monoxide production and oxidation kinetics during the decaying process of coal spontaneous combustion. Can. J. Chem. Eng. 2018, 96, 1752–1761. 10.1002/cjce.23119. [DOI] [Google Scholar]
  10. Zhang Y.; Shi X.; Li Y.; Wen H.; Huang Y.; Li S.; Liu Y. Inhibiting effects of Zn/Mg/Al layer double hydroxide on coal spontaneous combustion. J. China Coal Soc. 2017, 42, 2892–2899. [Google Scholar]
  11. Deng J.; Yang Y.; Zhang Y.-N.; Liu B.; Shu C.-M. Inhibiting effects of three commercial inhibitors in spontaneous coal combustion. Energy 2018, 160, 1174–1185. 10.1016/j.energy.2018.07.040. [DOI] [Google Scholar]
  12. Zhang Y.; Liu Y.; Shi X.; Yang C.; Wang W.; Li Y. Risk evaluation of coal spontaneous combustion on the basis of auto-ignition temperature. Fuel 2018, 233, 68–76. 10.1016/j.fuel.2018.06.052. [DOI] [Google Scholar]
  13. Dyrkacz G. R.; Bloomquist C. A. A. Binary solvent extractions of upper freeport coal. Energy Fuels 2001, 15, 1409–1413. 10.1021/ef010036s. [DOI] [Google Scholar]
  14. Petersen H. I. The petroleum generation potential and effective oil window of humic coals related to coal composition and age. Int. J. Coal Geol. 2006, 67, 221–248. 10.1016/j.coal.2006.01.005. [DOI] [Google Scholar]
  15. Petersen H. I.; Nytoft H. P. Oil generation capacity of coals as a function of coal age and aliphatic structure. Org. Geochem. 2006, 37, 558–583. 10.1016/j.orggeochem.2005.12.012. [DOI] [Google Scholar]
  16. Ibarra J.; Muñoz E.; Moliner R. FTIR study of the evolution of coal structure during the coalification process. Org. Geochem. 1996, 24, 725–735. 10.1016/0146-6380(96)00063-0. [DOI] [Google Scholar]
  17. Zhang Y. N.Study on the microcosmic characteristics and macro parameters in the process of coal oxidation and spontaneous combustion. Ph.D. Thesis, Xi’an University of Science and Technology, 2013. [Google Scholar]
  18. Yang S. Study on the active groups’ reaction activeness on self-ignition coal. Coal Mine. Modern. 2005, 65, 59–60. [Google Scholar]
  19. Li Z. Mechanism of free radical reactions in spontaneous combustion of coal. J. China Univ. Min. Technol. 1996, 25, 111–114. [Google Scholar]
  20. Retcofsky H. L.; Stark J. M.; Friedel R. A. Electron spin resonance in American coals. Anal. Chem. 1968, 40, 1699–1704. 10.1021/ac60267a011. [DOI] [Google Scholar]
  21. Ünal S.; Wood D. G.; Harris I. J. Effects of drying methods on the low temperature reactivity of Victorian brown coal to oxygen. Fuel 1992, 71, 183–192. 10.1016/0016-2361(92)90007-b. [DOI] [Google Scholar]
  22. Dai G. L.Comprehensive Experimental Study on Low Temperature Oxidation and Spontaneous Combustion of Coal; China University of Mining and Technology Press: Xuzhou, 2005; pp 45–53.
  23. Li C.; Takanohashi T.; Saito I.; Iino M.; Moriyama R.; Kumagai H.; Chiba T. The behavior of free radicals in coal at temperatures up to 300 °C in various organic solvents, using in situ EPR spectroscopy. Energy Fuels 2002, 16, 1116–1120. 10.1021/ef010296+. [DOI] [Google Scholar]
  24. Liu J.; Jiang X.; Shen J.; Zhang H. Influences of particle size, ultraviolet irradiation and pyrolysis temperature on stable free radicals in coal. Powder Technol. 2015, 272, 64–74. 10.1016/j.powtec.2014.11.017. [DOI] [Google Scholar]
  25. Liu J.; Jiang X.; Han X.; Shen J.; Zhang H. Chemical properties of superfine pulverized coals. Part 2. Demineralization effects on free radical characteristics. Fuel 2014, 115, 685–696. 10.1016/j.fuel.2013.07.099. [DOI] [Google Scholar]
  26. Xin H.-h.; Wang D.-m.; Qi X.-y.; Qi G.-s.; Dou G.-l. Structural characteristics of coal functional groups using quantum chemistry for quantification of infrared spectra. Fuel Process. Technol. 2014, 118, 287–295. 10.1016/j.fuproc.2013.09.011. [DOI] [Google Scholar]
  27. Zhang Y.; Yang C.; Li Y.; Huang Y.; Zhang J.; Zhang Y.; Li Q. Ultrasonic extraction and oxidation characteristics of functional groups during coal spontaneous combustion. Fuel 2019, 242, 287–294. 10.1016/j.fuel.2019.01.043. [DOI] [Google Scholar]
  28. Xu T.; Wang D.-M.; Xin H.-H.; Qi X. Y. Experimental study on the temperature rising characteristic of spontaneous combustion of coal. J. China Univ. Min. Technol. 2012, 29, 575–580. [Google Scholar]
  29. Mathews J. P.; Chaffee A. L. The molecular representations of coal—A review. Fuel 2012, 96, 1–14. 10.1016/j.fuel.2011.11.025. [DOI] [Google Scholar]
  30. Classification of Coals, 2005. ISO 11760-2005, ISO/TC27.
  31. Fei J.; Li W. Y.; Xie K. C. Research on coal structure using FTIR. J. China Univ. Min. Technol. 2002, 5, 362–366. [Google Scholar]
  32. Yu M. G.; Jia H. L.; Yu S. J.; Pan R. K. Calculation of micro-structure parameter of Wuda bituminous coal and relationship-analysis between coal structure and coal spontaneous combustion. J. China Coal Soc. 2006, 31, 610–614. [Google Scholar]
  33. Wang D. M.; Xin H. H.; Qi X. Y.; Dou G. L.; Zhong X. X. Mechanism and relationships of elementary reactions in spontaneous combustions of coal: The coal oxidation kinetics theory and application. J. China Coal Soc. 2014, 39, 1667–1674. [Google Scholar]
  34. Dack S.; Hobday M.; Smith T.; Pilbrow J. Free radical involvement in the oxidation of Victorian brown coal. Fuel 1983, 62, 1510–1512. 10.1016/0016-2361(83)90123-0. [DOI] [Google Scholar]
  35. Kudynska J.; Buckmaster H. A. Low-temperature oxidation kinetics of high-volatile bituminous coal studied by dynamic in situ 9 GHz c.w. e.p.r. spectroscopy. Fuel 1996, 75, 872–878. 10.1016/0016-2361(96)00014-2. [DOI] [Google Scholar]
  36. Liotta R.; Brons G.; Isaacs J. Oxidative weathering of Illinois No. 6 Coal. Fuel 1983, 62, 781–791. 10.1016/0016-2361(83)90028-5. [DOI] [Google Scholar]
  37. Cerny J.; Pavlikova H. Structural Analysis of Low-Rank-Coal Extracts and Their Relation to Parent Coals. Energy Fuels 1994, 8, 375–379. 10.1021/ef00044a013. [DOI] [Google Scholar]

Articles from ACS Omega are provided here courtesy of American Chemical Society

RESOURCES