As an index gas of spontaneous combustion, C2H4 has been used in several coal mines, but its adsorption on residual coal during the process from generation to sampling has an impact on evaluation. The molecular structure models of lignite and anthracite were constructed through molecular mechanics and dynamics, and the Grand Canonical Monte Carlo systems were used to simulate the adsorption of C2H4 and other common gases on lignite and anthracite at temperatures of 288.15–318.15 K. The order of adsorption capacity was CO2 > C2H4 > O2 > N2 in the lignite and anthracite molecular models within the low-pressure zone. When the pressure increases, the adsorption capacity of O2 and N2 exceeded that of C2H4. The C2H4 and CO2 have similar and concentrated adsorption sites, and it is easier to reach the limit adsorption capacity; O2 and N2 have similar adsorption sites, the number of sites is much more than that of C2H4 and CO2, and their limit adsorption capacities are larger than those of C2H4 and CO2. When C2H4 is monitored in a goaf, its concentration is not quite accurate because of the adsorption of residual coal, which is different from the detected temperature of the C2H4 concentration obtained from the experiment. This is of great significance to further experiments and more on-site data statistics to fix the problem for coal mines.

In the global energy system, fossil fuels account for 85% of the total energy supply, and coal is an important part of the consumption of them.1 A current mining challenge is the spontaneous combustion of coal in goafs. The residual coal exposed during the mining process will oxidize and accumulate heat due to air leakage from the working face, and eventually, spontaneous combustion may occur in the goaf2–4 and affect productivity. The toxic and harmful gases it produces will endanger the health of workers. Therefore, the early prediction and forecast is desirable to assist coal fire prevention and control.

In China, most mines use CO as an index gas to predict and forecast the spontaneous combustion in the goaf.5–7 However, with the improvement of mining technology, more vehicles in the intake tunnel exhaust fumes to the mine ventilation, causing the CO concentration to rise and making CO prediction inaccurate.8–10 Therefore, hydrocarbon gases have advantages and have been adopted by some coal mines to replace CO. During the oxidation of coal, the aliphatic side chains of molecules are broken to form alkanes, alkenes, and alkynes.11–13 Among them, the detection temperature range of C2H4 is narrow, and the initial generation temperature of C2H4 only needs to be determined through experiments. When C2H4 is produced and monitored by a duct monitoring system, it is determined that there is an area in the goaf that has reached the initial temperature, and then the corresponding treatment measures are employed.14–16 

However, the residual coal in the goaf will adsorb some gases, including C2H4, which will inevitably reduce the accuracy of evaluation. Therefore, it is necessary to evaluate the adsorption and diffusion behavior of C2H4 on coal molecules compared to other common gases to determine the adsorption capacity of C2H4 at different temperatures and pressures.

Coal is a porous material with a large specific surface area and pore volume, which provides porous cavities for gas adsorption. Gases present within coal are retained through physical adsorption. The adsorption capacity of gases in coal is mainly affected by rank,17 moisture content,18 pore structure,19,20 pressure, temperature,21 and functional groups.22 Many scholars have made detailed studies on coal pyrolysis or spontaneous combustion to generate C2H4 and coal adsorption of CH4 and CO2. Wu and Wu23 studied the generation rules of various index gases by using the temperature-programmed approach and obtained the prediction standard for the spontaneous combustion of the goaf with CO as the main component and a combination of multiple index gases; the initial generation temperature of each index gas corresponds to the increasing coal temperature within the goaf. Guo et al.24 improved the accuracy by using the ratio of different index gases. In addition, Deng et al.25 analyzed the relationship between the ratio of different index gases and temperature. As for the study of gas adsorption on the coal surface, Guan et al.26 verified that CO2 reached the adsorption limit capacity faster than CH4 by a volumetric adsorption apparatus and verified the adaptability of the Langmuir formula to the adsorption isotherm fitting. Mastalerz et al.27 analyzed the adsorption of N2 and CO2 on coal samples in different particle sizes and found that the optimal particle size for N2 and CO2 adsorption is 60 mesh. Busch et al.28 found that in the numerical modeling of CH4 and CO2, for moist coals, adsorption rates of both gases were reduced by a factor of more than 2 with respect to dry coals. Fitzgerald et al.29 studied the adsorption of CH4, N2, and CO2 mixed gas on coal and summarized the relationship between the molar ratio and uncertainty. Vandamme et al.30 provided a comprehensive framework to calculate the macroscopic strains induced by CO2 adsorption in a porous medium from the molecular level. Tang and Ripepi31 researched adsorption of supercritical CO2 on coal samples and found that the isosteric heat depends on both temperature and adsorption uptake.

Most of the literature studies of gas adsorption on coal focused on CH4, N2, H2O, and CO2, and the relationship of competitive adsorption was obtained by the experimental and molecular simulation methods.32–35 In addition, the adsorption characteristics of C2H4 on other materials have also been explored.36–38 As for the adsorption characteristics of C2H4 and other hydrocarbon index gases on coal, related studies are scarce.

Here, the adsorption capacity of C2H4 on coal molecules is calculated based on molecular mechanics and molecular dynamics, and the adsorption capacity of other common gases in coal mines is compared. The microscopic adsorption is linked with the macroscopic situation to provide theoretical support of adjustment for judging the temperature of the goaf through the coal mine index gas monitoring system. To compare the adsorption characteristics and adsorption capacity of C2H4 and other gases of coal, this paper establishes lignite and anthracite supercell models and uses Materials Studio software to achieve it. This research has the potential to impact the monitoring of spontaneous combustion.

In China, lignite and anthracite are representative. Lignite and sulfur-containing anthracite have risk of spontaneous combustion and conditions for producing C2H4. Here, by comparing the adsorption characteristics of C2H4 and other gases in lignite and anthracite, we discussed the governing factors for using C2H4 as an index gas. The basic unit models of lignite (C206H206N2O44) and anthracite (C258H172N4O12S) were quoted from the literature of Zhu et al.39 and Liu et al.40 Their models are reasonable and have a relatively close carbon content, so they were chosen for the simulation. The Amorphous Cell Tools, a module of Materials Studio, were used to generate unit supercells containing four molecules, fixed angle (α: 90°, β: 90°, and γ: 90°). Here, we adopted a regular supercell arrangement, which was helpful to observe the adsorption sites, because the molecular structure of coal has been optimized, and the adsorption sites of gases on the coal molecules will not change significantly. The regular supercell arrangement makes the law of different gases adsorption sites more obvious. The Forcite module of Materials Studio software was used to optimize and anneal the supercells, and the charge equilibration (QEq) was used to calculate the charge distribution. A CO2 probe with a radius of 1.6 Å was used to determine the difference in the pore structure, and the Connolly free volume and the surface area were obtained.

The adsorption of C2H4 and other gases on lignite and anthracite models is simulated at the temperatures of 288.15, 298.15, 308.15, and 318.15 K. The parameters of molar mass, critical temperature, critical pressure, and acentric factor are shown in Table I,41 and the conversion of pressure and fugacity is realized by the Peng–Robinson formula.42,43

TABLE I.

Physical properties of the gases.

Physical propertiesC2H4CO2N2O2
Molar mass (g/mol) 28.06 44.01 28.01 32.00 
Critical temperature (K) 282.35 304.13 126.19 154.58 
Critical pressure (MPa) 5.04 7.38 3.40 5.04 
Acentric factor 0.086 0.22 0.04 0.022 
Physical propertiesC2H4CO2N2O2
Molar mass (g/mol) 28.06 44.01 28.01 32.00 
Critical temperature (K) 282.35 304.13 126.19 154.58 
Critical pressure (MPa) 5.04 7.38 3.40 5.04 
Acentric factor 0.086 0.22 0.04 0.022 

The “fixed pressure” module in Sorption of Materials Studio was used to simulate the adsorption capacity of gases on the coal molecular supercells. Based on the GCMC (Grand Canonical Monte Carlo) method,44–46 the adsorption isotherm of each component gas under different temperature conditions was obtained by the Langmuir formula.47,48 The Langmuir fitting formula is

y=abx/(1+bx),
(1)

where a is the limit adsorption capacity when the pressure tends to infinity (mmol/g) and b is the adsorption constant (MPa−1).

This approach determines the adsorption position based on the annealing principle in which the Metropolis algorithm is selected.

The specific setting details in the “fixed pressure” module are shown in Table II.

TABLE II.

Parameter settings in the “fixed pressure” module of Materials Studio.

SettingParameterSettingParameter
Forcefield Dreiding Equilibration steps 106 
Charges QEq Production steps 2 * 106 
Electrostatic Ewald and group Exchange 0.39 
van der Waals Atom based Conformer 0.20 
Ewald accuracy 0.001 kcal/mol Rotate 0.20 
Cutoff distance 1.25 nm Translate 0.20 
SettingParameterSettingParameter
Forcefield Dreiding Equilibration steps 106 
Charges QEq Production steps 2 * 106 
Electrostatic Ewald and group Exchange 0.39 
van der Waals Atom based Conformer 0.20 
Ewald accuracy 0.001 kcal/mol Rotate 0.20 
Cutoff distance 1.25 nm Translate 0.20 

The unit of adsorption capacity obtained from the simulation results is (average molecules/cell), which should be converted to (mmol/g) by the following equation:41 

uptake(mmol/g)=LoadingmoleculesMrcell(g/mol)×103,
(2)

where Mrcell is the relative molecular mass of the adsorbent in the supercell.

The “locate” module in Sorption of Materials Studio was used to locate the adsorption positions and show them on graphs. The specific setting details in the “locate” module are shown in Table III.

TABLE III.

Parameter settings in the “locate” module of Materials Studio.

SettingParameterSettingParameter
Forcefield Dreiding Maximum loading steps 106 
Charges QEq Production steps 2 * 106 
Electrostatic Ewald Temperature cycles 
van der Waals Atom based   
SettingParameterSettingParameter
Forcefield Dreiding Maximum loading steps 106 
Charges QEq Production steps 2 * 106 
Electrostatic Ewald Temperature cycles 
van der Waals Atom based   

The basic unit models of lignite (C206H206N2O44) and anthracite (C258H172N4O12S) are shown in Fig. 1.

FIG. 1.

Basic unit of a coal molecule. (Atomic colors) C: gray; H: white; O: red; N: blue; and S: yellow. (a) Lignite (C206H206N2O44) and (b) anthracite (C258H172N4O12S).

FIG. 1.

Basic unit of a coal molecule. (Atomic colors) C: gray; H: white; O: red; N: blue; and S: yellow. (a) Lignite (C206H206N2O44) and (b) anthracite (C258H172N4O12S).

Close modal

As shown in Fig. 2, the lignite supercell size was a = 22.47 Å, b = 22.47 Å, and c = 44.93 Å; and the anthracite supercell size was a = 21.88 Å, b = 20.24 Å, and c = 49.19 Å. The Connolly free volume of a supercell with the best structure of lignite is 5755 Å3 and the surface area is 5023 Å2. The Connolly free volume of a supercell with the best structure of anthracite is 4155 Å3 and the surface area is 4456 Å2.

FIG. 2.

Four molecules in a supercell structure. The blue area is the pore structure; atomic colors: C: gray, H: white, O: red, N: blue, and S: yellow. (a) Lignite (C824H824N8O176) and (b) anthracite (C1032H688N16O48S4).

FIG. 2.

Four molecules in a supercell structure. The blue area is the pore structure; atomic colors: C: gray, H: white, O: red, N: blue, and S: yellow. (a) Lignite (C824H824N8O176) and (b) anthracite (C1032H688N16O48S4).

Close modal

The relationship between the fugacity and pressure of each gas is shown in Fig. 3, and the fugacity coefficient is the ratio of fugacity to pressure (the lines in the figures are drawn as guides to the eye). The fugacity of each gas is mainly affected by temperature and pressure. Under the same temperature condition, the fugacity coefficient decreases with the increase in pressure, while under the same pressure condition, the fugacity coefficient increases with the increase in temperature. In the pressure range of 0–10 MPa and the temperature range of 288.15–298.15 K, the fugacity coefficients of O2 and N2 change relatively small, basically between 0.9 and 1.0, while the fugacity coefficients of C2H4 and CO2 change greatly, between 0.4 and 1.0. The main reason is that the critical temperatures of C2H4 and CO2 are higher, which are 282.65 and 304.13 K, respectively, and the critical pressures are 5.1 and 7.4 MPa, respectively. Therefore, under the simulated temperature gradient, C2H4 enters a supercritical state when the pressure exceeds 5.1 MPa, which makes the fugacity coefficient decline faster. At 288.15 and 298.15 K, when the pressure exceeds 7.4 MPa, CO2 changes from gaseous to liquid, and the fugacity coefficient declines slower. At 308.15 and 318.15 K, when the pressure exceeds 7.4 MPa, CO2 enters the supercritical state, and the fugacity coefficient declines faster. The order of the fugacity coefficient of various gases affected by temperature and pressure is CO2 > C2H4 > N2 > O2.

FIG. 3.

The relationship between the fugacity coefficient of a gas and its pressure at different temperatures. (a) C2H4, (b) CO2, (c) N2, (d) O2, and (e) legend.

FIG. 3.

The relationship between the fugacity coefficient of a gas and its pressure at different temperatures. (a) C2H4, (b) CO2, (c) N2, (d) O2, and (e) legend.

Close modal

The Langmuir fitting curves of C2H4, CO2, N2, and O2 on the lignite supercell of 288.15–318.15 K and the pressure of 0–10 MPa are shown in Figs. 4(a)4(d), while the Langmuir fitting curves on the anthracite supercell are shown in Figs. 4(e)4(h). By comparing the N2/CO2 adsorption capacity with that proposed by lignite model authors, the results are close. Therefore, the optimization of the model is reasonable, further, which means that the adsorption of C2H4 under the same simulation parameters is reasonable and credible; for the anthracite model, the scholars only established a coal molecular model without performing related simulations. We performed simulations and compared the adsorption capacity results with that of lignite and found that the simulations were reasonable.39,40

FIG. 4.

Adsorption isotherms of CO, CO2, N2, and O2 at different temperatures. (a) C2H4 adsorbed on lignite, (b) CO2 adsorbed on lignite, (c) N2 adsorbed on lignite, (d) O2 adsorbed on lignite, (e) C2H4 adsorbed on anthracite, (f) CO2 adsorbed on anthracite, (g) N2 adsorbed on anthracite, (h) O2 adsorbed on anthracite, and (i) legend.

FIG. 4.

Adsorption isotherms of CO, CO2, N2, and O2 at different temperatures. (a) C2H4 adsorbed on lignite, (b) CO2 adsorbed on lignite, (c) N2 adsorbed on lignite, (d) O2 adsorbed on lignite, (e) C2H4 adsorbed on anthracite, (f) CO2 adsorbed on anthracite, (g) N2 adsorbed on anthracite, (h) O2 adsorbed on anthracite, and (i) legend.

Close modal

As shown in Fig. 4, whether on lignite or anthracite, under the same pressure condition, as the temperature increases, the adsorption capacity of the gases decreases. This is because the increase in temperature inhibits the adsorption of gases and promotes desorption. During the adsorption process, the adsorption capacity increases rapidly at low pressure. As the pressure increases, the adsorption capacity increases slowly and the adsorption isotherm gradually tends to be flat. These trends of C2H4 and CO2 are particularly obvious.

In the temperature and pressure range, comparing Figs. 4(a) and 4(e), it can be seen that the adsorption capacity of C2H4 in lignite is 0.418–2.121 mmol/g and that in anthracite is 0.779–1.786 mmol/g. Comparing Figs. 4(b) and 4(f), it can be seen that the adsorption capacity of CO2 in lignite is 0.448–3.212 mmol/g and that in anthracite is 1.011–2.462 mmol/g. Comparing Figs. 4(c) and 4(g), it can be seen that the adsorption capacity of N2 in lignite is 0.070–2.535 mmol/g and that in anthracite is 0.159–2.211 mmol/g. Comparing Figs. 4(d) and 4(h), it can be seen that the adsorption capacity of O2 in lignite is 0.080–3.179 mmol/g and that in anthracite is 0.195–2.592 mmol/g. Langmuir fitting parameters of adsorption isotherms at different temperatures on lignite and anthracite are shown in Tables IV and V, respectively. Here, the parameters a and b come from Eq. (1), a is the limit adsorption capacity when the pressure tends to infinity, while b is the adsorption constant, related to adsorption energy. Comparing the adsorption limit capacities (parameter a) of various gases from lignite and anthracite, it can be seen that the adsorption limit capacity of each gas in lignite is greater than that in anthracite. This is because the supercell of lignite has a larger Connolly free volume and surface area than that of anthracite, which gives gas molecules more adsorption sites. The parameter b is related to adsorption energy, the larger the parameter b is, the easier the gas reaches the adsorption limit capacity. Obviously, gases on anthracite are easier to reach the limit adsorption capacity than that on lignite. On the same coal, the order of reaching the adsorption limit capacity from easy to difficult is C2H4 > CO2 > N2 > O2.

TABLE IV.

Langmuir fitting parameters of adsorption isotherms at different temperatures on lignite.

GasTemperature (K)a (mmol/g)b (MPa−1)R2
C2H4 288.15 2.12 5.20 0.968 22 
298.15 2.06 3.64 0.973 77 
308.15 2.02 2.57 0.979 26 
318.15 1.94 2.34 0.991 82 
CO2 288.15 3.29 3.23 0.979 76 
298.15 3.28 2.35 0.991 17 
308.15 3.18 1.80 0.992 47 
318.15 3.10 1.41 0.992 51 
N2 288.15 3.34 0.32 0.998 37 
298.15 3.26 0.27 0.997 95 
308.15 3.22 0.23 0.999 14 
318.15 3.09 0.21 0.998 90 
O2 288.15 4.09 0.31 0.997 38 
298.15 3.88 0.29 0.998 84 
308.15 3.88 0.23 0.998 97 
318.15 3.95 0.18 0.998 43 
GasTemperature (K)a (mmol/g)b (MPa−1)R2
C2H4 288.15 2.12 5.20 0.968 22 
298.15 2.06 3.64 0.973 77 
308.15 2.02 2.57 0.979 26 
318.15 1.94 2.34 0.991 82 
CO2 288.15 3.29 3.23 0.979 76 
298.15 3.28 2.35 0.991 17 
308.15 3.18 1.80 0.992 47 
318.15 3.10 1.41 0.992 51 
N2 288.15 3.34 0.32 0.998 37 
298.15 3.26 0.27 0.997 95 
308.15 3.22 0.23 0.999 14 
318.15 3.09 0.21 0.998 90 
O2 288.15 4.09 0.31 0.997 38 
298.15 3.88 0.29 0.998 84 
308.15 3.88 0.23 0.998 97 
318.15 3.95 0.18 0.998 43 
TABLE V.

Langmuir fitting parameters of adsorption isotherms at different temperatures on anthracite.

GasTemperature (K)a (mmol/g)b (MPa−1)R2
C2H4 288.15 1.70 20.13 0.876 00 
298.15 1.67 13.27 0.966 42 
308.15 1.65 11.35 0.924 13 
318.15 1.54 9.29 0.916 01 
CO2 288.15 2.38 16.13 0.903 08 
298.15 2.33 11.86 0.920 17 
308.15 2.29 10.31 0.939 36 
318.15 2.26 6.92 0.929 77 
N2 288.15 2.33 0.97 0.988 17 
298.15 2.30 0.78 0.990 43 
308.15 2.27 0.64 0.993 68 
318.15 2.20 0.56 0.994 35 
O2 288.15 2.87 0.89 0.990 64 
298.15 2.82 0.70 0.989 77 
308.15 2.75 0.60 0.989 73 
318.15 2.62 0.55 0.992 15 
GasTemperature (K)a (mmol/g)b (MPa−1)R2
C2H4 288.15 1.70 20.13 0.876 00 
298.15 1.67 13.27 0.966 42 
308.15 1.65 11.35 0.924 13 
318.15 1.54 9.29 0.916 01 
CO2 288.15 2.38 16.13 0.903 08 
298.15 2.33 11.86 0.920 17 
308.15 2.29 10.31 0.939 36 
318.15 2.26 6.92 0.929 77 
N2 288.15 2.33 0.97 0.988 17 
298.15 2.30 0.78 0.990 43 
308.15 2.27 0.64 0.993 68 
318.15 2.20 0.56 0.994 35 
O2 288.15 2.87 0.89 0.990 64 
298.15 2.82 0.70 0.989 77 
308.15 2.75 0.60 0.989 73 
318.15 2.62 0.55 0.992 15 

The adsorbed gas molecules are all non-polar, so the adsorption capacity of each gas depends mainly on the critical temperature and critical pressure themselves. The pressure in the goaf is generally slightly higher than the atmospheric pressure. Therefore, this article focuses on the study of the adsorption laws and characteristics of C2H4 and other common gases of mine in the low-pressure stage. Taking the temperature condition of 288.15 K as an example, the adsorption capacity of each gas on lignite and anthracite is analyzed. In the lower pressure stage, it can be seen that the adsorption capacity of lignite (0–3 MPa) and anthracite (0–1.5 MPa) is positively related to the critical temperature. As shown in Table I, the higher the critical temperature, the greater the adsorption capacity. The order of adsorption capacity is CO2 > C2H4 > O2 > N2. Among them, the critical temperatures of CO2 and C2H4 are higher, which are closer to the simulated temperature, so they can reach the limit adsorption capacity faster, while the critical temperatures of N2 and O2 are lower and will not reach the limit adsorption capacity quickly like CO2 and C2H4. In the fitting results of Langmuir adsorption isotherms of lignite, O2 and N2 exceed C2H4 at 3 and 5 MPa, respectively, while in the fitting results of Langmuir adsorption isotherms of anthracite, O2 and N2 exceed C2H4 at 1.5 and 2.5 MPa, respectively, and the adsorption capacity of O2 gradually exceeds that of CO2 after 5 MPa. This is because the molecular model of anthracite contains the sulfur-containing functional groups. The O2 molecule has a relatively strong interaction with the C–S bond and the S–H bond. When O2 molecules are adsorbed on the surface of coal, there is a large reduction in the vibration frequency of the C–S bond and S–H bond, which can adsorb more O2 molecules.49 Therefore, the coal molecules with sulfur-containing functional groups are easy to adsorb more O2 molecules, and it is manifested as an increase in the limit adsorption capacity of O2.

Taking the adsorption of 288.15 K and 0.1 MPa as an example, the adsorption position of gas molecules on the lignite and anthracite was obtained. The regular arrangement of molecules in the supercell makes the comparison of the adsorption sites of different gases more obvious. Through the analysis and comparison of the adsorption sites of various gas molecules on lignite, it can be found from Fig. 5 that CO2 and C2H4 have relatively close concentrated adsorption sites, located in the areas where oxygen-containing functional groups are concentrated, while O2 and N2 also have relatively close concentrated adsorption sites, located in the areas with higher local density, and the unique adsorption sites of O2 and N2 are not reflected in the adsorption of CO2 and C2H4. Through the analysis and comparison of the adsorption sites of various gas molecules on anthracite, it can be found from Fig. 6 that CO2 and C2H4 have relatively close concentrated adsorption sites, located in the areas where oxygen-containing functional groups are concentrated, while O2 and N2 also have relatively close concentrated adsorption sites, located in the areas where sulfur-containing functional groups are concentrated. It can be seen that the adsorption sites of O2 and N2 are fundamentally different from those of CO2 and C2H4.

FIG. 5.

Adsorption sites of gases on lignite. (a) C2H4, (b) CO2, (c) N2, and (d) O2.

FIG. 5.

Adsorption sites of gases on lignite. (a) C2H4, (b) CO2, (c) N2, and (d) O2.

Close modal
FIG. 6.

Adsorption sites of gases on anthracite. (a) C2H4, (b) CO2, (c) N2, and (d) O2.

FIG. 6.

Adsorption sites of gases on anthracite. (a) C2H4, (b) CO2, (c) N2, and (d) O2.

Close modal

By using the Dreiding force field, there is only physical adsorption in the simulation. Through the observation of the simulation results, it is found that the isosteric heat is always less than 10 kcal/mol, further verifying that the form of adsorption of various gases is only physical adsorption.39,41

In the low-pressure zone, the adsorption capacity of gas molecules is relatively low, and the adsorption sites of each gas molecules are more obvious. The order of gas molecule size is C2H4 > N2 > O2 > CO2, and this is not reflected in the law of adsorption sites. Therefore, this phenomenon has nothing to do with the size of the molecular diameter and the size of the coal supercell voids. In fact, C2H4 and CO2 exhibit the same characteristics and N2 and O2 exhibit another. N2 and O2 are non-polar diatomic molecules without electric charges on atoms. The physical adsorption mainly depends on the van der Waals forces, while the C2H4 and CO2 have electric charges on atoms, and the physical adsorption is affected by electrostatic interactions and van der Waals forces. Part of the oxygen-containing functional groups of coal molecules have hydrogen bonds, such as –OH, and they are easier to adsorb CO2 and C2H4. Therefore, the adsorption of CO2 and C2H4 is relatively concentrated, which is the most obvious in lignite. It can be seen from Fig. 6 that CO2 and C2H4 are mostly adsorbed near the oxygen-containing functional groups. In addition, the critical temperatures of CO2 and C2H4 are higher, and they can maintain a higher density than N2 and O2, so their adsorption efficiency is higher, and more gas molecules can be adsorbed when the pressure is low.

The existence of these two different forms of adsorption sites makes the adsorption of CO2 and C2H4 in coal easier to reach the limit adsorption capacities. Although O2 and N2 have lower adsorption efficiency in the low-pressure zone, there are more available adsorption sites for adsorption (such as the adsorption site on the left side of the lignite) in the high-pressure zone, causing their limit adsorption capacities to be larger than that of CO2 and C2H4, that is, the adsorption sites of CO2 and C2H4 are more selective than N2 and O2.

  1. In the lignite and anthracite molecular models, within the low-pressure zone (lignite: 0–3 MPa and anthracite: 0–1.5 MPa), the order of adsorption capacity of C2H4 and other gases is CO2 > C2H4 > O2 > N2. As the pressure increases, the adsorption capacity of O2 and N2 has exceeded that of C2H4 (lignite: 3 and 5 MPa, respectively, and anthracite: 1.5 and 2.5 MPa, respectively).

  2. C2H4 and CO2 have similar and concentrated adsorption sites, and it is easier to reach the limit adsorption capacity; O2 and N2 have similar adsorption sites, the number of sites is much more than that of C2H4 and CO2, and their limit adsorption capacities are larger than those of C2H4 and CO2. Gases on anthracite are easier to reach the limit adsorption capacity than lignite. On the same coal, the order of reaching the adsorption limit capacity from easy to difficult is C2H4 > CO2 > N2 > O2.

  3. The simulation results show that under a low-pressure zone, the adsorption capacity of C2H4 in lignite and anthracite is very strong. Therefore, in the spontaneous combustion monitoring of goaf, when C2H4 is monitored, its concentration is not quite accurate because of adsorption of residual coal, which is different from the detected temperature of C2H4 concentration obtained from the experiment. Therefore, further experiments and more on-site data statistics are required for correction.

This study was financially supported by the National Natural Science Foundation of China (Grant Nos. 51574143 and 51974149).

The data that support the findings of this study are available from the corresponding author upon reasonable request.

1.
S. J.
Friedmann
, “
Geological carbon dioxide sequestration
,”
Elements
3
(
3
),
179
184
(
2007
).
2.
Y.
Liu
,
H.
Wen
,
J.
Guo
,
Y.
Jin
,
G.
Wei
, and
Z.
Yang
, “
Coal spontaneous combustion and N2 suppression in triple goafs: A numerical simulation and experimental study
,”
Fuel
271
,
117625
(
2020
).
3.
Y.
Zhang
,
K.
Niu
,
W.
Du
,
J.
Zhang
,
H.
Wang
, and
J.
Zhang
, “
A method to identify coal spontaneous combustion-prone regions based on goaf flow field under dynamic porosity
,”
Fuel
288
,
119690
(
2021
).
4.
Q.
Zuo
,
J.
Li
, and
Y.
Wang
, “
Distribution law for the danger area for spontaneous coal combustion in a dynamic goaf with low air leakage speed
,”
Therm. Sci.
2020
,
308
.
5.
D.
Burgess
and
H.
Hayden
, “
Carbon monoxide index monitoring system in an underground coal mine. [12 refs]
,”
Trans. Soc. Min. Eng. AIME
260
(
4
),
312
318
(
1976
).
6.
K.
Luft
, “
Infrared techniques for the measurement of carbon monoxide
,”
Ann. Occup. Hyg.
18
(
1
),
45
51
(
1975
).
7.
J. H.
Rowland
,
L.
Yuan
, and
R. A.
Thomas
, “
Evaluation of carbon monoxide and smoke sensors at a low ventilation velocity
,”
Min. Metall. Explor.
38
,
603
608
(
2020
).
8.
H.
Wen
,
Z.
Yu
,
S.
Fan
,
X.
Zhai
, and
W.
Liu
, “
Prediction of spontaneous combustion potential of coal in the gob area using CO extreme concentration: A case study
,”
Combust. Sci. Technol.
189
(
10
),
1713
1727
(
2017
).
9.
M.
Więckowski
,
N.
Howaniec
, and
A.
Smoliński
, “
Natural desorption of carbon monoxide during the crushing of coal simulating natural rock mass pressure
,”
Sci. Total Environ.
736
,
139639
(
2020
).
10.
J.
Zhang
,
J.
Wang
,
C.
Zhang
,
Z.
Li
,
J.
Zhu
, and
B.
Lu
, “
Molecular simulation of gases competitive adsorption in lignite and analysis of original CO desorption
,”
Sci. Rep.
11
(
1
),
11706
(
2021
).
11.
J.-L.
Li
,
W.
Lu
,
B.
Kong
,
Y.-J.
Cao
,
G.-S.
Qi
, and
C.-R.
Qin
, “
Mechanism of gas generation during low-temperature oxidation of coal and model compounds
,”
Energy Fuels
33
(
2
),
1527
1539
(
2019
).
12.
W.
Xu
and
A.
Tomita
, “
Effect of coal type on the flash pyrolysis of various coals
,”
Fuel
66
(
5
),
627
631
(
1987
).
13.
W.-C.
Xu
and
A.
Tomita
, “
The effects of temperature and residence time on the secondary reactions of volatiles from coal pyrolysis
,”
Fuel Process. Technol.
21
(
1
),
25
37
(
1989
).
14.
J.
Deng
,
J.
Zhao
,
Y.
Zhang
, and
R.
Geng
, “
Study on coal spontaneous combustion characteristic temperature of growth rate analysis
,”
Procedia Eng.
84
,
796
805
(
2014
).
15.
H.-Y.
Niu
,
X.-L.
Deng
,
S.-L.
Li
,
K.-X.
Cai
,
H.
Zhu
,
F.
Li
, and
J.
Deng
, “
Experiment study of optimization on prediction index gases of coal spontaneous combustion
,”
J. Cent. South Univ.
23
(
9
),
2321
2328
(
2016
).
16.
Y.
Xiao
,
Q.-W.
Li
,
J.
Deng
,
C.-M.
Shu
, and
W.
Wang
, “
Experimental study on the corresponding relationship between the index gases and critical temperature for coal spontaneous combustion
,”
J. Therm. Anal. Calorim.
127
(
1
),
1009
1017
(
2017
).
17.
Y.
Wang
,
D.
Liu
,
Y.
Cai
, and
X.
Li
, “
Variation of petrophysical properties and adsorption capacity in different rank coals: An experimental study of coals from the Junggar, Ordos and Qinshui Basins in China
,”
Energies
12
(
6
),
986
(
2019
).
18.
Z.-G.
Zhang
,
S.-G.
Cao
,
Y.
Li
,
P.
Guo
,
H.
Yang
, and
T.
Yang
, “
Effect of moisture content on methane adsorption-and desorption-induced deformation of tectonically deformed coal
,”
Adsorpt. Sci. Technol.
36
(
9–10
),
1648
1668
(
2018
).
19.
S.
Chen
,
S.
Tao
,
D.
Tang
,
H.
Xu
,
S.
Li
,
J.
Zhao
,
Q.
Jiang
, and
H.
Yang
, “
Pore structure characterization of different rank coals using N2 and CO2 adsorption and its effect on CH4 adsorption capacity: A case in Panguan syncline, western Guizhou, China
,”
Energy Fuels
31
(
6
),
6034
6044
(
2017
).
20.
B.
Nie
,
X.
Liu
,
L.
Yang
,
J.
Meng
, and
X.
Li
, “
Pore structure characterization of different rank coals using gas adsorption and scanning electron microscopy
,”
Fuel
158
,
908
917
(
2015
).
21.
Y.
Liu
,
Y.
Zhu
,
S.
Liu
,
W.
Li
, and
X.
Tang
, “
Temperature effect on gas adsorption capacity in different sized pores of coal: Experiment and numerical modeling
,”
J. Pet. Sci. Eng.
165
,
821
830
(
2018
).
22.
Y.
Dang
,
L.
Zhao
,
X.
Lu
,
J.
Xu
,
P.
Sang
,
S.
Guo
,
H.
Zhu
, and
W.
Guo
, “
Molecular simulation of CO2/CH4 adsorption in brown coal: Effect of oxygen-, nitrogen-, and sulfur-containing functional groups
,”
Appl. Surf. Sci.
423
,
33
42
(
2017
).
23.
Y.-g.
Wu
and
J.-m.
Wu
, “
Experimental study on significant gases of coal spontaneous combustion by temperature programmed (TP)
,”
Procedia Eng.
26
,
120
125
(
2011
).
24.
J.
Guo
,
H.
Wen
,
X.
Zheng
,
Y.
Liu
, and
X.
Cheng
, “
A method for evaluating the spontaneous combustion of coal by monitoring various gases
,”
Process Saf. Environ. Prot.
126
,
223
231
(
2019
).
25.
J.
Deng
,
L.
Liu
,
C.
Lei
,
C.
Wang
, and
Y.
Xiao
, “
Spatiotemporal distributions of the temperature and index gases during the dynamic evolution of coal spontaneous combustion
,”
Combust. Sci. Technol.
193
,
1679
1695
(
2019
).
26.
C.
Guan
,
S.
Liu
,
C.
Li
,
Y.
Wang
, and
Y.
Zhao
, “
The temperature effect on the methane and CO2 adsorption capacities of Illinois coal
,”
Fuel
211
,
241
250
(
2018
).
27.
M.
Mastalerz
,
L.
Hampton
,
A.
Drobniak
, and
H.
Loope
, “
Significance of analytical particle size in low-pressure N2 and CO2 adsorption of coal and shale
,”
Int. J. Coal Geol.
178
,
122
131
(
2017
).
28.
A.
Busch
,
Y.
Gensterblum
,
B. M.
Krooss
, and
R.
Littke
, “
Methane and carbon dioxide adsorption–diffusion experiments on coal: Upscaling and modeling
,”
Int. J. Coal Geol.
60
(
2–4
),
151
168
(
2004
).
29.
J. E.
Fitzgerald
,
Z.
Pan
,
M.
Sudibandriyo
,
R. L.
Robinson
, Jr.
,
K. A. M.
Gasem
, and
S.
Reeves
, “
Adsorption of methane, nitrogen, carbon dioxide and their mixtures on wet Tiffany coal
,”
Fuel
84
(
18
),
2351
2363
(
2005
).
30.
M.
Vandamme
,
L.
Brochard
,
B.
Lecampion
, and
O.
Coussy
, “
Adsorption and strain: The CO2-induced swelling of coal
,”
J. Mech. Phys. Solids
58
(
10
),
1489
1505
(
2010
).
31.
X.
Tang
and
N.
Ripepi
, “
High pressure supercritical carbon dioxide adsorption in coal: Adsorption model and thermodynamic characteristics
,”
J. CO2 Util.
18
,
189
197
(
2017
).
32.
Y.
Li
,
Z.
Yang
, and
X.
Li
, “
Molecular simulation study on the effect of coal rank and moisture on CO2/CH4 competitive adsorption
,”
Energy Fuels
33
(
9
),
9087
9098
(
2019
).
33.
S.
Wu
,
Z.
Jin
, and
C.
Deng
, “
Molecular simulation of coal-fired plant flue gas competitive adsorption and diffusion on coal
,”
Fuel
239
,
87
96
(
2019
).
34.
S.
Yu
,
J.
Bo
, and
L.
Fengjuan
, “
Competitive adsorption of CO2/N2/CH4 onto coal vitrinite macromolecular: Effects of electrostatic interactions and oxygen functionalities
,”
Fuel
235
,
23
38
(
2019
).
35.
Y.
Zheng
,
Q.
Li
,
G.
Zhang
,
Y.
Zhao
,
P.
Zhu
,
X.
Ma
, and
X.
Liu
, “
Effect of multi-component gases competitive adsorption on coal spontaneous combustion characteristics under goaf conditions
,”
Fuel Process. Technol.
208
,
106510
(
2020
).
36.
A. S.
Rad
, “
Adsorption of C2H2 and C2H4 on Pt-decorated graphene nanostructure: Ab-initio study
,”
Synth. Met.
211
,
115
120
(
2016
).
37.
M.
Valero
,
P.
Raybaud
, and
P.
Sautet
, “
Interplay between molecular adsorption and metal–support interaction for small supported metal clusters: CO and C2H4 adsorption on Pd4/γ-Al2O3
,”
J. Catal.
247
(
2
),
339
355
(
2007
).
38.
I.
van Zandvoort
,
J. K.
van der Waal
,
E.-J.
Ras
,
R.
de Graaf
, and
R.
Krishna
, “
Highlighting non-idealities in C2H4/CO2 mixture adsorption in 5A zeolite
,”
Sep. Purif. Technol.
227
,
115730
(
2019
).
39.
H.
Zhu
,
W.
Wang
,
Y.
Huo
,
X.
He
,
H.
Zhao
, and
H.
Wang
, “
Molecular simulation study on adsorption and diffusion behaviors of CO2/N2 in lignite
,”
ACS Omega
5
(
45
),
29416
29426
(
2020
).
40.
J.
Liu
,
Y.
Jiang
,
W.
Yao
,
X.
Jiang
, and
X.
Jiang
, “
Molecular characterization of Henan anthracite coal
,”
Energy Fuels
33
(
7
),
6215
6225
(
2019
).
41.
D.
Gao
,
L.
Hong
,
J.
Wang
, and
D.
Zheng
, “
Molecular simulation of gas adsorption characteristics and diffusion in micropores of lignite
,”
Fuel
269
,
117443
(
2020
).
42.
D. B.
Robinson
,
D.-Y.
Peng
, and
H.-J.
Ng
, “
Applications of the Peng-Robinson equation of state
,”
ACS Symp. Ser.
60
,
200
220
(
1977
).
43.
D. B.
Robinson
,
D.
Peng
, and
S. Y.
Chung
, “
The development of the Peng-Robinson equation and its application to phase equilibrium in a system containing methanol
,”
Fluid Phase Equilib.
24
(
1–2
),
25
41
(
1985
).
44.
H.
Aljamaan
,
M.
Al Ismail
, and
A. R.
Kovscek
, “
Experimental investigation and grand canonical Monte Carlo simulation of gas shale adsorption from the macro to the nano scale
,”
J. Nat. Gas Sci. Eng.
48
,
119
137
(
2017
).
45.
Y.
Liu
,
Y.
Zhu
,
W.
Li
,
J.
Xiang
,
Y.
Wang
,
J.
Li
, and
F.
Zeng
, “
Molecular simulation of methane adsorption in shale based on grand canonical Monte Carlo method and pore size distribution
,”
J. Nat. Gas Sci. Eng.
30
,
119
126
(
2016
).
46.
W.
Song
,
J.
Yao
,
J.
Ma
,
A.
Li
,
Y.
Li
,
H.
Sun
, and
L.
Zhang
, “
Grand canonical Monte Carlo simulations of pore structure influence on methane adsorption in micro-porous carbons with applications to coal and shale systems
,”
Fuel
215
,
196
203
(
2018
).
47.
S.
Hao
,
J.
Wen
,
X.
Yu
, and
W.
Chu
, “
Effect of the surface oxygen groups on methane adsorption on coals
,”
Appl. Surf. Sci.
264
,
433
442
(
2013
).
48.
T. C.
Ruppel
,
C. T.
Grein
, and
D.
Bienstock
, “
Adsorption of methane on dry coal at elevated pressure
,”
Fuel
53
(
3
),
152
162
(
1974
).
49.
C.
Deng
,
X.
Wang
,
J.
Wang
,
H.
Deng
, and
B.
Ye
, “
Physical adsorption mechanism of coal surface containing sulfur group adsorption to more oxygen molecule
,”
J. China Coal Soc.
33
(
5
),
556
560
(
2008
) (in Chinese).