Experimental Study of CO₂-ECBM by Injection Liquid CO₂

Abstract

Coal mine gas disasters have severely restricted production safety. Improving gas extraction efficiency can effectively reduce disasters. Scholars have confirmed that CO₂ successfully displaces coal seam CH₄. This study conducted displacement and in situ experiments and compared gas drainage under different injection pressures. The displacement experiments indicated that CH₄ production rates increased under increased pressures while the displacement ratios decreased. The pressure had a positive effect on sweep efficiency. The in situ experiment showed that CH₄ and CO₂ concentration trends in the inspection hole remained consistent. Through observing the data of the original and inspection holes, the average gas drainage concentration during low- and medium-pressure injections increased by 0.61 times and 1.17 times, respectively. The low-pressure average gas drainage scalar was increased by 1.08 times. During the medium-pressure injection, the average gas drainage purity increased by 1.94 times. The diffusion ranges of CO₂ under low- and medium-pressure injections were 20–25 m and 25–30 m, respectively. The sweep efficiency of medium-pressure injection was 26% better than that of the low-pressure injection, with average pressures of 2.8 MPa and 1.4 MPa, respectively, for sweep efficiency. This study proposes an effective method for improving coal mine gas drainage efficiency.

1. Introduction

Gas disasters are one of the most important factors affecting the safety of coal mining. Gas drainage is the most direct and effective method for controlling gas disasters and is also an effective way to obtain clean energy [1]. However, there are still some problems in coal mine gas extraction in low-permeability coal seams in China, such as a small influence range, difficult extraction, and fast attenuation [2,3]. Data statistics show that more than 95% of outburst and high-gas mines in China are in low-permeability coal seams, and the permeability is mostly 10⁻⁶–10⁻⁷ μm² [4]. The gas extraction technology for low-permeability coal seams at home and abroad focus on improving coal seam permeability and increasing gas desorption [5,6,7].

Domestic and foreign scholars have developed a variety of coal seam permeability enhancements and gas drainage technologies and have carried out field tests in many mining areas to improve coalbed methane recovery. These include pore fracture reconstruction technologies such as hydraulic fracturing, explosive blasting, shock wave fracturing, L-CO₂ blasting, and L-CO₂ fracturing [8,9,10,11,12,13,14], and gas desorption technologies such as gas injection, heat injection, and acoustic waves [15,16,17,18,19].

However, these technologies have drawbacks. For instance, hydraulic fracturing and hydraulic cutting require large amounts of water. During the extraction process, the water block effect significantly affects methane extraction. Detonation wave fracturing can result in the formation of a stress-concentrated area in the coalbed, thus increasing the possibility of accidents. Therefore, these technologies do not significantly improve the efficiency of coalbed methane recovery in coal seams. The gas phase displacement technology, which uses the competitive adsorption mechanism of adsorbed gas, increases the amount of gas desorption and improves the production of coalbed methane [20]. The study of gas displacement in coal seams is mainly concentrated on single or multiple gas injection experiments using CO₂ and N₂ gas, and the effect of gas displacement is evaluated by observing the replacement rate after gas injection [18]. The mechanism of gas phase displacement was revealed by studying the coal adsorption capacity to different gases and the competitive adsorption law of mixed gases [21,22].

In foreign countries, the first experimental study on coal gas displacement was conducted in a laboratory. CO₂ gas was injected into the coal core at a low pressure [23]. Gas injection was found to significantly improve the recovery rate of CH₄ through a comparison of the emission and natural desorption rates [24,25,26]. After injecting CO₂ into coal, the changes in the coal seam gas pressure and permeability were studied, and the key parameters in the process of methane replacement were obtained [27,28]. The injection of CO₂, N₂, and mixed gas for the enhanced coalbed methane recovery was also studied. The results revealed that the recovery efficiency of CH₄ by pure CO₂ was higher than the recovery efficiency by pure N₂ [29]. Mixed-gas injection reduces the efficiency of CH4 recovery. L-CO₂ is a cryogenic fluid with advantages such as a low viscosity, easy seepage diffusion, and high adsorption capacity. Injecting CO₂ into a coalbed at a constant low pressure improves the coalbed permeability under frost heaving and phase transformation forces generated by L-CO₂ [30,31,32].

Therefore, based on the low-viscosity permeability and displacement characteristics of L-CO₂, this study designs an engineering test of L-CO₂ displacing coal seam CH₄ technology [33,34]. Through the injection at different displacement pressures, the gas drainage concentration and purity after displacement were investigated, and the sweep efficiencies of different displacement pressures were compared and analyzed. This study provides a theoretical basis for L-CO₂ displacing coal seam CH₄ technology and new technical methods to improve gas drainage efficiency.

2. Theory and Experiment

2.1. Theoretical Analyses

According to the Langmuir adsorption theory, assuming that gas adsorption occurs on a monolayer solid surface, the proportion of adsorbent surface occupied θ is the coverage, and the ratio of the blank surface is 1 − θ [35].

$$\theta =\frac{\mathrm{b}P}{1+bP}$$

(1)

where θ is the coverage, b is the adsorption coefficient, and P is the equilibrium pressure. Assuming that a certain amount of CH₄ has been adsorbed by coal, the adsorption law after CO₂ injection can be expressed by a thermodynamic equation.

$$\frac{dx}{dt}=k{\theta }_{{\mathrm{CO}}_2}{\theta }_{{\mathrm{CH}}_4}=\frac{{\mathrm{kb}}_{{\mathrm{CO}}_2}{P}_{{\mathrm{CO}}_2}{b}_{{\mathrm{CH}}_4}{P}_{{\mathrm{CH}}_4}}{{\left(1+b_{{\mathrm{CO}}_2}{P}_{{\mathrm{CO}}_2}+b_{{\mathrm{CH}}_4}{P}_{{\mathrm{CH}}_4}\right)}^2}$$

(2)

where dx/dt is the desorption rate and ${\theta }_{{\mathrm{CO}}_2}$, ${\theta }_{\mathrm{C}{\mathrm{H}}_4}$ are the coverage of CO₂ and CH₄. The coal adsorption capacity to CO₂ is stronger than that of CH₄, and it can be concluded that ${\mathrm{b}}_{{\mathrm{CH}}_4}{P}_{{\mathrm{CH}}_4}\le {\mathrm{b}}_{\mathrm{C}{\mathrm{O}}_2}{P}_{\mathrm{C}{\mathrm{O}}_2}+1$. The desorption rate can be simplified as follows:

$$\frac{dx}{dt}\approx \frac{{\mathrm{kb}}_{{\mathrm{CO}}_2}{P}_{{\mathrm{CO}}_2}{b}_{{\mathrm{CH}}_4}{P}_{{\mathrm{CH}}_4}}{{\left(1+b_{{\mathrm{CO}}_2}{P}_{{\mathrm{CO}}_2}\right)}^2}$$

(3)

The CH₄ partial pressure remains constant, and with the increase in CO₂ partial pressure, the replacement rate will have a maximum value. The partial pressure of CO₂ is constant when the partial pressure of CH₄ decreases gradually, and the replacement rate range decreases linearly [36].

When the CH₄ extraction was carried out by bed drilling, the permeability of the coal seam in the vertical direction was lower than that along the coal seam’s horizontal direction. Therefore, the Darcy seepage flow along the bedding borehole by CO₂ injection can be expressed as follows [31]:

$$Q=A\frac{K}{\mu }\frac{\Delta P}{L_x}$$

(4)

where Q is the flow rate in m³/s, A is the area of the cross-section in m², and μ is the viscosity of the fluid (μPa · s). K is the coefficient of proportionality, called permeability, in μm². ΔP/L_x is the pressure loss when a fluid flows across a distance L in a porous medium, MPa/m. However, in situ test results have revealed that the efficiency of CH₄ extraction is not heavily dependent on the radius and length of the borehole. In the process of enhancing coalbed methane by injecting CO₂ into the coal seam, the influence range of CO₂ was primarily affected by pressure.

When injecting liquid CO₂ into the coal seam, it was affected by three comprehensive effects: phase change pressurization, low-viscosity permeability, and displacement desorption. Under the freezing or heating phase change pressurization, the expansion and extension of original fractures and the generation of new fractures increase permeability. The seepage and diffusion of CO₂ into the coal and the competitive adsorption with CH₄ at the corresponding adsorption sites are generated under a pressure difference and concentration gradient. Finally, under the partial pressure and concentration difference of CO₂ injection, CH₄ gas on the adsorption site in the coal matrix was replaced and displaced, so that it migrated and diffused along the coal seam gas seepage channel to the gas drainage borehole, thereby improving the efficiency of coal seam gas drainage [37,38].

2.2. Experiment of Displacing Coal Seam CH₄ by CO₂

Since the main role in the whole process is the driven influence and displacement of CO₂. Compared with the whole displacement process, the L-CO₂ phase transition time (about 120 min) is shorter, and the influence range is limited. CO₂ plays a more important role. Additionally, L-CO₂ was easy to transport in the underground test and could be stored in large quantities in the CH₄ displacement experiment. Therefore, the gas phase displacement process is considered in a laboratory displacement experiment.

(1)

Experimental preparation

The coal sample was taken from the #4 coal seam in the Mengcun coal mine of Binchang mining area in Shaanxi Province. A complete coal block of approximately 0.8 m × 0.8 m was taken from mining area, and 6 core samples of approximately 25 mm × 60 mm were chosen by the sampling drilling tool. The coal sample had a different microstructure than the pores and cracks of the coal seam. After the permeability test, three coal samples with similar permeabilities were selected. Therefore, the coal sample of the same coal block had little influence on the experimental results. The coal samples were placed into the gripper after drying, and the tightness of the gripper was checked before the experiment. CH₄ was injected into the coal samples by the pressure injection system to ensure that the coal samples fully absorbed CH₄, and then CO₂ was injected. After the pressure balanced out, the flow rate and gas composition of the output gas were monitored by adjusting the outlet pressure. The main parameters of the displacement experiments are listed in

Table 1. The main parameters of displacement experiment.

Numbering	Size of Coal Sample (mm)	CH4 Adsorption Pressure (MPa)	CH4 Adsorption Time (h)	Confining Pressure (MPa)	CO2 Injection Pressure (MPa)	Back Pressure (MPa)	Temperature (°C)
1#	25.1 × 60.0	1.0–2.0	16–24	8	2.0–3.0	1.0–1.5	30
2#	24.9 × 61.0	1.0–2.0	16–24	8	4.0–5.0	2.0–3.0	30
3#	25.0 × 59.0	1.0–2.0	16–24	8	6.0–7.0	4.0–5.0	30

. The pressure and temperature corresponding to the critical point of CO₂ are 7.38 MPa and 31.04 °C. To ensure the gas phase state of CO₂ during injection, the experimental temperature was set at 30 °C and the injection pressure was controlled between 2.0 MPa and 7.0 MPa.

(2)

Experimental structure

The experimental system for displacing CH₄ from the coal seam by injecting CO₂ is composed of CH₄, a CO₂ injection system, a temperature control system, a core holder, a pressure control system, a data acquisition system, and a gas collection system. The experimental setup is shown in Figure 1. The CO₂ injection system was pressurized using a booster pump with a maximum pressure of 15 MPa. The temperature control unit used a constant-temperature control box to adjust the ambient temperature of the holder. The instantaneous flows and cumulative flows of CH₄ and CO₂ were measured using a mass flowmeter. The data acquisition system collected the flow and pressure information in the injection process, the confining pressure of the holder, the back pressure of the outlet, and the flow and concentration of the outlet mixed gas. The gas acquisition system measured the instantaneous flow and cumulative flow of the mixed gas using a flow meter, and the concentration of the mixed gas was analyzed by gas chromatography [39].

(3)

Experimental procedure

• Vacuuming: The inlet stop valve was closed, and the vacuum pump was connected to the sample holder at the outlet. The outlet stop valve and vacuum pump were opened for 5–6 h.
• Setting confining pressure: The valve was closed at the outlet. The holder was applied with a confining pressure of 8.0 MPa.
• CH₄ adsorption equilibrium: The CH₄ cylinder was opened to set to a pressure of 1.5 MPa while opening the inlet valve. To ensure that the coal sample fully adsorbed the CH₄, the adsorption time was more than 24 h, and the changes in flow and pressure were monitored during the adsorption process.
• CO₂ injection: The CH₄ intake valve was closed, the CO₂ intake valve was opened, and the pressure was adjusted to the target value. After the pressure was balanced at both ends, the back pressure valve was adjusted, and the flow of the discharged gas was monitored and collected.
• Gas analysis: The outlet mixed gas was analyzed by gas chromatography.
• Pressure relief: The intake valve was closed. The internal pressure of the gripper was released by adjusting the back-pressure valve. The next set of tests were performed after the pressure was released.

With the increasing injection of CO₂ gas, CO₂ seeped from coal fractures and diffused into the coal matrix pores. Finally, CO₂ broke through the coal sample, and the concentration of CO₂ gradually increased, while the concentration of CH₄ gas began to decrease. Because the coal adsorption capacity to CO₂ was stronger than CH₄, CO₂ competed with CH₄ to adsorb on the pore surface, replaced the adsorbed CH₄ on the inner surface of the coal pore, and converted it into free CH₄. The migration and evolution process of CH₄ went through desorption–diffusion–seepage under the action of concentration and pressure gradient, then diffused and seeped from coal pores and coal fractures to the coal surface. The concentration changes of CH₄ and CO₂ are shown in Figure 2. The higher the pressure, the faster the decrease in the CH₄ concentration. When the concentration of CH₄ decreased to 5%, the injection pressures of 6.0 MPa, 4.0 MPa, and 2.0 MPa corresponded to the corresponding times of CH₄ change 3.8 h, 4.5 h, 5.1 h, respectively. Injection pressure also had a positive effect on the CO₂ concentration. When the concentration of CO₂ was increased to 95%, the corresponding times of CO₂ change were 3.5 h, 4.5 h, and 5.5 h at pressures of 6.0 MPa, 4.0 MPa, and 2.0 MPa, respectively. The cumulative injection amount of CO₂ was larger at higher pressures. A greater seepage force was provided by a higher injection pressure. Higher pressure provided a driving force for CH₄ migration in the coal sample and improved the sweep efficiency.

According to the experimental results in

Table 2. The experimental results of displacement of CH₄ by CO₂.

Order Number	Injection Volume/mL		Injection Pressure/MPa	Output Volume/mL		CO2 Storage/mL	Replacement Ratio	Sweep Efficiency/%
	CH4	CO2		CH4	CO2
1	779	4370	2.0	340	3060	1310	3.85	43.6
2	851	5192	4.0	462	3738	1454	3.15	54.3
3	900	6220	6.0	636	4664	1556	2.45	70.7

, the CH₄ production rates of coal samples under different pressures (2.0 MPa, 4.0 MPa, and 6.0 MPa) were 43.6%, 54.3%, and 70.7%, respectively. The CH₄ production rate is expressed as the output volume/injection volume. With the increase in pressure, the CH₄ production rate increased by 24.4% and 61.9%, respectively. The displacement ratios (CO₂ storage volume/CH₄ production volume) of CH₄ in the coal samples injected at different pressures were 3.85, 3.15, and 2.45, respectively. As the pressure increased, the displacement ratios decreased by 18.2% and 36.4%, respectively, and the sweep efficiency promoted by 12.5% and 62.2%, respectively.

Figure 3 shows the change in the cumulative displacement of CH₄ under different pressure conditions. In the early stage, cumulative growth showed a linear trend. Then, it slowly rose. The cumulative displacement increased with an increase in displacement pressure. The CO₂ diffused from the pore fractures to the micropores and competed with CH₄. The diffusion velocity was much smaller than the seepage velocity. A large amount of CO₂ was discharged from the coal before it entered the pores. As a result, the storage and output rates tended to flatten. Therefore, the adsorption and storage of CO₂ in the coal seam was faster and more effective in the early stages of displacement. The sweep effect tends to be stable, while the CH₄ production rate becomes small in later stages.

In the initial stage, the cumulative displacement of CH₄ increased rapidly and exhibited a linear growth trend. After reaching the critical time, the CH₄ desorption rate was smaller than the CO₂ displacement volume and then reached a constant value. Therefore, the cumulative amount of CH₄ displacement in the coal during the same period also changed slowly. At the beginning of the experiment, there was more free CH₄, which quickly flowed out of the coal under the displacement pressure. It can be seen that the initial CH₄ sweep efficiency is higher. After the critical time, the sweep efficiency gradually reached a constant value, and the concentration was stable at approximately 2.0%. With an increase in the displacement time, the adsorbed CH₄ in coal accounted for the majority, and there was very little free CH₄. The gas desorption rate tended to be constant, and the sweep efficiency did not change significantly.

2.3. In Situ Experiment

2.3.1. Experimental Design and System Process

This field industrial test was conducted in the return airway of the 401,101 working face in the Mengcun coal mine of Binchang mining area in Shaanxi Province. The #4 coal seam is a low metamorphic bituminous coal deposit with an average thickness of 13.0 m and an average dip angle of 3°, and it is a typical coal seam with high gas and low permeability. The predicted value of absolute gas emission in the mine is 110.5 m³/min, while the predicted value of relative gas emission is 8.1 m³/t. When gas extraction is carried out by dense drilling, problems such as high cost, long extraction time, and poor extraction effect must be overcome. To solve these problems, L-CO₂ displacing coal seam CH₄ technology was used to improve the efficiency of gas drainage and provide strong support for gas disaster prevention and mine safety production.

Two groups of injection holes and a group of contrast holes were designed in the L-CO₂ injection test; each group designed a L-CO₂ injection hole and six inspection holes. There were five contrast holes in the original area. The drilling design schematic is shown in Figure 4, and the drilling design parameters are listed in

Table 3. Drilling design parameter table of L-CO₂ injection system.

Boreholes	Depth (m)	Orientation (°)	Dip Angle (°)	Aperture (mm)	Sealing Length (m)
YZ	120	90	0.5–1	133	50
K1–K6	120	90	0.5–1	133	30
D1–D5	120	90	0–0.5	133	12

.

A process diagram of the L-CO₂ injection system is shown in Figure 5. It is composed of an L-CO₂ injection system, data acquisition system, gas drainage system, and borehole design. The L-CO₂ injection system consists of an L-CO₂ tanker, a low-temperature booster pump (with a maximum flow rate of 2000 L/h and maximum working pressure of 15.0 MPa), pressure-resistant transmission pipeline, stop valves, and relief valves. The data acquisition system includes pressure acquisition, flow acquisition, and drainage system monitoring (drainage flow, CO₂ concentration, and CH₄ concentration). The gas extraction system includes a gas extraction pump and extraction pipeline [40,41]. When the booster pump was used, the pipeline of the injection system was pre-cooled, and the temperature condition of the pump start was set to ≤−15 °C. In order to ensure that CO₂ was kept in the liquid state during the injection process, the pressure of the booster pump was never less than 3.0 MPa.

The success or failure of the experiment was determined based on the sealing quality of the borehole. The length of sealing in L-CO₂ displacing CH₄ experimental drilling involves several factors such as the pressure of the sealing section, the distance of the gas discharge belt from the coal wall, and the influence of air leakage from the hole. The sealing length was determined by combining data from the actual sealing experience on site with safety factors [42,43], which ensured the safety and reliability of the experiment. The sealing lengths of the injection hole, comparison hole, and original hole were 50 m, 30 m, and 12 m, respectively. A schematic diagram of the borehole sealing is shown in Figure 6.

2.3.2. Experimental Analysis

(1)

Pressure changes

The L-CO₂ injection tests were divided into two groups, low pressure and medium pressure. The low-pressure injection used L-CO₂ storage tanks to inject directly, while the medium-pressure injection utilized the L-CO₂ storage tank and booster pump. During the test, changes in the orifice pressure and extraction effect under different pressures were observed. The main technical indicators are listed in

Table 4. Main technical indicators of L-CO₂ injection.

Serial Number	Boost Time of Pressure/min	Pressure Fluctuation Range/MPa	Pressure Fluctuation time/min	Maximum Pressure/MPa	Cumulative Injection Volume/m3
1	23	1.3–1.6	110	1.52	5.9
2	17	2.5–3.2	60	3.16	5.8

, and the pressure trends during the injection process are shown in Figure 7.

The diagram shows that a tank injection with a maximum pressure of 2.50 MPa was used, and the pressure was maintained for approximately 90–110 min. When the system was cooled by gas injection for a certain time, liquid CO₂ was injected, and it reached its maximum value. With continuous injection, the pressure fluctuated around 1.5–1.6 MPa and decreased slowly. L-CO₂ was supercharged by a booster pump during medium pressure injection. Figure 7a shows that the pressure reached 2.60 MPa after 17 min, and the pressure continued rising after approximately 40 min. It reached a maximum value of 3.2 MPa and fluctuated around 3.0 MPa. The fluctuation time was approximately 25 min, after which the pressure decreased rapidly. Figure 7b shows that the pressure reached a maximum value of 3.20 MPa rapidly, fluctuated around 3.2 MPa for approximately 71 min, and then decreased when the pump stopped.

The pressure increase rate of the low-pressure injection was approximately 0.066–0.075 MPa/min, and that of the medium pressure injection was approximately 0.135–0.186 MPa/min. The pressure reduction rate during the low-pressure injection was approximately 0.013 MPa/min, and that of the medium-pressure injection was approximately 0.025–0.031 MPa/min. The increase rate and reduction rate of the medium-pressure injection were larger than those of the low-pressure injection. The effective injection time was approximately 100 min when the tank was used, while the effective injection time of the tank and booster pump was approximately 65 min. When the booster pump was used for injection, the effective injection time was reduced by 35 min. The increased displacement pressure improved the time efficiency, and the increase in pressure was beneficial to gas diffusion.

(2)

Analysis of displacement process

The relationship between the concentrations of CH₄ and CO₂ is shown in Figure 8. During the injection process, the change trend of the CH₄ concentration is consistent with that of the CO₂ concentration. The entire process can be divided into two stages: seepage and diffusion. The pressure gradient generated during the L-CO₂ injection and phase transition in seepage stage. The free CH₄ was driven out along the fractured channel. The initial concentration was high but decreased rapidly. At the same time, the displacement effect on the free gas was obvious, and the gas concentration was high in the initial stage. The diffusion stage occurred after the vaporization of L-CO₂ in the hole, and the concentration difference was generated by the gas. Adsorption CH₄ was replaced by CO₂ during diffusion. At that time, the gas diffused into the pores of the coal matrix through the concentration difference, and the displacement and desorption of the free CH₄ in the coal matrix and the adsorbed CH₄ were carried out. With the diffusion of CO₂, the displacement effect on a certain range of coal was gradually reduced, and the gas concentration also declined. When the concentration of CO₂ was approximately the same as that of the original area, this stage has ended. The CO₂ in the coal seam still played a role in displacement, and the gas remained at a higher concentration relative to the original area.

(3)

Seepage diffusion range for CO₂

The temperatures 3 m and 5 m from injection hole were measured by an infrared thermal imager under the medium pressure injection. The temperature decreased slowly as the distance from injection hole increased. As shown in Figure 9, when the pressure was between 2.5 MPa and 3.0 MPa, the CO₂ 3 m away from the injection hole was in the liquid phase, while that 5 m away was at the critical point of liquid–gas phase. It was concluded that the L-CO₂ phase transition range and seepage range is less than 5 m.

As shown in Figure 10, the CO₂ concentration changes in the inspection holes under low- and medium-pressure injections were investigated at different distances. The curve fitting formula for the CO₂ concentration change can be expressed as $y=y_0+A\cdot e^{R_0\cdot t}$. The fitting parameters are listed in

Table 5. The fitting parameters of CO₂ concentration during low-pressure injection.

	y0	A	B	R2	Distance/m
y1	0.51	6.56	−0.51	0.9679	5
y2	−0.53	5.55	−0.28	0.9742	10

and

Table 6. The fitting parameters of CO₂ concentration during medium-pressure injection.

	y0	A	B	R2	Distance/m
y1	−0.49	11.75	−0.38	0.9879	5
y2	0.02	10.69	−0.64	0.9737	10
y3	0.17	6.85	−0.52	0.9958	15
y4	0.14	18.40	−1.70	0.9997	20
y5	−1.15	2.97	−0.08	0.9818	25
y6	0.14	28.92	−2.15	0.9965	30

. At low-pressure injection, the R² values of Figure 10a at 5 m and 10 m from the injection hole were 0.9679 and 0.9742, respectively. The polynomial was used to fit 15 m to 30 m fitting curve, and the R² values of 15 m to 30 m were 0.9946, 0.9503, 0.9971 and 0.9970, respectively. After 15 m to 30 m, due to the uneven distribution of coal seam porosity and permeability, the pressure gradient of CO₂ during low-pressure injection is small; the concentration difference is small at the initial stage of diffusion, and the diffusion rate of CO₂ seepage is slow. With the passage of diffusion time and the increase in the concentration difference, the CO₂ concentration detected from 15 m to 30 m has an increasing trend compared with the initial stage, and gradually decreased with the decrease in concentration difference in the later stage.

For the ranges of 15 m, 20 m, 25 m, and 30 m from the injection hole, the concentration of CO₂ first increased and then decreased. Because the pressure gradient between the injection hole and the investigation hole was relatively small under low-pressure injection, the CO₂ seepage tended to be slow. With an increase in the total injection volume, the diffusion range and amount of CO₂ gradually increased under the effect of the CO₂ concentration gradient. In the medium-pressure injection, the variation trend of CO₂ concentration in each inspection hole in Figure 10b was in line with the exponential decreasing trend. The 25 m distance curve is slightly different from the other curves. The difference in the permeability and inhomogeneity of coal fracture development on both sides of the injection hole affected the seepage diffusion of CO₂. Under the medium-pressure injection, the CO₂ seepage diffusion velocity and concentration were higher. The original contrast-hole CO₂ concentration was 0.43%. The diffusion range of CO₂ under low-pressure injection was 20–25 m, while under medium pressure injection it was 25–30 m. Therefore, a higher displacement pressure can effectively promote the displacement of CH₄ by L-CO₂.

(4)

Analysis of the displacement effect

The average gas drainage concentration and drainage purity during low-pressure and medium-pressure injections are presented in Figure 11 and Figure 12, respectively. The average gas extraction concentration in the original area was 3.23%, and the average gas extraction purity was 0.00028 m³/min. The average gas drainage concentration of the monitoring borehole during low-pressure injection was 5.19%, and the average gas drainage purity was 0.00059 m³/min. The average gas extraction concentration during low-pressure injection was 1.61 times that of the original area, and the average gas extraction purity was 2.08 times that of the original area. The average gas extraction concentration of the monitoring borehole during medium-pressure injection was 7.02%, and the average gas extraction purity was 0.00083 m³/min. The average gas extraction concentration during medium-pressure injection was 2.17 times that of the original area, and the average gas extraction purity was 2.94 times that of the original area. The medium-pressure injection of average gas extraction concentration was 1.35 times of that of the low-pressure injection, and the medium-pressure injection of average gas extraction purity was 1.41 times of that of the low-pressure injection. The medium-pressure injection was more efficient.

(5)

Comparison of sweep efficiency

In the field test, to analyze the displacement effect under different pressures, the sweep efficiency was determined by comparing the total amount of displacement gas.

$$\eta =\frac{{\phi }_t\cdot {\rho }_t-{\phi }_0\cdot {\rho }_0}{{\phi }_t\cdot {\rho }_t}$$

(5)

where $\eta$ represents gas drainage efficiency, ${\phi }_t$ and ${\rho }_t$ are the gas drainage flow and gas drainage concentration after displacement, respectively; ${\phi }_0$ and ${\rho }_0$ are the gas drainage flow and gas drainage concentration before displacement, respectively.

The low- and medium-pressure gas drainage data can be calculated using Formula (5). The gas extraction efficiency in low-pressure injection ${\eta }_L$ was 52.4%, while the gas extraction efficiency in medium-pressure injection ${\eta }_M$ was 66.0%. The sweep efficiency of the medium-pressure injection (average pressure—2.8 MPa) was 26% higher than that of the low-pressure injection (average pressure—1.4 MPa).

In the displacement experiment, the variation law of CH₄ and CO₂ concentration under different pressure injection conditions was obtained, and the variation law of the concentration showed the rule of “alternating growth and decline”. At the same time, the injection pressure has a positive effect in the experimental process: the higher the pressure, the higher the displacement efficiency, and the greater the cumulative displacement. Combined with the laboratory results of CO₂ displacement coal seam CH₄, in the field test of liquid CO₂ displacement coal seam CH₄ under different pressure conditions, the injection pressure also has a positive effect on the improvement of displacement efficiency, influence range and gas extraction effect. Therefore, a CO₂ displacement experiment in the laboratory provides research foundation and theoretical basis for field tests.

3. Conclusions

(1)

The CH₄ production rates under different pressures (2.0 MPa, 4.0 MPa, and 6.0 MPa) were 43.6%, 54.3%, and 70.7%, respectively. With the increase in pressure, the CH₄ production rate increased by 24.4% and 61.9%, respectively. The displacement ratios of CH₄ in the coal samples injected with different pressures were 3.85, 3.15, and 2.45. As the pressure increased, the displacement ratios decreased by 18.2% and 36.4%, respectively, and the sweep efficiency increased by 12.5% and 62.2%, respectively. Increasing the pressure improved the CH₄ sweep efficiency and decreased the displacement ratio.

(2)

The maximum pressure of the low-pressure injection was 1.52 MPa, while the maximum pressure of the medium-pressure injection was 3.16 MPa. The pressure was 2.1 times that of the low-pressure injection. The pressure rise rate was 2.82 times that of low-pressure injection, and the pressure drop rate was 1.32 times that of the low-pressure injection. (3) Through a comparative analysis of the change trend of the CH₄ concentration and CO₂ concentration in the observation hole, it was determined that the change trend remained the same. The process of L-CO₂ displacing gas can be roughly divided into two stages: the seepage stage and the diffusion stage. In the seepage stage, CO₂ had a more obvious driving effect on free gas, and the gas concentration was relatively high in the initial stage. In the diffusion stage, CO₂ played a major role in the displacement and desorption of adsorbed gas, and the gas concentration gradually decreased. A higher displacement pressure effectively promoted the displacement of CH₄ by L-CO₂. The diffusion range of CO₂ under the low-pressure injection was 20–25 m, while under medium-pressure injection it was 25–30 m.

(3)

Inspection of the gas extraction field revealed that the concentration of CH₄ extracted increased from 3.23% to 5.19% after the low-pressure injection of L-CO₂ into the coal seam, increasing the concentration of gas extracted by 0.61 times. The pure flow of gas extracted increased from 0.028 m³/min to 0.059 m³/min, increasing the pure flow of gas extracted by 1.08 times. The concentration of CH₄ extracted increased from 3.23% to 7.02% after the medium-pressure injection of L-CO₂ into the coal seam, increasing the concentration of gas extracted by 1.17 times. The pure flow of gas extracted increased from 0.028 m³/min to 0.083 m³/min, increasing the pure flow of gas extracted by 1.94 times. The average gas drainage concentration of the medium-pressure injection was 1.35 times that of the low-pressure injection, and the average gas drainage scalar medium-pressure injection was 1.41 times that of low-pressure injection. The sweep efficiency of the medium-pressure injection (average pressure—2.8 MPa) was 26% higher than that of the low-pressure injection (average pressure—1.4 MPa). The overall efficiency of the medium-pressure injection was much higher.

Through this test, the displacement technology of coal seam CH₄ by injection of L-CO₂ proved to significantly improve gas drainage efficiency. An increase in pressure is beneficial for improving the sweep efficiency of L-CO₂. This field test only carried out displacement tests under two pressure conditions, and the experimental results have important significance for field gas drainage. Due to the influence of various mine production factors, it is impossible to carry out more displacement tests under pressure conditions and different flow conditions. As the test is difficult to implement in the field, it is still necessary to study displacement field tests under different pressure and different flow conditions in the future.