Experimental Study on the Effect of Unloading Paths on Coal Damage and Permeability Evolution

Abstract

Coal seam cavitation is one of the most effective techniques for gas disaster control in low-permeability coal. Due to the difference in cavitation method and process, the damage degree and fracture development range of the coal body around the cavern are greatly different, and the effect of increasing the permeability of the coal body is further changed. In order to further understand the permeability enhancement mechanism of cavitation technology on low-permeability coal and effectively guide engineering applications, this paper conducted experimental research on the unloading damage and permeability evolution characteristics of coal under different cavitation paths using a coal-rock “adsorption-percolation-mechanics” coupling test system. Through the analysis of coal strength and deformation characteristics, coal damage characteristics, and the evolution law of coal permeability combined with the macroscopic damage characteristics of coal, the strength degradation mechanism of unloaded coal and the mechanism of increased permeability and flow were revealed. The results show that unloading can significantly reduce the strength of coal, and the greater the unloading rate, the more obvious the reduction. The essence of this is that unloading reduces the cohesion and internal friction angle of coal—damage and breakage are the most effective ways to improve the permeability of the coal body. Unloading damaged coal bodies not only significantly improves the permeability of the coal body but also improves the diffusion ability of gas, and finally, shows a remarkable strengthening effect of gas extraction.

1. Introduction

The annual production of nearly 4 billion tons of coal has strongly supported the rapid development of China’s economy in recent years [1,2]. However, the strong demand for coal has led to the extension of mining at an average rate of 10–30 m per year [3,4]. Many mines now have depths exceeding 800 m, particularly in regions such as Shandong, Anhui, Heilongjiang, and Henan, where some mines exceed 1000 m. As mining depth increases, the coal seam’s in situ stress, gas pressure, and gas content increase sharply while permeability decreases significantly [5,6]. This “three highs and one low” situation has caused more and more non-outburst mines to become coal and gas outburst mines, posing severe safety challenges to coal mining enterprises [7,8,9,10]. As one of the world’s largest producers and consumers of coal, China heavily invests each year in managing coal mine gases. Despite significant achievements, gas accidents remain one of the primary causes of major casualties in coal mines [11,12].

Practical experience shows that gas extraction is the most effective measure to solve coal mine gas accidents and achieve energy conservation and environmental protection. However, more than 95% of high-gas coal seams and protruding coal seams in China are low-permeability coal seams, with a permeability of only 10⁻⁴~10⁻³ mD, which is 3~4 orders of magnitude lower than that in the United States [13,14,15]. The widespread low-permeability coal seams in gas extraction present the problems of low extraction efficiency and great difficulty in extraction. The high ground stress during deep digs into a coal mine can cause many permeability enhancement technologies applied in the shallow part of the mine to gradually fail [16,17,18]. The reason is that such measures have failed to effectively release the stress of the coal mass, resulting in the gradual closure of damaged cracks under the action of high-ground stress. Pressure relief and permeability enhancement technology is a method based on the relief of ground stress and the derivative cracks, which is a way to solve the low permeability problem of high-stress coal seams [19,20]. Some studies have proved that the relief of ground stress will be an effective means and a necessary path for managing gas disasters in deep coal seams [21,22].

The pressure relief and permeability enhancement technology mainly includes technologies such as mining protective layers, hydraulic cutting, and hydraulic cave-making. Among them, hydraulic cavitation technology has become more widely used for gas control in high-stress and low-permeability coal seams due to its simple operation and good pressure relief effect, coupled with continuous progress in equipment and theory [23,24,25]. In particular, mechanical cavitation technology has been successfully applied, and the defects of water jet coal destruction in submerged environments and hard coal seams have been overcome, leading to significant improvements in the efficiency of coal destruction and the successful application of cavitation technology in soft and hard coal [26,27]. Thus, the surrounding coal body undergoes the mechanical path of increasing vertical stress and decreasing horizontal stress because the essence of cavitation permeability enhancement is to relieve pressure and damage the coal mass around the cavern [28,29,30].

However, differences have gradually emerged between the methods and processes of cavitation (collectively referred to as cavitation paths) due to the differences between the coal destruction processes and techniques. For example, it is generally believed that under the conditions of the same water pressure and equipment, the diameter of the high-pressure water jet cavitation is positively correlated with the flushing time [31,32]. Under a specific cavitation diameter, the cavitation diameter can be gradually expanded by repeatedly pushing and pulling the drill rod for repeated flushing (as shown in Figure 1a), or the drill rod can be rotated for repeated flushing until the cavitation diameter reaches the target and then gradually advances forward (as shown in Figure 1b). The two water jet cavitation methods have different coal destruction sequences, resulting in different unloading paths for the coal mass around the cavern. Correspondingly, the mechanical cavitation technology cuts and breaks the coal with a mechanical knife. After the cavitation diameter is completed once, the length of the cavern is extended by pushing the drill rod, as shown in Figure 1c.

Compared with hydraulic coal destruction, mechanical coal destruction is highly efficient; that is, the coal mass around the cavitation experiences a higher unloading rate. In other words, different coal destruction efficiencies and coal destruction sequences cause the coal mass around the cavern to experience different unloading paths, which in turn have a significant difference in the degree of coal damage and the range of fracture development and ultimately show a significant effect of the cavitation method on the permeability enhancement effect of the coal mass.

Extensive research has been conducted on the damage of coal and rock during unloading confining pressure in recent years. Xue et al. [33] studied the effect of the unloading rate on the mechanical properties and permeability evolution (including energy evolution and fractal dimension) of coal and rock. Zhao et al. [34] studied the energy conversion and dissipation characteristics of rocks during different unloading confining pressure processes and obtained the effect of the unloading rate on both. Chen et al. [35] studied the effect of the unloading rate of confining pressure on the crack propagation characteristics of sandstone through experiments and believed that with the increase in the unloading rate of confining pressure, the crack volume expansion strain and crack propagation rate of sandstone would increase. Liang et al. [36] studied the post-peak mechanical characteristics and deformation characteristics of rock mass based on the demand for underground engineering support and believed that the rate of unloading confining pressure has a significant effect on the bearing capacity and deformation characteristics of rock mass. Wang et al. [37] studied the effect of the unloading rate on coal damage and permeability during unloading confining pressure based on the stress state of coal mass during mining. Zhang et al. [38] established a theoretical model of permeability evolution of gas-bearing coal considering the change of loading and unloading rate based on previous research and verified the effectiveness of the model through experimental data. Chen et al. [39] studied the damage and permeability evolution characteristics of coal bodies under stress relief conditions based on the application background of mining protective layers. These research results have proved that unloading has a significant impact on the strength, deformation and permeability of coal and rock mass. They also proved theoretically that pressure relief has obvious significance for gas extraction but did not provide an explanation as to how pressure relief is more beneficial. In addition, they did not clearly explain the evolution law of local pressure relief damage of a coal body around cavitation and cannot effectively guide the application of cavitation technology.

In order to further understand the permeability enhancement mechanism of cavitation technology on low-permeability coal and effectively guide engineering applications, this paper is based on the effect of different cavitation methods on the permeability enhancement effect of coal in field applications and condenses the idea of conducting different unloading process and unloading rate tests to study the effect of unloading behavior on coal damage and permeability enhancement effect. On this basis, the mechanism of cavitation paths’ effect on the unloading permeability enhancement effect is revealed. The research results help strengthen the understanding of the mechanism of cavitation technology for coal body pressure relief and permeability increase. The stress evolution characteristics and permeability increase characteristics of coal after cavitation are clarified, which is very helpful for guiding site construction and improving gas extraction efficiency.

2. Materials and Equipment

2.1. Coal Sample Preparation and Experimental System

The coal samples were taken from the E_9-10 coal seam of Pingmei No. 8 Mine in Pingdingshan City, Henan Province. The coal seam thickness is 2.5~2.7 m, the structure is uniform, and the Proctor coefficient is about 0.5. Large pieces of coal with a complete structure were selected from the underground site and sent to the laboratory for coal sample preparation. According to the standard methods for engineering rock tests, large coal blocks were processed into standard cylindrical specimens with dimensions of φ50 mm × 100 mm. Intact and undamaged specimens were selected and encapsulated for testing, as shown in Figure 2a.

The experimental system used a coal-rock “adsorption–percolation–mechanics” coupling test system, including mechanical testing, permeability testing, acoustic emission monitoring, and temperature control modules, as shown in Figure 2b. The mechanical testing module can perform conventional uniaxial and triaxial loading tests on standard specimens. The permeability testing module measures the permeability of standard coal samples under different stress states and temperatures. The acoustic emission monitoring module analyzes the damage characteristics of standard coal samples during mechanical tests. The system can apply axial loads up to 600 kN with control precision ≥ ±1% and confining pressure up to 60 MPa with control precision ≥ ±1%. The strain gauge is series non-contact micro displacement sensor, which is an axial range of 25.0 mm, a radial range of 5.0 mm, and a linear error ≤0.25%. Permeability measurement uses the transient method with helium as the non-adsorptive gas source, and the metering pump’s maximum driving pressure is 40 MPa. The acoustic emission module uses the Micro-II acoustic emission acquisition system (Version number: TEST0000/2014) produced by MISTRAS Group, Inc. Products & Systems Division Princeton Junction, West Windsor Township, NJ, USA, including a 24-channel high-speed data acquisition system that receives acoustic signals at frequencies up to 0.1 μs and monitors parameters such as energy, amplitude, waveform, and count of acoustic emission signals. At the same time, according to the difference of signals received by multiple monitoring probes, the 3D location of the fracture in the sample is realized, and then the damage and fracture situation inside the sample and the macro failure characteristics are studied. In order to detect a weak break signal, the probe and the sample are closely fitted, and the probe surface is coated with a coupling agent and then contacted with the sample wall and tightened by a strong spring ring.

Before the experiment, strain sensors and acoustic emission monitoring probes were fixed on standard coal samples. The strain sensors include axial and radial strain gauges, and acoustic emission monitoring probes are evenly arranged at both ends of the coal samples, with three probes at 120° intervals on the upper and lower ends. The prepared coal samples are fixed in the testing chamber to conduct mechanical and permeability tests.

2.2. Experimental Scheme

To study the effect of unloading paths on coal damage and permeability evolution characteristics, the following three types of experiments were designed: conventional tri-axial loading experiment (I), low-rate unloading confining pressure and loading experiment (II) and high-rate unloading confining pressure and loading experiment (III), respectively, to carry out the mechanical damage and permeability evolution experiments of coal samples under three mechanical paths. (I) The loading rate is 50 N/s, and the experiment concludes after the coal sample is damaged while loading continues for an additional 3~5 s. (II) After the confining pressure reaches the target value, the axial load is gradually increased. When the axial load is equal to the confining pressure ${(\sigma }_1=2{\sigma }_3)$, the confining pressure is unloaded. To ensure that the axial pressure loading rate and the confining pressure unloading rate are the same, the axial pressure loading rate and the confining pressure unloading rate are 50 N/s and 25 N/s, respectively. The experiment concludes after the coal sample is damaged, with loading continuing for an additional 3~5 s. Experiment (III) has the same preparation stage as Experiment (II) but with an axial loading rate and confining pressure unloading rate of 100 N/s and 50 N/s, respectively. Similarly, the experiment ends 3 to 5 s after the coal sample is damaged. All three types of experiments are stress-controlled and are conducted with 5 MPa, 10 MPa, and 15 MPa confining pressures. During the tests, acoustic emission monitoring is used to detect damage signals in the coal samples. The specimen parameters and experimental conditions are listed in

Table 1. Coal sample specifications and experimental parameters.

Test Classification	Coal Samples S.N.	Height/mm	Diameter/mm	Confining Pressure/MPa	Axial Pressure/KN	Loading Method
I	1-1	100.0	48.5	5	Load to damage	Constant confining pressure and axial pressure loading
	1-2	100.08	48.5	10
	1-3	100.1	48.5	15
II	2-1	100.1	48.6	5	Load to ${\sigma }_1=2{\sigma }_3$, then Unload confining pressure and load to damage	Confining pressure unloading at 25 N/s, Axial pressure loading at 50 N/s
	2-2	100.05	48.5	10
	2-3	100.1	48.6	15
III	3-1	100.1	48.5	5	Load to ${\sigma }_1=2{\sigma }_3$, then Unload confining pressure and load to damage	Confining pressure unloading at 50 N/s, Axial pressure loading at 100 N/s
	3-2	100.1	48.5	10
	3-3	100.08	48.6	15

.

3. Results and Analysis

3.1. Coal Strength and Deformation Characteristics of Different Unloading Paths

Different confining pressures are applied first, and then the loading tests are carried out on the coal samples as per the experimental scheme. During the loading process, the coal samples are continuously compressed and deformed until they become unstable and damaged. The relationship between the axial strain ${\epsilon }_1$, radial strain ${\epsilon }_2$, and volumetric strain ${\epsilon }_v$ of the coal samples and the stress is plotted, with the deviatoric stress (${\sigma }_1$–${\sigma }_3$) as the ordinate and the strain as the abscissa, and the deviatoric stress–strain curve of the coal sample is obtained (where compression is a positive value and expansion is a negative value).

The stress–strain characteristics of the coal samples under different confining pressures at the same unloading rate are shown in Figure 3. Also shown are the conventional triaxial load-in, confining pressure unloading at 25 N/s, and loading and confining pressure unloading and loading at 50 N/s. As can be seen from Figure 3a, when the confining pressure is fixed at 5 MPa, 10 MPa, and 15 MPa, the peak deviatoric stress of the coal samples is 21.99 MPa, 31.22 MPa, and 50.23 MPa, respectively, the axial strain (mm) is 0.01709, 0.05687, and 0.07262, respectively, the radial strain (mm) is −0.00932, −0.03181, and −0.05818, respectively, and the volumetric strain (mm) is −0.00155, 0.00675, and −0.04373, respectively. The results show that except for volumetric strain, which first increases and then decreases with the increase of confining pressure, the strength, axial strain, and radial strain of the coal samples exhibit a significant increasing trend. As shown in Figure 3b, for the initial confining pressures of 5 MPa, 10 MPa, and 15 MPa with a confining pressure unloading rate of 25 N/s, the peak deviatoric stresses of the coal samples are 12.88 MPa, 22.95 MPa, and 32.99 MPa, respectively. The axial strains (mm) are 0.01534, 0.02345, and 0.01723, respectively, the radial strains (mm) are −0.02206, −0.03659, and −0.00933, respectively, and the volumetric strains (mm) are −0.02878, 0.04974, and −0.00142, respectively. The overall trend indicates that as the initial confining pressure increases, the strength, axial strain, radial strain, and volumetric strain of the coal samples also increase. However, after the unloading point, the stress–strain curve shows a significant downward trend, indicating a notable decrease in the strength of the coal samples. As shown in Figure 3c, for initial confining pressures of 5 MPa, 10 MPa, and 15 MPa with a confining pressure unloading rate of 50 N/s, the peak deviatoric stresses of the coal samples are 10.01 MPa, 17.83 MPa, and 20.53 MPa, respectively. The axial strains (mm) are 0.01471, 0.02094, and 0.02717, respectively, the radial strains (mm) are −0.01505, −0.01416, and −0.02102, respectively, and the volumetric strains (mm) are −0.0154, 0.00738, and −0.01488, respectively. The patterns observed are similar to those in Figure 3b, but due to the higher unloading rate, the coal samples quickly reach their ultimate strength and damage after unloading. From Figure 3, it can be seen that regardless of whether unloading occurs, the ultimate damage strength of the coal samples significantly increases with increasing confining pressure. However, unloading significantly reduces the ultimate strength of the coal samples.

In order to further compare the effect of unloading rate on coal strength, the stress–strain characteristics of coal samples under the same confining pressure and different unloading rates were analyzed, as shown in Figure 4. It contains an initial confining pressure of 5 MPa, an initial confining pressure of 10 MPa, and a loading and initial confining pressure of 15 MPa. It can be seen from the figure that when the confining pressure is 5 MPa, the peak point deviatoric stress of the coal sample under the conditions of constant confining pressure, 25 N/s unloading confining pressure and 50 N/s unloading confining pressure are 21.99 MPa, 12.88 MPa and 10.01 MPa, respectively, which are 41.42% and 54.47% lower than the peak point strength of 5 MPa constant confining pressure; when the confining pressure is 10 MPa, the peak point deviatoric stress of the coal sample under the three unloading rates are 32.33 MPa, 22.95 MPa, and 17.83 MPa, respectively, which are 29.01% and 44.84% lower than the peak point strength of 10 MPa constant confining pressure; similarly, when the confining pressure is 15 MPa, the peak point deviatoric stress of the coal sample under the three unloading rates are 50.23 MPa, 32.99 MPa, and 20.53 MPa, respectively, which are 34.32% and 59.12% lower than the peak point strength of 15 MPa constant confining pressure. It shows that under different unloading rates, the ultimate strength of the coal sample decreases with the increase of the unloading rate, and the greater the unloading rate, the more obvious the attenuation of the ultimate strength of the coal sample. In addition, from the evolution trend of the stress–strain curve, it can be seen that the slope of the strain curve drops suddenly after the unloading point, and the sudden drop trend slows down with the increase of the confining pressure. The reason for the sudden drop in the slope of the strain curve is that the unloading causes the coal sample’s ability to resist deformation to suddenly decrease. However, with the increase of the confining pressure, the confining pressure has a relatively high value at the beginning of unloading, and the coal sample still has a strong ability to resist deformation, but it is reduced compared with the constant confining pressure. The changes in the curves after the peak point are similar. The curves of the coal samples at different unloading rates basically fall rapidly after the peak point, especially under low confining pressure conditions. This shows that under the condition of unloading confining pressure, the friction force in the fracture surface decreases significantly as the confining pressure decreases after the coal sample is damaged so that the damaged coal mass almost loses its bearing capacity. In general, when the unloading rate is large, the strain value required to destroy the coal mass is smaller, and the ultimate strength is lower.

In order to further analyze the intrinsic mechanism of the effect of unloading rate and initial confining pressure on the damage and destruction of coal samples, the peak stress and initial confining pressure of coal samples under different unloading conditions were statistically analyzed, and the corresponding strength parameters were regressed and calculated. The fitting of peak stress and initial confining pressure data under different unloading behaviors is shown in Figure 5. The peak stress and initial confining pressure data under different unloading behaviors are fitted to obtain the slope and intercept of the lines using the least squares method. Based on the physical meaning of the slope and intercept of the lines and the Mohr–Coulomb strength criterion, the cohesion and internal friction angle of the coal samples for the corresponding paths can be calculated. According to the fitted data in Figure 5, it can be obtained:

Under the condition of constant confining pressure and axial pressure loading, the relationship between the peak stress and initial confining pressure of the coal sample is:

$${\sigma }_1=6.240+3.824{\sigma }_3$$

(1)

When confining pressure unloading at (25 N/s) and axial pressure loading to the coal sample, the relationship between the peak stress and the initial confining pressure is:

$${\sigma }_1=4.430+2.841{\sigma }_3$$

(2)

When confining pressure (50 N/s) unloading and axial pressure loading to the coal sample, the relationship between the peak stress and the initial confining pressure is:

$${\sigma }_1=3.803+2.192{\sigma }_3$$

(3)

The linear regression results and strength parameters of coal sample experimental data under different unloading conditions are shown in

Table 2. Calculation results of strength parameters of different unloading conditions.

Stress Path	Intercept	Slope	Cohesion/MPa	Internal Friction Angle/°	R2
Constant confining pressure and axial pressure loading	6.240	3.824	1.595	35.83	0.978
Confining pressure unloading at 25 N/s and axial pressure loading	4.430	2.841	1.314	28.64	0.999
Confining pressure unloading at 50 N/s and axial pressure loading	3.803	2.192	1.284	21.93	0.986

. It can be seen from the table that the fitting accuracy of the three groups of data is high. The results show that compared with the constant confining pressure, the cohesion and internal friction angle calculated for the coal sample under unloading confining pressure is smaller. According to the Mohr–Coulomb criterion, it can be known that internal friction and cohesion constitute the shear strength of the coal mass. Therefore, the cohesion and internal friction angle comprehensively reflect the strength characteristics of the coal mass. It shows that the mechanical essence of unloading confining pressure making the coal mass more easily damaged and destroyed is that unloading confining pressure reduces the cohesion and internal friction angle of the coal mass, and the cohesion and internal friction angle of the coal mass decreases more obviously with the increase of the unloading rate of confining pressure.

3.2. Coal Damage Characteristics in Different Unloading Paths

In order to study the effect of different unloading paths on the damage characteristics of coal mass during destruction, acoustic emission (AE) was used to monitor the AE energy and AE count of the coal sample destruction process. Where AE energy refers to the acoustic emission energy detected during the experiment, and AE count refers to the number of acoustic emissions during the experiment. Both reflect the damage situation inside the coal sample monitored at a certain moment during the test, while AE cumulative energy and AE cumulative count reflect the cumulative damage situation inside the coal sample before a certain moment.

The acoustic emission signals of coal samples obtained during the mechanical experiment were processed and analyzed to obtain the acoustic emission AE counts (as shown in Figure 6) and AE energy (as shown in Figure 7) of coal samples under different unloading paths. The number of experimental coal samples corresponds to the number in the picture. As can be seen from Figure 6, the AE cumulative counts of all coal samples during the mechanical experiment showed a trend of first slowly increasing, then sharply increasing, and finally steadily increasing. Combined with the axial stress curve of the coal sample, it was found that the AE count increased slowly during the compression deformation stage, and the AE count increased sharply before the coal sample was damaged until the AE count incidence rate decreased after the coal sample was damaged and a stable increase appeared. In addition, by comparing the acoustic emission AE counts of coal samples with different unloading paths, the AE counts of coal samples loaded under constant confining pressure increased more slowly, and the AE counts of coal samples under confining pressure unloaded at 25 N/s and 50 N/s increased faster, and the time for the coal samples to reach destruction due to unloading was faster. For example, when the initial confining pressure was 5 MPa, the time for the coal samples to reach destruction under constant confining pressure, 25 N/s unloading confining pressure, and 50 N/s unloading confining pressure was 1200 s, 300 s, and 200 s, respectively. In addition, the final AE cumulative counts of coal samples under different unloading paths have no obvious regularity, which may be due to the heterogeneity of the coal sample itself.

As can be seen from Figure 7, the AE energy of the coal sample is basically the same as the AE counting law, but the AE cumulative energy has obvious regularity. Under the same initial confining pressure, the AE cumulative energy decreases with the increase of the unloading rate when the coal sample is damaged. It is confirmed that unloading reduces the energy required for damage and destruction of the coal sample, and the greater the unloading rate, the more obvious the reduction.

In order to quantitatively analyze the damage characteristics of coal samples under different unloading paths, the AE cumulative counts and AE cumulative energies of each group of coal samples were statistically analyzed, as shown in Figure 8. Under the same unloading path, the AE cumulative counts and AE cumulative energies of coal samples both showed an increasing trend with the increase of confining pressure. Under the same initial confining pressure, except for a few discrete points, the cumulative AE (Acoustic Emission) counts and cumulative AE energy of the coal samples generally show a pattern where constant confining pressure is greater than that at 25 N/s unloading rate and greater than that at 50 N/s unloading rate. This proves that unloading conditions can reduce the strength of the coal, and the greater the unloading rate, the more significant the strength reduction.

3.3. Effect of Different Unloading Behaviors on Coal Permeability Evolution

In order to obtain the effect of different unloading paths on the damage and permeability enhancement of coal, the permeability of the coal samples was measured during the experiment. Before the coal samples reached their peak strength, pulse gas was released at randomly selected test points to measure the permeability of the coal samples. After the coal samples reached their peak strength, one or more measurements were continued according to their bearing capacity. The permeability measurement results of each group of coal samples were matched with the stress–strain curve, as shown in Figure 9. The number of experimental coal samples corresponds to the number in the picture. The initial permeability of the coal samples and the permeability at the time of damage are shown in

Table 3. Sudden increases in permeability of coal mass after damage under different unloading behaviors.

Test Classification	Sample Serial Number	Confining Pressure/MPa	Initial Value/mD	Damage Value/mD	Sudden Increase by Times
I	1-1	5	0.064	7.625	119.1
	1-2	10	0.022	1.655	75.2
	1-3	15	0.0082	0.712	86.8
II	2-1	5	0.042	12.953	308.4
	2-2	10	0.028	7.632	272.6
	2-3	15	0.0063	1.153	183
III	3-1	5	0.046	15.668	340.6
	3-2	10	0.036	11.336	314.9
	3-3	15	0.00506	1.735	342.9

. It can be seen from the experimental results that during the full stress–strain process of the coal samples, the permeability first decreased and then increased. For example, when the initial confining pressure is 5 MPa, the initial permeabilities of the coal samples under constant confining pressure, 25 N/s unloading confining pressure, and 50 N/s unloading confining pressure are 0.064 mD, 0.042 mD, and 0.046 mD, respectively. The minimum permeabilities of the coal samples measured during the loading process are 0.003 mD, 0.0065 mD, and 0.0018 mD, respectively. As the accumulated permeability of the coal samples due to plastic damage gradually increases, the permeability shows a jump growth after reaching the peak stress point. The permeabilities at the time of instability and damage are 7.625 mD, 12.953 mD, and 15.668 mD, respectively.

Similarly, when the initial confining pressure is 10 MPa and 15 MPa, the permeability of the coal samples changes in the same way, and they all experience a process of first decreasing and then increasing. This is because at the beginning of the experiment, the load on the coal sample increased, and the internal cracks were compressed, resulting in a decrease in permeability. As the load increased, the coal sample gradually produced plastic strain and damage, the internal cracks of the coal sample increased, and the permeability began to rise slowly; when the strength limit was reached, the coal sample was damaged, macro cracks were generated, and the permeability suddenly increased. In addition, under the condition of constant confining pressure, the permeability of the coal sample showed a downward trend after the peak point. This is because the existence of confining pressure limits the continued increase of macro damage cracks and instead gradually shrinks under the action of axial load. This shows that high stress leads to a significant decrease in coal permeability, and damage is the most effective way to increase permeability.

As can be seen from Table 3, when the confining pressure is 5 MPa, 10 MPa, and 15 MPa, the permeability of the coal sample after instability and damage in the constant confining pressure experiment increased by 119.1 times, 75.2 times, and 86.8 times, respectively, compared with the initial state; the permeability of the coal sample after instability and damage in the 25 N/s unloading confining pressure experiment increased by 308.4 times, 272.6 times, and 183 times, respectively; the permeability of the coal sample after instability and damage in the 50 N/s unloading confining pressure experiment increased by 340.6 times, 314.9 times, and 342.9 times, respectively. It reflects that under constant confining pressure or a low unloading rate, the permeability enhancement effect after coal mass damage is significantly affected by the initial stress. The initial stress is relatively large, and the permeability enhancement effect is relatively low, but the unloading condition can significantly reduce the effect of the initial stress on the permeability enhancement effect. This may be because the cracks after coal mass damage at a high unloading rate will not reclose under stress. It can be inferred that unloading damage can significantly improve the permeability of low-permeability coal seams, and the greater the unloading rate, the better the permeability enhancement effect.

3.4. Macroscopic Damage Characteristics of Coal Mass under Different Unloading Behaviors

The macroscopic damage mode of coal samples is influenced by their physical and chemical properties and the mechanical loading pattern. Numerous experiments have confirmed that coal rock under confining pressure conditions will only undergo shear damage, whereas, under uniaxial compression conditions, it will undergo tensile damage along the axial direction, which is more complex in form. Comparing the macroscopic damage modes of coal samples under different mechanical paths helps to gain a deeper understanding of the influence of mechanical paths on the damage characteristics of coal. The macroscopic damage modes of the coal samples from each group after the experiments are shown in Figure 10. The number of experimental coal samples corresponds to the number in the picture.

As can be seen from the figure, under the constant confining pressure and axial pressure mechanical path, the coal sample damage is all in the form of shear damage. When the confining pressure is 5 MPa, the coal sample has only one macroscopic fracture surface. When the confining pressure increases to 10 MPa, in addition to a main shear fracture surface, the coal sample also produces multiple associated small fracture surfaces around it; that is, the coal sample begins to show multiple damage. When the confining pressure is 15 MPa, the coal sample produces multiple independent macroscopic fracture surfaces. This shows that when the confining pressure is further increased, the damage form of the coal sample is obviously different from that at low confining pressure. When the high confining pressure coal sample is damaged, it presents typical multiple shear fractures. This is because of the significant confining pressure that creates higher levels of friction on both sides of the coal sample fracture surface, which in turn leads to secondary fracture surfaces or independent fracture surfaces on both sides of the fracture surface.

There is also a certain regularity in the damage of coal samples under unloading confining pressure conditions. Under the conditions of 5 MPa confining pressure, the coal sample damage forms shear damage and tensile damage when the confining pressure is unloaded at 25 N/s, while the form of coal sample damage is tensile damage that only occurs under uniaxial compression when the confining pressure is unloaded at 50 N/s. Under the conditions of 10 MPa confining pressure and 15 MPa confining pressure, the damage forms of coal samples under 25 N/s unloading confining pressure and 50 N/s unloading confining pressure are basically consistent with those under constant confining pressure of 5 MPa and 10 MPa, respectively. The macroscopic damage forms of coal samples show that the damage forms of coal samples under the mechanical path of unloading confining pressure and loading axial pressure are downgraded compared with conventional triaxial loading; that is, the damage forms of coal samples under unloading confining pressure and loading axial pressure are similar to those of conventional triaxial under a lower level of constant confining pressure-loading.

3.5. Discussion

According to the results of the mechanical strength characteristics analysis of coal samples under different unloading paths, the coal damage and destruction criterion can be understood from the perspective of stress Mohr’s circle as follows (as shown in Figure 11): under the path of constant confining pressure and loading axial pressure, the minimum principal stress remains unchanged while the maximum principal stress continues to increase. When the stress circle formed by the maximum principal stress and the minimum principal stress is tangent to the strength curve, the coal is damaged, while the intercept and slope of the strength curve formed under the path of unloading confining pressure and loading axial pressure are smaller than those under the path of constant confining pressure and loading axial pressure, and the minimum principal stress continues to decrease due to the unloading of the confining pressure. As the maximum principal stress increases, the strength of the coal is quickly reached; that is, the stress circles formed by the blue dotted line and the pink dotted line are tangent to the corresponding strength curve. In addition, as the unloading rate increases, the coal is more likely to reach the ultimate strength. It can be explained that the unloading behavior is equivalent to changing the mechanical properties of the coal and reducing its strength.

The permeability evolution during the full stress–strain process of the coal sample shows a law of first decreasing and then increasing. The stress–strain and permeability evolution trends during the experiment are simplified and combined with the acoustic emission monitoring results, which can be described as shown in Figure 12. The acoustic emission signal shows that the coal sample damage is a cumulative process, and when the damage exceeds the influence of fracture compression, the seepage gradually increases. As shown in the figure, during the experiment, as the load on the coal sample increases, the interior is gradually compacted. The stress–strain evolution of this process corresponds to the AC stage in the figure, showing that the coal sample is compressed but has not been damaged. The compression crack causes the permeability to gradually decrease. The permeability evolution corresponds to the EF stage in the figure. Afterward, the coal sample enters the plastic deformation stage, and the coal sample begins to produce plastic damage, but the coal sample at this time does not show macroscopic damage and still has good integrity. The stress–strain evolution corresponds to the CD stage in the figure. The gradual increase in new cracks causes the permeability of the coal sample to rebound and gradually increases with the accumulation of damage. The permeability evolution is shown in the FH stage of the figure. When the pressure on the coal sample reaches the limit, instability and damage occur, and the permeability of the coal sample increases in a jump-like manner after the peak point, as shown in the HG stage in the figure. The main reason for the jump-like increase in permeability at this stage is that macro cracks appeared in the coal sample, and the unloading damage after the peak point caused the cracks between the coal matrix to expand further.

According to the experimental results, the permeability will increase in a jump-like manner as the coal mass is damaged during the constant confining pressure test, but the permeability will increase earlier during the unloading confining pressure test, and the permeability will be greater after the coal mass is broken. In addition, the permeability increases more with the increase in the unloading rate. In the figure, T1 is the constant confining pressure loading process, T2 and T3 are the unloading confining pressure loading processes, but the unloading rate of T3 is greater.

As a typical dual-porosity medium, the cubic structure model established by simplifying the fracture network and coal mechanism has been widely recognized and applied. Combined with the study of the mechanical permeability evolution law of coal samples, the mechanism of coal unloading damage permeability enhancement can be described as follows (as shown in Figure 13): unloading increases the crack opening between the coal matrix units and increases the gas flow channels in the cracks, which results in an increase in the permeability of the coal. At the same time, the reduction in confining pressure greatly reduces the ultimate strength of the coal, making the coal more susceptible to damage and destruction. The continuous damage causes the internal matrix of the coal to be further divided into smaller matrix units by the new cracks, and the diffusion path of methane molecules adsorbed in the coal matrix into the cracks becomes smaller; that is, the desorbed methane molecules are more likely to enter the cracks, showing an increase in the diffusion capacity of the coal. Therefore, under the action of coal unloading and damage, the flow enhancement effect of gas extraction is achieved, which reduces the difficulty of coal seam gas extraction and improves the extraction efficiency.

4. Conclusions

In order to explore the effect of the unloading path on coal damage and permeability evolution, experiments with different loading methods and unloading rates were designed. The strength, deformation, and damage characteristics of coal and the evolution of coal permeability under different unloading paths were studied. Combined with the macroscopic damage characteristics of coal, the strength degradation mechanism of unloaded coal and the mechanism of increased permeability and flow were revealed. The research results further clarify the permeability enhancement mechanism of low-permeability coal seam cavitation technology and provide help in guiding on-site engineering practice. The main conclusions are as follows:

(1)

When the initial confining pressure was 5 MPa, 10 MPa, and 15 MPa, the peak differential stress of the coal samples in the constant confining pressure test was 21.99 MPa, 32.33 MPa, and 50.23 MPa, respectively. The peak differential stress of the coal samples in the unloading confining pressure (25 N/s) test was 41.4%, 29.0%, and 34.3% of that in the constant confining pressure and loading axial pressure path, respectively, indicating that the unloading confining pressure significantly reduced the strength of the coal mass. The peak differential stress of the coal samples in the 50 N/s unloading confining pressure test was 77.7%, 77.6%, and 62.2% of that in the 25 N/s unloading confining pressure test, respectively, indicating that the higher the unloading rate, the smaller the peak strength when the coal mass is damaged, that is, the high unloading rate is more likely to cause the destruction of the coal mass. The essence is that the unloading confining pressure reduces the cohesion and internal friction angle of the coal mass.

(2)

The results of acoustic emission monitoring show that the statistics of AE cumulative counts and AE cumulative energy show that the greater the unloading rate of confining pressure, the smaller the energy required for coal sample destruction, and the process of mechanical destruction of coal mass is the result of continuous accumulation of internal damage. This reflects that the unloading environment can significantly reduce the strength of coal mass, and the greater the unloading rate, the more obvious the strength reduction.

(3)

Permeability experiments show that at the initial confining pressures of 5 MPa, 10 MPa, and 15 MPa, the permeability of coal samples under constant confining pressure tests increased by 119.1 times, 75.2 times, and 86.8 times, respectively, before and after damage. For the unloading confining pressure (25 N/s) experiments, the permeability of the coal samples increased by 308.4 times, 272.6 times, and 183 times, respectively, before and after damage. This indicates that the unloading path results in a more significant increase in permeability after coal damage. For the unloading confining pressure experiments at 50 N/s, the permeability of the coal samples increased by 340.6 times, 314.9 times, and 342.9 times, respectively, before and after damage. This shows that the higher the unloading rate, the more significant the increase in permeability after coal damage. Essentially, the unloading damage to the coal reduces the paths for adsorbed gas to diffuse into the fractures, thereby increasing the channels for gas flow.

(4)

The results show that the coal mass is more easily damaged and has a better permeability increase under high-speed pressure relief. In engineering practice, mechanical cavitation technology or hydraulic cavitation with one reaming method should be preferred, as it will obtain better gas extraction efficiency under the same cavitation chamber and lower the risk of coal after pressure relief.