Introduction

As a by-product gas in coal seams, coalbed methane (CBM) has emerged as an important unconventional natural gas (Taheri et al., 2017), which is also relevant to underground mining disasters such as gas explosions (Tong et al., 2018; Wang et al., 2020) and coal and gas outbursts (Wang et al., 2013; Du et al., 2020). CBM production from shallow minable coal seams has become prominent for resource utilization demands and mine safety (Karacan et al., 2011), and underground drainage with boreholes has become an enforcement measure for mines with high CBM.

The borehole drainage of CBM is the most efficient method for underground mine gas prevention and control (Zhou et al., 2016). There are two types of extraction boreholes, namely crossing and bedding boreholes (Wang et al., 2014). Where drilling crossing boreholes must have a service gallery in the rock formation under the coal seam, it is a time-consuming and costly method (Liu et al., 2020). Compared to the crossing boreholes method, drainage CBM with bedding borehole has become the preferred method for coal seams that contain large amounts of CBM because of its advantage of time saving by eliminating the service gallery construction (Li et al., 2020b). At present, major underground drilling methods include water-based, air-based, and dry drilling methods with spiral drill pipes (Han et al., 2018; Li et al., 2020a). However, it is a challenge to use these methods in tectonically developed coal seams where abnormal in situ stress developed CBM accumulation (Tu et al., 2019; Cheng & Pan, 2020). Although drilling in complex formation is common in surface drilling projects and although several methods have been developed to overcome the adverse factors of drilling, it is difficult to use these methods directly because the majority of underground boreholes are horizontal or inverted holes. The formed borehole wall for these holes is in a free state and does not restrict the pore pressure and in situ geo-stress, leading to difficulties in overcoming borehole creep and collapse caused by the reserved CBM pressure and in situ geo-stress (Zhai et al., 2015; Zhao et al., 2020). Thus, coal production is typically delayed due to CBM extraction limitations (Zhou et al., 2016).

In this paper, we propose an underground pressurized drilling method to ensure the smooth drilling of boreholes in coal seams. As indicated in Figure 1, the drilling system consists of a slush pump, a pressurized slush tank, a borehole mouth-mounted sealing device, pressure-controlled solid–liquid separator, and underground drill rig. The drilling fluid is pressurized using a slush pump (to inhibit gas gushing into the borehole and to equalize the in situ formation stress), a borehole mouth-mounted sealing device, and a pressure-controlled solid–liquid separator during the drilling process.

Figure 1
figure 1

Sketch of pressurized drilling method

Full size image

Drilling fluid plays a key role in the drilling process. Based on different formation conditions, researchers have developed various drilling fluids that can inhibit wellbore instability and lost circulation effectively during oil and gas development, such as high-performance water-based drilling fluids, which can maintain stability and inhibit shale dispersion at high temperatures (Akpan et al., 2019), drilling fluid with added boron nitride nanoparticles, which has good rheological properties in low temperature condition and is used in natural gas hydrate drilling (Maiti et al., 2021), and fuzzy ball drilling fluid, which is an environment-friendly drilling fluid used to control CBM wellbore collapse (Zheng et al., 2018a; Zhang et al., 2019). Compared to other types of drilling fluids, water-based drilling fluids are used diffusely in CBM drilling because their raw materials are inexpensive, environment-friendly, and safe. The fluid rheology, inhibition, and plugging performance of water-based drilling can be adjusted by adding additives such as polymers, surfactants, and sealing materials (Chen et al., 2019; Sun et al., 2020).

However, drilling fluid can cause serious reservoir damage (Ni et al., 2019). Previous studies on oil and gas drilling indicate that drilling fluid infiltration into a reservoir can cause particle aggregation and blockage, polymer gel blockage, water sensitivity damage, and water locking damage, which block the effective seepage channel and influence the migration of oil and gas (Zhao et al., 2019; Wang & Zhou, 2020; Seright & Brattekas, 2021). Compared to conventional oil and gas reservoirs, coal has characteristics of low original porosity, micropore development, low permeability, and strong adsorption capacity (Huang et al., 2022). Meanwhile, coal is a typical stress sensitive material whose permeability would significantly reduce with stress compression (Zheng et al., 2020). Researchers are committed to optimizing the drilling fluid formula to control lost circulation and reduce the damage rate (Kang et al., 2014; Freire Soares et al., 2018; Lyu et al., 2019). Several methods have been proposed to evaluate reservoir damage accurately (Okere et al., 2020; Liu et al., 2022). However, there has been only limited research on permeability damage mechanism and evolution characteristics of a coal seam, although this is extremely important for reducing coal seam damage.

Therefore, we attempted to investigate the permeability damage mechanism and evolution of coal seams induced by drilling fluid. First, the core flow test method was used to evaluate the extent of permeability damage. The extents of water sensitivity, water locking, and solid invasion damages were analyzed by scanning electron microscopy (SEM), nuclear magnetic resonance (NMR) spectroscopy, and other test methods. The effect of infiltration time was also discussed. Finally, the time-dependent stress sensitivity of contaminated coal was studied and the effects of confining and gas pressures on contaminated coal were compared.

Materials and Methods

Specimen Preparation

The coal samples used in this study were bituminous coal from Shenmu in Shaanxi Province. The coal samples were processed into cylinders each with diameter of 50 mm and height of 100 mm. The longitudinal wave velocity of each sample was tested using an ultrasonic detector to ensure the uniformity of the coal samples. Coal samples with similar P-wave characters were used. The pretreatment scheme is presented in Table 1 and Figure 2. Coal samples were divided into eight types. The natural coal samples were raw coal pillars without any pretreatment. The raw coal samples were dried at 378.15 K for 24 h to obtain dry coal samples. The pretreatment method of drilling fluid pollution and saturated water coal samples was as follows. The dried coal samples were soaked in the fluid to emulate the fluid filtration process with 2 MPa fluid pressure for predetermined times. The weights of the coal samples before and after pretreatment were recorded to calculate their moisture contents.

Table 1 Coal sample preparation details
Full size table
Figure 2
figure 2

Coal sample preparation

Full size image

Water-based drilling fluids have several advantages including low cost and stability. Moreover, water supply to coal mines is usually available. Therefore, water-based drilling fluids are the preferred drilling medium in the pressurized drilling method. We chose polymer additives, solid-phase materials as drilling mud material mixed with water to prepare the drilling fluids. The additive materials contained mainly 8 wt.% bentonite, 0.5 wt.% carboxymethyl cellulose (CMC), 0.1 wt.% potassium polyacrylate (K-PAM), and 0.5 wt.% sodium carbonate (Na2CO3) (Fig. 3). The performance of the drilling fluid was tested per the API test standards (Saleh, 2022). The density, apparent viscosity, and filter loss of the drilling fluids were 1.06 g/cm3, 26 mPa s, and 6.8 ml, respectively.

Figure 3
figure 3

Drilling fluid

Full size image

Experimental Methodology

Coal Samples Analysis Experiments

The proximate analysis confirmed that the moisture content (Mad) was 6.48%, ash content (Aad) 7.60%, volatile content (Vad) 33.47%, and fixed carbon content (FCad) 52.45%. X-ray diffraction (XRD) analysis using a D8 Advance X-ray diffractometer (Bruker Company, Germany) was performed to determine the mineral composition of the coal samples, and the results indicated that the main minerals of the coal were kaolinite (7%), chlorite (1%), illite (2%), quartz (6%), feldspar (2%), calcite (2%), and amorphous (80%).

Contact angle is a key parameter for evaluating the capillary pressure and relative permeability of porous sample, and the contact angle was tested by a JC-PHa contact angle meter. SEM was used to analyze the surface microstructure of the raw and contaminated coal samples. A low-field NMR Carr–Purcell–Meiboom–Gill pulse sequence was used to measure the fluid relaxation process in the samples with an external magnetic field by a MacroMR12-150H-I NMR system (Suzhou Niumag Analytical Instrument Corporation), and the relaxation time (T2) distribution spectrum was obtained as well. Given that a coal skeleton has no relaxation signal, the fluid relaxation characteristics of the saturated water coal samples could reflect their pore distribution (Liang et al., 2020). The pore distribution was calculated using Eqs. S1 – S3 (Supplementary Information). The water distribution of the coal samples at different drilling fluid pollution times was analyzed by NMR imaging.

Furthermore, using a briquetting machine, approximately 70 g of pulverized coal with particle sizes in the range of 60–80 mesh and water was mixed, forming coal briquettes with diameter of 50 mm and height of 25 mm. The resulting cylindrical coal briquettes were immersed in drilling fluid after installing strain gauges, and the strain evolution within 24 h was recorded.

Gas Permeability Test Methodology

To assess the permeability damage characteristics of the drilling fluid in coal seams, a self-designed rock seepage device was used to test the changes in permeability (Fig. 4). Pretreated coal samples were placed in the core holder and high-pure nitrogen was injected from the inlet after confining pressure and axial pressure was applied. Under four different confining pressure conditions (ranging from 2 to 5 MPa), the permeability was tested with gas pressures ranging from 0.2 to 1.6 MPa. Gas flow was recorded at the outlet when it was stable. Permeability was calculated based on Darcy’s law (Eq. S4, Supplementary Information).

Figure 4
figure 4

Gas permeability testing diagram

Full size image

Results and Discussion

Figure 5 shows the permeability of coal samples with 3 MPa confining pressure. The gas permeability of four coal samples with the same gas pressure demonstrated the following trend: CND > CN0 > CNW > CN24. Among these, the gas permeability of samples CND and CN0 decreased initially with increase in pore pressure; it was directly proportional to pore pressure when it exceeded 1 MPa. Gas permeability of coal is related closely to the processes of gas flow and gas adsorption. At the low pore pressure stage, the gas adsorption capacity of coal was enhanced with increase in gas pressure, which caused the adsorption expansion of the coal matrix and narrowing of the seepage channel. Moreover, the influence of the Klinkenberg effect on gas flow in the coal samples increased gradually with decrease in gas pressure (Zou et al., 2016; Bobo et al., 2019; Meng et al., 2021). Consequently, the gas permeability of the coal samples tended to decrease with increasing pore pressure. However, with increment of gas pressure, the gas adsorption capacity of the coal saturated gradually, and the influence of the Klinkenberg effect was weakened. Afterward, the decrease in effective stress and expansion of seepage space were the main reasons for the increase in permeability.

Figure 5
figure 5

Gas permeability of coal samples

Full size image

However, unlike samples CND and CN0, the gas permeabilities of samples CNW and CN24 increased as a function of pore pressure because the preferential adsorption of water reduced coal matrix swelling due to gas adsorption. Thus, the permeability of the coal samples was mainly influenced by the reduction in effective stress and by the expansion of seepage channel. To describe the permeability damage of the coal samples, Dk was defined as the permeability damage rate; it can be calculated using Eq. S5 (Supplementary Information).

Figure 6 displays the gas permeability and permeability damage rate of four coal samples at 1 MPa gas pressure, with the permeability of sample CND as the initial value. The permeability of the coal samples was correlated negatively with the confining pressure for the seepage space reduced with increasing of confining pressure. Compared with dry coal samples, the average permeability damage rate of coal samples CN0 and CNW increased by 20.55% and 71.79%, respectively. Correspondingly, the moisture contents of samples CN0 and CNW increased from 0% to 3.58% and 6.51% (Table 1), respectively, indicating that water saturation was a major contributor to the gas permeability of the coal samples. Two possible reasons for this damage are water sensitivity and water locking damages (Li et al., 2022).

Figure 6
figure 6

Permeability damage rate of coal samples with different confining pressures

Full size image

Water Sensitivity Damage of Coal

The proximate analysis indicated that ash content of the coal was 7.6%. Based on XRD analysis, clay minerals (kaolinite, chlorite, and illite) were found to account for 10% of coal samples. The clay minerals could experience hydration, expansion, and dispersion in water, thus increasing gas flow resistance, which would imply that the coal had a certain water sensitivity.

Figure 7 displays the volumetric strain of a briquette sample at different contamination times. The process of volumetric strain evolution can be divided into three stages. In the first stage (t < 1 h), when the drilling fluid was injected into the beaker, the volumetric strain of the briquette sample was negative due to the irregular deformation caused by the difference in infiltration rate. In the second stage (1 h < t < 5 h), the volumetric strain of the briquette sample increased rapidly due to water swelling of the coal matrix and clay minerals when the drilling fluid infiltrated the briquette sample rapidly. Then, as mud cake was formed and a water-saturated state was reached gradually, the evolution process of volumetric strain entered the third stage (t > 5 h), in which the volumetric strain increased slowly and was stabilized gradually. The evolution process of the volumetric strain with drilling fluid infiltration time was similar to that of the moisture content of the coal samples. Water swelling of the coal matrix and clay minerals, which was related to water saturation, caused the narrowing and blockage of pores and fissures, thus reducing the effective seepage channel of the coal samples. These microstructure evolutions led to the decline of gas permeability.

Figure 7
figure 7

Volumetric strain change with contamination time in drilling fluid

Full size image

Water Locking Damage of Coal

Water locking damage occurs because when water is adsorbed by coal, a concave surface is formed at the gas/water interface in the pore throat, generating additional capillary pressure in the coal as water saturation increases. Capillary resistance is inversely proportional to pore radius and contact angle and it is directly proportional to surface tension (Eq. S6, Supplementary Information). Measuring the pore distribution and contact angle of the coal samples can reflect the capillary pressure.

As a prominent nondestructive testing method, NMR is used widely for analyzing coal microstructure (Zheng et al., 2018b; Zhang et al., 2022). Based on Eq. S3 (Supplementary Information), pore size was correlated positively with T2, and the T2 distribution can reflect the pore characteristics of coal samples. Figure 8a displays the T2 spectrum of the coal samples. Only one evident peak was observed in the T2 spectrum. The T2 value peak was between 0.5 and 0.7 ms, indicating that the coal micropores were more developed (Yao et al., 2010; Yao & Liu, 2012). To quantify the pore distribution of the coal samples, their pore size distributions were calculated based on Eq. S3 (Supplementary Information), where ρ = 5.4 nm/ms, Fs = 2, and the conversion coefficient C = 10.8 nm/ms (Xie et al., 2015). The porosity of the coal samples was obtained by correcting the value of T2. The results are presented in Figure 8b.

Figure 8
figure 8

NMR test results of coal samples

Full size image

Based on pore diameter (d) and the coal pore classification method proposed by Hodot (1966), the coal pores can be classified into micropores (d ≤ 10 nm), transitional pores (10 nm < d ≤ 100 nm), mesopores (100 nm < d ≤ 1000 nm), and macropores pores (d > 1000 nm). As indicated in Figure 8b, the porosity of the coal was 14.59% and the proportions of porosity in the four pore ranges were as follows: micropores (50.85%) > transitional pores (38.15%) > mesopores (10.54%) > macropores pores (0.46%). The NMR results indicated that the coal pores were predominantly micropores and transitional pores. Moreover, the average contact angle between the coal rock and water was 51.62° (Fig. 9), and the wettability was water-wetting (Li et al., 2022). Therefore, the water lock effect could substantially reduce the permeability of the coal samples.

Figure 9
figure 9

Contact angle of coal

Full size image

Figure 10 displays the evolution of gas flow rate with gas pressure under 2 MPa confining pressure, and the gas flow rate was found to be correlated positively with gas pressure. With increasing gas pressure, the evolution curves changed gradually from a nonlinear concave curve to a linear curve. No clear gas flow was observed in sample CN24 when gas pressure was less than 0.6 MPa, indicating that flow channel blockage was related to critical gas pressure. Correspondingly, with increasing gas pressure, additional blocked flow channels were reopened.

Figure 10
figure 10

Gas flow rate and gas breakthrough pressures of coal samples

Full size image

Based on the evolution characteristics of the gas flow rate, the gas breakthrough pressure (Pb) is defined as the intersection point of the linearly extending evolution curve at the high gas pressure and abscissa axis (Xu et al., 2019). Gas breakthrough pressure is the critical pressure when the gas can overcome the internal resistance of the coal and form a continuous flow. This is a key parameter that reflects coal microstructure, which correlates with the stress state of the coal and developing degree, shape, and connectivity of the pores and fissures (Wu et al., 2020). Figure 10 displays the gas breakthrough pressures of the coal samples, which were in the following order: CND (1.077 MPa) < CN0 (1.094 MPa) < CNW (1.145 MPa) < CN24 (1.248 MPa). The connectivity of the pores and fissures in the coal samples decreased and the gas flow resistance increased due to the aggravation of water locking damage and water sensitivity damage with moisture content, leading to the increase in gas breakthrough pressure. In addition, the microflow channels in the coal samples were blocked by bentonite particles and polymer molecules, and the flow channel inlet was blocked by mud cake, leading to further increase in gas breakthrough pressure.

Solid Invasion Damage of Coal

As indicated in Table 1, the moisture content of sample CN24 was close to that of sample CNW. However, the average permeability damage rate of sample CN24 was 93.46%, which was greater than that of sample CNW. This was because the bentonite particles in the drilling fluid were deposited on the coal surface and invaded into the coal samples during the filtration process, which blocked part of the pores and fissures. Although the plugging of the seepage channel reduced the leakage of drilling fluid, it caused a decline in the coal sample permeability (Al Jaberi et al., 2021).

Figure 11 displays SEM images of the raw and contaminated coal samples. After immersion of the coal samples in drilling fluid, the microflow channels in the coal samples were blocked by bentonite and polymer molecules, and filter cake blocked the flow channel inlet, leading to further permeability decline (Lei et al., 2017). As indicated in Figure 6, the permeability damage rate of sample CN24 increased by 21.67% relative to sample CNW. However, this only accounted for 23.19% of the total permeability damage of sample CN24. This result confirmed that water sensitivity and water locking damages were the main reason of coal permeability decrease.

Figure 11
figure 11

SEM images of coal samples

Full size image

Effect of Infiltration Time

During drilling, the permeability damage of the contaminated coal seam could be aggravated with the extension of drilling fluid infiltration time. The process of drilling fluid infiltration into a coal seam under a drilling fluid pressure of 2 MPa within 24 h was emulated in a laboratory, and the permeability of the coal samples was tested. Based on Eq. S5 (Supplementary Information), the permeability damage rate was calculated with the permeability of coal sample CN0 as the initial value.

As indicated in Figure 12, the gas permeabilities of the coal samples under 1 MPa gas pressure and 3 MPa confining pressure were taken as examples. The permeability damage of the coal samples was most significant in the first 6 h. Compared with the gas permeability of sample CN0, that of sample CN6 reduced to 0.012 × 10−3μm2, corresponding to a relative decrease in 85%. When the infiltration drilling fluid time was 24 h, the gas permeability decreased from 0.08 × 10−3μm2 to 0.006 × 10−3μm2, and the permeability damage rate was 92.5%. Correspondingly, the gas permeability of sample CN24 decreased by 0.0057 × 10–3 μm2 and the permeability damage rate increased by 7.14% compared with that of sample CN6. This result indicated that the gas permeability of the contaminated coal samples decreased gradually with the extension of drilling fluid infiltration time. When the drilling fluid infiltration time was between 3 and 6 h, the gas permeability and permeability damage rate curves had an evident inflection point. Subsequently, the permeability stabilized gradually with time.

Figure 12
figure 12

Permeability damage rate based on considerations of time

Full size image

The permeability damage rates of the contaminated coal samples decreased with increase in gas pressure because the seepage channel plugging could be relieved with increasing gas pressure. As the gas pressure increased from 0.6 to 1.6 MPa, the change in permeability damage rate of coal sample CN1 decreased from 81.51% to 40.97%, whereas that of sample CN24 merely decreased from 95.17% to 86.76%. These results indicated that the damage degree of the coal samples was aggravated with increased soaking time in the drilling fluid. Consequently, it was more difficult to improve the pore connectivity, and the permeability of coal samples changed marginally with the gas pressure.

The infiltration of the drilling fluid filtrate and blockage of the seepage channels by polymer and bentonite particles were the reasons for the coal permeability damage. As shown in Figure 13, the water distribution in the coal samples was obtained using magnetic resonance imaging (MRI). Within the red circle, the blue area represents an anhydrous coal matrix, whereas the other area’s color gradually changes from blue to red with increase in moisture content. The blue area decreased and the dark areas with irregular bands expanded as contamination time increased, indicating that the mud filtrate diffused gradually in the coal samples and occupied the seepage space, which could reduce the gas permeability of the coal sample. Therefore, we should reduce drilling time as much as possible during pressurized drilling of gas drainage boreholes in gas-bearing coal seams. Effective measures of leakage prevention should be taken to reduce the pollution range of the drilling fluid. After pressurized drilling, the long-term contaminated section of the borehole was the main scope of implementing permeability improvement technology.

Figure 13
figure 13

MRI of water distribution in coal samples

Full size image

Stress Sensitivity of Contaminated Coal

The distribution characteristics of pores and fractures in coal determine its permeability. Moreover, the microstructure of coal is extremely sensitive to stress. However, the infiltration of drilling fluid could influence the microstructure of coal and then influence its stress sensitivity. Therefore, the evolution characteristics of stress sensitivity of contaminated coal were studied. Cp was defined as the stress sensitivity coefficient to describe the change rate of permeability with increasing stress (Eq. S7, Supplementary Information).

As indicated in Figure 14, the permeability of the coal samples decreased with increase in confining pressure. The stress sensitivity coefficient (Cp) under low confining pressure was greater than that under high confining pressure. This is because the original pores and fractures in the coal samples could be regarded as weak interlayers, which could be the first to deform and reduce the seepage channels with increasing confining pressure (Kumar et al., 2018). Under low confining pressure, the anti-deformation ability of the original pores and fractures was low, whereas the compressibility of the fractures was strong. However, with increase in confining pressure, the influence of compaction forced the mineral particles and coal matrix to contact gradually, thus improving the anti-deformation ability of the fractures. Therefore, the gas permeability as well as the stress sensitivity of coal samples decreased with increase in confining pressure.

Figure 14
figure 14

Gas permeability variation with confining pressure

Full size image

Table 2 presents the average stress sensitivity coefficient of the coal samples when the confining pressure increased from 2 to 5 MPa. Overall, the stress sensitivity coefficient of the contaminated coal samples decreased with infiltration time, indicating that the compressibility of fractures was weakened. Cf was defined as cleat compressibility, and its relationship with permeability and confining pressure is (Wu et al., 2021):

$$k = k_{0} e^{{( - 3C_{f} (\sigma - \sigma_{0} ))}} ,$$
(1)

where k is gas permeability (10−3μm2), k0 is initial permeability (10−3μm2), Cf is cleat compressibility, σ is confining pressure (MPa), and \(\sigma_{0}\) is initial confining pressure (MPa).

Table 2 Permeability damage with confining pressure and fitting results
Full size table

Equation 1 was used to fit the permeability evolution curve with the confining pressure. The fitting results are displayed in Figure 14 and Table 2. The permeability damage rate and Cf of sample CN0 were smaller than those of samples CN1 and CN3 because hydration reduced the deformation resistance of the coal samples. However, with increasing drilling fluid infiltration time, the Cf of the contaminated coal samples decreased, indicating that the compressibility of the fractures decreased. Two possible reasons can explain this phenomenon. Firstly, the coal sample creep and compaction caused by drilling fluid pressure could have improved the deformation resistance of the fractures with time. Secondly, the pores and fractures of the coal samples were filled with drilling fluid filtrate. Moreover, the drilling fluid filtrate took much work to compress. Nevertheless, it could provide resistance to pore and fracture deformation under increasing confining pressure. Correspondingly, the permeability damage rate of the coal samples caused by the increase in confining pressure decreased with drilling fluid infiltration time.

The permeability of the coal samples changed with the infiltration of drilling fluid and was influenced by the gas and confining pressures (Zhang et al., 2021). Therefore, effective stress was used to represent the comprehensive influence of the gas and confining pressures, thus:

$$\sigma_{e} = \sigma_{c} - \frac{1}{2}(P_{1} + P_{2} ),$$
(2)

where σe is effective stress (MPa), σc is confining pressure (MPa), and P1 and P2 are gas pressure (MPa) at inlet and outlet, respectively.

Figure 15 displays the evolution curve of gas permeability with effective stress. In the first stage, the confining pressure was fixed, and the gas pressure increased, corresponding to a decrease in effective stress. With increasing gas pressure, additional blocked flow channels were reopened. Correspondingly, the permeability of the contaminated coal samples increased with decreasing effective stress. The permeability of sample CN0 decreased with decreasing effective stress at low gas pressures due to the combination of gas adsorption and the Klinkenberg effect. In the second stage, the gas pressure was fixed and the confining pressure was increased. The permeability of the coal samples decreased with increasing effective stress because the original pores and fractures in the coal samples could be compressed and seepage channels reduced with increasing confining pressure. Compared with confining pressure, coal permeability was more sensitive to changes in effective stress induced by gas pressure because the increased confining pressure only influenced the fracture opening. Conversely, the increased gas pressure not only expanded the pores and fractures but also alleviated the blockage of the seepage channels (Xu et al., 2019).

Figure 15
figure 15

Gas permeability variation with effective stress.

Full size image

Arguably, gas pressure influenced the connectivity of the pores and fractures more likely than confining pressure, especially for coal samples with long drilling fluid infiltration time. The reduction of gas pressure could cause a significant decline in gas drainage efficiency in contaminated coal seams during gas drainage. Therefore, anti-seepage measures should be taken to reduce the drilling fluid pollution degree and scope during drilling. It is necessary to employ pressure-releasing technology and take permeability enhancement measures after drilling, which could reduce effective stress, promote gas desorption, and improve the coal seam permeability.

Conclusions

In this study, we evaluated the infiltration of drilling fluid that caused permeability damage for gas-bearing coal seams. The coal seam damage mechanism was analyzed. The influence of effective stress and pollution time on the coal permeability evolution was investigated. This study provides a theoretical basis for the coal seam protection design of the underground pressurized drilling methods. Based on the research undertaken, the following conclusions can be drawn.

  1. 1.

    The permeability of coal samples was closely related to moisture content because the gas flow channel was narrowed and blocked by water-sensitive and water lock damages. Bentonite and polymer molecules occupied the gas flow channel, which decreased in the pore connectivity and permeability of the coal samples.

  2. 2.

    The permeability damage increased as a function of drilling fluid infiltration time. Moreover, the permeability damage mainly occurred in the first 6 h of pollution and then it tended to stabilize.

  3. 3.

    The permeability of the coal samples was correlated negatively with the confining pressure. However, the influence of the confining pressure weakened with the extension of infiltration time because of the reduction of the fracture compressibility.

  4. 4.

    The permeability of the coal samples decreased with increase in effective stress. Based on the evolution of permeability with effective stress, gas pressure was more favorable in ameliorating the permeability of the contaminated coal sample.