Investigation into the pore structures and CH₄ adsorption capacities of clay minerals in coal reservoirs in the Yangquan Mining District, North China

3.1 Sample collection and preparation

One floor sample (XJ-15-di) and two parting samples (XJ-15-g1 and XJ-15-lvshi) were collected from the No. 15 coal seam of the Xinjing Coal Mine at a depth of 530 m. One coal sample (YQ-15-c) and one parting sample (YQ1-15-g2) were collected from the No. 15 coal seam of the Guoyang First Coal Mine at a depth of 495 m. Additionally, one roof sample (YQ5-15-ding) was collected from the No. 15 coal seam of the Guoyang Fifth Coal Mine at a depth of 510 m. The mineralogical compositions of the parting, roof, and floor samples were expected to be similar to those of the minerals in the coal sample, but the proportions of different minerals in the parting, roof, and floor samples covered a wider range than those of the coal sample. All samples were used in the calculation of the CH₄ adsorption capacities of different minerals in the coal reservoir, rather than using the same species of pure minerals collected from other strata, as reported by Tang and Fan [10], and Fan et al. [13]. All samples were collected from the fresh working face of the coal mines. Approximately 2 kg of each sample was collected and placed in plastic bags immediately after collection to avoid possible contamination and oxidation. According to the Chinese Standard Method GB/T 19560-2008 [25], all the samples were crushed and ground to <2 mm and then sieved using standard sieves to obtain powder samples with particle sizes ranging from 0.18 mm (80 mesh) to 0.25 mm (60 mesh) for subsequent analysis.

3.2 X-ray diffraction analysis

The minerals in the parting, roof, floor, and coal samples were analyzed using X-ray diffraction (XRD). Non-oriented powder XRD patterns were obtained using an X-ray powder diffractometer (D8 ADVANCE, Germany). The operating conditions for which were as follows: power − 2.2 kW; scanning time − 4 min; step size −0.019651°; fixed div slit − 0.6 mm; fixed det slit − 10.5 mm; fixed anti slit − 6.76 mm, and 2θ interval – 7–45°. The XRD pattern of each sample was examined using Quan software, developed by Lin [26], to quantitatively calculate the proportion of each mineral phase identified via XRD. This was performed according to the methods outlined by Chung [27,28,29] and the Chinese Petroleum and Gas Industry Standard Method SY/T 5163-2010 [30].

3.3 Analysis of organic matter content

The XRD results indicate that carbonate minerals were absent in all the samples (Section 4.1). The organic carbon contents of the roof and floor samples were determined according to the Chinese Geology and Mineral Resource Standard Method DZ/T 0279.27-2016 [31]. The organic carbon contents of the coal and parting samples were determined according to the Chinese Standard Method GB/T 476-2008 [32]. Yu [33] reported that the organic carbon portion of the total organic matter in the Chinese low-volatile bituminous coal and anthracite is approximately 90%. Therefore, the organic matter contents of the coal, roof, floor, and parting samples in the present study were calculated as follows: Organic matter = Organic carbon/90%.

3.4 CH₄ isothermal adsorption experiment

Methane isothermal adsorption analysis is a classical method for determining the adsorbed methane content under the conditions of gradually increasing pressure and constant temperature. In this study, the experiments were conducted under dry conditions at a pressure of <9 MPa and a temperature of 35°C. The calibration gas was helium (He; 99.999 mass%), and the carrier gas was nitrogen (N₂; 99.999 mass%). Before the experiments were conducted, the coal, parting, roof, and floor samples (approximately 20 g) were degassed, at approximately 100°C, under a vacuum for approximately 8 h. The CH₄ isothermal adsorption experiments were conducted using a high-pressure high-temperature adsorption analyzer (H-Sorb 2600, China) made by the Gold APP Instruments Corporation in China. According to the Chinese Standard Method GB/T 19560-2008 [25], the Langmuir equation, which is a model describing the monolayer adsorption state of CH₄ on porous materials, was used to calculate the CH₄ adsorption content of the samples and is expressed as follows:

The Langmuir volume of the samples (V _L) represents the saturated CH₄ adsorption amount, p represents the experimental pressure, V represents the CH₄ adsorption content corresponding to p, and b represents the adsorption constant. The V _Ls reflects the CH₄ adsorption capacities of the samples.

3.5 Low-temperature N₂ adsorption experiment

The low-temperature N₂ adsorption and desorption analysis (77.35 K at 101.3 kPa) is a common method used to characterize porous materials [34,35]. The N₂ adsorption and desorption curves of the coal, parting, roof, and floor samples were obtained using a specific surface area and pore size analyzer (V-Sorb 2008P, China) made by the Gold APP Instruments Corporation in China. Before the experiment was carried out, all samples (1–2 g for each sample) were degassed in a vacuum, at approximately 200°C, for approximately 8 h to remove adsorbed moisture and volatile matter. Porous materials can be classified into three categories according to their size: micropores (<2 nm), mesopores (2–50 nm), and macropores (>50 nm). The specific surface areas of some micropores (1–2 nm), mesopores (2–50 nm), and macropores (50–200 nm) were calculated using the Brunauer–Emmett–Teller (BET) method [36]. The BET equation is as follows:

where p _o represents the saturated steam pressure of N₂ at a temperature of 77.35 K; p represents the absolute pressure of N₂, p/p _o represents the relative pressure, and c represents the BET equation comprehensive constant. The fractions of the specific surface areas contributed by the micropores (1–2 nm) were calculated using the t-plot method [37]. The fractions of the specific surface areas contributed by the meso- and macropores (50–200 nm) were calculated by subtracting the micropore surface area from the BET surface area.

The pore volumes of the micropores, with a pore size of 1–2 nm, were calculated using N₂ adsorption data according to the density functional theory (DFT) method [38]. Furthermore, the pore volumes of the meso- and macropores, with a 50–200 nm pore size, were calculated using the Barrett–Joyner–Halenda (BJH) method [36,39]. The BJH method is more suitable for calculating the pore volumes of meso- and macropores, based on the Kelvin equation, as follows:

where γ represents the surface tension of liquid nitrogen, V _m represents the molar volume of liquid nitrogen, r represents the droplet radius of liquid nitrogen, R represents the gas constant, and T represents the temperature in Kelvin. The DFT method is more suitable for calculating the pore volumes of micropores. The total pore volumes of the coal, parting, roof, and floor samples represent the sum of the volumes of the 1–2 nm micropores, mesopores, and 50–200 nm macropores. The pore size distributions (PSDs) of the samples were obtained using the DFT model [35,38].

4.1 Sample compositions

The sample compositions, including the contents of organic matter and inorganic mineral matter, are shown in Table 1. The parting samples had a much higher content of organic matter (32.7% on average) than those of the roof (0.5%) and floor (2.9%) samples. The minerals in the parting, roof, and floor samples were significantly different. The parting samples mainly contained illite (33.2% on average), followed by kaolinite (18.0%), and quartz (16.1%). The minerals in the roof and floor samples were mainly kaolinite (72.7%), followed by quartz (18.7%), and mixed-layer illite-smectite (6.9%). The coal sample had an organic matter content of 88.6%, and the XRD results indicated that the minerals in the coal sample were mainly illite. This result is consistent with that of previous studies on the mineral compositions of coal, parting, roof, and floor samples in the North China Coal Basin [9,40].

4.2 CH₄ isothermal adsorption

The CH₄ isothermal adsorption results for the coal, parting, roof, and floor samples are shown in Figure S1, Table 2, and Table S1. The CH₄ adsorption capacities of the samples varied significantly. The coal sample had the highest V _L of 29.57 cm³/g, which was followed by the parting samples (10.50 cm³/g on average), and roof and floor samples (1.64 cm³/g on average). The organic matter in the samples was strongly correlated with V _L (Figure S2, r = 0.99), indicating that organic matter was the main adsorbent of CH₄. Although the adsorption capacities of the samples showed a decreasing trend with various minerals (r = −0.99), this does not mean that the minerals had no adsorption capacity. It has been reported that the adsorption capacity of clay minerals is lower than that of organic matter [10,41,42].

4.3 Low-temperature N₂ adsorption

4.3.1 N₂ adsorption curves

The low-temperature N₂ adsorption and desorption curves for the coal, parting, roof, and floor samples are shown in Figure 2. According to the classification proposed by Brunauer et al. [43], all the samples have Ⅱ-type adsorption curves characterized by an inverse-S shape, indicating the existence of both mesopores and macropores [44]. Notably, there were differences between these adsorption curves. The adsorption content of N₂ at a very low relative pressure (p/p _o <0.01) increased significantly as organic matter decreased and inorganic mineral matter, especially clay minerals, increased (Figure S3). This indicates that clay minerals have more micropores in the range of 1–2 nm than organic matter, which results in a significant increase in the adsorption content of N₂ at a very low relative pressure (p/p _o <0.01). This indicates the presence of different pore structures between the organic matter and inorganic minerals. The pore structures of both the organic matter and clay minerals vary widely. The pore structure of organic matter is mainly dependent on factors such as its type and maturity, and those of clay minerals are mainly associated with factors such as their species and formation processes [12,45,46,47]. At medium relative pressure (p/p _o = 0.3–0.8), the N₂ adsorption contents of all the adsorption curves increased slowly, but stably, owing to the multilayer adsorption of N₂ in the meso- and macropores. The N₂ adsorption contents increased sharply at high relative pressure (p/p _o = 0.9–1.0), close to the saturated vapor pressure, which is mainly attributed to N₂ capillary condensation (Figure 2 [48]). Capillary condensation during adsorption, and evaporation during desorption in some meso- and macropores can cause hysteresis loops to appear between the adsorption and desorption curves, the shapes of which can reflect the pore types of porous media (Figure 2 [49,50]). The desorption curve was forced to coincide with the adsorption for all samples at a p/p _o of 0.45–0.5, indicating the collapse of the hemispherical meniscus of condensed nitrogen in the mesopores of <4 nm [14,47,51]. According to their shapes, the hysteresis loops are divided into four types by the IUPAC, which represent different types of pores [44]. All the samples have a hysteresis loop characterized by the H3 type, indicating the dominance of slit-shaped and wedge-shaped pores.

Figure 2

Low-temperature N₂ adsorption and desorption isotherms of coal, parting, roof, and floor samples.

4.3.3 Pore volume and pore size distribution

The pore volumes of the coal, parting, roof, and floor samples, calculated using the DFT and BJH methods, are shown in Table 4. XJ-15-di had the highest meso- and macropore volumes of 0.0289 cm³/g, and XJ-15-lvshi had the highest 1–2 nm micropore volume of 0.0010 cm³/g. The mesopore, macropore, and 1–2 nm micropore volumes of the coal sample were the lowest. The micropore volume of the Yangquan coal-measure shale was much higher than the 1–2 nm micropore volume of the parting, roof, and floor samples [52]. Based on the Yangquan coal-measure shale, the <1 nm micropore volumes of the parting, roof, and floor samples, which varied from 0.0033 to 0.0064 cm³/g, are shown in Table 4. The <1 nm micropore volume of the Yangquan coal was 0.072 cm³/g that was cited from Reference [53], as shown in Table 4. The <1 nm micropore volume was much higher than the 1–2 and 2–200 nm micropore volumes, indicating a dominant pore size of <1 nm in the Yangquan coals [53]. The total pore volumes of the parting, roof, and floor samples were strongly correlated with the total specific surface area (r = 0.88), which supports the calculations of the specific surface areas and pore volumes. Although the total specific surface area of the coal sample was much higher than those of the parting, roof, and floor samples, the total pore volume of the former had the same order of magnitude with the latter samples. This was mainly attributed to their different pore structures, and the fact that the coal sample had more micropores.

Table 4

Pore volumes of meso- and macro-pores and micro-pores of the coal, parting, roof, and floor samples

Sample	Pore volume (cm3/g)				Citation
	2–200 nm	1–2 nm	<1 nm	Total
XJ-15-di	0.0289	0.0009	0.0034	0.0332	Present study
XJ-15-g1	0.0115	0.0004	0.0039	0.0158
XJ-15-lvshi	0.0159	0.0010	0.0033	0.0202
YQ1-15-g2	0.0100	0.0004	0.0039	0.0143
YQ5-15-ding	0.0164	0.0008	0.0035	0.0207
YQ-15-c	0.0041	0.0001	0.072	0.0762
XJ1	0.001		0.072	0.073	[53]
YQ-04-10	0.0367	0.0068		0.0435	[52]
YQ-04-13	0.0288	0.0029		0.0317
YQ-04-15	0.0347	0.0025		0.0372
YQ-04-17	0.033	0.0050		0.038
YQ-04-19	0.0381	0.0013		0.0394
YQ-04-22	0.0448	0.0070		0.0518

The <1 nm micro-pore volumes of the parting, roof and floor samples were calculated according to Reference [51]. The <1 nm micro-pore volume of the coal sample was calculated according to Reference [52].

The PSDs of the samples, calculated using the DFT model, are shown in Figure 3. The PSD patterns contained different peaks representing different pore size groups, and some peaks overlapped with each other. To quantify the pore size groups of the samples, a deconvolution method was applied to determine the positions (mean size), areas (pore volume), and widths (range) of the peaks in the PSD patterns of the samples, using a commercially available data-processing program [14,58] (OriginPro software). A normal distribution was used to describe the PSD of each pore size group. The pore sizes of the samples were mainly distributed in the range of <6 nm, and three to five peaks were identified in each PSD pattern (Figure 3 and Table 5).

Figure 3

PSDs and pore size groups identified by the deconvolution method. Kao, kaolinite; Ilt, illite; Ilt/Sme, mixed-layer illite smectite; Sme, smectite.

Table 5

Positions, half widths, and areas of the peaks identified in the PSD patterns of the coal, parting, roof, and floor samples

Sample	Group 1 (Ilt/Sme)			Group 2 (Ilt)			Group 3 (Kao)
	Position	Half width	Area	Position	Half width	Area	Position	Half width	Area
XJ-15-di	1.03	0.41	0.00034				1.34	0.43	0.00073
XJ-15-g1				1.36	0.28	0.00019
XJ-15-lvshi				1.2	0.29	0.00039	1.4	0.31	0.00057
YQ1-15-g2				1.36	0.28	0.00019
YQ5-15-ding	1.08	0.34	0.00032				1.39	0.44	0.00071
YQ-15-c

Sample	Group 4			Group 5			Group 6
	Position	Half width	Area	Position	Half width	Area	Position	Half width	Area
XJ-15-di	1.74	0.37	0.00024	2.98	1.25	0.00061	4.75	1.49	0.00027
XJ-15-g1	1.62	0.6	0.00042	2.95	1.12	0.00051	4.72	1.59	0.00031
XJ-15-lvshi	1.74	0.31	0.00037	2.89	1.29	0.00060	4.62	1.28	0.00026
YQ1-15-g2	1.67	0.34	0.00019	3.04	1.07	0.00032	4.84	1.4	0.00021
YQ5-15-ding	1.78	0.31	0.00015	2.88	1.41	0.00050	4.75	1.36	0.00023
YQ-15-c	1.62	0.57	0.00009	2.51	0.88	0.00007	4.95	0.88	0.00003

Position represents mean pore size (nm) and area represents pore volume (cm³/g).

5.1 Comparison between methane adsorption capacities and compositions

The CH₄ adsorption capacities of the coal, parting, roof, and floor samples reflected the sum of the CH₄ adsorption capacities of the various components. Organic matter has a much higher CH₄ adsorption capacity than inorganic mineral matter, clay minerals have higher CH₄ adsorption capacities than other non-clay minerals, and the CH₄ adsorption capacities of various clay minerals are also significantly different [41,42]. Of all the clay minerals, Ca-smectite has the highest CH₄ methane adsorption capacity, followed by illite, kaolinite, and Na-smectite. Meanwhile, chlorite has the lowest CH₄ adsorption capacity [12,13,42]. In the present study, an overdetermined equation was established as follows:

where A represents the coefficient matrix of the sample compositions, x represents the unknown vector of the V _L values of various components in the samples, and b represents the vector of the V _L values of the samples. The least-squares method was used to solve this equation, which may be expressed as follows:

The calculation results (significant, P < 0.05) indicate that the organic matter had the highest V _L value of 33.80 cm³/g. However, the V _L values of the minerals were much lower, and quartz (−3.50 cm³/g) and illite (−1.40 cm³/g) even had negative values. The CH₄ adsorption capacity of the organic matter was much higher than that of the inorganic clay minerals. The inorganic clay minerals showed a blocking effect on the organic matter in the coal, parting, roof, and floor samples, which are the mixtures of organic matter and inorganic clay minerals. Therefore, the CH₄ adsorption capacity of each mineral calculated using equation (5) can be considered the sum of the net CH₄ adsorption capacity and the blocking effect (negative effect) on organic matter. It has been reported that the CH₄ adsorption capacity of quartz is very limited, close to 0, and the negative value (−3.50 cm³/g) calculated using equation (5) is mainly attributed to the blocking effect of quartz on organic matter. If the blocking effects of the clay minerals are similar to those of quartz, the net CH₄ adsorption capacities of the clay minerals can be calculated by subtracting the blocking effect from the results of equation (5). The net methane adsorption capacities of mixed-layer illite-smectite, kaolinite, and illite were 4.47, 5.81, and 2.08 cm³/g, respectively (Table 6). Compared with other studies on the CH₄ adsorption capacities of various clay minerals [10,12,13,41], the kaolinite in the present study had a higher V _L value than those of other studies, whereas the CH₄ adsorption capacities of illite were similar. Kaolinites with different formation processes have different pore structures [12,59]. The kaolinite in the parting, roof, and floor of the Yangquan coal seam is a detrital mineral with a terrigenous origin [9]. Erosion during mechanical transport, from the sediment source region (the Yinshan Oldland) to the peat mire, resulted in kaolinite with a relatively well-developed pore structure. It has been reported that detrital kaolinite has a higher CH₄ adsorption capacity than kaolinite of a syngenetic origin [9,12]. The mixed-layer illite-smectite in the present study had an adsorption capacity between smectite and illite, depending on the ratio of the layers of smectite to illite. The mixed-layer illite-smectite in the roof and floor samples had a smectite layer proportion (S%) of 15%. Moreover, if the illite layers have a similar CH₄ adsorption capacity to that of the illite in the parting samples, the V _L of the smectite layers can be calculated to be approximately 18.01 cm³/g, which is higher than that of the other clay minerals and that of the smectite reported in other studies [12,13,42] (Table 6). The Ca-smectite has a much higher CH₄ adsorption capacity than that of Na-smectite, kaolinite, illite, and chlorite. This is because the interlayer distance of Ca-smectite is approximately 0.51 nm higher than the molecular diameter of CH₄ (0.38 nm), whereas the interlayer distances of the other clay minerals (0.01 nm kaolinite, 0.04 nm K-illite, 0.07 nm NH₄-illite, 0.30 nm Na-smectite, close to 0 nm for chlorite) are lower than the molecular diameter of CH₄ [21,41,42]. This results in the surface area of the smectite including not only external surface area (50 m²/g [57]) but also including internal area (750 m²/g [57]) different from other clay minerals. The existence of an internal surface area contributed to the higher CH₄ adsorption capacity of the Ca-smectite. The relatively higher CH₄ adsorption capacity of the smectite layers in the mixed-layer smectite-illite in the present study is inferred to be associated with the Ca-smectite.

Because of the significantly higher specific surface area, the organic matter in the samples had a much higher CH₄ adsorption capacity than the inorganic clay minerals. For the coal sample, the organic matter contributed to the CH₄ adsorption capacity, and its contribution to the total CH₄ adsorption capacity was as high as 99.3%. For the parting, roof, and floor samples, such as XJ-15-lvshi, both the organic matter (49.3%) and clay minerals (50.7%) contributed to the total CH₄ adsorption capacity.

5.2 Comparison between PSDs and compositions

The PSD patterns of the coal, parting, roof, and floor samples contained different peaks representing different pore size groups, which corresponded to different types of porous materials. For example, the pore size group with a mean size of 1.03–1.08 nm was identified in the PSD patterns of XJ-15-di and YQ5-15-ding but was absent in the PSD patterns of the other samples. Correspondingly, mixed-layer illite-smectite was present in XJ-15-di and YQ5-15-ding but was absent in the other samples. It was inferred that the mixed-layer illite-smectite contributed to the pore size group with 1.03–1.08 nm as the mean size range. According to the comparison between the PSDs and the compositions of the samples, it can be seen that some pore size groups and some clay minerals were present and absent simultaneously (Figure 3 and Table 5). It is suggested that these clay minerals corresponded to different pore size groups. The pore size group with 1.34–1.40 nm as the mean size range (referred as Group 3) was mainly associated with the kaolinite, and its pore volume (peak area) was positively correlated with the kaolinite content, although there were only three points (Figure 4 and Table 5). The pore size group with 1.20 nm to 1.36 nm as the mean size range (referred as Group 2) was mainly associated with illite, and its volume was positively correlated with the illite content, as expected (Figure 4 and Table 5). The mixed-layer illite-smectite had a pore size group with a mean size range of 1.03–1.08 nm (referred as Group 1; Figure 4 and Table 5). The three pore size groups with mean size ranges of 1.62–1.78, 2.51–3.04, and 4.62–4.95 nm (referred as Groups 4, 5, and 6, respectively) were present in all the PSD patterns. Notably, the pore volume of Group 4 first increased and then decreased with increasing OM content (Figure 5). According to the report of Feng et al. [60] on the pore structure of organic matter in the terrestrial shale of North China, this pore size group may occur in the organic matter that developed along the interface between inorganic minerals and organic matter. They also determined that this pore size group is mainly associated with the shrinkage of organic matter during maturation. The pore volumes of Groups 5 and 6 showed no significant variation with organic matter content (Figure 5), indicating that these two pore size groups were developed in both organic matter and inorganic clay minerals. In the PSD patterns of XJ-15-di and YQ5-15-ding, which contained mixed-layer illite-smectite, the curves showed an increasing trend as the pore size decreased when the pore diameter was <1 nm (Figure 3). It is inferred that for XJ-15-di and YQ5-15-ding, there is a peak centered at approximately 0.51 nm, which represents the interlayer spaces of the smectite layers.

Figure 4

Variations of the pore volumes of Groups 1, 2, and 3 along with the contents of mixed-layer illite-smectite, illite, and kaolinite. Kao, kaolinite; Ilt, illite; Ilt/Sme, mixed-layer illite smectite; Sme, smectite; V, pore volume.

Figure 5

Variations of the pore volumes of Groups 4, 5, and 6 along with the organic matter content OM, organic matter content.

5.3 Influence of diagenetic conversion on the CH₄ adsorption capacity

The CH₄ adsorption capacities of various clay minerals are mainly influenced by their pore structures, which in turn are further influenced by their species and formation processes. In the present study, the pores with a diameter of <6 nm were the major contributors to the CH₄ adsorption capacities of the clay minerals. The <6 nm pore volumes of the clay minerals were the sum of the 1–6 nm pore volumes, which included Groups 1, 2, 3, 4, 5, and 6, and the <1 nm pore volume. Each pore volume was calculated as follows: the pore volumes of Groups 1, 2, and 3 in the mixed-layer illite-smectite, illite, and kaolinite, respectively, were calculated based on the regression equations in Figure 4 when the content of each clay mineral was 100%; the pore volume of Group 4 was 0 when organic matter was absent; the pore volumes of Groups 5 and 6 were calculated as the averages of the parting, roof, and floor samples; and the < 1 nm pore volume was calculated similarly to the average of the parting, roof, and floor samples. The < 6 nm pore volume of the smectite layers in the mixed-layer illite-smectite was calculated according to the < 6 nm pore volume of illite and proportion of the smectite layer. The results for the <6 nm pore volumes of kaolinite, illite, mixed-layer illite-smectite, and smectite layers were 0.00531, 0.00519, 0.00525, and 0.00559 cm³/g, respectively. A comparison between the CH₄ adsorption capacities and < 6 nm pore volumes of the clay minerals showed that the CH₄ adsorption capacity increased significantly as the < 6 nm pore volume increased (Figure 6).

Figure 6

Variations of CH₄ adsorption capacities and <6 nm pore volumes of the clay minerals in the process of diagenetic conversion. Kao, kaolinite; Ilt, illite; Ilt/Sme, mixed-layer illite-smectite; Sme, smectite.

The diagenetic conversion had a significant influence on the pore structures of the clay minerals in this study. The clay minerals in the Yangquan coal reservoir mainly undergo two conversion processes during coalification: the conversion of kaolinite to illite and the illitization of mixed-layer illite-smectite [9]. The action of both processes resulted in a decrease in the CH₄ adsorption capacity (Figure 7).

Figure 7

Diagram for the conversion of kaolinite to illite.

The illite in the intra-seam partings was mainly converted from detrital kaolinite at approximately 150°C during coalification [9,21]. Importantly, illite is a 2:1 type clay mineral, whereas kaolinite is a 1:1 type. The conversion of kaolinite to illite was mainly attributed to the incorporation of an Si–O tetrahedron [21]. This caused the blocking and filling of the illitic pores inherited from kaolinite. Due to this blocking and filling effect, Group 3 of the kaolinite, which had a pore size of 1.34 nm to 1.40 nm, was reduced to Group 2 of the illite (1.20–1.36 nm). Therefore, the illite had a lower CH₄ adsorption capacity, lower <6 nm pore volume, and smaller pore size than kaolinite (Figure 6 and Table 4). In addition, the interlayer cations of the illite in the present study were mainly composed of NH₄ ⁺ rather than K⁺ [9], and NH₄ ⁺ and CH₄ are isoelectronic species with similar physical properties. In light of this, it would be beneficial to study whether the presence of NH₄ ⁺ influences the CH₄ adsorption capacity of illite.

Compared with illite, smectite has crystal lattice defects, which result in the presence of a greater internal surface area and micropore volume [57]. The mixed-layer smectite-illite in the intra-seam partings was gradually converted to illite during coalification. Meanwhile, the crystal lattice defects of the smectite layers in the mixed-layer smectite-illite gradually decreased. It is reasonable to believe that the illitization of the mixed-layer smectite-illite results in a decrease in the internal surface and micropore volume, as well as further decreases in the CH₄ adsorption capacity.