Comparative Study on Pore Characteristics and Methane Adsorption Capacity of the Lower Silurian Longmaxi Shales with Different Lithofacies

Abstract

In this study, shale samples with diverse lithofacies from the Lower Silurian Longmaxi Formation in the Fuling Field were investigated to evaluate the variations in pore characteristics and methane adsorption capacity (MAC) of different shale lithofacies. A set of experiments were performed, such as total organic carbon (TOC) content, X-ray diffraction (XRD), field emission–scanning electron microscopy (FE-SEM), low-pressure gas (CO₂/N₂) adsorption, and high-pressure methane adsorption. Combined with TOC content and mineral composition, three types of shale lithofacies were identified, including organic-rich (OR) argillaceous-rich siliceous (S-3) shale lithofacies, organic-moderate (OM) argillaceous/siliceous mixed (M-2) shale lithofacies, and organic-lean (OL) siliceous-rich argillaceous (CM-1) shale lithofacies. Through detailed comparative analyses, we found that OR S-3 shales possess the maximum TOC content, the most developed heterogeneous organic micro-mesopores, the largest pore volume (PV), and the highest pore surface area (PSA); consequently, they display the strongest MAC. Conversely, OL CM-1 shales have the lowest TOC content and the highest clay content, and thus the smallest PSA and the poorest methane adsorption performance. In conclusion, considering the excellent gas storage potential, sustained shale gas production, and brittle response to hydraulic fracturing, OR S-3 shales are superior to shale gas exploration and exploitation compared with OM M-2 and OL CM-1 shales.

1. Introduction

In recent years, shale gas has been commercially developed and has become one of the most important energy resources in North America and China due to advanced horizontal drilling and hydraulic fracturing techniques [1,2,3,4]. Distinguished from conventional gas resources, shale gas is unique and stored in shale reservoirs in three main mechanisms: (1) free gas in pores and fractures, (2) adsorbed gas on pore surfaces of organic matter and clay minerals, and (3) a minor fraction of dissolved gas in formation water and liquid hydrocarbons [5]. Consequently, the nanometer- to micrometer-scale pores, together with the fractures, serve as the essential storage spaces, and migration pathways are critical to the productivity of shale gas reservoirs [6,7,8]. Determining pore characteristics (including pore types and pore structures such as pore volume (PV), pore surface area (PSA), and fractal dimension (FD)) [9,10,11], and understanding geological controls on pore development are fundamental to evaluating the potential of shale gas resources [12,13,14].

Lithofacies represent a rock or rock assemblage formed under specific sedimentary conditions, resulting from comprehensive effects of various physical, chemical, and biological factors in complex sedimentary environments. It is the material manifestation of a sedimentary environment and preserves valuable mineralogy, geochemistry, petrology, and petrophysical information of rocks [15]. Currently, the variations in pore development in shales among different lithofacies have been reported by some studies [16,17,18], and meaningful discoveries have been achieved [19,20,21]. Several scholars preliminarily suggested that siliceous lithofacies exhibit large pore spaces and that it is the most favorable lithofacies type for shale gas storage [22,23,24]. However, comparative studies on the pore fractal characteristics of shales with different lithofacies are relatively rare.

Previous studies have shown that adsorbed gas constitutes a significant portion of total gas content, ranging from 20 to 85% in the five United States shale formations of Lewis, Barnett, New Albany, Antrim, and Ohio [5], and exceeding 50% for Longmaxi shales in China [25]. Nevertheless, the factors affecting the methane adsorption capacity (MAC) of shales, especially the effects of the shale pore fractal characteristics on the MAC, are ambiguous. In addition, the relationship between MAC and shale lithofacies remains poorly reported, and the variations in MAC for shales with different lithofacies are not well understood. It is necessary to thoroughly discuss the influencing factors and differential characteristics of MAC among different shale lithofacies.

Employing integrated experiments of organic geochemistry, mineral composition, field emission–scanning electron microscopy (FE-SEM), low-pressure gas adsorption, and high-pressure methane adsorption, this study systematically investigates the pore characteristics (including pore types, PV, PSA, and FD) and MAC of different shale lithofacies. The key objectives of this study are to (1) identify the types of shale lithofacies; (2) analyze the variations in pore types, pore structures, and MAC across different shale lithofacies; and (3) discuss the dominant factors controlling pore development and MAC of different shale lithofacies.

This study focuses on the reservoir and gas-storage characteristics across different shale lithofacies, as well as the dominant factors controlling the formation of high-quality shale lithofacies, which is of great significance for well location optimization and shale gas exploration and development.

2. Geological Setting and Samples

The Fuling Shale Gas Field is located in the eastern Sichuan Basin, southwest China (Figure 1a,b). Its structural framework is controlled by three fault systems trending in the NE, NS, and NW directions, which subdivide the Fuling Field into several secondary structural units, including the Jiangdong slope, Jiaoshiba anticline, Zilichang anticline, Baitao syncline, Pingqiao anticline, and Baima syncline (Figure 1c). The shale gas wells of JYA and JYB studied in this paper are situated in the main part of the Jiaoshiba block (Figure 1c). The Jiaoshiba block is a gentle box-shaped anticline and marked by flat strata and undeveloped faults [26]. The fault densities in this area are generally less than 0.02/km², and the fault scales are also small, with the largest fault distance less than 50 m [27]. The relatively tectonically stable background provides an ideal research area to eliminate the influence of external structural deformation and focus on the discussion of internal lithofacies on the control of shale pore development and MAC.

During the Late Ordovician to Early Silurian, the Fuling area was characterized by a low-energy and hypoxia deep-water shelf to shallow-water shelf environment, with deposits of Upper Ordovician Wufeng–Lower Silurian Longmaxi Formation shales [28,29]. Based on shale lithology, color, and biological fossil assemblages, the Wufeng–Longmaxi Formation shales can be divided into three intervals from bottom to top: the first member from Longmaxi Formation shales (Long 1 shales, including Wufeng Formation shales), the second member from Longmaxi Formation shales (Long 2 shales), and the third member from Longmaxi Formation shales (Long 3 shales). Among them, the dark radiolarian and graptolite Long 1 shales are the main target interval of shale gas production [30].

Typical marine shale samples in this study within the Lower Silurian Longmaxi Formation were selected from wells JYA and JYB in the Jiaoshiba block. These collected shale samples (standard core cylinders with a diameter of 10 cm) are approximately evenly distributed at the first member of the Longmaxi Formation, covering the variability in TOC content, mineral composition, and lithofacies (Figure 2). Detailed geological attributes of these samples regarding burial depth, TOC content, mineral composition, and lithofacies are provided in

Table 1. Basic geological parameters of the studied shale samples.

Well	Sample ID	Depth (m)	TOC (%)	Minerals (%)						Lithofacies
				Quartz	Feldspar	Calcite	Dolomite	Pyrite	Total Clays
JYA	JYA-1	2271.72	0.71	34.9	5.1	0.0	0.0	2.0	58.0	OL CM-1
	JYA-2	2277.59	0.50	37.3	6.2	0.0	1.8	2.0	52.7	OL CM-1
	JYA-3	2293.85	1.31	34.5	6.7	4.2	3.9	0.0	50.7	OL CM-1
	JYA-4	2307.17	1.67	36.6	9.9	4.1	0.0	4.7	44.7	OM M-2
	JYA-5	2310.84	1.81	35.7	9.7	2.6	9.6	4.7	37.7	OM M-2
	JYA-6	2315.71	1.36	31.0	12.3	4.9	8.9	2.8	40.1	OM M-2
	JYA-7	2330.97	3.20	42.3	8.9	3.7	5.2	7.7	32.2	OR S-3
	JYA-8	2341.34	2.56	49.0	9.3	7.4	3.0	5.4	25.9	OR S-3
	JYA-9	2350.30	3.85	50.1	7.2	2.7	3.6	11.1	25.3	OR S-3
JYB	JYB-1	2523.35	0.41	31.3	5.7	0.0	0.0	1.7	61.3	OL CM-1
	JYB-2	2538.95	0.41	27.0	6.1	0.0	1.9	1.7	63.3	OL CM-1
	JYB-3	2543.47	0.98	32.2	8.1	2.2	0.0	1.9	55.6	OL CM-1
	JYB-4	2548.08	1.48	31.5	5.0	5.3	9.6	1.6	47.0	OM M-2
	JYB-5	2566.09	1.36	37.4	8.2	3.9	6.1	3.0	41.4	OM M-2
	JYB-6	2572.28	1.41	35.9	9.7	5.6	6.1	1.9	40.8	OM M-2
	JYB-7	2598.12	2.71	46.6	7.4	8.7	3.7	5.8	27.8	OR S-3
	JYB-8	2610.63	3.93	54.4	8.2	2.6	5.3	5.4	24.1	OR S-3
	JYB-9	2617.79	5.27	58.4	3.5	3.2	4.9	5.5	24.5	OR S-3

.

3. Materials and Methods

3.1. TOC Content

The TOC content was measured with a CS844 carbon/sulfur analyzer (Leco Corporation, St Joseph, MI, USA). Prior to the analysis, the studied samples of about 10 g were crushed to a powder of 200 mesh and were then treated with hydrochloric acid to eliminate any inorganic carbonates. The TOC values were derived from the amounts of CO₂ released during sample combustion.

3.2. X-Ray Diffraction (XRD)

The mineral composition of the studied samples was determined by a D/max-2600 X-ray diffractometer (Rigaku Corporation, Tokyo, Japan) at a working voltage of 40 kV and current of 30 mA with Cu Kα radiation (λ = 0.154 nm). Powdered shale samples with 200 mesh of about 2 g were scanned using a step size of 4°/min from 3° to 85°. Semi-quantitative analysis of relative mineral percentages was conducted based on the area under the curve for the major peaks of each mineral.

3.3. Field Emission–Scanning Electron Microscopy (FE-SEM)

The microscopic pore type, morphology, size, and number of studied shale samples was observed using a Zeiss Merlin Compact FE-SEM (Carl Zeiss AG, Oberkochen, Germany). Prior to the experiment, rectangular shale samples 10 mm × 8 mm × 5 mm in size were first mechanically grinded, followed by argon ion polishing with a LEICA EM RES102 system (Leica Corporation, Wetzlar, Germany) to produce a flat surface. Secondary electron imaging was subsequently employed to capture high-resolution images of the polished sample surfaces.

3.4. Low-Pressure Gas Adsorption

Low-pressure CO₂ and N₂ adsorption analyses were conducted to measure the pore structure of the studied shale samples using a Quantachrome Autosorb-iQ3 fully automatic specific surface area and porosity analyzer (Quantachrome Instruments, Boynton Beach, FL, USA). Before the experiment, about 2 g of the powdered samples (60–80 mesh) were oven-dried and outgassed at 110 °C for 12 h in a vacuum tube. The CO₂ adsorption measurements were maintained at temperature of 273.15 K, while the N₂ adsorption–desorption isotherms were obtained at temperature of 77.35 K. The relative pressure (P/P₀) applied for CO₂ and N₂ adsorption was 4 × 10⁻⁴–3 × 10⁻² and 0.005–1, respectively. CO₂ adsorption data were interpreted with the Density Functional Theory (DFT) model to calculate the volume (V) and surface area (SA) of micropores (MIPs, <2 nm), and N₂ adsorption data were interpreted with the Barrett–Joyner–Halenda (BJH) method to calculate the V and SA of mesopores (MEPs, 2–50 nm) and macropores (MAPs, >50 nm).

3.5. Fractal Theory

Fractal theory has been widely employed to describe the porous structure and surface irregularity of shales [11,13,14,15]. FD (D), ranging from 2 to 3, is a petrophysical parameter to characterize the pore surface roughness and pore structure complexity of porous materials. The larger the FD, the greater the complexity and the stronger the heterogeneity of the pores. Based on the N₂ adsorption data, we could effectively calculate the FD using the Frenkel–Halsey–Hill (FHH) model as follows [31,32]:

$$\ln \left(\mathrm{V}\right)=\left(\mathrm{D}-3\right)\left[\mathrm{l}\mathrm{n}\left(\mathrm{l}\mathrm{n}{\displaystyle \frac{{\mathrm{P}}_0}{\mathrm{P}}}\right)\right]+\mathrm{C}\mathrm{o}\mathrm{n}\mathrm{s}\mathrm{t}\mathrm{a}\mathrm{n}\mathrm{t}$$

(1)

where V (cm³/g) is the adsorbed gas volume at the equilibrium pressure P, P (MPa) is the equilibrium pressure, P₀ (MPa) is the saturation vapor pressure, and D is the FD.

3.6. High-Pressure Methane Adsorption

3.6.1. Excess Methane Adsorption

High-pressure methane isothermal adsorption experiments were performed on a gravimetric isothermal adsorption apparatus (ISOSORP-HP II Static (Rubotherm GmbH, Bochum, Germany)) with maximum test pressure and a temperature of 35 MPa and 200 °C, respectively. Powdered samples (60–80 mesh) of about 20 g were first outgassed in a vacuum sample cell at 110 °C for 12 h to fully remove moisture and impure gases. High-purity methane (99.99%) was introduced into the sample cell as the adsorbent. Adsorption experiments were conducted over a pressure range from 0 to 24 MPa at a constant temperature of 30 ± 0.1 °C. The methane excess adsorption was quantified using the routine methods reported in the literature [33,34].

3.6.2. Calculation of Absolute Methane Adsorption

The isothermal adsorption experiment could only obtain the excess adsorption capacity (V_exc) [33,34,35,36]. The absolute adsorption capacity (V_abs) was acquired through the transformation of excess adsorption capacity according to the adapted Langmuir equation [37,38]:

$${\mathrm{V}}_{\mathrm{a}\mathrm{b}\mathrm{s}}={\mathrm{V}}_{\mathrm{L}}{\displaystyle \frac{\mathrm{P}}{\mathrm{P}+{\mathrm{P}}_{\mathrm{L}}}}$$

(2)

$${\mathrm{V}}_{\mathrm{e}\mathrm{x}\mathrm{c}}={\mathrm{V}}_{\mathrm{a}\mathrm{b}\mathrm{s}}\left(1-{\displaystyle \frac{{\mathsf{\rho }}_{\mathrm{g}\mathrm{a}\mathrm{s}}}{{\mathsf{\rho }}_{\mathrm{a}\mathrm{d}\mathrm{s}}}}\right)={\mathrm{V}}_{\mathrm{L}}{\displaystyle \frac{\mathrm{P}}{\mathrm{P}+{\mathrm{P}}_{\mathrm{L}}}}\left(1-{\displaystyle \frac{{\mathsf{\rho }}_{\mathrm{g}\mathrm{a}\mathrm{s}}}{{\mathsf{\rho }}_{\mathrm{a}\mathrm{d}\mathrm{s}}}}\right)$$

(3)

where V_abs (cm³/g) is the absolute adsorption capacity of methane; V_exc (cm³/g) is the excess adsorption capacity of methane; V_L (cm³/g) is the Langmuir volume, representing the maximum MAC; P_L (MPa) is the Langmuir pressure, representing the pressure at half the amount of the Langmuir volume; and ρ_gas (g/cm³) and ρ_ads (g/cm³) are the density of free-phase methane and adsorbed-phase methane, respectively.

At low-pressure subcritical conditions, the density of free-phase methane is much less than the density of adsorbed-phase methane and can be ignored, and the absolute adsorption capacity closely approximates the excess adsorption capacity. In contrast, at high-pressure supercritical conditions, the density of free-phase methane increases rapidly with increasing pressure, and pronounced divergence exists between the absolute and excess adsorption capacities. The excess adsorption capacity begins to decline after the maximum volume [33,34,35,36], and the “negative adsorption” phenomenon appears [37,38].

4. Results

4.1. Shale Lithofacies

The TOC values of the analyzed shales from the Longmaxi Formation in wells JYA and JYB range from 0.50% to 3.85% and 0.41% to 5.27%, with an average of 1.89% and 2.00%, respectively (Table 1). The mineral composition is predominantly composed of quartz and clay minerals. In well JYA, the quartz and clay mineral contents range from 31.0% to 50.1% and 25.3% to 58.0%, with an average of 39.0% and 40.8%, respectively (Table 1). In well JYB, the quartz and clay mineral contents range from 27.0% to 58.4% and 24.1% to 63.3%, with an average of 39.4% and 42.9%, respectively (Table 1). Overall, both the TOC and quartz contents of these shales exhibit an increasing trend with depth, whereas the clay mineral content shows a decreasing tendency with depth (Table 1).

According to the ternary diagram composed of carbonate (calcite and dolomite), siliceous (quartz and feldspar), and clay minerals, shale lithofacies can be categorized into four types: calcareous shale lithofacies, siliceous shale lithofacies, argillaceous shale lithofacies, and mixed shale lithofacies. Each lithofacies type can be further subdivided into four specific lithofacies [22,39]. Three main types of shale lithofacies were identified in the study area based on XRD experimental data, including siliceous-rich argillaceous shale lithofacies (CM-1), argillaceous/siliceous mixed shale lithofacies (M-2), and argillaceous-rich siliceous shale lithofacies (S-3) (Figure 3a). Figure 3b shows the TOC content distribution across these three main shale lithofacies. The TOC content for the CM-1 shales varies from 0.41% to 1.31%, with an average value of 0.72%. The TOC content for the M-2 shales varies from 1.36% to 1.81%, with an average value of 1.52%. The TOC content for the S-3 shales varies from 2.56% to 5.27%, with an average value of 3.59%. Integrating the TOC content and mineral composition, we further divided shale lithofacies into organic-lean (OL) CM-1 shale lithofacies, organic-moderate (OM) M-2 shale lithofacies, and organic-rich (OR) S-3 shale lithofacies. OL CM-1 shales are mainly developed at the upper part of the Long 1 Formation (Figure 2). OM M-2 shales are mainly located at the middle part of the Long 1 Formation (Figure 2). OR S-3 shales are mainly distributed at the lower part of the Long 1 Formation (Figure 2).

4.2. Pore Characteristics

4.2.1. Pore Types

In recent studies, many scholars have successfully recognized shale pore types based on pore morphology and its interrelation with mineral particles. Shale pores can be divided into organic matter pores, mineral matrix pores in the form of interparticle pores and intraparticle pores, and micro-fracture pores [6,7,8].

The formation of organic matter pores is attributed to the generation and expulsion of hydrocarbons during the thermal evolution of organic matter [1,6,40], and they are widely distributed in OR S-3 and OM M-2 shales (Figure 4d–i). These organic matter pores are usually nanometer scale and exhibit bubble-like or elliptical shapes (Figure 4g–i).

Interparticle pores exist between adjacent mineral particles and typically display triangular to polygonal shapes, generally with a large in size of several micrometers (Figure 4a,c). Intraparticle pores are commonly identified as intercrystalline pores within pyrite framboids and dissolution pores within unstable minerals. Pyrite intercrystalline pores are visible across all shale lithofacies (Figure 4b–d,h), some of which filled with organic matter to form secondary organic pores. Dissolution pores (including some ring-like grain marginal dissolution micro-fractures) are produced through dissolving unstable minerals such as calcite and feldspar by organic acidic fluids during hydrocarbon generation [41]. These dissolution pores are typically circular to elliptical and most abundant in OR S-3 shales (Figure 4g–i).

Interlayer micro-fractures formed along the cleavage planes of oriented clay platelets due to dehydration shrinkage and mineral-phase transition during diagenesis [42,43] represent the dominant micro-fracture pores in the studied shales. These micro-fractures are parallel to each other, with lengths extending up to several micrometers, and are abundantly developed in OL CM-1 and OM M-2 shales (Figure 4a–f).

4.2.2. Isotherms of CO₂ Adsorption

Figure 5a–c present CO₂ adsorption isotherms of samples from different shale lithofacies. These CO₂ adsorption isotherms correspond to type I according to the IUPAC classification [44], indicating MIPs within the shales. The maximum adsorption (i.e., the amount of adsorption at maximum relative pressure) is greatest in OR S-3 shales, ranging from 1.76 cm³/g to 2.37 cm³/g, with an average of 2.03 cm³/g. The maximum adsorption for OM M-2 shales varies from 1.48 cm³/g to 2.09 cm³/g, with an average of 1.72 cm³/g. OL CM-1 shales have the lowest CO₂ adsorption, lying between 0.96 cm³/g and 1.96 cm³/g, with an average of 1.55 cm³/g.

4.2.3. Isotherms of N₂ Adsorption and Desorption

N₂ adsorption–desorption isotherms of shale samples with different lithofacies are shown in Figure 5d–f. These isotherms for N₂ adsorption–desorption are classified as type IV based on the IUPAC classification [45]. The adsorption branches display reversed S-shaped features marked by a sharp uptake in adsorption amount at both low (<0.05) and high (>0.95) relative pressures. The typical capillary hysteresis loops appear in the desorption process to be a mixture of types H2 and H3, indicative of the superposition of ink-bottle-shaped and slit-shaped pore types in the shales [46]. The main pore types developed in the OL CM-1 shales are clay cleavage micro-fractures, which contribute to the slit-shaped pores, whereas the OR S-3 shales exhibit composite pore shapes and develop some ink-bottle-shaped pores, which are consistent with the primary pore types of organic matter and dissolution pores. In addition, the average N₂ adsorption amounts at the maximum relative pressure for OR S-3, OM M-2, and OL CM-1 shales are 16.64 cm³/g, 13.07 cm³/g, and 12.47 cm³/g, respectively.

4.2.4. PV and SA

The PV and SA of different shale lithofacies are summarized in

Table 2. Pore volumes (PVs) of the studied shale samples.

Sample ID	PVs							Lithofacies
	MIP (cm3/g)	Percentage (%)	MEP (cm3/g)	Percentage (%)	MAP (cm3/g)	Percentage (%)	TP (cm3/g)
JYA-1	0.00575	27.91	0.00800	38.81	0.00686	33.28	0.02062	OL CM-1
JYA-2	0.00665	28.04	0.00914	38.55	0.00792	33.41	0.02370	OL CM-1
JYA-3	0.00898	31.09	0.01400	48.46	0.00591	20.45	0.02889	OL CM-1
JYB-1	0.00415	23.09	0.00561	31.22	0.00821	45.69	0.01797	OL CM-1
JYB-2	0.00730	28.33	0.00895	34.73	0.00952	36.94	0.02577	OL CM-1
JYB-3	0.00838	34.56	0.01067	44.00	0.00520	21.44	0.02425	OL CM-1
JYA-4	0.00669	28.94	0.01037	44.86	0.00606	26.20	0.02312	OM M-2
JYA-5	0.00775	32.01	0.01058	43.66	0.00589	24.32	0.02422	OM M-2
JYA-6	0.00745	30.50	0.01015	41.55	0.00683	27.95	0.02442	OM M-2
JYB-4	0.00920	33.43	0.01165	42.33	0.00667	24.24	0.02752	OM M-2
JYB-5	0.00761	31.51	0.01012	41.90	0.00642	26.58	0.02415	OM M-2
JYB-6	0.00664	29.38	0.00849	37.57	0.00747	33.05	0.02260	OM M-2
JYA-7	0.01038	32.28	0.01417	44.07	0.00760	23.65	0.03216	OR S-3
JYA-8	0.00916	28.91	0.01319	41.63	0.00933	29.46	0.03168	OR S-3
JYA-9	0.01022	31.82	0.01443	44.93	0.00747	23.25	0.03211	OR S-3
JYB-7	0.00811	26.65	0.01158	38.05	0.01074	35.29	0.03043	OR S-3
JYB-8	0.00881	30.83	0.01373	48.04	0.00604	21.13	0.02858	OR S-3
JYB-9	0.00958	32.30	0.01482	49.97	0.00526	17.73	0.02966	OR S-3

and

Table 3. Pore surface areas (PSAs) of the studied shale samples.

Sample ID	PSAs							Lithofacies
	MIP (m2/g)	Percentage (%)	MEP (m2/g)	Percentage (%)	MAP (m2/g)	Percentage (%)	TP (m2/g)
JYA-1	16.683	75.60	5.149	23.33	0.235	1.06	22.068	OL CM-1
JYA-2	19.141	74.30	6.366	24.71	0.254	0.99	25.761	OL CM-1
JYA-3	25.572	71.84	9.803	27.54	0.222	0.62	35.597	OL CM-1
JYB-1	11.642	77.41	3.131	20.82	0.266	1.77	15.039	OL CM-1
JYB-2	20.196	76.65	5.873	22.29	0.278	1.06	26.347	OL CM-1
JYB-3	23.722	76.37	7.163	23.06	0.175	0.56	31.060	OL CM-1
JYA-4	18.885	72.62	6.902	26.54	0.218	0.84	26.006	OM M-2
JYA-5	22.415	75.69	6.979	23.57	0.222	0.75	29.616	OM M-2
JYA-6	21.084	75.28	6.696	23.91	0.229	0.82	28.009	OM M-2
JYB-4	26.581	76.45	7.967	22.91	0.223	0.64	34.771	OM M-2
JYB-5	22.032	75.66	6.874	23.60	0.215	0.74	29.121	OM M-2
JYB-6	19.184	76.17	5.769	22.90	0.234	0.93	25.187	OM M-2
JYA-7	29.400	74.75	9.663	24.57	0.267	0.68	39.330	OR S-3
JYA-8	25.602	74.08	8.654	25.04	0.306	0.89	34.562	OR S-3
JYA-9	30.060	75.38	9.540	23.92	0.279	0.70	39.879	OR S-3
JYB-7	23.542	74.16	7.906	24.90	0.297	0.94	31.745	OR S-3
JYB-8	26.435	74.54	8.818	24.87	0.210	0.59	35.463	OR S-3
JYB-9	27.062	75.97	8.375	23.51	0.186	0.52	35.623	OR S-3

and Figure 6a,b, and show obvious distinctions. The PV of OR S-3 shales is the largest (Figure 6a), with an average total pore (TP) V of 0.03077 cm³/g (Table 2). In comparison, the PV in OM M-2 shales is in the middle (Figure 6a), and the average TP V is 0.02434 cm³/g (Table 2), while OL CM-1 shales exhibit the smallest PV (Figure 6a), with an average TP V of 0.02353 cm³/g (Table 2). Similarly, OR S-3 shales possess the largest PSA (Figure 6b), and the average TP SA is 36.100 m²/g (Table 3). OM M-2 shales have the middle PSA (Figure 6b), with an average TP SA of 28.785 m²/g (Table 3), whereas OL CM-1 shales hold the smallest PSA (Figure 6b), and the average TP SA is 25.979 m²/g (Table 3).

The proportional distributions of PV and SA across MIPs, MEPs, and MAPs are depicted in Table 2 and Table 3 and Figure 6c,d. For OR S-3 shales, MEPs are the main contributor to PV, accounting for an average of 44.45%, followed by MIPs, with a mean value of 30.46%, while MAPs are the most undeveloped ones, with a percentage of 25.09% (Figure 6c, Table 2). The development pattern of the PV proportion in OM M-2 shales is similar to that in OR S-3 shales (Figure 6c), where the average percentage of PV for MIPs, MEPs, and MAPs is 30.96%, 41.98%, and 27.06%, respectively (Table 2). For OL CM-1 shales, although MEPs dominate the PV, different from OR S-3 and OM M-2 shales, the proportion of MAPs increases and exceeds that of MIPs (Figure 6c). Specifically, the average PV percentage of MIPs, MEPs, and MAPs is 28.84%, 39.29%, and 31.87%, respectively (Table 2). MIPs provide the main PSA and substantially exceed the contributions from MEPs and MAPs (Figure 6d). The average proportion of MIP SA in OR S-3, OM M-2, and OL CM-1 shales is 74.81%, 75.31%, and 75.36%, respectively (Table 3).

4.2.5. FD

The FHH regression analysis plots of lnV versus ln(ln(P₀/P)) from N₂ adsorption data of four representative shale samples are presented in Figure 7. As displayed in Figure 7, there are two distinct segments with different slopes at the P/P₀ intervals of 0–0.45 and 0.45–1. Two fractal dimensions (FDs), D₁ and D₂, were calculated from the two linear segments at the P/P₀ intervals of 0–0.45 and 0.45–1, respectively (

Table 4. FDs of the studied shale samples.

Sample ID	D1 (P/P0 < 0.45)	Coefficient (R2)	D2 (P/P0 > 0.45)	Coefficient (R2)	Lithofacies
JYA-1	2.5755	0.9992	2.7663	0.9978	OL CM-1
JYA-2	2.5948	0.9980	2.7980	0.9912	OL CM-1
JYA-3	2.5994	0.9964	2.8335	0.9934	OL CM-1
JYB-1	2.5339	0.9988	2.6786	0.9929	OL CM-1
JYB-2	2.6094	0.9973	2.7822	0.9902	OL CM-1
JYB-3	2.6122	0.9932	2.8300	0.9933	OL CM-1
JYA-4	2.5771	0.9967	2.8025	0.9979	OM M-2
JYA-5	2.6332	0.9924	2.8201	0.9981	OM M-2
JYA-6	2.6196	0.9949	2.8086	0.9989	OM M-2
JYB-4	2.6313	0.9917	2.8330	0.9974	OM M-2
JYB-5	2.6007	0.9954	2.8141	0.9984	OM M-2
JYB-6	2.6049	0.9958	2.8028	0.9935	OM M-2
JYA-7	2.6194	0.9922	2.8293	0.9959	OR S-3
JYA-8	2.6057	0.9922	2.8063	0.9953	OR S-3
JYA-9	2.6222	0.9917	2.8217	0.9964	OR S-3
JYB-7	2.5878	0.9928	2.8085	0.9884	OR S-3
JYB-8	2.6343	0.9879	2.8361	0.9825	OR S-3
JYB-9	2.6482	0.9881	2.8925	0.9573	OR S-3

). D₁ represents the pore surface roughness or irregularity, and D₂ reflects the pore structure complexity [11,13,14,15].

The OL CM-1 shales have the smallest values of FDs D₁ and D₂ (Figure 6e,f), ranging from 2.5339 to 2.6122 and 2.6786 to 2.8335, with an average of 2.5875 and 2.7814, respectively (Table 4). The FDs D₁ and D₂ of the OM M-2 shales are moderate and range from 2.5771 to 2.6332 and 2.8025 to 2.8330, with average values of 2.6111 and 2.8135, respectively (Table 4). However, the OR S-3 shales show the maximum FD values of D₁ and D₂ (Figure 6e,f), ranging from 2.5878 to 2.6482 (average 2.6196) and 2.8063 to 2.8925 (average 2.8324), respectively (Table 4). The highest values of FD suggest that the pores in the OR S-3 shales are the most heterogeneous.

4.3. Methane Adsorption Isotherms

Methane adsorption isotherms of the shale samples from various lithofacies are shown in Figure 8. Across the measured pressure range, the excess adsorption isotherms initially rise rapidly with the increasing pressure, reaching a maximum when the pressure is about 8–10 MPa, after which they begin to slowly decrease as the pressure continues to increase. It is obvious that “negative adsorption” appears at a high pressure (Figure 8a–c).

Although the adsorption isotherms of shales from different lithofacies are similar in shape, the MAC is pretty different. The excess adsorption isotherms are well fitted by the adapted Langmuir equation, with the correlation coefficients (R²) exceeding 0.99, and the derived Langmuir volumes (V_L) are presented in Figure 8. As can be seen, the average V_L is 1.76 cm³/g for OL CM-1 shales, compared to 2.76 cm³/g for OM M-2 shales and 4.65 cm³/g for OR S-3 shales. These results indicate that OR S-3 shales possess the largest MAC, whereas OL CM-1 shales have the smallest value.

In addition, the absolute adsorption isotherms are calculated and exhibited in Figure 8d–f. The absolute adsorption capacity of methane increases sharply at a low-pressure range (<6 MPa), followed by a slow increase at a high-pressure stage. Notably, the absolute adsorption amount is significantly higher than the excess adsorption amount, and this difference becomes more pronounced with increasing pressure.

5. Discussion

5.1. Controlling Factors of Pore Development

5.1.1. Effect of Organic Matter on Pore Development

The internal components of shale (both organic matter and mineral composition) serve as the fundamental factors that determine the variations in pore development characteristics in shales across different lithofacies.

TOC content exhibits a strong positive relationship with the volumes (Vs) of both MIPs and MEPs (Figure 9a,b). However, there is no clear correlation between TOC content and MAP V (Figure 9c). Similar to the case of PV, evidently positive relationships have been found between TOC content and MIP and MEP surface areas (SAs) (Figure 10a,b), but TOC content and MAP SA is poorly correlated (Figure 10c). These findings suggest that organic matter is the main contributor to the development of MIPs and MEPs [15,19], which aligns with the results of previous studies on Wufeng–Longmaxi shales [22,24,47]. A large number of bubble-like and elliptical organic pores are developed in organic matter after intensive thermal maturation and hydrocarbon expulsion, which are mainly of MIP and MEP scales (Figure 4g–i). These organic micro-mesopores are the main contributors to PV and SA development, as shown by the apparent positive relationships between TOC content and both TP V and SA (Figure 9d and Figure 10d).

In addition, Figure 11 shows the relationships between TOC content and FDs (both D₁ and D₂). Positive relationships clearly exist between the TOC content and FDs, indicating that higher organic matter content contributes to more irregular and complex pore networks. These positive relationships have also been confirmed by the Lower Cambrian Niutitang shales in the Sichuan Basin [48] and Upper Triassic Yanchang shales in the Ordos Basin [49]. Organic-rich shales have more heterogeneous organic micro-mesopores, likely resulting in larger FDs and more complicated pore networks [48,49].

5.1.2. Effect of Mineral Composition on Pore Development

Strong negative correlations are observed between total clay content and the Vs of MIPs and MEPs (Figure 12a,b), as well as the correlations between total clay content and the SAs of MIPs and MEPs (Figure 13a,b). Regarding MAPs, both PV and SA tend to decrease with an increase in total clay content. However, the situation changes when total clay content exceeds 50%, with the V and SA of MAPs increasing with the increase in total clay content (Figure 12c and Figure 13c). As indicated, clay minerals play a more substantial role in the development of MAPs instead of MIPs and MEPs. This observation is consistent with previous studies that clay mineral composition rather than organic matter serves as the main factor controlling MAP development [15,19,22]. Abundant interlayer cleavage micro-fractures developed in the clay-rich shales are the fundamental factor for the increase in MAP V and SA (Figure 4a–c). Nevertheless, due to the low compressive strength of clay minerals, shales with high clay mineral content lack the protective framework provided by rigid quartz particles (Table 1) and are more susceptible to being compacted during the diagenesis process, resulting in a reduction in or even closure of clay mineral pores [18,50]. In general, clay minerals have a negative effect on pore structure parameters (Figure 12d and Figure 13d).

Furthermore, Figure 14 represents an obvious negative relationship between clay content and FDs (both D₁ and D₂), implying that increased clay content reduces the heterogeneity of pores. The lower FD values of clay-rich shales are possibly due to the development of clay-associated macropores or microfractures and to the decrease in complex organic micro-mesopores [48,49].

5.1.3. Conceptual Model for Development of Pores in Different Shale Lithofacies

Based on the comparative analysis of pore types and quantitative evaluation of pore structures, an idealized conceptual model for the development of pores among shales of different lithofacies is proposed (Figure 15).

Figure 15a shows the pore development in OL CM-1 shales, which are the shales rich in clay minerals and poor in siliceous minerals and displaying the lowest TOC abundance (Table 1). The existence of organic pores is limited, with only a small number of pores appearing in the residual organic matter dispersed among clay particles. Instead, the interlayer cleavage micro-fractures between oriented clay platelets formed during clay diagenesis constitute the predominant pore type of OL CM-1 shales. These micro-fracture systems serve as channels connecting pores and macro-fractures, effectively improving the gas flowing capacity of shales [51,52]. Figure 15c illustrates the development of pores in OR S-3 shales. These shales feature a high content of siliceous minerals and low content of clay minerals and also present a high TOC abundance (Table 1). Unlike the dispersed distribution of organic matter in OL CM-1 shales, organic matter particles in OR S-3 shales are usually largely spongy. Abundant organic pores formed by the generation and expulsion of hydrocarbon fluids during the thermal evolution of kerogen [1,6,40] dominate the pore type in these organic-rich shales. A single organic matter sponge with micrometer scale can contain hundreds of pores. Meanwhile, these organic pores are well preserved against compaction due to the rigidity of siliceous minerals [24,53] and exhibit perfect roundness (Figure 4g–i and Figure 15c). Such organic pores form efficient pore networks [6,7,9,54], contributing to the total porosity and gas storage of organic-rich shales around the world [55,56,57,58]. Additionally, the dissolution pores produced by organic acidic fluids dissolving unstable minerals contribute another significant pore type for OR S-3 shales (Figure 15c). Figure 15b represents the development of pores in OM M-2 shales, which are the shales characterized by moderate values of clay and siliceous minerals and also by intermediate TOC abundance (Table 1). The coexistence of organic pores and clay cleavage micro-fractures developed within scattered organic matter and clay particles becomes the distinctive pore structure of OM M-2 shales, though in lesser abundance than that of corresponding pores in OR S-3 and OL CM-1 shales, respectively.

In terms of pore structures, the development of MIPs, MEPs and FDs is primarily controlled by the organic matter, while clay minerals play an important role in the development of MAPs. Consequently, OR S-3 shales have the largest TP and fractional pore (particularly MIP and MEP) Vs and SAs and the most complicated pore networks (Figure 6), attributed to having the highest TOC content. In contrast, OL CM-1 shales have a relatively developed MAP V but the smallest PSAs (Figure 6) due to having the highest clay mineral content and lowest TOC abundance.

5.2. Controlling Factors of MAC

5.2.1. Effect of Organic Matter on MAC

The MAC of shales with different lithofacies is closely related to both organic matter and inorganic composition, as well as the pore systems they controlled.

Organic matter contains numerous nanoscale organic pores, which have larger internal SAs and greater adsorption energies compared with large pores, and it is widely considered a critical factor controlling the MAC of shales [59,60]. As can be seen from Figure 16, methane adsorption parameters (both the Langmuir V_L and the maximum V_exc) are positively correlated with the TOC content. Similar correlations have been reported in many other shales, such as the Lower Cretaceous Fort St John shales and Buckinghorse shales [61,62], Lower Jurassic Gordondale shales [40], Upper Devonian Woodford shales [59], Lower Cambrian Niutitang shales, Middle Cambrian Arthur creek shales, Lower Silurian Longmaxi shales and Upper Permian Dalong shales [37,63], Upper Cambrian–Lower Ordovician Alum shales and Lower Toarcian Posidonia shales [34,36], and Upper Triassic Yanchang shales [64]. This widespread correlation is explained as the formation of a large number of organic micro-mesopores after the generation and expulsion of hydrocarbons at the mature to over-mature thermal stages of organic-rich shales (Figure 4g–i, Figure 9a,b and Figure 10a,b). These organic micro-mesopores provide abundant PSAs (Figure 10d), thereby increasing the sites for methane retention and enhancing the MAC of shales.

5.2.2. Effect of Mineral Composition on MAC

In general, in addition to organic matter, mineral composition—in particular, clay minerals such as montmorillonite, illite, illite–smectite mixed layer, kaolinite, and chlorite—have been reported to be associated with small pores with effective radii of 1–2 nm, supplying a certain number of adsorption sites and contributing to the MAC of shales [40,65,66]. To evaluate whether clay minerals contribute to the MAC of the studied shales, the Langmuir V_L and the maximum V_exc are plotted against the total clay content and displayed in Figure 17. Apparently, a negative correlation is identified between total clay content and MAC. This result suggests that the adsorption of clay minerals is very limited, despite clay minerals being the dominant component of the Longmaxi shales (Table 1). This finding contrasts with certain previous studies [67,68], in which the MAC of shales increases with total clay content. In fact, shales with higher clay content in the study area have a poorer MAC due to their concomitant lower TOC content (Table 1), which further supports the interpretation that organic matter is the primary factor controlling the MAC of these shales. This kind of manifestation was also discovered by Wang et al. (2013) [63] and Tian et al. (2016) [38]. Furthermore, the abundant interlayer cleavage micro-fractures developed in clay-rich shales (Figure 4a–c) are the main providers of MAPs (Figure 12c), while the contribution of MAPs to PSA is quite restricted (Figure 6b,d). In addition, the pores associated with clay minerals are likely destroyed by compaction due to the low mechanical strength of clay minerals [69,70]. All of the above reasons collectively account for the negative impact of clay minerals on the adsorption of shales in the study area.

5.2.3. Effect of Pore Structure on MAC

As discussed above, the influences of organic matter and mineral composition on shale MAC can be fundamentally attributed to the control of the pore system on the MAC of shales. Both the Langmuir V_L and maximum V_exc are positively correlated with the pore structure parameters of MIPs and MEPs (MIP V, MIP SA, MEP V, and MEP SA), but are not significantly correlated with the pore structure parameters of MAPs (MAP V and MAP SA) (Figure 18), indicating that the adsorption of methane molecules predominantly occurs in MIPs and MEPs. The adsorption process of gas in porous shale reservoirs reflects the interaction between the pore framework and gas molecules and is governed by the van der Waals force, which can be strengthened because of the proximity of the pore walls [71]. As a result, with the increased adsorption energies, the smaller pores have greater adsorption potential compared with the larger pores and could preferentially adsorb more gas. Additionally, smaller pores are the main contributor to the PSAs (Figure 6), and thus could provide substantially more adsorption sites for gas molecules.

Moreover, when the methane adsorption parameters (Langmuir V_L and Maximum V_exc) were plotted against the FDs (D₁ and D₂), positive correlations were found to exist between the MAC and FDs (Figure 18g,h). The higher values of FDs indicate more organic micro-mesopores with rough surfaces and complex structures developed in the shales (Figure 9, Figure 10 and Figure 11), at which more abundant active sites are available for methane adsorption. The coupling parameters constructed by the FD and PSA have superior potential to predict the MAC of shales.

From the comprehensive analysis presented above, it can be concluded that OR S-3 shales have the highest TOC content, well-developed heterogeneous organic micro-mesopores, and the largest PSA; consequently, they have the strongest MAC (Figure 19). In contrast, OL CM-1 shales, characterized by the lowest TOC content and the highest clay mineral content, have the worst methane adsorption performance (Figure 19).

6. Conclusions

This work comprehensively investigated the pore characteristics and MAC of different shale lithofacies from the Longmaxi Formation in Fuling Field, Sichuan Basin, using a series of related experimental methods, such as TOC content, XRD, FE-SEM, low-pressure CO₂/N₂ adsorption, and high-pressure methane adsorption. Positive correlations between TOC content and FDs (D₁ and D₂), as well as Vs and SAs of MIPs, MEPs, and TPs, suggest that heterogeneous organic micro-mesopores are the primary contributors to pore development. The cleavage micro-fractures developed in clay minerals are the main controlling factor for the development of MAPs. However, clay-rich shales are more susceptive to mechanical compaction because of the low siliceous mineral content, resulting in a reduction in or even closure of clay pores and a negative impact of clay minerals on the TP development. MAC increases with the increase in TOC content, FDs (D₁ and D₂), and micro- and mesopore Vs and SAs, while it decreases with the increase in clay mineral content, indicating that heterogeneous organic micro-mesopores function as the primary adsorption sites and promote the MAC of shales. OR S-3 shales have the most developed heterogeneous organic micro-mesopores, the largest PV, the highest PSA, and thus the strongest MAC. Considering the excellent storage space, sustainable gas production, and favorable brittleness to hydraulic fracturing, OR S-3 shales are the most promising reservoir for shale gas exploration and development.

More importantly, FDs can serve as effective bridges to establish the relationships between pore heterogeneity and gas adsorption behavior, thereby improving the prediction of the MAC of shales based solely on the TOC content. FDs are valuable parameters for evaluating the resource potential of shale reservoirs. However, unlike the relatively stable tectonic background of North America, marine shale formations in southern China have experienced multiple tectonic activities and complex structural deformation after their original deposition. Thus, discussing the comprehensive effects of external structural deformation and internal lithofacies on shale pore properties and methane adsorption behavior based on shale gas wells from varied structural settings is a potential topic of further study.