Classification of Pore-fracture Combination Types in Tectonic Coal Based on Mercury Intrusion Porosimetry and Nuclear Magnetic Resonance.

Differences in content, distribution, and connectivity of pores and fractures with different sizes in coal lead to different modes of gas migration. An accurate classification of pore-fracture combination types in coal can lay a foundation for studying gas migration. High-pressure mercury intrusion and nuclear magnetic resonance (NMR) experiments were conducted on coal samples collected from the Changping coal mine in Jincheng City, Shanxi Province, and Pingdingshan no. 4 mine in Pingdingshan City, Henan Province, China. The fractal dimensions of pores with different sizes were calculated using the Menger model. By combining the data with T2 spectra obtained by NMR, critical values for distinguishing diffusion pores from seepage pores-microfractures were determined. In addition, the main parameters affecting development of diffusion pores and seepage pores-microfractures and pore-fracture connectivity were analyzed, and a comprehensive evaluation index system for pores and fractures was established by selecting eight indices. Based on the method combining the analytical hierarchy process with multiparameter superposition, a method for determining critical values, establishing the evaluation index system, and classifying pore-fracture combination types was formed. The pore-fracture combination types in the test coal samples were classified according to the experimental data. The results indicate that the critical values for distinguishing diffusion pores from seepage pores-microfractures based on fractal dimensions obtained through mercury intrusion porosimetry and T2 spectra obtained by NMR are 72 nm and 2.5 ms, respectively. The studied coal samples can be classified into three combination types, separately characterized by high diffusivity and permeability and poor pore-fracture connectivity; low diffusivity, high permeability, and good pore-fracture connectivity; and low diffusivity and permeability and good pore-fracture connectivity. In the coal samples from the Changping coal mine, diffusion pores and seepage pores-microfractures are developed, while the connectivity between pores and fractures is poor. The coal samples from Pingdingshan no. 4 mine have undeveloped diffusion pores and seepage pores-microfractures but good connectivity between pores and fractures. The research results provide a method for classifying pore-fracture combination types in coal samples taken from different regions.

Introduction

Coal is a multiscale pore–fracture medium. The content, distribution, and connectivity of pores and fractures with different sizes are different, which leads to different modes of gas migration. Accurately classifying pore–fracture combination types in coal can lay a foundation for studying gas migration. Scholars have characterized pore–fracture structures with different sizes by using various methods, such as high-pressure mercury intrusion porosimetry (MIP), low-temperature liquid nitrogen adsorption, nuclear magnetic resonance (NMR), and scanning electron microscopy (SEM). For high-pressure MIP, by analyzing types of mercury injection and ejection curves and combining the data with fractal theory,[1] critical values for distinguishing diffusion from seepage pores are separately determined and connectivity between pores and fractures measuring more than 7.2 nm is evaluated.[2−5] The liquid nitrogen adsorption method describes and characterizes the distribution of pores with a size of 2–100 nm and adsorption characteristics of pores to nitrogen by analyzing morphologies of hysteresis loops on adsorption and desorption curves and calculating fractal dimensions.[6−9] As for the NMR method, by measuring 1H signals of fluid in rock and coal and plotting T2 spectra, the types of pore structures in coal were classified. It is generally considered that three spectral peaks with relaxation times at 0.5–2.5, 20–50 ms, and greater than 100 ms separately represent micropores, mesopores, and macropores and microfractures.[10−13] The SEM observation method can be used to directly observe the morphologies of nano-/micropores on the surfaces of coal samples[14,15] but cannot characterize the three-dimensional (3-D) structure of coal.

To overcome the deficiency of the single test method in the pore size measurement range and characterization, some researchers have modified and superimposed the test results obtained by various methods and jointly characterized the pores with different sizes in coal.[16−19] They thereby focus on the characterization of pores and fractures with different sizes but rarely classify the combinations thereof. By collecting the coal samples from the Changping coal mine in Shanxi province and Pingdingshan no. 4 mine in Henan province, high-pressure MIP and NMR experiments were conducted and an evaluation index system established for diffusion pores and seepage pores–microfractures. Moreover, a three-step method for classifying the pore–fracture combination types in coal was proposed, thus providing a method for classifying the pore–fracture combination types in coal in different regions.

Research Methods

There are diffusion pores and seepage pores–microfractures in coal, and gas migrates in pores and fractures with different sizes in different modes. Gas in diffusion pores mainly migrates by diffusion, while gas in seepage pores–microfractures mainly migrates by non-Darcy flow with a starting pressure gradient or Darcy flow.[20,21] To classify pore–fracture combination types, we first conducted NMR and high-pressure MIP tests on the same coal sample. The critical values for distinguishing diffusion pores from seepage pores–microfractures were determined based on T2 spectra yielded by NMR and fractal dimensions obtained by MIP. Furthermore, the valuation index system for pores and fractures with different sizes was built and the pore–fracture combination types were classified.

Determination Methods for Critical Values for Diffusion Pores and Seepage Pores–Microfractures

NMR-based Determination Method for the Critical Value for Distinguishing Diffusion Pores from Seepage Pores–Microfractures

The coal samples were saturated for 12 h with water and then tested by NMR to obtain T2 spectral curves under different pore sizes in coal. The morphologies of T2 spectral curves reflect the distribution of pores and fractures with different sizes. The smaller the T2 value, the smaller the corresponding pore size. The area enclosed by peak values represents the proportion of pores of such a size, and the width of each peak indicates the quality of sorting of such pores and fractures. Moreover, the number of peaks denotes the degree of connectivity between pores and fractures of different sizes. Theoretically, there should be three peaks in a T2 spectrum obtained by NMR on the coal samples, separately corresponding to diffusion pores, seepage pores, and microfractures.[11] Based on T2 values, the critical value for distinguishing diffusion pores from seepage pores–microfractures could be determined.

MIP-based Determination Method for the Critical Value for Diffusion Pores and Seepage Pores–Microfractures

The critical values for distinguishing diffusion pores from seepage pores–microfractures in the coal samples can be classified by differences in the fractal dimensions of pores of different sizes. When testing pores with the MIP, the relationship between pore size and pressure satisfies the Washburn equation,[22] namelywhere, p(r) and r indicate the applied pressure (MPa) and pore radius (nm) of coal, respectively; δ represents the surface tension of mercury (480 dyne/cm); θ denotes the contact angle between mercury and the solid surface (140°).

By substituting these parameters into Formula , the following formula can be obtained

Differentiating both sides of Formula , the following formula can be obtainedThat is

In the experiment, the total pore volume under a given pressure is equal to the mercury volume injected into the pore, namelywhere, V and V( represents the total pore volume under a given pressure (cm3/g) and the mercury volume injected into the pore (cm3/g), respectively.

The idea of Menger sponge can be used to characterize the pore structure of coal,[23] namely

By simultaneous use of Formulae , 4–6, the following formula can be obtained

By taking logarithms on both sides of Formula , the following formula can be obtainedwhere, D represents the fractal dimension of pores in the coal samples.

By plotting log[dV(/dP(r)] and log P(r), the slope K can be obtained, giving the fractal dimension as follows

By using this method, the relationship between log[dV(/dP(r)] and log P(r) can be deduced. This indicates that the pore structure in the coal samples changes abruptly at a certain inflection point. The research demonstrates that the mode of gas migration also changes.[1] The pore radius corresponding to the inflection point of the two line segments is taken as the critical value for distinguishing diffusion pores from seepage pores–microfractures in coal.

Establishment of the Evaluation Index System

The difficulty of gas migration and production is jointly determined by development degrees and connectivity of diffusion pores and seepage pores–microfractures. The main parameters affecting the development of diffusion pores and seepage pores–microfractures and pore–fracture connectivity were analyzed, and we established the associated evaluation index system.

Selection of Parameters for Evaluating the Degree of Development of Diffusion Pores and Seepage Pores–Microfractures

After determining the critical values for distinguishing diffusion pores from seepage pores, the porous volumes of diffusion pores and seepage pores–microfractures can be separately obtained while performing MIP tests on coal samples. The larger the volume of diffusion pores (or seepage pores–microfractures), the more numerous the corresponding pores in the coal samples, indicating that diffusion pores (or seepage pores–microfractures) in the coal samples are more developed. Therefore, the porous volume becomes an important parameter with which the degree of development of diffusion pores can be evaluated (or seepage pores–microfractures). The mercury ejection efficiency can be calculated based on mercury injection and ejection curves in the MIP test. The lower the mercury ejection efficiency is, the wider the hysteresis loop between the mercury injection and ejection curves and the more numerous the open pores; otherwise, there are more semiclosed pores. The fractal dimension reflects the complexity of pore size distribution. The smaller the fractal dimension, the simpler the pore size distribution. By separately calculating mercury ejection efficiencies and fractal dimensions of diffusion pores and seepage pores–microfractures, the structural characteristics and complexity of diffusion pores and seepage pores–microfractures can be characterized.

When measuring T2 spectra of the coal samples by NMR, after determining the critical relaxation time of diffusion pores and seepage pores–microfractures, their percentages of porous volumes can be determined according to the morphology of the T2 spectra. Moreover, the relative amounts of diffusion pores and seepage pores–microfractures can be characterized, thus revealing the degree of development of diffusion pores (or seepage pores–microfractures).

Selection of Parameters for Evaluating Connectivity between Pores and Fractures

According to distribution of pores and fractures with different sizes, the uniformity of pore–fracture distribution in the coal samples can be evaluated. When other conditions are the same, the more uniform the pore–fracture distribution, the higher the connectivity. A characteristic structural coefficient is the reciprocal of the product of the coefficient of variation of pores and fractures and structural coefficient, and the smaller its value, the poor the pore–fracture structure and connectivity in the coal samples. Mercury saturation is the ratio of the maximum amount of mercury injected in the coal samples per unit mass to the total porosity. The larger the value, the more the connected pores. The more continuous the spectral peaks in morphologies of T2 spectra obtained by NMR, the better the connectivity of diffusion pores, seepage pores, and fractures in the coal samples. Therefore, the distribution uniformity, characteristic structural coefficient, mercury saturation, and morphology of T2 spectra can be regarded as important parameters for evaluating connectivity between pores and fractures.

In accordance with the above parameters, the parameters including porous volume, percentage of porous volume, mercury ejection efficiency, and fractal dimension of pores were selected to evaluate the degree of development of diffusion pores and seepage pores–microfractures. Parameters, such as pore–fracture distribution uniformity, characteristic structural coefficient, mercury saturation, and the type of T2 spectra were selected for comprehensive evaluation of the connectivity between pores and fractures. Thus, a comprehensive evaluation index system for diffusion pores and seepage pores–microfractures was established, as shown in Figure .

Figure 1

Comprehensive evaluation index system for diffusion pores and seepage pores–microfractures.

Classification Method for the Pore–Fracture Combination Types

After establishing the evaluation index system, based on the high-pressure MIP and NMR test results, thresholds of the parameters were determined by combining with other research results and using the analytical hierarchy process. Different thresholds were scored and the pore–fracture combination types were finally evaluated. The classification method for the pore–fracture combination types is shown in Figure .

Figure 2

Classification method of pore–fracture combination types.

Results and Discussion

Determination of the Critical Values for Distinguishing Diffusion Pores from Seepage Pores–Microfractures

Classification of Diffusion Pores and Seepage Pores–Microfractures Based on NMR

Based on the NMR test data, T2 spectra of the coal samples were plotted. By combining the spectra with the correspondence between peaks and pore–fracture systems in different positions, diffusion pores and seepage pores–microfractures were classified (Figure ).

Figure 3

Critical values of diffusion pores and seepage pores–microfractures of each coal sample based on the T2 spectrum.

As shown in Figure , T2 spectra of different coal samples are different, but the positions of peaks on each spectrum are similar. The T2 spectrum corresponding to diffusion pores is generally in the range of 0.05–2.5 ms, while that corresponding to seepage pores–microfractures is generally larger than 2.5 ms, which matches previous research results.[8] Therefore, in the following comprehensive evaluation of pore–fracture systems, 2.5 ms is taken as the critical value for distinguishing diffusion pores from seepage pores–microfractures. Furthermore, the spectral area enclosed by the spectral peaks and changes in morphology of T2 spectra were used to characterize the percentage of porous volumes of diffusion pores, seepage pores, and microfractures and the connectivity between pores and fractures.

Determination of the Critical Value for Distinguishing Diffusion Pores from Seepage Pores–Microfractures Based on the MIP

For the eight test samples in the MIP experiment, ln p and ln(Dv/dp) were fitted (Figure ).

Figure 4

Fractal dimension calculation results of diffusion pores and seepage pores of each coal sample.

As shown in Figure , the critical values for distinguishing diffusion pores from seepage pores–microfractures in different coal samples were determined. To be specific, the critical values in coal samples C1-1, C1-2, and C1-3 are 67–72 nm, while those in coal samples C2-1, C2-2, and C2-3 range from 60 nm to 75 nm. Moreover, the critical values in samples P-1 and P-2 are between 66 and 89 nm. Although the critical values for distinguishing diffusion pores from seepage pores in the test coal samples are slightly different, the overall change is small. The average, 72 nm, is taken as the critical value for distinguishing diffusion pores from seepage pores–microfractures in the studied coal samples.

Comprehensive Evaluation of Pores and Fractures

Evaluation Results of Development Characteristics of Diffusion Pores and Seepage Pores–Microfractures

Porous Volumes of Diffusion Pores and Seepage Pores–Microfractures

Based on the MIP test data and Formula , porous volumes of diffusion pores and seepage pores–microfractures were calculated by combining with the critical values for distinguishing diffusion pores from pores–microfractures determined by fractal dimensions with the MIP, that iswhere, Vk and Vs represent the porous volumes (cm3/g) of diffusion pores and seepage pores, respectively; V and m separately represent the porous volume (cm3) when the pore throat radius is r and mass (g) of the coal sample.

The calculated porous volumes of diffusion pores and seepage pores–microfractures in the coal samples are listed in Table .

Table 1

Calculation Results of the Evaluation Parameters of Diffusive Pores and Seepage Pores–Microfracturesa

	volumes of pores (cm3/g)		percentages of porous volumes (%)		mercury ejection efficiency (%)		fractal dimensions
sample number	A	B	A	B	A	B	A	B
C1-1	0.0130	0.0272	98.361	1.639	70.769	4.412	3.697	2.457
C1-2	0.0136	0.0323	95.123	4.877	66.176	8.359	3.532	2.428
C1-3	0.0130	0.0296	99.207	0.793	70.769	4.392	3.650	2.412
C2-1	0.0147	0.0410	97.242	2.758	61.905	7.561	3.603	2.527
C2-2	0.0112	0.0336	89.361	10.639	74.107	4.167	3.581	2.367
C2-3	0.0117	0.0369	98.174	1.826	57.265	6.775	3.467	2.405
P-1	0.0152	0.0013	28.979	71.021	92.291	73.917	3.874	2.820
P-2	0.0162	0.0014	28.697	71.303	85.585	84.997	3.760	2.800

In the table, A represents the diffusion pores and B represents the seepage pores–microfractures.

Percentages of Porous Volumes of Diffusion Pores and Seepage Pores–Microfractures

The area enclosed by different peaks of pores and fractures on T2 spectral curves represents the proportion of such pores or fractures, from which, their percentage porous volumes arewhere, Vi denotes the percentage of porous volume; Mi and Mt are the spectral area corresponding to diffusion pores or seepage pores–microfractures in the T2 spectrum and the total spectral area, respectively.

The calculation results of percentages of porous volumes of diffusion pores and seepage pores–microfractures in the coal samples are summarized in Table .

Mercury Ejection Efficiency

The maximum mercury injection and ejection during the MIP test were obtained, and the mercury ejection efficiency is given bywhere, η, v2, and v1 represent the mercury ejection efficiency, the mercury ejection (cm3/g), and the maximum mercury injection (cm3/g), respectively.

The calculated mercury ejection efficiencies of various coal samples are listed in Table .

Fractal Dimensions of Diffusion Pores and Seepage Pores–Microfractures

In accordance with Formula and mercury injection data under different injection pressures, the fractal dimensions of diffusion pores and seepage pores in the test samples were calculated (Figure ). Seepage pores–microfractures with large pore sizes have typically fractal features, and the fractal dimension varies between 2 and 3. Moreover, the degree of fitting exceeds 90%. The degree of fitting for smaller diffusion pores is relatively low and the fractal dimension is between 3 and 4. The reason for this is that the coal matrix is compressed and deformed under high-pressure mercury injection, so the fractal dimension exceeds 3, but it is still an effective index for characterizing pore characteristics.[22,24] The calculated fractal dimensions of diffusion pores and seepage pores–microfractures in various coal samples are listed in Table .

Evaluation Results of Connectivity Between Diffusion Pores and Seepage Pores–Microfractures

Calculation Results of Pore–fracture Distribution Uniformity

Volumes of pores of different sizes as measured by MIP change with the pore-throat radius. Pore distributions in the coal samples can be classified into three morphologies, namely, unimodal, crescent, and stepped shapes, as shown in Figure .

Figure 5

Pore distribution classes in the coal samples.

In coal samples with a crescent-shaped pore distribution, both diffusion pores and microfractures are developed, while seepage pores are not developed. The peak value of diffusion pores is more than 1/3 of the microfractures, and the uniformity of the pore–fracture distribution is medium. Developed microfractures and fewer diffusion pores and seepage pores are found in coal samples with a unimodal pore distribution. The peak value of diffusion pores is less than 1/3 of the microfractures, and the uniformity of the pore–fracture distribution is the lowest. In the coal samples with the stepped pore distribution, diffusion pores are the most developed, followed by seepage pores, while microfractures are the least developed. The overall pore–fracture distribution is uniform.

Characteristic Structural Coefficient

The characteristic structural coefficient is the reciprocal of the product of the coefficient of variation Dr and structural coefficient φp. The coefficient of variation (relative sorting coefficient) Dr can be expressed as followswhere, Dr indicates the coefficient of variation and is dimensionless; Sp and r represent the sorting coefficient and pore–throat radius (μm) at a certain point, respectively; r̅ and ΔS denote the average pore–throat radius (μm) and mercury saturation (%) corresponding to r in a certain interval, respectively.

The structural coefficient φp can be expressed as followswhere, φp, φ, and K denote the structural coefficient (dimensionless), porosity (%), and air permeability (μm2), respectively.

The characteristic structural coefficient Csc can be expressed aswhere, Csc denotes the characteristic structural coefficient.

The calculated characteristic structural coefficients of various coal samples are listed in Table .

Table 2

Calculation Results of Connectivity Parameters of Diffusion Pores and Seepage Pores–Microfractures

sample number	distribution uniformity	coefficient of variation	structural coefficient	characteristic structural coefficient	maximum mercury injection (cm3/g)	density (g/cm3)	porosity (%)	mercury saturation (%)	types of T2 spectra
C1-1	crescent-shaped	59.574	2.928	0.006	0.0402	1.46	7.01	83.762	unimodal
C1-2	crescent-shaped	64.726	1.033	0.015	0.0459	1.42	7.44	87.644	discontinuous bimodal
C1-3	crescent-shaped	75.707	1.684	0.012	0.0426	1.48	7.39	85.343	unimodal
C2-1	unimodal	88.731	0.038	0.297	0.0557	1.39	8.38	92.413	unimodal
C2-2	unimodal	86.344	0.019	0.610	0.0448	1.41	7.32	86.279	discontinuous bimodal
C2-3	unimodal	83.601	0.011	1.087	0.0486	1.43	7.75	89.625	unimodal
P-1	stepped	94.176	0.298	0.036	0.0165	1.35	2.61	85.487	continuous bimodal
P-2	stepped	87.748	0.021	0.543	0.0176	1.33	2.74	85.281	continuous bimodal

Mercury Saturation

The maximum mercury injection and porosity during the MIP test were obtained, and the mercury injection saturation was calculated thuswhere, C, v1, and ρ denote the mercury injection saturation (%), the maximum mercury injection (cm3/g), and density (g/cm3) of coal, respectively.

The calculated mercury saturations of various coal samples are listed in Table .

Types of T2 Spectra

The width of the NMR spectral peak indicates the quality for sorting of a certain type of pores, and the number of peaks represents the connectivity of pores of different sizes in coal and rock. According to the distribution morphologies of NMR spectral peaks of the studied coal samples, the T2 spectra of the coal samples were classified into three types: unimodal, discontinuous bimodal, and continuous bimodal (Figure ). The classification of the types of T2 spectra of the coal samples is summarized in Table .

Comprehensive Evaluation Index System for Diffusion Pores and Seepage Pores–Microfractures

The weights of the parameters when evaluating the development of diffusion pores and seepage pores–microfractures could be determined by combining with the principle for obtaining the parameters by utilizing the fuzzy comprehensive evaluation method. According to the distribution characteristics of the parameters of the studied coal samples and development characteristics of pore–fracture structures in the other areas,[25] the parameters were classified into good, medium, and poor ones, to which scores 3, 2, and 1 were assigned. The established comprehensive evaluation index system for diffusion pores and seepage pores–microfractures in tectonic coal is demonstrated in Table .

Table 3

Comprehensive Evaluation Index System for Diffusion Pores and Seepage Pores–Microfractures

			evaluation grades and scores
evaluation types	evaluation indexes	weight	good (3)	medium (2)	poor (1)	evaluation principle
the development of diffusion pores	volumes of pores (cm3/g)	0.30	>0.015	0.013–0.015	<0.013	according to the scores and weights of each parameter, the total score of no less than 2 points is high diffusion/high seepage/good connectivity, scores less than 2 points indicate low diffusion/low seepage/poor connectivity
	percentages of porous volumes (%)	0.25	>90	50–90	<50
	mercury ejection efficiency (%)	0.30	<70%	70–80%	<80%
	fractal dimensions	0.15	<3.5	3.5–3.7	>3.7
the development of seepage pores–microfractures	volumes of pores(cm3/g)	0.40	>0.03	0.01–0.03	<0.01
	percentages of porous volumes (%)	0.10	>50	10–50	<10
	mercury ejection efficiency (%)	0.25	<50%	50–70%	<70%
	fractal dimensions	0.25	<2.4	2.4–2.6	>2.6
connectivity between pores and fractures	distribution uniformity	0.20	stepped	crescent-shaped	unimodal
	characteristic structural coefficient	0.20	>0.5	0.1–0.5	<0.1
	mercury saturation(%)	0.30	>90	85–90%	<85%
	types of T2 spectra	0.30	continuous bimodal	discontinuous bimodal	unimodal

Classification of the Pore–fracture Combination Types in Tectonic Coal

Based on the established comprehensive evaluation index system for diffusion pores and seepage pores–microfractures in tectonic coal and the parameters of the coal samples tested by NMR and high-pressure MIP tests, the pore–fracture combination types of the studied coal samples were classified, as displayed in Table .

Table 4

Classification Results of the Combination Types of Diffusion Pores and Seepage Pores–Microfracturesa

sample number		evaluation indexes	C1-1	C1-2	C1-3	C2-1	C2-2	C2-3	P-1	P-2
evaluation types	the development of diffusion pores	volumes of pores (cm3/g)	0.0130	0.0136	0.0130	0.0147	0.0112	0.0117	0.0152	0.0162
		percentages of porous volumes (%)	98.361	95.123	99.207	97.242	89.361	98.174	28.979	28.697
		mercury ejection efficiency (%)	70.7692	66.1765	70.7692	61.9048	74.1071	57.265	92.2909	85.5848
		fractal dimensions	3.6970	3.5323	3.6497	3.6033	3.5807	3.4666	3.8739	3.7595
	the development of seepage pores–microfractures	volumes of pores (cm3/g)	0.0272	0.0323	0.0296	0.0410	0.0336	0.0369	0.0013	0.0014
		percentages of porous volumes (%)	1.639	4.877	0.793	2.758	10.639	1.826	71.021	71.303
		mercury ejection efficiency (%)	4.4118	8.3591	4.3919	7.561	4.1667	6.7751	73.9171	84.9969
		fractal dimensions	2.4570	2.4277	2.4118	2.5265	2.3666	2.4045	2.8203	2.7995
	connectivity between pores and fractures	distribution uniformity	crescent-shaped	crescent-shaped	crescent-shaped	unimodal	unimodal	unimodal	stepped	stepped
		characteristic	0.006	0.015	0.012	0.297	0.610	1.087	0.036	0.543
		structural coefficient	83.762	87.644	85.343	92.413	86.279	89.625	85.487	85.281
		mercury saturation (%)	unimodal	discontinuous bimodal	unimodal	unimodal	discontinuous bimodal	unimodal	continuous bimodal	xontinuous bimodal
evaluation results	types of diffusion pores		high diffusion	high diffusion	high diffusion	high diffusion	low diffusion	high diffusion	low diffusion	low diffusion
	types of seepage pores–microfractures		high seepage	high seepage	high seepage	high seepage	high seepage	high seepage	low seepage	low seepage
	types of connectivity		poor connectivity	poor connectivity	poor connectivity	poor connectivity	good connectivity	poor connectivity	good connectivity	good connectivity
	comprehensive evaluation results		X	X	X	X	Y	X	Z	Z

In the table, X represents the pore–fracture combination type of high diffusivity and permeability and poor pore–fracture connectivity, Y represents the pore–fracture combination type of low diffusivity, high permeability, and good pore–fracture connectivity, and Z represents the pore–fracture combination type of low diffusivity and permeability and good pore–fracture connectivity.

There are three pore–fracture combination types in the studied coal samples with following characteristics:

High Diffusivity and Permeability and Poor Pore–Fracture Connectivity

In the coal samples, large diffusion pores, and seepage pores, or microfractures are developed, while the lack of seepage pores of medium sizes leads to poor connectivity between pores and fractures of different sizes. A reservoir with such a pore–fracture combination type has a large porosity but average permeability, and gas migration is difficult.

Low Diffusivity, High Permeability, and Good Pore–Fracture Connectivity

In the coal samples, diffusion pores are undeveloped, while seepage pores–microfractures are well developed. Pores and fractures with different sizes are well connected. Gas production in a reservoir with such a pore–fracture combination type is mainly limited by diffusion velocity.

Low Diffusivity and Permeability and Good Pore–Fracture Connectivity

Although diffusion pores and seepage pores–microfractures in the coal samples have good connectivity, both of them are undeveloped. A reservoir with such a pore–fracture combination type shows poor overall permeability, and gas migration is difficult.

Conclusions

Based on the NMR test and MIP experiment, the parameters including porous volume, percentage of porous volume, mercury ejection efficiency, and fractal dimension of pores were selected to evaluate the degree of development of diffusion pores and seepage pores–microfractures. Parameters, such as pore–fracture distribution uniformity, characteristic structural coefficient, mercury saturation, and type of T2 spectra were selected for comprehensive evaluation of the connectivity between pores and fractures. Thus, a comprehensive evaluation index system for diffusion pores and seepage pores–microfractures was established.

A three-step method for determining critical values, establishing the evaluation index system, and classifying pore–fracture combination types was formed for classifying the pore–fracture combination types in coal; the study has important guiding significance for the difference of gas migration under different combination types.

The studied coal samples can be divided into three types by using the method of pore–fracture combination-type division, namely, the combination type with high diffusivity and permeability and poor pore–fracture connectivity, the combination type with low diffusivity, high permeability, and good pore–fracture connectivity and the combination type with low diffusivity and permeability and good pore–fracture connectivity, which lays a theoretical foundation for revealing the difference of gas migration in the coal.

Experimental Section

Experimental Samples

The coal samples used in the experiments were taken from the Changping coal mine, Jincheng city, Shanxi province, and Pingdingshan no. 4 mine, Pingdingshan city, Henan province, China. Blocks measuring 250 mm × 250 mm × 150 mm were collected from the Changping coal mine at positions with distances from the fault less than 300 m and more than 300 m and Pingdingshan no. 4 mine (one from each site). They were numbered C1 (there were 2–9 face cleats and 1–4 butt cleats, per 50 mm length of the coal sample, all isolated), C2 (there were 3–13 face cleats and 2–9 butt cleats, per 50 mm length of the coal sample, reticular and isolated combined) and P (there were 2–12 face cleats and 1–7 butt cleats, per 50 mm length of the coal sample, all isolated). Three, three, and two coal columns measuring Φ 20 mm × 20 mm were separately drilled in blocks C1, C2, and P, which were numbered: C1-1, C1-2, C1-3, C2-1, C2-2, C2-3, P-1, and P-2.

Equipment

A Meso MR23-060H-I low-field NMR instrument with a resonant frequency of 21.67568 MHz in Henan Polytechnic University (Jiaozuo City, Henan Province, China) was used. For the instrument, the magnet temperature was maintained at 32 ± 0.1 °C and the magnetic field intensity was 0.5 T. An Auto Pore IV 9505 automatic mercury porosimeter was used in the high-pressure MIP experiment. The instrument was suitable for measuring pore size distribution, total porous volume, total specific surface area of pores, and the pore size in the range of 5.5 nm–360 μm.