1 Introduction
Coal pores are spaces in the coal matrix that are not filled with solids; this makes them a type of complex porous media (
Meyers, 1982;
Fu et al., 2007;
Liu et al., 2020a). Coalbed methane is usually stored in coal pores in an absorbed state. It is important to analyze the pore characteristics of coal which forms a vital basis for analyzing the reserves, flowability, and recoverability of CBM (
Fu et al., 2017;
Liu et al., 2017;
Wang et al., 2017;
Liu et al., 2018). The pore structure of coal affects pore size, reservoir connectivity, and the adsorption, diffusion, and penetration of CBM in coal reservoirs and can, to some extent, further influence its recoverability (
Clarkson and Bustin, 1999;
Yao and Liu, 2012;
Wang et al., 2019). Structure properties, coal roof lithology and hydrogeology conditions have important influences on gas accumulation, preservation and enrichment (
Meng et al., 2014). The reservoir pressure is a key factor of CBM productivity (
Meng et al., 2020). The effect of coal metamorphism on the occurrence and enrichment of CBM has been the focus of several studies (
Cai et al., 2016;
Li et al., 2016,
2017;
Zhao et al., 2016). The impact of coal deformation on the recoverability of CBM is also of importance since coal deformation influences coal porosity and permeability (
Zhu et al., 2016;
Fu et al., 2017;
Hou et al., 2017;
Zhao et al., 2017a). As a common component of low-rank coal seams, ash has a considerable impact on the pore structure of coal seams. Ash in coal originates from residues resulting from the complete combustion of minerals in coal and is an important parameter for the assessment of coal quality (
Zhou et al., 2011). A large volume of reports and studies exist on the pore structure of coal seams. Of these, a few focus on the impact of ash on pore structure and use analytical tools like low-temperature nitrogen adsorption, NMR, and argon ion polishing and imaging. The data suggests that in the case of low ash content, ash particles generally fill in pore throats and micropores, resulting in a decrease in fractal dimensions and an increase in heterogeneity (
Zhang et al., 2019a;
Wang, 2020;
Jia et al., 2021). As compared with the samples which are not de-ashed, the de-ashed samples feature increased pore quantity, pore volume, pore-specific surface area, pore surface fractal dimensions, percentage of micropores, and total pore quantity; they also have a decreased mean pore diameter (
Song et al., 2014). For example, the lignite and long flame coals with medium ash content and high content of volatiles in the Bayanhua sag in the Erlian Basin show considerable resource potential (
Yao et al., 2020)
Yao et al. (2020) discovered that lignite and long flame coals with medium ash content and high volatile content have mainly developed in the Bayanhua sag in the Erlian Basin, showing considerable resource potential. Based on this, they further summarized the key geological elements and resource potential assessment methods of hugely-thick low-rank coal reservoirs. Previous studies merely incorporated a brief analysis of the ash yield as one of the factors. They lack a more comprehensive, systematic study on the impact of the ash yield on the porosity of pores and fissures.
Theoretical and experimental methods, including the fluid injection method, the spectrum testing method, and the image analysis method are applied in China and around the world to target the pore structure characteristics of coal (
Li, 2018). Among them, the commonly used fluid injection method includes mercury injection, gas adsorption (N
2 and CO
2), and low-field NMR. In general, the CO
2 (or low-temperature N
2) adsorption method, combined with mercury injection, is mostly employed to quantitatively characterize the pores and fissures in coal reservoirs, to explore the distribution characteristics of pore specific surface area, pore volume, and pore diameter, and to study the fractal characteristics of pores and fissures of different sizes based on different mathematical models (
Yao et al., 2008, 2009a;
Jiang et al., 2011;
Clarkson et al., 2013;
Liu et al., 2020a;
Yang et al., 2020).
Yao et al. (2010),
Yao and Liu (2012) first established the conversion relationship between T2 relaxation time and the porosity and pore-diameter distribution of coal reservoirs by analyzing NMR relaxation time. These results proved the feasibility and effectiveness of the application of NMR in characterizing pores and fissures of coal reservoirs (
Zheng et al., 2018;
Kang et al., 2019;
Zhang et al., 2019b). Despite its wide application in characterizing pores and fissures of coal reservoirs, the fluid injection method can only be used to test open pores. The limitations of experimental methods do not allow the characterization of closed pores. In recent years, multiple non-destructive and highly-efficient non-fluid-injection testing techniques have been gradually applied to the quantitative characterization of pores and fissures in coal reservoirs, such as small-angle neutron scattering (SANS), small-angle X-ray scattering (SAXS), focused ion beam-scanning electron microscope (FIB-SEM), and micro-computed tomography (
μ-CT). Owing to the low requirements on coal samples, the SAXS and SANS techniques are suitable for fragile samples that are tough to acquire through coring; these methods are also nondestructive to samples during testing (
Sakurovs et al., 2012;
Clarkson et al., 2013;
Cai et al., 2014;
Zhao et al., 2014;
Okolo et al., 2015;
Liu et al., 2020b). FIB-SEM and
μ-CT technology allow continuous scanning and 3D reconstruction, which are conducive to the study of pores and fissures in coal reservoirs. They effectively reveal the 3D continuous changes of the pores and fissures, and can also reflect the characteristics of strong heterogeneity within the pores and fissures (
Watanabe et al., 2011;
Golab et al., 2013;
Jing et al., 2016,
2017;
Li et al., 2017;
Liu et al., 2017;
Zhou et al., 2017;
Jiu et al., 2021;
Wang and Mao, 2021). For the image analysis technique, the pores and fissures in coal reservoirs are observed under a microscope, followed by image acquisition and the analysis of the size, shape, connectivity, and surface morphology characteristics of pores and fissures. Image analysis techniques mainly include scanning electron microscopy (SEM), field emission scanning electron microscopy (FESEM), transmission electron microscopy (TEM), and atomic force microscopy (AFM) (
Harris and Yust, 1976;
Li et al., 2012;
Liu et al., 2014,
2021;
Nie et al., 2015;
Zhao et al., 2017b;
Huang and Liu, 2019). Individual methods differ in the scale of pore characterization due to different testing principles and media. Since pores in coal reservoirs cannot be completely characterized using a single method, it is necessary to combine the aforementioned testing methods to comprehensively characterize the pore–fissure system and to study the degree of coal metamorphism and pore characteristics.
The Erlian Basin is characterized by widely developed coal seams, ubiquitous gas logging anomalies, and rich resources of CBM, which make it a favorable region for low-rank CBM exploration in China. The Lower Cretaceous coal seams in the Huolinhe and Jiergalangtu depressions on the eastern end of the Erlian Basin are similar in burial depth, metamorphic grade, tectonics but differ greatly in pore structure and the recoverability of CBM. In this study, representative coal samples were collected from the Jiergalangtu and Huolinhe Depressions in the Erlian Basin. They were used to study the impacts of the ash yield on the characteristics of coal pore structures and to analyze the restrictions of the recoverability on CBM through macroscopic and microscopic structure observations, observation under a scanning electron microscope, vitrinite reflectance tests, low-temperature N2 adsorption, NMR, and micron-CT analysis of the samples. The purpose is to further clarify the impact of ash on the pore structure of coal and to provide references for research and production— related to CBM.
2 Geological setting
The Erlian Basin is a Mesozoic–Cenozoic coal-bearing rift valley that developed within the folded basement of the Hercynian geosyncline in the Tianshan–Xingmeng Fold Belt. It has undergone five periods of basic tectonic evolution, including the fault depression during the Early–Mid Jurassic, tectonic reversion during the Late Jurassic (accompanied with intense volcanism), massive fault depression in the early Early Cretaceous, tectonic the reversion in the late Early Cretaceous, and uplift since the Late Cretaceous, resulting in the formation of 47 coal-bearing depressions in total (
Sun et al., 2017). In the Erlian Basin, the depressions are relatively independent of each other and have independent sedimentary systems while their sediments have multiple provenances and contain coarse debris, as a result of their proximity to source areas. Two coal-bearing formations, the Mid-Lower Jurassic Alatanheli Group and the Hantala Formation of the Lower–Cretaceous Bayanhua Group, developed in the periods of initial tension cracking and shrinkage during the tectonic evolution of the basin. Of these two formations, the latter is the main coal-bearing strata in the Erlian Basin (
Sun et al., 2017). This paper focuses on the Huolinhe Depression in the northeastern part of the Erlian Basin and the Jiergalangtu Depression in the center (Fig.1).
The Huolinhe Depression is a long and narrow coal-bearing basin that spreads in the NE strike. It is approximately 60 km long and 9 km wide on average, covering an area of 540 km2. Its coal-bearing strata formed in the Lower Cretaceous, and the coal seam depth is generally less than 1000 m, with proven coal reserves of 1.31 × 1010 t. The Jiergalangtu sag, which is located in a suburb of Xilinhot City, is a dustpan-shaped NE-strike depression that is faulted in the north-west and overlapped in the south-east. It has a length of approximately 67 km, a width of 7–20 km, and an area of 1100 km2. The coal-bearing strata in the Jiergalangtu Depression include Mid and Lower Jurassic coal seams with depth generally less than 1500 m, as well as Lower Cretaceous coal seams with depth generally less than 1000 m.
The coal seams in the Huolinhe Depression are generally made-up of less than 0.5% vitrinite; they are mainly composed of lignite. The coal seams in the Jiergalangtu Depression are mainly composed of lignite too, with generally 0.32%–0.47% vitrinite. Developed under the control of plutonic metamorphism, the metamorphic grade of coal in the Jiergalangtu Depression is mainly controlled by the depth of its coal seams. The deeper the coal seam, the bigger the Ro (Fig.2). As the Cretaceous coal seams in the Huolinhe and Jiergalangtu Depressions are similar in depth, their metamorphic grades are somewhat similar.
3 Methods
3.1 Sample collection
To assess the characteristics of pore structures of coal reservoirs with different ash yields, 13 mines located in the Lower Cretaceous of the Huolinhe and Jiergalangtu depressions were selected for underground observations and to collect samples. During sampling, the different impact levels of tectonism and magmation on samples were taken into consideration. Twelve representative coal samples that are similar in both structural features and depth were selected to be measured for their ash yield. The samples were subsequently analyzed with SEM, micron-CT, NMR, and low-temperature N2 adsorption experiments. Data from the experimental analysis of the 12 samples are shown in Tab.1.
3.2 Experimental tests
Analytical techniques like SEM, micron-CT scanning, NMR, and low-temperature N2 adsorption are used to analyze the characteristics of micropore structures, fissure development, movable fluid pores, and adsorption–desorption of coal in the Erlian Basin. These experiments were conducted at laboratories of the PetroChina Institute for Exploration and Development in Langfang, China. SEM and micron-CT analysis were conducted using the environmental scanning microscope ESEM2020 and the micron-CT analyzer MICRO XCT-400, respectively to qualitatively determine how the pores in coal rocks formed.
The NMR testing experiment used a low-field NMR core analyzer RecCore-2500 to analyze the overall porosity of the coal samples. Nearly 9 cm3 of fresh samples were taken from larger coal samples and placed in a vacuum oven and dried at 65°C until their weight was constant, after which they were placed in a saturated solution KCl for over 24 h. The samples were then centrifuged at high speed and at 200 Psi for 1.5 h, and then tested by NMR; each sample was tested twice. The T2 relaxation spectrum under saturated water reflects the distribution of pore sizes throughout the sample. The differences between the T2 relaxation spectrum under the state of saturated water and that under the state of bound water can reflect the quantity of the movable fluid pores in the sample in water.
For the low-temperature N2 adsorption experiment, the low-temperature liquid nitrogen absorption Tristar 23020 was used. Samples were ground and sieved to 60‒80 meshes. Two samples of 2 g each were dried in an oven at a constant temperature of 105°C. After 3 h, the samples were moved to a vacuum chamber and degassed for 24 h. During the experiment, saturated pressure conditions at a temperature of 77.35 K were maintained. High-purity liquid nitrogen was used as the adsorption medium to conduct the adsorption-desorption experiment under different relative pressures.
4 Results
4.1 Analysis of pore types
Six samples with different ash yields were observed by SEM (Fig.3). However, only the characteristics of medium and large pores are observable as a result of the limited resolution of the SEM. There are mainly two types of coal pores in the study area: residual issue pores with regular shapes and coal-rock plant tissue pores with irregular shapes. Cell-cavity tissue pores, tissue pores, and plant cell residual pores mainly develop in xylem reserved in fusinites and semi-fusinites. The elliptical cellular structures are mainly xylem. Their longitudinal sections are irregular under the microscope, with some deformation affecting a small quantity of cell cavities. Based on pore diameter, and in accordance with the Hodot pore-size classification standard, coal pores can be classified into two types. The first type is transition pores-micropores with a pore radius less than 100 nm, which occur in large quantities except at the surface and make up 26% of the total surface porosity on average. The second type is the medium-large pores with a pore radius greater than 100 nm, which occur in relatively smaller quantities but make up the majority (74% on average) of the surface porosity. The pores of this type have a pore diameter of 4 μm–800 nm generally and a certain degree of connectivity. Comparing the SEM images of JM3 and JM2 shows that JM2 with low ash yield has better-developed medium-large pores and better connectivity.
The grid patterns of pores and fissures in coal are clear in the three samples scanned with micron–CT (Fig.4); fissures are mostly in the shape of grids or branches— the higher the ash yield, the more severely the fissures are filled with minerals. In Fig.4, the micron-CT scan of JM2 samples, with 15.23% ash yield, shows a large volume of massive open fissures and a very small volume of filled fissures. Based on the developed pattern, the fissures are divided into three types. The first type is the nearly-horizontal bedding fissures, which extend along the surface of the bedding plane and are narrow and not filled (Fig.4). The second type is structural fissures generated under the impact of local tectonic stress; they occur perpendicular to the bedding surface and extend like planar branches with somewhat varying widths— generally, main structural fissures are wider compared to the considerably narrow branch fissures which thin out gradually; these fissures are not filled (Fig.4). Lithogenetic evolutionary fissures are the third type of fissures formed by coal in the process of thermal evolution; these form small planar grids, which in section show multiple small fissures forming some complex flower- or X-shaped patterns; in comparison with the two aforementioned types of fissures, lithogenetic evolutionary fissures are very small in volume, and appear blue in CT images (Fig.4). Combining these results with those from the SEM experiment suggests that bedding and structural fissures have better connectivity and can form effective seepage passages favoring the seepage of CBM. The difference in development between the fissures in JM2 and JM3 is that JM2, with a low ash yield, has more developed fissures. This indicates that ash yield has some impact on the brittleness of coal; lower ash yields suggest relatively brittle coal which is more vulnerable to fissuring, compared to less brittle samples, at the same stress levels.
4.2 Nuclear magnetic resonance analysis
The relationship between relaxation time and pore diameter is obtained by performing mercury penetration and NMR experiments on samples. Pressure values of the Hg injection are first converted into pore diameter to plot the probability distribution frequency curve of Hg saturation versus pore diameter, which is used, together with the curve from the NMR experiment, to establish the relationship between pore diameter and relaxation time through the deformation and stretching of the horizontal and vertical coordinates (Fig.5). That is, the transverse relaxation time corresponds to pore sizes 0.1 μm and 1 μm is 2.5 ms and 20 ms, respectively.
In combination with the results from this NMR test (Fig.6(a)), coal pores are divided into three types by pore diameter—micro-small pores with diameter less than 0.1 μm, medium pores with diameter more than 0.1 μm but less than 1 μm, and large pores with diameter more than 1 μm and fissures.
Experimental results from the six samples suggest that the T2 spectrum of coal pores is characterized by a single peak; larger peak areas correspond to better developed pores. For the spectrum peaks of these six samples, the T2 relaxation time mainly ranges between 0.5 and 10 ms, the peak appears close to 3 ms, the signal intensity of the peaks develops between 600 and 1600, and the span is larger. This suggests that, in the study area, coal pores are uneven in diameter and there is considerable variation in their pore diameter. Micropores, small pores, medium pores, and fissures are well developed in the study area. The difference in the NMR and SEM experimental results stems mainly from the difference in their measurement precision. The NMR precision is higher and can better reflect the impact of micro-small pores.
The movable fluid pores are mainly medium in size, with a lower limit of 25 ms; the approximate pore diameter is 1 μm (Fig.6(b)). Large pores and fissures have good movability but fissures have limited volume and do not constitute the main reservoir space. Micro-small pores are mainly adsorptive pores, occupied by bound water and poor in movability. By comparing samples JM1 and HM3, it is found that their lower limits of pore movability are similar, but the volume of movable fluid pores varies greatly. For samples JM1, the T2 spectrum from NMR shows large porosity, and so the volume of corresponding movable fluid pores is large.
Experimental results show that the curves of samples with a high ash yield have low signal intensity and the peak value shifts toward the left (closer to the graphical origin) compared to the samples with low ash yield. This indicates an increase in the number of micro-small pores and reinforces that coal seams with high ash yield have more developed micro-small pores.
4.3 Analysis of N2 adsorption curves
Low-temperature N2 adsorption mainly reflects the distribution of micro-small pores; it is not effective (i.e., shows weak adsorption) for medium and large pores. The liquid nitrogen adsorption curves (Fig.7) suggest that mainly two types of pores develop in coal reservoirs— semi-open pores and open gas pores. Open gas pores are characterized by an adsorption curve that is relatively independent of the desorption curve; however, once the relative pressure is more than 0.8, both curves rise steeply. This is the case for samples with lower ash yield, such as JM4, HM4, and HM5; such pores play a more crucial role in spreading, adsorbing, and desorbing CBM. Semi-open pores are mainly characterized by a desorption curve that generally coincides with the adsorption curve, at a relative pressure of 0–0.3. This indicates that at low-pressure conditions, such pores are in the semi-closed condition; when the relative pressure is more than 0.3, the desorption curve separates from the adsorption curve, indicating the pores begin to open and morph into open gas pores—like the samples with higher ash yield, JM6, and JM8. Erlian Basin.
Converting the low-temperature N2 adsorption curves into pore diameter distribution shows that when the micro-small pore distribution of samples JM6 and JM8 have high ash yields, micropores are more significant in number; however, in the case of the micro-small pore distribution of samples JM4, JM5, and HM4 which have low ash yields, small pores dominate (Fig.8).
5 Discussion
5.1 Classification of coal reservoir types
The water content and the content of volatile components in coal seams in the Erlian Basin do not change much; they range between 12.66%–20.22% and 41.16%–50.19%, respectively. The ash yield of coal has a more significant variation (9.74%–41.47%), based on which coal seams in the study area can be divided into three types: low-ash-bearing coal (Type I, ash yield < 10%), ash-bearing coal (Type II, ash yield: 10%–25%) and ash coal (Type III, ash yield: 25%–50%).
5.2 Analysis on pore structure of coal with different ash yields
Ash is the inorganic content of coal. Inorganics, when filled in coal pores and fissures, lower the porosity and permeability of coal. Its presence can impact the properties of coal reservoirs significantly (
Liu et al., 2021). In the study area, the coal seams vary greatly in ash yield. This makes the ash yield a vital and basic parameter that affects the conditions of physical properties as well as the quality of coal reservoirs. It is important to investigate the impact of the ash yield on the quality of coal reservoirs.
Experimental results suggest the impact of ash yield on porosity can be estimated statistically (Fig.9). Data from different experiments follow the same basic trend, which is that the ash yield results in decreased porosity—both porosity and fissure porosity reduce. This indicates that the ash yield not only has an impact on pores but also has an effect on fissures, with the impact on fissures smaller than that on pores (Fig.9(a)). SEM and NMR experiments show that in the case of samples with ash yield less than 30%, porosity decreases quickly, whereas in the case of ash yield more than 30%, porosity changes (reduces) more slowly (Fig.9(a)). The impact of the ash yield on pores of different sizes is also analyzed. It shows that ash yield strongly affects micro-small pores and, to a lesser extent, medium-large pores—both types of pores get smaller. However, it has a greater impact on medium-large pores; it reduces some medium-large pores to micro-small pores. This is also reflected in the gradual increase in the number of micro-small pores compared to medium-large pores as the ash yield increases (Fig.9(b)).
Although the increased ash yield affects fissures, it has a greater effect on the movable fluid porosity (Fig.10). μ-CT images show that, in the study area, structural fissures are vertical or near-vertical to the coal seam interface while bedding fissures run parallel or nearly parallel to the coal seam interface. The porosity of both types of fissures can be approximated using their 3D volumes picked up from CT scanning. As the ash yield increases, the porosity of both structural fissures and bedding fissures decreases. Ash yield has a larger effect on the latter primarily because ash deposition is also a product of sedimentation, and was likely contemporaneous with the deposits that evolved into coal seams. As a result, ash content increases significantly at the bedding surface (Fig.10(a)). As the ash yield increases, the movable fluid porosity decreases. This is because the ash causes the medium-large pores which are the main flowable pores to be converted into micro-small pores which are predominantly non-flowable, adsorptive pores (Fig.10(b)).
The N2 adsorption experiment analyses the impact of ash yield on the volume of adsorbed and desorbed gas (Fig.11). The results show that the ash yield has a greater impact on the adsorbed gas volume, especially in the case of low ash yield; this indicates that ash yield impacts the gas saturation of coal (Fig.11(a)). Ash yield has a small impact on the desorbed gas volume. This is attributed to the impact of the conditions for the low-temperature N2 adsorption experiment. The curves from the experiment (Fig.7) show that as N2 does not reach saturation in the experiment, it has a larger impact on the desorbed gas volume (Fig.11(b)).
5.3 Impact of coal with different ash yields on recoverability of CBM
Factors like porosity, gas content, depth, and stratum pressure are typically considered in the evaluation of coal seams, while factors like the ash yield of coal and the recoverability of CBM are not as well studied. This study established the ash yield of coal as a critical, basic indicator of the quality of coal reservoirs which exerts an obvious controlling effect on porosity, pore structure, gas content, movable fluid porosity, and the volume of adsorbed and desorbed gas of CBM. Considering this, the ash yield of coal can be used as an indicator and can be combined with coal metamorphism to develop, in the future, a coal classification scheme that will play an important role in controlling the development of CBM.
As suggested by the presence of many well-developed medium-large pores, high porosity, well-developed bedding and structural fissures, high permeability and movable fluid porosity, and a large volume of adsorbed, desorbed, and recoverable gas, the analytical results from this paper show that low-ash-bearing coal is a kind-of-reservoir with the best recoverability. In comparison, ash-bearing coal has moderately-well-developed medium-large pores, porosity, bedding and structural fissures, permeability, moderate movable fluid porosity, and a moderate volume of adsorbed, desorbed, and recoverable gas. These characteristics make ash-bearing coal a kind-of-reservoir with good recoverability but less efficient than low-ash-bearing coal. Ash coal has fewer medium-large pores, low porosity, mainly micro-small pores, poorly-developed bedding and structural fissures which are mostly filled, low permeability, less movable fluid porosity, and lower volumes of adsorbed, desorbed, and recoverable gas. Therefore, ash coal is a kind-of-reservoir with poor recoverability.
6 Conclusions
1) For coal pores in the Erlian Basin, the micro-small pores significantly outnumber the medium-large pores. However, the latter accounts for much of the surface porosity and have good connectivity. In the Erlian Basin, the movable fluid pores are mainly medium-large pores with a lower pore-size limit of 25 ms and an approximate pore diameter of 1 μm. Large pores and fissures have good movability, but fissures have limited volume and do not constitute the main reservoir space.
2) In the Erlian Basin, the ash yield of the coal shows a significant variation (9.74%–41.47%), based on which coal seams in the study area are divided into three types: Type I is low-ash-bearing coal with ash yield < 10%, Type II is ash-bearing coal with ash yield between 10%–25%, and Type III is ash coal with ash yield between 25%–50%.
3) Ash is the inorganic content in coal which, when filled in coal pores and fissures, lowers its porosity and permeability; its presence can impact the properties of coal reservoirs significantly. Thus, as a kind-of-reservoir low-ash-bearing coal in the Erlian Basin has the best recoverability for CBM. The recoverability of ash-bearing coal is not as efficient as the former, but is moderately better than that of ash coal, which has poor recoverability.
PDF
(4232KB)
