https://orcid.org/0009-0007-2777-1584
Figures
Abstract
In order to study gas adsorption performance effected by micro-rock in coal matrix, dual energy X-ray CT was used to calculate the density change of coal core before and after gas injection, which can obtain the CO2 concentration (adsorption amount) in coal core, and reveal the uneven characteristics of CO2 adsorption by minerals. The research results show that coal core in the region where minerals exist has a higher density, while the density of the coal matrix is smaller. Regions with higher coal density (more minerals) have weaker adsorption capacity for CO2. The CO2 concentration in coal core decreases approximately linearly along the axial direction of coal core from the gas injection end to the outlet section. The average voxel density is basically the same at the same coal core section before and after gas injection, which indicates that the coal matrix recovers and approaches the original state after desorption. However, before and after gas injection, the average voxel density of coal core varies greatly, and the frequency variation and average deviation factor of different voxel densities vary greatly, which indicates that the adsorption of CO2 by coal core is extremely uneven.
Citation: Jing G, Zhao Q, Jin Z (2025) Experiment and analysis on the heterogeneity of CO2 adsorption-transport by microrocks in coal matrix. PLoS ONE 20(3): e0314162. https://doi.org/10.1371/journal.pone.0314162
Editor: Kang Wang, Shenyang Jianzhu University, CHINA
Received: January 7, 2024; Accepted: November 5, 2024; Published: March 6, 2025
Copyright: © 2025 Jing et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: All relevant data are within the manuscript and its Supporting Information files.
Funding: The author(s) received no specific funding for this work.
Competing interests: The authors have declared that no competing interests exist.
1 Introduction
In the process of CO2 enhanced coalbed methane (ECBM) production/storage, injecting CO2 into coal seams will not only increase the pressure gradient to drive methane flow, but also replace the methane adsorbed in the coal matrix by reducing the partial pressure of methane in the gas mixture [1,2]. The adsorption capacity of coal for gases can not only be used to predict the maximum possible volume of gas stored in coal, but also to predict the amount of gas released from coal during production. Numerous scholars have studied the gas adsorption capacity of coal using coal powder [3,4]. Although these studies provide a relatively fast method for estimating coal storage capacity, in practical situations, coal is not only in a three-dimensional stress state, but also does not exist in powder form [5]. Moreover, a large number of experiments have shown that coal has a much greater adsorption capacity for CO2 than methane, and coal seams have enormous potential for sequestering carbon dioxide [6,7]. Therefore, it is necessary to study the adsorption capacity and diffusion effect of CO2 in coal matrix under stress, which is also an important parameter for the technical and economic indicators of CO2 storage in coal reservoirs [8].
Research has shown that coal contains a large amount of minerals, and the gas permeation diffusion process in the coal matrix also exhibits anisotropic characteristics, while gas adsorption also exhibits non-uniformity [9–11]. The mineral content in coal ranges from 10% to 25%, with some low rank coal having a mineral content of up to 40% [12,13]. The mineral composition in coal is relatively complex, mainly consisting of vitrinite, inertinite, and chitin. The distribution of minerals in coal not only affects the pore structure, but also affects the gas adsorption ability of coal [14–16]. For the ability of minerals to adsorb and store gas in coal, most scholars believe that the Langmuir volume of the same coal quality decreases with the increase of ash content. The pore characteristics of coal are analyzed, minerals fill a certain proportion of micropores in coal, thus reducing the specific surface region of coal, and thus reducing the adsorption amount of gas. At the same time, minerals fill the cracks in the coal matrix, thus reducing the permeability of gas in the coal seam [17–20].
Regarding the adsorption mechanism of coal on gases, some scholars have used X-ray scanning to observe the transportation and storage of gases in the coal core during CO2 injection, and analyzed the adsorption rates of different coal on gases through CT image processing [21]. Smith [22] measured the adsorption capacity and diffusion rate of unconfined coal cores using environmental pressure gravity technology; The results indicate that CT can not only provide qualitative measurements of spatial changes in CO2 concentration within the core, but also provide quantitative and accurate measurements of CO2 adsorption. Liu et al. [23] quantitatively analyzed the damage of power ultrasound stimulation on multi-scale pores and cracks in coal based on gas adsorption and CT image 3D reconstruction. Li et al [24] quantitatively characterized the physical properties of coal using non-destructive low field nuclear magnetic resonance (NMR) and X-ray computed tomography (X-CT). The results indicate that with the increase of stress, the volume of mesopores, macropores, and cracks sharply decreases. Zhou et al. [25] used the method of SEM-EDS and micro CT scanning to observe the microstructure deformation of middling coal during methane adsorption desorption cycle. The results indicate that the deformation of coal during methane adsorption is mainly caused by the expansion and compression stress between different regions. The deformation of clay minerals is mainly influenced by the deformation of their adjacent coal matrix, and residual deformation in coal mainly exists in areas with strong density non-uniformity. Zhou et al. [26] conducted extensive research on the deformation of the microstructure during coal adsorption of methane and found the localized density homogenization effect of the microstructure during coal adsorption of methane. By establishing a new coal structure model, the mechanism of local density homogenization effect was analyzed, and the local density homogenization equation for ethane adsorption in coal was derived. Yu et al. [27] used digital image subtraction technology to extract the structural characteristics of pores and fractures and fluid distribution. Explore the fluid ascent behavior in a single capillary and the flow characteristics within the pore fracture network, in order to reveal the mechanism by which pore fractures affect fluid migration during self suction. Although these studies analyzed gas adsorption, deformation, and damage in coal, they did not investigate the non-uniformity of gas adsorption and transport process.
In order to study the influence of micro rocks (minerals) in coal on its diffusion adsorption, dual energy X-ray CT was used to measure the gas transport and storage in coal core during the CO2 injection process, so as to obtain the temporal variation of CO2 concentration in coal core. Moreover, the density variation characteristics of different regions affected by minerals and the non-uniformity of CO2 adsorption transport were compared and analyzed. In sections 2, 3, and 4, the experimental equipment, methods, and experimental processes are described. The experimental analysis content is in Section 5: (1) Analysis of density differences in coal-body regions; (2) Analysis of liquid nitrogen adsorption and pore structure differences in coal-body regions; (3) The spatiotemporal variation characteristics of coal core gas adsorption under different injection pressures; (4) Characteristics of uneven changes in gas adsorption-desorption of coal core; (5) Heterogeneous changes in gas adsorption-desorption of coal core.
2 Coal samples and experimental system
The experimental coal sample was taken from the raw coal block of the No. 21 coal seam in Jiulishan Coal Mine, Jiaozuo. The HZ-15 electric corer (Fig 1A) was used to core the raw coal block (Fig 1B) with a diameter of 25mm (Fig 1C), and the coal pillar was cut and polished using a TCHR-II cutting mill (Fig 1D) to produce a coal core with a length of 50mm, as shown in Fig 1E.
Put three coal samples into a drying oven, set the temperature to 40°C, dry for 3 days, and then put them into a closed bag for backup. The weight losses of the three coal samples are 0.55%, 0.52%, and 0.57%, respectively. This indicates that during the drying process at 40°C, only the external moisture of the coal sample is removed, and its inherent moisture is difficult to remove.
The Phoenix v|tom|x s industrial CT scanning system introduced from the key laboratory of Henan Polytechnic University is adopted, as shown in Fig 2. The industrial CT scanning system mainly consists of X-ray source, turntable, coal sample gas injection, coal sample clamping system, X-ray detector, and computer data processing unit.
3 Testing methods
When X-ray passes through an object, it will produce complex physical processes such as Photoelectric effect and Compton effect. Part of the X-ray is reflected, scattered and absorbed by the substance, and the intensity of the X-ray is attenuated [28]. The linear attenuation coefficient of a material corresponds to the density of the material. The linear attenuation coefficient of a sample composed of various substances is equal to the sum of the product of the linear attenuation coefficient of the different components and the mass percentage.
Due to the uneven composition of coal, the attenuation of X-rays varies after passing through different regions, and some photons cannot reach the flat panel detector. Some photons can reach the detector through an object and be visualized by a computer controlled system. If the CT value is converted into a certain proportion of grayscale values, the corresponding CT image can be obtained. The bright colors in the CT image represent the high-density regions of the coal body, while the dark colors represent the low-density regions.
When scanning a cylindrical coal core, a digital image is produced. These images reflect the value of the attenuation coefficient in the axial plane (i.e., the “slice”) of the object of study. Attenuation depends on electron density, but can usually be calibrated by conventional weight density. If only one energy level is used, the only quantitative information obtained is the local value of the linear attenuation coefficient, which may not be sufficient to analyze the interface density of the “slice” in detail [29]. Therefore, dual energy scanning technology was used in this experiment [30].
The density of the scanning material is a linear function of the low energy [CTlow] and high energy [CThigh] CT values, and its relationship can be expressed as [31].
Here, the CT values of three reference materials (Table 1) are used to determine the coefficients a, b, and c in the CT value calculation equation. Furthermore, using these coefficients, the density of the coal core under vacuum and the density of the coal core during CO2 adsorption can be calculated. When the density values at each time and pressure are obtained, the volume and porosity of the coal sample can be obtained, and thus the CO2 adsorption amount can be calculated.
Based on calibration measurements (see Table 1) and Eq 1, three Equations for water, silica, and aluminum can be obtained:
(2)
According to Eq (1) and the CT values measured at 80 kV and 130 kV, the calculated values of parameters a, b, and c for mass density are: a = 2.71 g/cm3, b = -0.81 g/cm3, and c = 1206.23 g/cm3, respectively.
4 Testing process
The rock gripper can apply radial confining pressure (N2 pressurization) along the coal core and inject CO2 at one end. In this experiment, CO2 pressure is applied, and the other end is connected with the atmosphere, so that the CO2 adsorption kinetics can be observed at different positions (section slices) along the whole length of the coal core. The test steps are as follows (room temperature):
- The assembled core holder is installed on the test turntable in the industrial CT scanning system. The coal core is scanned into 2mm thick slices at different locations, and each measured CT number represented 0.25×0.25×2.0mm three-dimensional pixels. 6 scans are carried out along the axial direction (Fig 3).
- Confining pressure is provided by injecting N2 via pressure regulator. During each pressure period of the experiment, the pressure remains constant.
- Use a vacuum pump to vacuum the coal core until the internal pressure is below 20Pa.
- Under an effective confining pressure of 1.5MPa, the coal core was scanned and vacuumed at two energy levels of 130 kV and 80 kV.
- CO2 is injected into one end of the coal core. Due to the small size of the coal core, the volume of CO2 in the cylinder is relatively large. Therefore, constant CO2 pressure is also maintained at the injection inlet. In the gas injection test, the pressure difference between the confining pressure and the inlet CO2 pressure remain constant at 1.5MPa.
- At CO2 injection pressures of 1.0 MPa (confining pressure of 2.5 MPa), 2.0 MPa (confining pressure of 3.5 MPa), 3.0 MPa (confining pressure of 4.5 MPa), and 4.0 MPa (confining pressure of 5.5 MPa), the coal core at each section position is repeatedly scanned at the set CO2 permeation and adsorption time, and the continuous injection time is 150h.
- After CO2 injection (pressure is 4.0 MPa) is completed, the coal core is desorbed for 50 hours, and the coal core is repeatedly scanned at each slice position.
- The average density of each section at each time is calculated from its average CT value (Eq 1) and the a, b, and c values. After CO2 injection, the density of each slice is subtracted from the average density of the corresponding slice under vacuum, which is the density increment of the slice caused by CO2 adsorption. The average CO2 density of these slices (unit: mass of CO2 per unit volume of slice).
- The analysis is performed using the VoxelCalc NDT and Image J software processing packages, and color images are generated to illustrate the spatial distribution of density in each slice before or after adsorption occurred.
5 Test results and analysis
5.1 Coal core density after vacuum extraction
After vacuuming the coal core, the density changes of each cross-section of the coal core obtained through scanning calculation are shown in Fig 4. Baby blue indicates that the density of coal matrix is small, while red indicates that the coal contains minerals with high density, as shown in Fig 5 (regions A and B in Fig 4). From Fig 4, it can be seen that the upper left and middle regions of each slice exhibit a higher density of coal specimens (influenced by minerals). The macerals in these regions are closely mixed and mainly composed of three components (vitrinite, sporolite and inertinite). In Fig 4A, sporolite and inertinite dominate the region as consistent bands, and vitrinite zone alternates with sporolite and inertinite mixed zone; The vitrinite in Fig 4B is mixed with meshed inertinite. The average bulk density of coal specimens calculated based on CT data is ρ = 1.194 g/cm3, and the density of coal samples calculated based on mass and volume is 1.21g/cm3, with an error less than 1.40%.
As can be seen from Fig 5, the core is not uniform in each slice and along the core axis. In order to compare and analyze the non-uniformity on the coal core slice, two regions A and B are selected on the coal core slice (as shown in Fig 4). The density of each slice of coal core in these two regions (before vacuuming and gas injection) is calculated according to the CT values of regions A and B. Fig 5 shows the axial change curve of the average density of regions A and B and different slices along the coal core. There is a big difference between the coal density and the average slice density in region A and B. The coal core density in region A is higher than the average slice core density, but its change along the axial direction of the coal core tends to the change rule of the average density. The core density in region B is always lower than that in region A, and it fluctuates greatly along the core axis.
Coal samples are collected from Zone A and Zone B of Slice 1 for liquid nitrogen adsorption and mercury intrusion tests to obtain their pore structure characteristics. The liquid nitrogen adsorption/desorption curves and mercury intrusion/desorption curves in coal samples from regions A and B are shown in Fig 6. From Fig 6A, it can be seen that the adsorption and desorption curves exhibit an inverted S-shape, belonging to the type II adsorption isotherm. When the relative pressure is less than 0.5 (p/p0<0.5), the adsorption capacity slowly increases, the adsorption isotherm protrudes upwards, and the first adsorption layer roughly disappears. As the relative pressure increases, the second adsorption layer begins to form on the first adsorption layer. The relative pressure (p/p0) increases from 0.5 to 0.95, and the adsorption curve sharply rises until the vapor pressure approaches saturation (1>p/p0>0.95) before the adsorption saturation phenomenon occurs. For the adsorption of low-temperature liquid nitrogen in coal, single-layer adsorption occurs under surface tension under low pressure (p/p0<0.5); Under medium pressure (0.5<p/p0<0.95), multi-layer adsorption occurs under the action of van der Waals forces; Under high pressure (p/p0>0.95), capillary condensation occurs. The mercury intrusion and mercury removal curves of coal samples in regions A and B are shown in Fig 6B. The mercury injection curve is S-shaped, with a convex curve at low pressure and a concave curve at high pressure. The cumulative amount of mercury entering the coal samples in Region A and Region B gradually increases. Compared with the mercury intrusion hysteresis loop of Region A coal sample, the mercury intrusion hysteresis loop of Region B coal sample is wider, and the pores are mainly open pores.
Liquid nitrogen and mercury intrusion test curves of coal samples (a) Liquid nitrogen test curve (b) Mercury intrusion test curve.
According to liquid nitrogen adsorption/desorption curves, the specific surface areas of micropores, mesopores, and macropores in coal sample are shown in Fig 7A. The specific surface areas of micropores, mesopores, and macropores in region A coal sample are 0.0605m2/g, 0.4725m2/g, and 0.1611m2/g, respectively, with a total specific surface area of 0.6941m2/g, which is smaller than the specific surface areas of micropores, mesopores, and macropores in region B coal sample of 0.5354m2/g, 1.687m2/g, and 0.6833m2/g, and a total specific surface area of 2.9057m2/g. According to the mercury intrusion/mercury removal curves, the pore volumes of micropores, mesopores, and macropores in coal sample are shown in Fig 7B. The pore volumes of micropores, mesopores, and macropores in region A coal sample are 0.0035cm3/g, 0.0191cm3/g, and 0.0139cm3/g, respectively, with a total pore volume of 0.0365cm3/g, which is smaller than the pore volumes of micropores, mesopores, and macropores in region B coal sample of 0.0074cm3/g, 0.055cm3/g, and 0.038cm3/g, and a total pore volume of 1.004cm3/g. This is mainly due to the high content of minerals (such as inertinite and spore stones) in Region A, while the mineral content in Region B is relatively low. Therefore, the specific surface area and pore volume of Region A coal sample are smaller than those of Region B coal sample, which will lead to differences in the adsorption capacity of Region A and Region B coal samples.
Specific surface area and pore volume of coal samples (a) Specific surface area (b) Pore volume.
5.2 CO2 adsorption-migration characteristics under different injection pressures
In this experiment, CO2 in the reference tank and auxiliary pipeline (Fig 2) is a high-pressure gas source at the end face of coal core, which forms gas pressure difference with the inside of coal body. Under the pressure difference, the gas penetrates into coal body, which is similar to the coal permeability test [32]. This process is different from the instantaneous release of high-pressure CO2 to fracture coal body, which is an extremely slow process [33]. Therefore, the damage effect of high-pressure gas source on coal body can be ignored.
CO2 was injected at 1.0 MPa (first injection pressure, confining pressure of 2.5 MPa), 2.0 MPa (second injection pressure, confining pressure of 3.5 MPa), 3.0 MPa (third injection pressure, confining pressure of 4.5 MPa), and 4.0 MPa (fourth injection pressure, confining pressure of 5.5 MPa). After 150 hours of injection, the spatial distribution of CO2 density in coal core after CO2 adsorption (average profile value) is shown in Fig 8 (scan positions 1–6 in Fig 3).
As shown in Fig 8, the coal core slices farther away from the injection inlet contain less CO2. In addition, it can be seen that the CO2 adsorbed at the coal core slices is uneven, and the upper left part of each slice absorbs less CO2. This is because the upper left part of coal core has more minerals and weaker CO2 adsorption capacity. Meanwhile, it can also be seen from Fig 9 that the higher the CO2 injection pressure, the higher the CO2 density at the same position in coal core after the injection is completed (150 hours). That is to say, the higher the CO2 injection pressure, not only increases the adsorption capacity of coal for CO2, but also promotes the migration of CO2 in coal core (from the inlet end to the outlet end). From Fig 9, it can also be seen that the CO2 infiltration-diffusion-adsorption process in coal core has not reached equilibrium within 150 hours. Moreover, the further away from the gas injection end, the smaller the amount of CO2 adsorbed, that is, the smaller the volume of CO2 adsorbed per unit mass of coal core.
Relationship between CO2 concentration and injection time in different slices under different pressures (a) 1.0MPa (b) 2.0MPa (c) 3.0MPa (d) 4.0MPa.
Fig 10 shows the average CO2 concentration of coal core slices at different injection pressures (150 hours). CO2 concentration decreases almost monotonically along the coal core axis from the injection end to the outlet end. Fig 11 shows the apparent adsorption isotherms of different slices obtained from CT data. For slice 1 with the closest diffusion of CO2 injection, when the injection pressure is above 3.0 MPa, the CO2 adsorption capacity tends to stabilize approximately with increasing adsorption pressure, indicating that adsorption equilibrium may have been reached at the position of slice 1. At other locations, as the injection pressure increases, the CO2 concentration (adsorption capacity) also gradually increases, indicating that CO2 adsorption at this location has not yet reached equilibrium.
In summary, under constant temperature conditions, the influence of injection pressure is mainly reflected in two aspects:
- Adsorption performance: Gas adsorption capacity in coal satisfies the Langmuir adsorption isotherm, which means that the higher the pressure, the greater the adsorption capacity of gas in coal. Therefore, in this experiment, it was observed that the CO2 concentration (adsorbed CO2) at coal core slice position was higher, as shown in Fig 8. The higher the injection pressure, the higher the CO2 concentration at the same coal core position with the same injection time (red represents CO2 concentration, and the higher the pressure, the greater the significance of red color in the same slice).
- Gas permeability performance: The driving force for gas permeation in coal cores mainly comes from the gas pressure difference, and the larger the pressure difference, the faster the gas permeation. The higher the gas pressure at the injection end of coal core, the more significant the pressure difference formed in coal core, which can quickly drive gas to penetrate into the another end of coal core. In this process, the adsorption amount (CO2 concentration) is larger in coal core at the same slice location at the same time, the CO2 concentration in coal core slice in Fig 8 is more significant in red, and the CO2 concentration value in coal core in Fig 9 is larger. Moreover, under different injection pressures and with the same injection time, the CO2 concentration along the coal core axis shows a decreasing trend. Although this is a sign that gas in coal body has not reached adsorption saturation, the CO2 concentration value clearly indicates that when the injection pressure is higher, the CO2 adsorption amount is greater farther away from the injection end of coal core, which indicates that CO2 permeation in coal core is faster.
5.3 CO2 concentration in coal core after gas injection/desorption
The spatial distribution of CO2 density in the coal core after CO2 adsorption at 4.0MPa (injection pressure) and injection time of 150h is shown in Fig 6. The further away from the injection port the coal core slice contains less CO2, and there is no adsorption equilibrium at this time. In addition, it can be seen that the CO2 adsorbed at the coal core slices is uneven, and the upper left part of each slice adsorbs less CO2. This is because the upper left part of the coal core has more minerals and weaker CO2 adsorption ability.
Taking the injection pressure of 4.0 MPa as an example, the spatial distributions of CO2 density in coal core are shown in Fig 12 after injection time of 150 hours and desorption time of 50 hours. From Fig 12, it can be seen that the coal core slices farther away from the injection inlet contain less CO2, and at these locations, they have not yet reached adsorption equilibrium. In addition, it can be seen that the CO2 adsorbed on the coal core slices is uneven, and the CO2 adsorbed on the upper left part of each slice is relatively small. This is because the upper left part of coal core has more minerals and weaker CO2 adsorption ability. After 50 hours of desorption, there is still a certain amount of CO2 in coal core, because during the desorption process, CO2 moves slowly in coal core and has not been completely desorbed. But overall, the closer to the exit end, the smaller the CO2 concentration and the more significant the decrease. From Fig 12, it can also be seen that the average CO2 concentration of the slice decreases almost monotonically along the coal core axis from the injection end to the outlet end. Meanwhile, CO2 concentration is uneven in each slice of coal core. CO2 concentration adsorbed is more significant in the lower right area of the slice, and less in the upper left area. This is because there are more minerals in the upper left area of the coal core, and its adsorption performance is weak.
In order to compare and analyze the non-uniformity of CO2 adsorption on coal core slices, two Regions A and B were selected on the coal core slices. Based on the CT values of Region A and B after adsorption and desorption, the difference in coal core density between these two areas can be calculated to obtain the CO2 adsorption concentration for each slice.
Fig 13 shows the axial variation of CO2 concentration in Regions A and B, as well as the entire slice, after 150 hours of gas injection (4.0 MPa) and 50 hours of desorption. These curves illustrate the non-uniformity of CO2 concentration in coal core. CO2 concentration in Region A is much lower than the average CO2 concentration in this slice, which also confirms the visual observation results in the pictures. However, CO2 concentration in Region B is much higher than that in Region A, usually higher than the average CO2 concentration. This is because there are more minerals in the upper left area of coal core, and its adsorption performance is weak. It was also observed that the higher the coal density at the slice, the lower the CO2 concentration. From Fig 13, it can also be seen that after 50 hours of desorption, the overall CO2 concentration in coal core decreases, and the average concentration approximately linearly decreases along the axial direction of coal core. For local areas, CO2 concentration in Region B decreases particularly significantly along the axial direction of coal core (with a large gradient), while CO2 concentration in Region A is lower, and CO2 concentration is smaller after desorption.
5.4 Characteristics of uneven adsorption of coal cores
When a small region inside the coal core is selected for analysis, a very significant density change (non-uniformity) is observed compared to the average value of the coal core specimen. The calculated CO2 concentration in the high-density region of the coal core is significantly lower than the average CO2 adsorption amount in the slice. Here, a CT instrument was used to detect core heterogeneity at a single 3D pixel (voxel) scale (hexahedron of 0.25×0.25×2.0mm size per 3D pixel) with approximately 9900 voxels in each slice. Taking section 1 as an example, the frequency and cumulative percentage of coal core density at different voxels before gas injection, after gas injection and after desorption are shown in Fig 14.
In Fig 14, the voxel density of coal core slice 1 before gas injection is mainly concentrated between 1.16g/cm3 and 1.25g/cm3, with an average voxel density of 1.195g/cm3 and an average skewness factor (asymmetry measure) of 0.67 (completely symmetric distribution, skewness factor of 0). After gas injection, the voxel density of coal core slice 1 is mainly concentrated between 1.20g/cm3 and 1.30g/cm3, with an average voxel density increasing to 1.23g/cm3 and an average skewness factor of 0.88. After desorption, the voxel density of coal core slice 1 is mainly concentrated between 1.15g/cm3 and 1.28g/cm3, with an average voxel density of 1.199 mg/cm3, which is close to returning to the pre gas injection state (difference of 0.34%), and the average skewness factor of distribution is 0.82. Although the difference in coal voxel density before and after CO2 injection and desorption indicates that a certain amount of CO2 remains in the coal core, the frequency distribution and average voxel density of the coal core after desorption are close to that before gas injection.
At the same time, it can be seen from Fig 14 that the density of different voxels in slice 1 changes significantly, which also reflects the extremely uneven adsorption of CO2 by the coal core. The relationship between the voxel density difference caused by adsorption after CO2 injection (voxel density after adsorption minus initial voxel density) and the initial coal density is shown in Fig 15. Fig 15 shows that the low-density macerals (coal matrix) absorb more CO2 than the high-density macerals (minerals), which is also consistent with the results shown in Figs 12 and 13.
6 Conclusion
Using dual energy X-ray CT, the density change of coal core before and after gas injection was calculated, and the CO2 concentration in coal core was obtained. The influence of minerals in coal matrix on coal density was analyzed, and the characteristics of uneven adsorption of minerals on CO2 were revealed. The conclusions are as follows:
- Due to the mineral composition in the coal core, the density distribution of the coal core has certain differences (uneven density). The coal core in the region where the mineral exists has higher density, while the density of the coal matrix is smaller. The error between the density of coal calculated by CT data and the density of coal core calculated by mass/volume method is less than 1.50%.
- The average CO2 concentration of the slices increases with time, with higher CO2 concentration near the gas injection end and lower CO2 concentration far from the gas injection end. That is, CO2 concentration decreases almost monotonically along coal core axis from gas injection end to outlet end. Meanwhile, the higher the CO2 injection pressure, the higher the CO2 density at the same location within the same injection time.
- The farther away from the injection inlet, the less CO2 the coal core slices contain, and the CO2 adsorbed by each slice of the coal core is also uneven. The higher the coal density (with more minerals) at the slice, the weaker the CO2 adsorption ability.
- When the coal core is not fully desorbed (CO2 migrates slower in the coal core and takes longer), the closer it is to the outlet section, the smaller the CO2 concentration, and the more significant the decrease. The average CO2 concentration in the slice decreases approximately linearly along the axial direction of the coal core. For local regions, the higher the mineral content (higher coal density), the more significant the CO2 concentration changes along the axial direction of the coal core.
- The average voxel densities of coal core slice 1 before and after gas injection and desorption are 1.195g/cm3, 1.23g/cm3, and 1.199 mg/cm3, respectively, and the average skewness factors of distribution are 0.67, 0.88, and 0.82, respectively. The significant changes in different voxel densities before and after gas injection indicate the extreme non-uniformity of CO2 adsorption by the coal core. There is little difference in the changes in different voxel densities before and after gas injection and desorption, indicating that the coal matrix recovers and approaches its original state after desorption.
Supporting information
S1 Text. Minerals in high density region of coal core.
https://doi.org/10.1371/journal.pone.0314162.s001
(XLSX)
S2 Text.
Liquid nitrogen and mercury intrusion test curves of coal samples (a) Liquid nitrogen test curve. Liquid nitrogen and mercury intrusion test curves of coal samples (b) Mercury intrusion test curve.
https://doi.org/10.1371/journal.pone.0314162.s002
(ZIP)
S3 Text.
Specific surface area and pore volume of coal samples (a) Specific surface area. Specific surface area and pore volume of coal samples (b) Pore volume.
https://doi.org/10.1371/journal.pone.0314162.s003
(ZIP)
S4 Text.
Relationship between CO2 concentration and injection time in different slices under different pressures (a) 1.0MPa. Relationship between CO2 concentration and injection time in different slices under different pressures (b) 2.0MPa. Relationship between CO2 concentration and injection time in different slices under different pressures (c) 3.0MPa. Relationship between CO2 concentration and injection time in different slices under different pressures (d) 4.0MPa.
https://doi.org/10.1371/journal.pone.0314162.s004
(ZIP)
S5 Text. Changes of CO2 concentration along the axial direction of coal core after gas injection.
https://doi.org/10.1371/journal.pone.0314162.s005
(XLSX)
S6 Text. Apparent sorption isotherms for various slices.
https://doi.org/10.1371/journal.pone.0314162.s006
(XLSX)
S7 Text. Relationship curves between CO2 concentration and coal core length for complete sections and two regions.
https://doi.org/10.1371/journal.pone.0314162.s007
(XLSX)
S8 Text. Voxel density of coal sample slice 1 before and after CO2 injection and desorption.
https://doi.org/10.1371/journal.pone.0314162.s008
(XLSX)
S9 Text. Increase in voxel density of slice 1 before and after CO2 injection into coal core.
https://doi.org/10.1371/journal.pone.0314162.s009
(XLSX)
References
- 1. Safaei-Farouji M., Misch D., Sachsenhofer R.F., Rauscher M., Kostoglou N. (2024). From abandoned mines to carbon sinks: Assessing the CO2 storage capacity of Austrian low-rank coal deposits[J]. International Journal of Coal Geology, 286: 104495.
- 2. Guang W.F., Zhang Z.Y., Zhang L., Ranjith P.G., Hao S.P., Liu X.Q. (2023). Confinement effect on transport diffusivity of adsorbed CO2–CH4 mixture in coal nanopores for CO2 sequestration and enhanced CH4 recovery [J]. Energy, 278: 127929.
- 3. Almolliyeh M., Tripathy S., Sadasivam S., Masum S., Thomas H.R. (2023). Effect of Physical Nature (Intact and Powder) of Coal on CO2 Adsorption at the Subcritical Pressure Range (up to 6.4 MPa at 298.15 K)[J]. ACS Omega, 8, 7070–7084.
- 4. Wang J.B., Chen J., Liu J.Z., Liu H., Wang M.X., Cheng J. (2022). Synergistic effects of mixing waste activated carbon and coal in co-slurrying and CO2 co-gasification[J]. Powder Technology, 395, 883–892.
- 5. Wang K., Fu Q.C., Zhang X., Jia H.Y. (2021). Experimental Investigation on Strain Changes during CO2 Adsorption of Raw Coal Sample: Temperature and Effective Stress[J]. Energies, 14(3), 717.
- 6. Wu M., Qin Y., Zhang Y.Y., Zhu S.F., Zhang G.C., Lan F.J., et al. (2023). Influence Factors and Feasibility Evaluation on Geological Sequestration of CO2 in Coal Seams: A Review[J]. ACS Omega, 8(19): 16561–16569.
- 7. Liu S.Q., Liu T., Zheng S.J., Wang R., Sang S.X. (2023). Evaluation of carbon dioxide geological sequestration potential in coal mining area[J]. International Journal of Greenhouse Gas Control, 122: 103814.
- 8. Ji B.N., Pan H.Y., Pan M.Y., Zhou Y.X., Pang M.K., Wang K., et al. (2024). Molecular simulation of free CO2 injection on the coal containing CH4 structure and gas replacement[J]. Fuel, 377, 132789.
- 9. Zhou A.T., Du C.A., Tian J., Xu Z.Y., Wang D.X., Wang K. (2023). Experimental study on coal deformation induced by gas adsorption-instantaneous pressure relief under unconstrained stress state with different pore structures[J]. Energy, 276, 127480.
- 10. Li J.H., Li B.B. (2021). Modeling of anisotropic coal permeability under the effects of matrix-fracture interaction[J]. Journal of Natural Gas Science and Engineering, 93, 104022.
- 11. Zhou D., Feng Z.C., Zhao D., Zhao Y.S., Cai T.T. (2017). Experimental meso scale study on the distribution and evolution of methane adsorption in coal[J]. Applied Thermal Engineering, 112, 942–951.
- 12. Zhou S.D., Liu D.M., Cai Y.D., Wang Y.J., Yan D.T. (2021). Mineral Characteristics of Low-Rank Coal and the Effects on the Micro- and Nanoscale Pore-Fractures: A Case Study from the Zhundong Coalfield, Northwest China[J]. Journal of Nanoscience and Nanotechnology, 21(1), 460–471.
- 13. Suyatno S., Ghazidin H., Prabowo P., Kuswa F.M., Putra H.P., Ruhiyat A.S., et al. (2024). Investigation of Slagging-Fouling and Mineral Contents in Low-Rank Coals for Improved Combustion Performance[J]. Combustion Science and Technology, 1–21.
- 14. Wang Y.Z., Deng H.C., Wang Z.G., Wang X.D., Cao Q., Cheng D., et al. (2023). Characteristics and Factors Influencing Pore Structure in Shale Oil Reservoirs of Different Lithologies in the Jurassic Lianggaoshan Formation of the Yingshan Gas Field in Central Sichuan Basin[J]. Mineral, 13(7), 958.
- 15. Hou Y.D., Jia Y.Z., Ge Z.L., Zhou Z., Zhang D., Fu W.Y., et al. (2024). Mineralogical control on the dissolution induced microscopic pore alternations of high-rank coal during CO2 sequestration[J]. Geoenergy Science and Engineering, 234, 212614.
- 16. Jiang W.P., Zhang Q., Zeng F.G. (2020). Influence characteristics and its mechanism of minerals in coal on methane generation and storage[J]. Journal of China Coal Society, 45 (S2): 904–911.
- 17. Xu Y., Lun Z.M., Hu Z.C., Guo J.Q., Zhou X., Wang H.T., et al. (2023). Alterations in pore and fracture structure of coal matrix and its surrounding rocks due to long-term CO2-H2O-rock interaction: Implications for CO2-ECBM[J]. Fuel, 352, 129104.
- 18. Kang J.Q., Fu X.H., Cheng M., Zhang B.X., Qi Q., Chen Z.Y. (2024). The Difference between Stress Deformation and Adsorption Deformation of Coal Reservoirs and Its Role in Permeability Prediction[J]. ACS Omega, 9(22), 23998–24008
- 19. Huang Q., Du Z.G., Liu H.W., Niu Q.H., Fang H.H., Yang J.Y., et al. (2023).Investigation of cleat and micro-fracture and its aperture distribution in the coals of different ranks in North China: Relative to reservoir permeability[J]. Frontiers in Earth Science, 10:1048042.
- 20. Wang B., Wu Y.T., Li G.F., Yang J.S., Zhao Y., Wang M.Z., et al. (2022). Laboratory study of fracture permeability of mineral-filled coal from Fanzhuang Block, southern Qinshui Basin, China[J]. Journal of Petroleum Science and Engineering, 208, 109799.
- 21. Yi C., Chen J.B., Han W., Li Z.Y., Hai H., Suo G.X. (2023). The Influence Law and Mechanism of CO2 Injection Pressure on Coal Rock Pore-Fracture Characteristics[J]. Chemistry and Technology of Fuels and Oils, 59(3), 638–646.
- 22.
Smith D.H., Jikich S.A., Seshadri K. (2007). Carbon Dioxide Sorption Isotherms and Matrix Transport Rates for Non-Powdered Coal[J]. Proc., International Coalbed Methane Symposium, Tuscaloosa, Alabama, USA.
- 23. Liu P., Nie B.S., Zhao Z.D., Zhao Y.L., Li Q.G. (2023). Characterization of ultrasonic induced damage on multi-scale pore/fracture in coal using gas sorption and μ-CT 3D reconstruction[J]. Fuel, 332: 126178.
- 24. Li S., Tang D.Z., Xu H., Yang Z. (2012). Advanced characterization of physical properties of coals with different coal structures by nuclear magnetic resonance and X-ray computed tomography[J]. Computers & Geosciences, 48, 220–227.
- 25. Zhou D., Feng Z.C., Zhao D., Zhao Y.S., Cai T.T. (2017). Experimental study of meso-structural deformation of coal during methane adsorption-desorption cycles[J]. Journal of Natural Gas Science and Engineering, 42: 243–251.
- 26. Zhou D., Liu Z.X., Feng Z.C., Shen Y.X., Wu Y.C. (2020). The study of the local area density homogenization effect of meso-structures in coal during methane adsorption[J]. Journal of Petroleum Science and Engineering, 191: 107141.
- 27. Yu Y.B., Wang J.H., Cui W.T., Gao C.W., Chen Y.T., Yuan Z.H. (2024). Imbibition characteristics and mechanisms of coal using digital image subtraction technique based on Micro-CT[J]. Fuel, 365: 131240.
- 28. Voegeli W., Takayama H., Liang X.Y., Shirasawa T., Arakawa E., Kudo H., et al. (2024). Multibeam X-ray tomography optical system for narrow-energy-bandwidth synchrotron radiation[J]. Applied Physics Express, 17(3), 032002.
- 29. Sakurai H., Hoshi K., Watabe D., Suzuki K., Hasebe Y., Suzuki S., et al. (2024). Local state of charge measurement on a Li battery by X-ray photon counting computed tomography[J]. Nuclear Instruments & Methods In Physics Research Section B-beam Interactions with Materials and Atoms, 557, 165537.
- 30. Ren J.R., Zhang W.K., Wang Y.Z., Liang N.N., Wang L.Y., Cai A.L., et al. (2024). A dual-energy CT reconstruction method based on anchor network from dual quarter scans[J]. Journal of X-ray Science and Technology, 32(2), 229–252.
- 31. Meng Q.Q., Li J., Hu B.R., Zhang X.B., Wang S.C., Shi X.M., et al. (2024). Investigation of optimal combination of monochromatic image of dual-energy CT system for proton range calculation[J]. Radiation Physics and Chemistry, 217, 111526.
- 32. Wang H, Yue G.W, Yue J.W, Li M.M, Zheng X.J. (2019). Permeability of coal associated with cleat and bedding structure: measurement and modeling[J]. Energy Sources, Part A: Recovery, Utilization, and Environmental Effects,
- 33. Wang K, Pan H.Y, Fujii Y. (2024). Study on energy distribution and attenuation of CO2 fracturing vibration from coal-like material in a new test platform[J]. Fuel, 356:129584.