Natural Maturation-Induced Changes in Molecular Structure and Associated Micropores of Kerogen in Shale: Implications for Geological Storage of Carbon Dioxide

Abstract

Kerogen micropores in shale are crucial for gas sorption; nevertheless, their formation and evolution have been insufficiently studied. To qualitatively and quantitatively analyze changes in the molecular structure and associated micropores of kerogen throughout natural maturation, a series of experiments were performed on collected kerogen samples with varying maturity levels. The results demonstrated that the evolution of the chemical structure and associated micropores of kerogen can be divided into three stages during thermal evolution. When Ro < 1.43%, micropores are primarily controlled by the aliphatic composition of kerogen, resulting in a reduction in the micropore volume with decreasing numbers of aliphatic functional groups. When 1.43% < Ro < 1.93%, the micropore structure is primarily governed by the aromatic composition of kerogen and the micropore volume exhibits a gradual increment with increasing aromatic ring size. When Ro > 1.93%, the micropore volume rapidly increases with increasing aromatic composition of kerogen, and the aromatic rings become more orderly. Moreover, molecular simulations also show that the adsorption curves of CO₂ and CH₄ are type I isotherms, and that the adsorption capacity of CO₂ is greater than that of CH₄ under the same conditions, which indicates that the microscopic adsorption mechanism of CH₄ and CO₂ is gas competitive adsorption. This research offers new perspectives on the formation and evolution of kerogen micropores, and the carbon sequestration process.

1. Introduction

The environmental problems caused by excessive CO₂ emissions are seriously threatening the survival of all humanity [1]. The shale and coal reservoirs have been proposed as potential targets for long-term CO₂ sequestration [2,3]. Moreover, injecting CO₂ into shale reservoir also could displace adsorbed CH₄, thereby enhancing shale gas recovery (CO₂-ESGR) [4,5]. Micropores related to kerogen in shale are important components of shale pore systems and the primary sites for CH₄ desorption and CO₂ sequestration, with its structure mainly determined by the chemical structure evolution of kerogen [6,7,8,9]. However, there remains insufficient knowledge of the detailed changes in the molecular structure and associated micropores of kerogen during the maturation process, as well as the competitive adsorption mechanism of CH₄ and CO₂ in kerogen.

Various analytical techniques have been used to elucidate the changes in the kerogen structure, including Fourier-transform infrared (FTIR) and ¹³C solid-state nuclear magnetic resonance (¹³C SSNMR) spectroscopy [10,11,12,13,14,15]. The relative abundance of various functional groups that include carbon, hydrogen, and oxygen can be detected in FTIR spectra (e.g., -C-H, O-H, C=C, and C=O). Moreover, ¹³C SSNMR generally provides information regarding the bond environment and connectivity of aromatic carbons, aliphatic carbons, and their subgroups [10,11,14]. High-resolution transmission electron microscopy (HRTEM) is another practical technique that provides quantitative and qualitative details regarding the aromatic composition of kerogen. Compared with other spectral analysis methods, HRTEM can be used to directly observe the nanoscale structure of carbonaceous materials; hence, it is extensively utilized in the examination of molecular structure alterations in kerogen during thermal evolution [6,16,17,18]. In recent years, scientists have come to understand that kerogen comprises different large molecular components, leading to the use of diverse methods to examine the alterations in the molecular composition of kerogen [10,11,15]. Kerogen-associated micropores in shale reservoirs are important for shale gas/oil adsorption and migration [19,20,21,22], significantly contributing to the total pore volume and total specific surface area [23,24]. Moreover, these micropores are directly related to thermal maturity levels and are controlled by the type and molecular structure of kerogen [6,7,25]. However, the relationships between thermal maturity levels, molecular structure, and kerogen-associated micropores during thermal evolution remain insufficiently understood.

Previous researchers have always conducted physical experiments to investigate the competitive adsorption behavior of CO₂ and CH₄ under in situ conditions [4,5]; however, these experiments do not reflect the conditions of real shale gas reservoirs. Considering that commercial shale gas reservoirs are always buried at more than 2000 m depth, previous studies on CH₄ and CO₂ comparative adsorption are not sufficient to accurately, directly, and comprehensively represent the competitive adsorption behavior in practice [26,27]. Nowadays, researchers mainly use molecular simulations to study competitive adsorption and have achieved fruitful results [1,3]. It is worth mentioning that, although many efforts have been made to ensure the rationality of kerogen molecular structure, it is necessary to validate these structures by comparing simulated adsorption isotherms with experimental data. In fact, the microscopic adsorption mechanisms of CH₄ and CO₂ in relation to gas competitive adsorption behavior on different mature kerogens have not been investigated.

Herein, we collected kerogen samples of varying maturity levels and qualitatively and quantitatively analyzed the chemical structure via FTIR, ¹³C SSNMR, and HRTEM. And CO₂ adsorption experiments were conducted to characterize the micropore properties of the kerogen samples. Based on the collected data, changes in the molecular and pore structures of kerogen during thermal evolution were comprehensively analyzed. Finally, molecular simulations were implemented using the Materials Studio software 2020 (MS simulation, Hohhot, China, Inner Mongolia University of Technology) to investigate the microscopic mechanisms of CH₄ and CO₂ comparative adsorption. This research offers new perspectives on CO₂ injection into shale gas reservoirs to improve CH₄ recovery and the carbon sequestration process.

2. Samples and Methods

2.1. Sample Preparation

To enhance experimental comparability, seven shale samples with varying maturity levels were collected from the Nanjiang Formation (NJ-1) in the upper Cretaceous (located in Songliao Basin), the Dalong Formation (DL-1) in the lower Permian, the Longmaxi Formation (LJ-1, HP-17, ZP-16, BL-26) in the lower Silurian, and the Lujiaping Formation (LT-25) in the lower Cambrian (located in Sichuan Basin) (Figure 1). The purple–red and green continental mudstones from the NJ Formation were deposited in deep and semi-deep lacustrine facies during the depressional stage of the basin [28], and a major sea-level transgression stage led to the deposition of organic-rich, black, marine shale in deep-water shelf facies in the DL, LMX, and LJP Formations [29,30,31,32]. The shale samples exhibited total organic carbon levels of > 2.00%, with some reaching as high as 10%. In addition, thermal maturity levels significantly differed among samples, with vitrinite reflectance (R_o) values ranging from 0.89% to 3.03%. Samples from the NJ and DL Formations had low maturity levels, while those from the LMX and LJP Formations had high maturity levels. Kerogen extraction from these shale samples was performed using chemical techniques described by Lis et al. [33].

2.2. Raman Spectroscopy

RS analysis of the isolated kerogen samples was performed as described in Song et al. in the 45–4500 cm⁻¹ region with a 532 nm wavelength and 2 s integration time using a Sentera Raman spectrometer (Bruker, Berlin, Germany) [34]. The laser was adjusted to a power of 5 milliwatts to avoid the occurrence of coking [35]. The experiment was repeated five times to ensure the representative results, and R_o values were calculated from RS spectra using the method proposed by Liu et al. [36].

2.3. FTIR Spectroscopy

FTIR experiments were performed on a Vertex 80v FTIR device (Bruker, Xuzhou, China, China University of Mining and Technology). Prior to analysis, each kerogen sample was pulverized to a fine powder (<200 mesh) and combined with KBr (60 mg) for sample preparation. Then, the blend was positioned on a vacuum-sealed disk measuring approximately 10 mm in diameter. The FTIR spectrum of the disk sample was recorded at a resolution of 0.06 cm⁻¹ in the 500–4000 cm⁻¹ wavenumber range.

2.4. Solid-State ¹³C SSNMR

¹³C SSNMR analysis was performed using an Avance III 400 spectrometer (Bruker) with operating frequencies of 400 MHz 1H and 100 MHz ¹³C. Each kerogen powder sample (<200 mesh) was loaded into a dual-resonance probe with a 4 mm ZrO₂ rotor and centrifuged at 5 kHz. The ¹³C cross-polarized magic-angle spinning (CP-TOSS/MAS) technique was employed to obtain the kerogen structure and suppress sidebands [37,38]. The SSNMR experiment was carried out at a speed of 4 kHz and contact time of 1 ms.

2.5. HRTEM

HRTEM has the capability to visualize the multiscale organization of kerogen at the nanometer level. In this study, HRTEM tests were carried out using a G2 F20 Field Emission Electron Microscope (Tecnai, Hillsboro, OR, USA) equipped with an advanced aberration corrector and CCD camera. Prior to observation, the TEM grid was prepared as follows: The kerogen powder was crushed and dispersed in absolute ethanol, and the resulting suspension was placed on a high-quality 200-mesh copper microgrid and left to dry. In order to avoid possible interference caused by the support film of quasi-amorphous carbon and overlapping kerogen particles, the particles were imaged in a thin area of <10 nm.

Following the procedure proposed by Liu et al. [23], image analysis was performed on several representative HRTEM images of each kerogen sample. Adobe Photoshop software 2020 (Adobe Systems Incorporated, San Jose, CA, USA) was utilized to process the image by adjusting the threshold to reduce the background noise. Subsequently, the lattice fringe coordinate data were extracted using Mapgis 6.7 software (China University of Mining and Technology, Xuzhou, China), enabling the calculation of fringe length and fringe angle distribution. Over the past decade, Mathews and coworkers have developed, adapted, and applied many advanced HRTEM lattice fringe analysis techniques [16,39,40,41,42], with the present study referring to the assignments of different lattice fringes proposed by these researchers. More detailed information about these assignments can be found in the study by Van Niekerk and Mathews [40].

2.6. Low-Pressure CO₂ Physisorption Analyses

Low-pressure gas physisorption analysis is commonly used to evaluate the pore structure of shale reservoirs [19,43]. Following the procedure outlined by Shang et al. [44], after initial evacuation under high vacuum at approximately 80 °C for >10 h, the CO₂ adsorption isotherms of isolated kerogen samples were measured at 0 °C (273 K) using an ASAP 2020 HD88 analyzer (Micromeritics Instrument Corp., Norcross, GA, USA). To investigate the distribution of micropore size in the kerogen samples, the CO₂ isotherm data were analyzed using the density functional theory (DFT) approach [45,46].

2.7. Molecular Simulation

In this work, we selected three kerogen samples of different maturity levels and established the three-dimensional molecular structure based on derived structure parameters from the FTIR spectroscopy and ¹³C NMR spectroscopy using the method proposed by Liu et al. [47]. We then simulated the isothermal adsorption behavior of CO₂ and CH₄ in the kerogen model use the Sorption module in MS software 2020. Regarding the simulation details, the grand canonical Monte Carlo (GCMC) simulations with a COMPASS force field were used. For each test condition, we selected ultra-fine quality and performed 1 × 10⁶ equilibrium steps to ensure the adsorption state, while adopting 1 × 10⁷ steps in the production stage to quantify gas adsorption capacity, with Fugacity steps set to 10. Forcefield assigned was used to calculate charges (the charge distribution of the molecular force field). Atom based was chosen to calculate electrostatic and van der Walls forces, but due to different molecular polarities, Ewald and Group was chosen for calculating electrostatic forces when adsorbing CO₂. The pressure and temperature ranges were 0.001–15 MPa and 0–100 °C, representing the conditions of shale gas reservoirs.

3. Results and Discussion

3.1. Changes in the Molecular Structure of Kerogen

RS is widely recognized for its ability to analyze the chemical composition of geological carboniferous substances [13,17,35]. Thermal evolution is often considered to have a strong correlation with Raman parameters (i.e., separation between the D and G bands), which is particularly suitable for evaluating the maturity levels of organic matter in shale with R_o = 0.50–3.50%. Accordingly, RS was employed herein to determine the maturity levels of various kerogen samples. The Raman spectra of kerogen samples exhibited the typical presence of D1 and G broad bands, located at approximately 1345 and 1590 cm⁻¹, respectively (Figure 2). Separation between the D and G bands gradually increases during thermal evolution, which is the basic principle underlying the use of the extent of separation to calculate maturity levels, as proposed by Liu et al. [36]. According to the thermal evolution stages proposed by Ko et al. [21], the NJ-1 and LJ-1 samples were in the low-maturity stage, while the LJ-1, HP-17, ZP-16, BL-26, and LT-25 samples were in the over-mature stage.

Figure 3 shows the FTIR spectra and ¹³C SSNMR curves of kerogen samples with different maturities. Several typical absorbance peaks exist in the FTIR spectra of the kerogen samples: aliphatic -C-H at 2920 cm⁻¹, 2850 cm⁻¹ and 1450 cm⁻¹; aromatic C=C at 1700 cm⁻¹ and 1600 cm⁻¹: and aromatic out-of-plane -C-H at 700–900 cm⁻¹ (Figure 3a). For the kerogen samples with low maturity (NJ-1 and DL-1), peaks in the region 2920 cm⁻¹ and 2850 cm⁻¹ are presented as these kerogen samples contain some aliphatic side chains.

In addition, the ¹³C SSNMR spectra show that for the NJ-1 samples, the area of peaks in the aliphatic parts are greater than those in the aromatic parts, which also implies that more aliphatic carbons exist within the kerogen structure. With the continuous hydrocarbon generation (R_o from 1.0% to 3.03%), the intensities of the aliphatic -C-H (2920 cm⁻¹, 2850 cm⁻¹ and 1450 cm⁻¹) peak in the FTIR spectra, and the aliphatic parts (0–90 ppm) in the ¹³C SSNMR spectra nearly disappeared, which implies that a large amount of aliphatic functional groups and oxygen-containing groups were tailored downward from the kerogen macromolecular structure.

¹³C SSNMR parameters were calculated through the peak fitting method [38,48]. Based on Figure 4, different types of carbon atoms in the molecular structure of kerogen usually evolve in a two-stage process during thermal evolution, and the jump point is about 1.43% R_o. It is clear that aliphatic carbon parameters ${\mathrm{f}}_{\mathrm{al}}^{\mathrm{H}}$ (fraction of aliphatic CH₄ or CH₂) and ${\mathrm{f}}_{\mathrm{al}}^{*}$ (fraction of aliphatic CH₃ or quaternary carbon) both decrease first and then vary slightly with an increase in maturation, suggesting that aliphatic structures fall off from the chemical structure of kerogen during the thermal evolution.

Herein, we innovatively introduced the use of HRTEM to quantitatively characterize the aromatic rings of the kerogen structure, and combined it with spectral analyses to probe alterations in the kerogen molecular structure during natural evolution. The kerogen structure evolution can be divided into three stages based on the aforementioned results for R_o, ranging from 0.89% to 3.03% (Figure 5). For 0.89%< R_o < 1.43%, the aliphatic chains, naphthalene, and 2 × 2 rings experience rapid cracking from the kerogen structure, which was evidenced by a rapid decrease in the intensity of peaks at 20–50 ppm in the ¹³C SSNMR spectra and at 2850 and 2920 cm⁻¹ in the FTIR spectra (Figure 3a). These trends were also supported by the HRTEM images in which the percentages of smaller aromatic rings greatly decreased within this interval, indicating the breakage of naphthalene and 2 × 2 rings from the kerogen structure (Figure 4). Previous studies have also confirmed that the significant decrease in aliphatic carbon content during this stage (0.89% < R_o < 1.76%) is accompanied by the generation of substantial amounts of C₁₄₊ oils and C_6–14 hydrocarbons [49,50].

For 1.43% < R_o < 1.93%, small aromatic ring-connected aliphatic structures formed into 3 × 3 rings through aromatization, as evidenced by the HRTEM images showing that the percentages of small-size aromatic rings (naphthalene and 2 × 2 rings) continued to decrease while that of 3 × 3 rings kept increasing. Meanwhile, the aliphatic carbon content continued to decrease, according to the SSNMR data, indicating a shedding of the aliphatic side chains at this time (Figure 3b). Previous studies also observed that kerogen molecular composition, structure, and properties change significantly when Ro = 1.8~2.0% and that the aliphatic components are mainly methyl groups connected to aromatic rings, with only small quantities of aliphatic chains [7,8,51].

For 1.93% < R_o < 3.03%, the hydrocarbon generation potential of kerogen was nearly nonexistent, but the aromatic carbon content varied dramatically. In this maturation period, 3 × 3 rings formed large aromatic rings through condensation, and their orientation was significantly enhanced. The HRTEM images also demonstrated that the 3 × 3 ring content declined, while that of larger aromatic rings (>3 × 3 rings) increased. In fact, Hou et al. believed that the kerogen structure evolution at this stage can be interpreted as molecular structural adjustment between aromatic rings. Hence, based on the above discussion, we speculate that kerogen slowly transforms into a turbostratic-like structure.

3.2. Relationship Between Chemical Structure Changes and Micropore Evolution During Maturation

Figure 6 illustrates the CO₂ adsorption isotherms and pore volume change curves of kerogen samples of various maturities. It was found that the amounts of adsorbed CO₂ differ among different kerogen samples (Figure 6a). LT-25 (R_o = 3.03%) has the largest CO₂ adsorption amount, and NJ-1 (R_o = 0.89%) has the smallest CO₂ adsorption amount. In reality, the relationship between CO₂ adsorption amount and maturity was relatively complex, especially in the mature stage (R_o = 0.89~1.93%), which is closely related to the differential development of organic matter pores at various stages. Moreover, the variations in micropore volume were plotted against Ro, as shown in Figure 6b. From R_o = 0.89% to 3.03%, the micropore evolution can be divided into three stages: <1.43%, 1.43–1.93%, and 1.93–3.03%. When R_o < 1.43%, micropore volume decreases during the thermal evolution. When R_o = 1.43% to 1.93%, micropore volume gradually increases with increasing R_o. When R_o = 1.93% to 3.03%, micropore volume increases rapidly with increasing R_o. Moreover, it can be found that the micropore volumes of NJ-1 were relatively small and did not follow the trend of the first stage, perhaps due to differences in sources of organic matter [19,25,52,53].

To analyze the relationship between micropore parameters and kerogen composition, the variations in carbon atom content were plotted against the micropore parameters (Figure 7). Relatively few of the collected samples possessed a high aliphatic carbon content; hence, the correlation was not significant at high aliphatic carbon content (>50%) (Figure 6a). However, Duan et al. observed that aliphatic carbon content primarily controls kerogen-associated micropores within this interval, indicating that massive shedding of the aliphatic structures would lead to a significant reduction in the number of micropores [7]. When R_o was >1.43%, the aliphatic carbon content of kerogen was reduced, thus weakening the controlling effect of aliphatic structures on micropores, while the influence of other components on micropores was strengthened. These findings aligned with the outcomes of prior research conducted on coal and other kerogen samples [6,14].

Variations in the aromatic carbon content and micropore parameters of kerogen samples with different maturity levels are shown in Figure 7b. When the aromatic carbon content was relatively low (few data), the correlation between aromatic carbon content and micropore parameters was not significant, but previous studies have observed that micropore volume decreases with increasing aromatic carbon content at this stage [7,24]. Since kerogen molecules are mainly composed of aliphatic carbons, the influence of aromatic structures on micropore parameters was masked by the significant loss of aliphatic structures. The aromatic carbon content continuously increased as thermal evolution continued, leading to changes in micropore volume that could be categorized into two distinct stages. During the first stage (aromatic carbon content of 45–58%), the micropore volume of kerogen generally increased with increasing aromatic carbon content, suggesting that these micropores were initially controlled by aromatic structures. In the second stage (aromatic carbon content of 58–70%), the micropore volume significantly increased with increasing aromatic carbon content, which may have been closely related to the formation of larger aromatic rings via the condensation of 3 × 3 rings.

Aromatic carbons comprise three distinct categories: alkylated aromatic carbons (${\mathrm{f}}_{\mathrm{a}}^{\mathrm{S}}$), aromatic bridgehead carbons (${\mathrm{f}}_{\mathrm{a}}^{\mathrm{B}}$), and protonated aromatic carbons (${\mathrm{f}}_{\mathrm{a}}^{\mathrm{H}}$). Figure 8a–c show the direct relationship between micropore volume and protonated aromatic carbons, particularly when the aromatic structures comprised >50% of the overall composition. In contrast, there was not a clear relationship between micropore volume and alkylated aromatic carbon and aromatic bridgehead carbon contents (Figure 8a,b). In fact, low-maturity kerogen with a high alkylated aromatic carbon content usually also contained a large number of aliphatic structures, suggesting that aliphatic structures continued to mainly control the micropore volume at this stage. This may be responsible for the lack of correlation between alkylated aromatic carbon content and micropore volume. As thermal evolution progressed, the percentage of protonated aromatic carbons continuously increased. When the protonated aromatic carbon percentage was >30%, micropore volume increased with increasing protonated aromatic carbon percentage (Figure 8c), indicating that protonated aromatic carbons contributed to increased micropore abundance.

3.3. Microscopic Mechanism of Methane Displacement by Carbon Dioxide

In the past, some studies have found that kerogen mainly contains various elements such as carbon, hydrogen, oxygen, nitrogen and sulfur [53,54]. However, some scholars have also pointed out that the content of nitrogen and sulfur elements in the kerogen of marine shale in southern China is relatively low, and can even be ignored [10,11,12,13]. In our study, based on ¹³C NMR and FTIR data, the carbon chains and chemical structure fragments of kerogen samples were combined and analyzed. The ¹³C NMR spectrograms of the model were calculated and plotted, using CHNMRPRO software, and compared with experimental values until they were almost identical. The average molecular structures of three shale kerogen samples were constructed, and their average molecular structure chemical formulas were as follows: NJ-1, C₁₅₃H₂₄₂O₃₀; LJ-1, C₁₆₉H₁₆₄O₃₁; and BL-26, C₂₁₀H₁₁₈O₄₂ (Figure 9a–c). For the low-maturity kerogen sample, the main form of carbon content was long aliphatic rings, while the high-maturity kerogen sample was mainly composed of aromatic cluster structures with more than five aromatic rings, which is consistent with the findings reported in Section 3.1.

Due to the predominance of micropores in shale OM pores, in order to characterize the pore structure of shale kerogen as much as possible, the size of the kerogen unit cell model constructed should also conform to the micropore pore size. In this study, the Amorphous Cell module in Materials Studio software 2020 was used to construct kerogen unit cell models of NJ-1, LJ-1, and BL-26, from 20, 12, and 9 kerogen molecules, with side lengths of 39.3, 35.7, and 39.6 nm, respectively. Subsequently, the Forcite module was used to optimize the geometric configuration of the model, with Task selected as Geometry Optimization, Quality set to Ultra fine, Max. iterations set to 20,000 times, and Forcefield selected as COMPASS. Unreasonable parameters such as key length and key angle in the initial model were corrected to obtain the model with the lowest local energy. Finally, the molecular structure was annealed using the Forcite module, where Task was selected as Anneal, Quality was selected as Ultra fine, Annealing cycles were set to 20 times, initial temperature was set to 300 K, and the highest temperature in a single temperature cycle was set to 800 K. Forcefield was selected as COMPASS, and Ensemble was selected as the NVT. By annealing, the model is freed from local energy minima, resulting in a more stable global minimum energy structure. Based on the established three-dimensional macromolecular model (Figure 10), CO₂ molecules were used as probes to calculate the internal pores of the kerogen samples. Moreover, the pore volume and specific surface area of NJ-1, LJ-1, and BL-26 were 6.40 ų and 31.28 Ų, 37.71 ų, and 194.26 Ų, and 110.50 ų and 318.05 Ų, respectively. The pore volume and specific surface area increased with increasing maturity, which is consistent with the findings reported in Section 3.2. In addition, previous studies have also observed similar changes and proposed that the macromolecular structure of kerogen can control the development of micropores [6,8].

In the actual process of CO₂-ESGR, there is competitive adsorption of CO₂ and CH₄ on the kerogen surface [1,2,3,4]. In our study, we simulated the adsorption characteristics of CO₂ and CH₄ under different temperature and pressure conditions using the three-dimensional macromolecular structure of kerogen. (Figure 11). Compared to the CO₂ adsorption experiment in Figure 5, the CO₂ adsorption capacity in this simulation experiment is significantly higher. This is because the experimental value is set at a lower pressure, which may not have reached saturation adsorption. Another reason is that simulation experiments are based on the adsorption capacity under ideal conditions, while actual experiments are influenced by factors such as sample purity.

Both the adsorption curves of CO₂ and CH₄ can be categorized as type I isotherms; specifically, the adsorption capacity initially increases significantly, and then slightly increases under high pressure. In addition, the amount of CO₂ adsorbed by kerogen is several times that of CH₄ under the same experimental conditions; hence, the adsorbed CH₄ in shale is displaced by CO₂ and transformed into free CH₄ in the displacement process. Actually, previous researchers have also observed similar results and proposed that they are related to the pore structure and surface chemistry groups of kerogen, such as the aromatic rings and oxygen-containing functional groups [54,55,56,57,58,59,60,61,62]. In our study, we have observed that the as the number of micropores and specific surface area of kerogen increase during the thermal evolution process, the adsorption capacity of CO₂ and CH₄ gradually increases, which also indicates that micropores in kerogen are the key control for the displacement behavior (Figure 11). In addition, some studies have observed that aromatic rings of coal have a stronger ability to adsorb CO₂ than CH₄, and attribute this phenomenon to the difference in molecular configuration and charge distributions, specifically because the electrons and orbits of CO₂ molecules are more likely to interact with the electrons of aromatic rings [57,62]. Other studies also have observed that the polarity of oxygen-containing functional groups can change the charge distribution of molecules, increase the surface polarity of kerogen, enhance the electrostatic interaction with CO₂, and reduce the dispersion force with CH₄ [56,59]. Therefore, before the implementation of a CO₂-ESGR project, the pore and molecular structure of the kerogen should be comprehensively considered.

4. Conclusions

Here, the changes in the molecular structure and related micropores of kerogen during natural maturation were investigated. The following conclusions can be drawn:

(1)

For R_o <1.43%, the kerogen structure is mainly composed of aliphatic chains, naphthalene (56.3–10.1%), 2 × 2 rings (26.4–41.1%), and 3 × 3 rings (6.7–41.4%). For 1.43% < R_o < 1.93%, the proportions of smaller structures (naphthalene and 2 × 2 rings) continuously decrease, while those of 3 × 3 rings and larger structures increase. For R_o >1.93%, over-mature kerogen has lower proportions of naphthalene and 2 × 2 rings and higher percentages of larger structures, such as 4 × 4 and 5 × 5 rings. In addition, the orderliness of aromatic stripes gradually increases with increasing maturity levels.

(2)

The molecular structure evolution of kerogen controls the development of its micropores. When 0.89% < R_o < 1.43%, the micropore volume is significantly reduced as aliphatic chains, naphthalene, and 2 × 2 rings rapidly crack from the kerogen structure. When 1.43% < R_o < 1.93%, the micropore volume gradually increases with increasing maturity levels because small aromatic ring-connected aliphatic structures form 3 × 3 rings through aromatization. When 1.93% < R_o < 3.03%, 3 × 3 rings form larger aromatic rings through condensation and their orientation is significantly enhanced (reaching 80% of the major direction), leading to rapid growth in micropore volume and specific surface area.

(3)

Both the adsorption curves of CO₂ and CH₄ are type I isotherms; specifically, the adsorption capacity initially increases significantly and then slightly increases under high pressure. The amount of CO₂ adsorbed by kerogen is several times that of CH₄ under the same experimental conditions; hence, the adsorbed CH₄ in shale is displaced by CO₂ and transformed into free CH₄ in the displacement process. In addition, functional groups are also important factors affecting the process of CO₂ displacement of CH₄.