CO2 pre-fracturing technology represents a novel approach to CO2 fracturing, effectively enhancing carbon capture efficiency in shale reservoirs while increasing the recovery of hydrocarbon resources. This study systematically quantifies the impact of water injection on the pore structure and permeability of shale samples saturated with pre-injected CO2. Based on X-ray diffraction (XRD) and low-temperature nitrogen adsorption (LT-NA) results, the dominant mechanism of the swelling-chemical coupling effect in shale property is clarified. Pre-injected CO2 can effectively mitigate the impact of water absorption. After a reaction time of 7 days, the permeability of the CO2-involved hydrated shale samples is four times that of samples without CO2. Nevertheless, the influence of subsequently injected water on shale permeability persists, resulting in an 80% reduction in shale permeability. XRD and LT-NA results indicate that the swelling-chemical coupling effect is the dominant factor in shale property variation during CO2 pre-fracturing. In the initial stage, the decline in calcite and clay mineral content is significant, and chemical dissolution dominates the change in pore structure. As the reaction progresses, the intensity of chemical reactions weakens, and clay mineral swelling becomes the primary factor affecting the shale properties. In this stage, K+ generated from original minerals effectively inhibits clay mineral swelling. Therefore, the swelling-chemical coupling effects should be comprehensively considered during the CO2 pre-fracturing process, and an appropriate soaking time should be selected to maximize CO2 storage efficiency and oil and gas production.

The relentless global increase in energy demand, driven by population growth, industrialization, and technological advancements, has necessitated the exploration of more efficient and sustainable methods for extracting hydrocarbons from unconventional reservoirs. Unconventional reservoirs, such as shale gas and oil, are characterized by their low-permeability and complex geology, presenting unique challenges for hydrocarbon extraction.1–4 Traditional hydraulic fracturing has been instrumental in unlocking these unconventional reservoirs by creating fractures through high-pressure injection of water, sand, and chemicals.5,6 However, hydraulic fracturing has raised concerns related to water usage, necessitating the exploration of alternative, more sustainable methods.7 In recent years, CO2 based fracturing method has emerged as a promising and environmentally sustainable solution for enhancing hydrocarbon recovery (HER) from unconventional reservoirs.8 This approach not only addresses the challenges posed by traditional hydraulic fracturing but also makes significant strides toward reducing greenhouse gas emissions through carbon capture and storage (CCS) practices.

However, CO2 poses certain challenges in fracturing due to its low viscosity and limited sand-carrying capacity. Therefore, the CO2 pre-fracturing technique, composed of CO2 and water-based fracturing fluids, has been proposed to address the aforementioned issues. The CO2 pre-fracturing method primarily involves two stages: (1) Pre-injection of CO2: Initially, liquid CO2 is injected into the reservoir, reaching a supercritical state underground (7.38 MPa, 31.1 °C). The introduction of CO2 effectively reduces the breakdown pressure and creates a complex fracture network, thereby increasing the stimulated reservoir volume (SRV). (2) Injection of water-based fracturing fluid: Following the CO2 injection, water-based fracturing fluid is introduced into the reservoir to further extend the fractures. Subsequently, proppants are added to the water-based fracturing fluid to enhance fracture conductivity. Due to the unique physical and chemical properties of CO2, pre-injected CO2 offers several advantages that can enhance oil and gas production. CO2 has been applied as an enhanced hydrocarbon recovery agent for decades.9,10 The injection of CO2 into unconventional reservoirs promotes increased oil and gas recovery by leveraging several distinct mechanisms, including swelling effect, hydrocarbon mobility enhancement, and pressure maintenance Moreover, the pre-injected CO2 can effectively facilitate the flowback of subsequent injected water-based fracturing fluids, enhancing flowback efficiency.11,12 The injection of high-pressure CO2 into the shale reservoir is conducive to the efficient sequestration of CO2 and mitigates the environmental impact of the greenhouse effect.13–15 

The low viscosity of CO2 enables it to infiltrate deep into shale reservoirs, where it engages in extensive reactions with the shale matrix, resulting in substantial alterations to shale matrix characteristics, including pore distribution, permeability, and wettability.16 These changes not only affect the production of hydrocarbons trapped in the shale but also hold significant importance for the efficient storage of CO2 in the shale formation. Zhou et al.17 conducted a comprehensive analysis of the alterations in shale pore structure induced by CO2–water–shale reactions under varying reaction pressures. Their findings revealed that as pressure increased, the specific surface area (SSA) and pore volume (PV) of shale samples gradually decreased, while the average pore size increased, resulting in enhanced pore connectivity. In addition, additives in water-based fracturing fluid also affect the degree of water-rock reaction. Lu et al.18 postulated that the network structure of polymers in slickwater could diminish pore connectivity, impeding the dissolution and penetration of CO2. In fact, the adsorption of both CO2 and water within shale pores has been shown to cause shale swelling in several studies.19–21 However, the influence of external water on the pore structure of CO2-saturated shale is unclear and requires further investigation.

The shale permeability post-CO2 treatment exhibits varying trends, influenced by factors such as mineral composition and formation conditions.22–24 Throughout the CO2 pre-fracturing process, CO2 initially interacts with the matrix near the fractures and diffuses into the depths of the shale formation. Subsequently, water-based fracturing fluid is introduced, further engaging with the CO2-saturated shale. This process leads to alterations in the pore structure and mineral composition of the shale matrix within the SRV, resulting in changes in shale permeability. It is imperative to comprehend the variations in shale matrix permeability under different durations of water exposure. Currently, the primary mechanisms for assessing shale permeability rely on both steady-state and unsteady-state theories.25 Tian et al.22 carried out research on changes in permeability and porosity in dry and wet shale subjected to CO2 treatment, using the Darcy equation based on steady-state methodology to ascertain alterations in shale permeability. Compared to the steady-state method, the pulse decay permeability (PDP) and pressure transmit (PTT) devices for permeability testing based on the non-steady-state method require shorter durations, leading to higher efficiency. Zou et al.23 conducted research on changes in shale permeability under the influence of brine and CO2. Their findings demonstrated that increasing soaking time, environmental temperature, and reaction pressure positively contributed to enhanced shale permeability. Lufeng et al.26 conducted a study on the formation damage caused by fracturing fluid invasion in the tight reservoir, and a PTT instrument was applied to evaluate the variation of permeability. Liang et al.27 used a PTT device to evaluate sandstone permeability under different working fluids treatment.

It is evident that the shale property changes under the influence of CO2 and water are relatively complex.17,18,22,28–30 Understanding the changes in shale matrix permeability and pore structure during the CO2 pre-fracturing process is crucial for CO2 geo-storage and enhancing hydrocarbon recovery (EHR). Currently, the properties of shale directly exposed to CO2 have been extensively studied. However, during the CO2 pre-fracturing process, the evolution pattern of matrix properties within the SRV has not been adequately explored, giving rise to several unresolved issues: (1) CO2 is initially injected into the shale bed, followed by water-based hydraulic fluid. The permeability and pore structure changes of the shale under such a saturation sequence are not fully understood. (2) Nitrogen or argon was widely used as the permeability test media, and the sample was heated to remove the residual water. However, temperature variation may introduce uncertainties in shale permeability tests.31,32 (3) The underlying mechanisms governing the profound changes in shale properties post-fracturing remain unclear. Alterations in shale permeability and pore structure stem from the synergistic interplay of multiple mechanisms. The predominant mechanisms at different reaction times require further investigation to achieve a comprehensive understanding. In this study, a pressure conductivity apparatus is employed to investigate the variations in shale matrix permeability, and all test samples were not treated for water removal. The role of coupled swelling-chemical effects in the evolution of pore structure was elucidated in depth by XRD and low-temperature nitrogen adsorption (LT-NA) experiments. This exploration aimed to uncover the changing of shale pore structure during CO2 pre-fracturing, providing a theoretical reference for CO2 sequestration and EHR.

This section offers a comprehensive account of the sample preparation methods employed to cater to diverse experimental needs. A series of tests, encompassing pressure transmit tests, XRD tests, LT-NA experiments are conducted to thoroughly characterize the samples in terms of permeability, mineral composition, and pore structure.

The experimental samples were sourced from the Jimsar shale reservoir in Xinjiang, extracted at a depth of 2674–2679 m. These samples were collected directly from the wellhead and subsequently processed into various sizes to suit specific experimental requirements. For pressure transfer tests, the rock samples were precisely drilled to form standard rock columns measuring 5 cm in length and 2.54 cm in diameter. To measure changes in permeability, these rock columns were transformed into core thin sections by embedding them in epoxy resin. The rock samples were then placed in molds, filled with epoxy resin. After waiting for 3 days, the samples were cut into thin slices with a thickness of 10 mm using wire-cutting technology (Fig. 1). For XRD and LT-NA tests, shale samples were initially crushed into pieces ranging from 10 to 20 mm, awaiting subsequent reaction with water or CO2. After the reaction, the samples were further crushed to facilitate XRD and LT-NA experiments (Fig. 2). It is worth note that the experimental conditions are above 7.38 MPa and 31.1 °C, and CO2 is under supercritical state during the experiment.

FIG. 1.

Shale slice prepared for pressure transmit test.

FIG. 1.

Shale slice prepared for pressure transmit test.

Close modal
FIG. 2.

Sample preparation and pretreatment.

FIG. 2.

Sample preparation and pretreatment.

Close modal

During the CO2 pre-fracturing process, CO2 is initially saturated with the shale matrix near the fractures and gradually diffuses deeper into the reservoir. Subsequently, water-based fracturing fluid is introduced into the fractures. At this point, the CO2-saturated shale matrix within the SRV interfaces with water, initiating a sequence of intricate physicochemical reactions. In order to scrutinize the role of CO2 in this process, two distinct sets of experiments were executed in this section: One group was performed in a saturated CO2 environment and the other in an environment without CO2. The shale hydration experiment with the absence of CO2 serves as a control group, helping to discern the alteration in shale properties under the influence of CO2. First, the reaction of shale with water in the absence of CO2 was conducted, subject to varying reaction durations of 3, 5, and 7 days. The second set of experiments was conducted within the CO2 reaction system, which consists of a water bath temperature control system, with a temperature controlling accuracy of ± 0.05 °C, a high-temperature and high-pressure reaction vessel, a vacuum pump, and a gas booster pump. The experimental steps are as follows:

  1. The shale samples were immersed in the saturating fluid, subjected to vacuum, and then pressurized to 15 MPa to ensure complete saturation of the samples.

  2. Place the sample in the high-temperature and high-pressure reaction vessel, and evacuate.

  3. Place the reaction vessel in the water bath temperature control system and heated to 50 °C. Inject CO2 into the reaction vessel using the gas booster pump until the pressure reaches 10 MPa, waiting for 2 h.

  4. The water was injected into the reaction vessel, and the reaction proceeded for 3, 5, and 7 days, respectively.

  5. The treated samples were removed from the reaction vessel. The shale partial was further crushed into finer powder to meet the requirement of XRD and LT-NA, and the shale slice was used to conduct a pressure transmit test (Figs. 1 and 2).

1. Permeability determination—pressure transmit test

Shale samples exhibit extremely low permeability, making steady-state permeability testing time-consuming, inefficient, and of limited precision. In this study, based on an unsteady-state theory, a pressure conduction apparatus is employed to test the changes in liquid permeability of the shale sample. This method involves recording the pressure variations upstream and downstream of a shale core thin section, establishing a dimensionless pressure-time relationship, and thereby reflecting the changes in shale permeability under different treatment methods.

The system comprises an ISCO pump, a vacuum pump, a specialized core holder, an intermediate container, a thermostatic chamber, and a pressure collection system (Fig. 3). The experimental procedure is as follows:
FIG. 3.

Schematic diagram of the pressure conduction system.

FIG. 3.

Schematic diagram of the pressure conduction system.

Close modal
  1. The thin sections are placed in the core holder, and the thermostatic chamber is preheated to 50 °C for 24 h. After that, the system is evacuated.

  2. The switches T1, T2, and T3 are turned on, and the liquid within the intermediate containers is drawn into the core holder. Once the pressure at both ends of the shale slice has reached equilibrium, data recording begins

  3. The downstream switch T2 is turned off, and the upstream pressure of the shale core is increased to 0.35–0.5 MPa. The changes in pressure at both ends are recorded, and the test concludes once the downstream pressure stabilizes.

According to the transient pressure model as follows, core permeability can be calculated as:
k = μ LVC A d γ d t , γ = ln P P L , t P P i ,
where k is the calculated permeability, nD; μ is the liquid viscosity, 1.4 mPa s; L is the thickness of the shale slice, 1 cm; V is the volume of the downstream room, 11.3 cm3; C is the compression coefficient of the water, 0.0009 MPa−1; A is the cross area of the shale, 4.9 cm2; γ is the dimensionless pressure; P is the upstream pressure, MPa; P L , t is the downstream pressure at time t; and P i is the initial downstream pressure.

2. Mineral composition identification—XRD

X-ray diffraction is a nondestructive and highly accurate technique for mineral identification and crystallography. Each mineral in the sample has a unique crystal lattice structure, and x-rays will interact with each mineral differently. The mineral component can be specified by comparing it to a database of known patterns.33 The shale sample was crushed into 200 mesh to conduct XRD test, and the Rigaku D/Max-2500/PC-type X-ray diffractometer equipment was used to determine the changes in shale mineral composition.

3. Pore structure characterization—LT-NA

Shale matrix contains a significant number of nanoscale pores, which contribute significantly to its porosity and specific surface area. Low-temperature nitrogen gas adsorption, by measuring the quantity of nitrogen gas adsorbed, offers valuable insights into the material structural and textural properties. Prior to testing, the samples needed to be prepared to a particle size of 60–80 mesh for accurate assessment of shale pore distribution, and Belsorp-Max II equipment was used to characterize the pore size distribution of the Jimsar shale.

The shale slice before and after water–CO2 treatment is shown in Fig. 4, revealing significant changes. However, as the cemented epoxy resin is almost non-porous, the corrosion occurs only on the surface and does not affect the experimental results. The pressure transfer test results are depicted in Figs. 5–8. The left panel illustrates the variations in upstream and downstream pressures, where the black curve represents the upstream pressure, regulated by the ISCO pump throughout the experimental process. The colored curve depicts the variations in the downstream pressure of the shale sample under different water treatment durations. The higher upstream pressure drives fluid flow toward the downstream, resulting in an increase in downstream pressure. After a certain period, it reaches a stable state. Due to the semipermeable membrane effect of low-permeability rocks, there is still a slight difference between the downstream pressure and the upstream pressure in the stable state.34,35

FIG. 4.

Shale slice before and after water–CO2 treatment.

FIG. 4.

Shale slice before and after water–CO2 treatment.

Close modal
FIG. 5.

Upstream, downstream pressure and dimensionless pressure of shale sample under different hydration times.

FIG. 5.

Upstream, downstream pressure and dimensionless pressure of shale sample under different hydration times.

Close modal
FIG. 6.

Upstream, downstream pressure and dimensionless pressure before and after CO2 treatment for 3 days.

FIG. 6.

Upstream, downstream pressure and dimensionless pressure before and after CO2 treatment for 3 days.

Close modal
FIG. 7.

Upstream, downstream pressure and dimensionless pressure before and after CO2 treatment for 5 days.

FIG. 7.

Upstream, downstream pressure and dimensionless pressure before and after CO2 treatment for 5 days.

Close modal
FIG. 8.

Upstream, downstream pressure and dimensionless pressure before and after CO2 treatment for 7 days.

FIG. 8.

Upstream, downstream pressure and dimensionless pressure before and after CO2 treatment for 7 days.

Close modal

As shown in Figs. 5–8, the downstream pressure equilibrium time can provide initial insights into shale slice permeability. For untreated shale samples, the stabilization time remains consistently uniform, ranging from 2.55 to 2.81 h. This uniformity can be attributed to all shale samples being extracted from the same rock column, resulting in minimal heterogeneity within the shale column and a comparable initial permeability. In contrast, the stabilization time for CO2 and water-treated shale samples exhibits a wide range. As the reaction time increases, the equilibrium time experiences an upward trend. In the absence of CO2, downstream pressure reaches equilibrium in 8.36 h after 3 days of water treatment. Thin sections subjected to 5 and 7 days of reaction display equilibrium times of 15.83 and 23.88 h, respectively. For the shale hydration experiments involving CO2, the downstream stabilization time ranges from 8.36 to 14.88 h. Whether CO2 is present or not, the equilibrium time for the hydrated shale samples exhibits a significant increase, with this pronounced increase signifying a reduction in shale permeability.

The right column of Figs. 5–8 illustrates the relationship between dimensionless pressure (denoted as γ) and time at various water treatment durations. The permeability of shale samples was determined using a linear fitting of γ against time, with the slope (dγ/dt) employed for permeability calculations. The summarized permeability results for shale samples are presented in Table I. It is evident that as water treatment time increases, the permeability of thin sections decreases. The slope of untreated shale ranges from 2.58 × 10−4–2.98 × 10−4 with permeability between 75 and 86.5 nD. After 3 days of water treatment, permeability was reduced by 62% to 32.54 nD. Samples treated for 5 and 7 days exhibit permeabilities of 6.91 and 3 nD, corresponding to a 91% and 96% decrease in liquid-measured permeability. The data indicates that increasing treatment time results in a steady reduction in permeability, with the most significant change occurring during the initial 3 days of the reaction.

TABLE I.

Fitting slope of dimensionless pressure and calculated permeability.

Reaction time (day) Untreated shale CO2 + water treated Water treated
Slop of PPT (1/s) Permeability (nD) Slop of PPT (1/s) Permeability (nD) Slop of PPT (1/s) Permeability (nD)
−2.98 × 10−4  86.59  −1.47 × 10−4  42.71  −1.12 × 10−4  32.54 
−2.58 × 10−4  74.96  −5.14 × 10−5  14.93  −2.38 × 10−5  6.91 
−2.73 × 10−4  79.32  −5.46 × 10−5  15.86  −1.03 × 10−5  3.00 
Reaction time (day) Untreated shale CO2 + water treated Water treated
Slop of PPT (1/s) Permeability (nD) Slop of PPT (1/s) Permeability (nD) Slop of PPT (1/s) Permeability (nD)
−2.98 × 10−4  86.59  −1.47 × 10−4  42.71  −1.12 × 10−4  32.54 
−2.58 × 10−4  74.96  −5.14 × 10−5  14.93  −2.38 × 10−5  6.91 
−2.73 × 10−4  79.32  −5.46 × 10−5  15.86  −1.03 × 10−5  3.00 

For the CO2 involved cases, the slope of the shale samples was −3.23 × 10−5∼−7.83 × 10−5. Accordingly, the permeability was 42.71, 14.93, and 15.86 nD, which also decreased with longer water treatment duration. However, the permeability reduction of the CO2 involved cases is much smaller, with a 51% decrease after 3 days of water treatment. After 5 days, permeability decreases to 14.93 nD, which is 20% of the original permeability. Extending the reaction time has a relatively minor impact on permeability, indicating that the influence of CO2 on shale permeability is primarily observed within the first 5 days, with limited effects on permeability beyond that point at the current sample scale (Fig. 9).

FIG. 9.

Shale permeability under different treatment conditions.

FIG. 9.

Shale permeability under different treatment conditions.

Close modal
Shale is a type of sedimentary rock primarily composed of fine-grained mineral particles, with a composition that typically includes clay minerals, quartz, and a variety of accessory minerals. The mineral composition of shale varies across different regions due to distinct depositional environments. In the Junggar Basin, shale formation during the Early to Middle Permian period took place in a gently sloping, saline lake basin, leading to a relatively higher quartz content in the shale36,
CO 2 + H 2 O H 2 CO 3 H + + HCO 3 2 H + + CO 3 2 ,
(1)
CaCO 3 ( calcite ) + 2 H + Ca ( 2 + ) + CO 2 + H 2 O ,
(2)
Na 2 Al 2 Si 6 O 16 ( albite ) + 2 H + + H 2 O kaolinite + 2 Na + + 4 quartz ,
(3)
KAl 2 ( Si 3 , Al 10 ( OH ) 2 ( Illite ) + 1 . 1 H + 0.77 kaolinite + 0 . 6 K + + 0 . 25 Mg ( 2 + ) + 1.2 quartz + 1 . 35 H 2 O ,
(4)
2 KAlSi 3 O 8 ( K feldspar ) + 2 H + + H 2 O 2 K + + Al 2 Si 2 O 5 ( OH ) 4 ( kaolinite ) + 4 SiO 2 ( quartz ) ,
(5)
Chlorite + 16 H + 5 Fe 2 + + 2 . 3 Al 3 + + 3 SiO 2 + 12 H 2 O .
(6)
The interaction between CO2 and water leads to the formation of carbonic acid [Eq. (1)], creating an acidic environment within porous media, which may induce many reactions in shale minerals, shown in Eqs. (2)–(6). Elevated pressure enhances CO2 dissolution, increasing H+ ionization and lowering pH. Saturated carbonic acid water is known to have a pH of 3.2 at 10 MPa.23 These conditions initiate substantial reactions with carbonate and silicate minerals,37–39 which induce mineral composition change of shale, as detailed in Table II. Notably, the group of clay minerals includes illite-smectite, chlorite-smectite, chlorite, and illite. Other minerals encompass certain components that are less abundant and do not vary significantly in content, like glauberite, gibbsite, and pyrite. As the reaction progresses, the quartz content gradually increases. CO2–water interactions can potentially lead to hydrolysis and carbonation reactions with carbonate and clay minerals within the shale.40,41 Hydrogen and hydroxide ions generated through water ionization infiltrate mineral lattices, replacing cations or anions, ultimately leading to mineral decomposition. As shown in Eqs. (4) and (6), illite, and chlorite can react with carbonates to generate quartz, consequently elevating the quartz mineral content. Figure 10 illustrates a slight increase in quartz content after 3 days of reaction, during which clay mineral content slightly decreases. After 5 days of reaction, a more substantial increase in quartz content is observed, accompanied by a significant decrease in clay mineral content. At this stage, the reactions involving illite and chlorite significantly contribute to the increase in quartz content. The chemical properties of quartz remain relatively stable, and it exhibits minimal participation in chemical reactions under weakly acidic conditions. Meanwhile, the decline in the content of other minerals may also contribute to the increase in quartz content. For dolomite minerals, their content remains relatively stable after seven days of reaction, hovering around 10%. Notably, calcite content experiences a noticeable decrease after 3 days, with only minor reductions thereafter, suggesting that calcite dissolution mainly occurs within the first three days. In comparison to other minerals, calcite reacts more rapidly with carbonates and is preferentially consumed, followed by feldspar and other minerals.29,42
TABLE II.

Mineral compositions of the shale before and after treatment.

Reaction time, day Quartz Clay mineral Dolomite Calcite K-feldspar Others
30.9  23.4  10.7  10.4  5.4  19.2 
32.4  22.7  10.7  8.7  4.3  21.2 
38.3  19.3  11.7  8.6  2.1  20 
38.7  18.9  10.4  8.1  2.2  21.7 
Reaction time, day Quartz Clay mineral Dolomite Calcite K-feldspar Others
30.9  23.4  10.7  10.4  5.4  19.2 
32.4  22.7  10.7  8.7  4.3  21.2 
38.3  19.3  11.7  8.6  2.1  20 
38.7  18.9  10.4  8.1  2.2  21.7 
FIG. 10.

Mineral composition of shale with different CO2-water treatment duration.

FIG. 10.

Mineral composition of shale with different CO2-water treatment duration.

Close modal

1. Low-pressure N2 gas adsorption–desorption isotherm

The LT-NA adsorption–desorption isotherms of the shale samples are shown in Fig. 11. According to the International Union of Pure and Applied Chemistry (IUPAC) classification, the isotherms of the Jimsar shale samples exhibit characteristics of type-IV isotherms.43 At low relative pressures (P/P0 < 0.45), nitrogen molecules adsorb as a monolayer on the pore surfaces due to the relatively high surface energy of micropores. In this stage, the adsorption capacity experiences a gradual increase with the growing pressure. As the relative pressure rises (0.45 < P/P0 < 0.9), the adsorption shifts from primarily monolayer to multilayer adsorption. This transition results in a slight acceleration in the adsorption, manifesting as a concave shape in the isotherm curve. As relative pressure continues to increase (P/P0 > 0.9), gas experiences capillary condensation within mesopores and macropores, leading to a rapid upsurge in adsorption.44 Based on the desorption isotherms, it is evident that there is a pronounced hysteresis loop, which can be utilized to evaluate the pore morphology in shale. With the shape of the loops classified as type H2(b), suggests that the pore connectivity in the shale sample resembles that of a wide-necked ink bottle.43 Furthermore, the shape of the hysteresis loops remains unchanged before and after the reaction, indicating that the pore structure of the shale has not undergone substantial alterations.

FIG. 11.

N2 adsorption–desorption isotherms of shale with different reaction durations.

FIG. 11.

N2 adsorption–desorption isotherms of shale with different reaction durations.

Close modal

The maximum nitrogen adsorption capacity at P/P0 can serve as an initial gauge of the shale adsorption capability. For the untreated sample, the maximum adsorption capacity is 17.76 cm3/g. In the absence of CO2, the shale adsorption capacity significantly decreases with longer hydration duration. After three days of hydration reactions, the maximum adsorption capacity of the shale sample decreased to 15.21 cm3/g. Subsequently, at five days and seven days, it further declined to 14.57 and 13.99 cm3/g, respectively. In cases involving CO2 reactions, the maximum adsorption capacity at three different reaction durations is as follows: 17.63, 16.53, 16.39 cm3/g, also signifying a diminishing trend in adsorption capacity with prolonged exposure time. However, in an environment devoid of CO2, the decreasing trend in adsorption capacity becomes more pronounced. This indicates that CO2 has the capability to mitigate the decline in adsorption capacity caused by shale hydration (Fig. 12). This observation aligns with the similar patterns observed in permeability changes. Nevertheless, it is essential to note that the variation in adsorption capacity does not exhibit a direct correlation with permeability. Shale adsorption capacity is influenced by a variety of factors, with the most significant contributions stemming from mesopores and micropores. Over time, changes in shale porosity occur, which will be discussed in Sec. III C 2 addressing variations in pore structure parameters.

FIG. 12.

Adsorption capacity of shale under different treatment methods.

FIG. 12.

Adsorption capacity of shale under different treatment methods.

Close modal

2. Pore structure parameter analysis from LT-NA

Based on the multilayer physical adsorption model of BET and the BJH equation, additional processing and analysis were performed on the LT-NA data, resulting in a comprehensive set of pore structure parameters.

The BET adsorption theory is rooted in the Langmuir adsorption theory, which assumes the thermodynamic and kinetic uniformity of adsorption sites. However, the BET model is particularly applicable within the range of 0.05 < P/P0 < 0.35. This range is chosen to focus the BET model analysis on conditions where multilayer adsorption is predominant, and avoid the effects of monolayer adsorption and capillary condensation. The BET model can be expressed as
p V p 0 p = C 1 V m C × p p 0 + 1 V m C ,
where V m is the monolayer adsorption volume, P is adsorption equilibrium pressure, P0 is the saturation vapor pressure, and C is a constant related to the heat of adsorption.
After that, the adsorption surface area is expressed as
S = V m 22 400 N A σ m ,
where S is the total surface area of the adsorbent, σ m is the cross-sectional area of the adsorbent molecule, and N A is the Avogadro constant.
According to the BJH equations, the determination of pore size distribution using gas adsorption relies on the principles of capillary condensation and volume equivalent substitution. This approach considers the liquid nitrogen volume filling the measured pore as an equivalent to the pore volume. According to capillary condensation theory, the range of pore diameters capable of undergoing capillary condensation varies at different pressure conditions. As the pressure increases, the pore radius at which condensation can occur also increases. For a specific P/P0, there exists a critical pore radius, which can be expressed as45 
ln p p 0 = 2 γ m RT ρ × 1 r c ,
where γ is the surface tension of liquid N2, M is the molar mass of N2, ρ is the density of liquid N2, r is the capillary pore radius corresponding to p/p0, and c is the adsorption layer thickness

Prior to delving into the analysis of pore distribution characteristics, it is imperative to present an overview of the mechanisms influencing shale pore structure. First, exposure to CO2 and water induces a series of physical changes in shale. The adsorption of CO2 and water causes the swelling of the shale matrix, leading to the conversion of some large pores into mesopores or micropores.19,21,46 Simultaneously, the shale matrix contains a significant amount of organic matter such as kerogen, which features abundant micropores. These organic components can be extracted by CO2, resulting in the disappearance of some micropores.47,48 Additionally, in a water-containing environment, CO2 dissolves in water, forming carbonic acid that initiates chemical reactions with specific rock minerals. Under weakly acidic conditions, clay and carbonate minerals are consumed, leading to the formation of new minerals such as quartz and kaolinite, and transforming micropores into mesopores or macropores.37,49,50 The interaction among CO2, water, and shale is a complex process where alterations in permeability, pore structure, and mineral composition result from the synergistic effects of various physical changes and chemical reactions. However, within these mechanisms, one or more mechanisms may dominate changes in shale permeability and pore structure. A comprehensive analysis, considering mineral composition, permeability, and pore distribution, is necessary for a thorough understanding of these changes.

The pore structure distribution (PSD) of the shale samples is illustrated in Figs. 13 and 14. As shale hydration progresses, noticeable differences in PSD are observed under different environmental conditions. In the case without CO2, the most significant decline in the curve occurs after the third day, with subsequent changes being relatively minor. In contrast, in the presence of CO2, there is a slight increase after three days of reaction, followed by the most pronounced decrease on the fifth day. Moreover, irrespective of whether CO2 is involved in shale hydration reactions, there is an obvious peak located in the pore size range of 25.25–40.03 nm. Compared to mesopores and macropores, micropores are more significantly affected by hydration, with micropores nearly disappearing after a 7-day hydration period.

FIG. 13.

Pore-size distributions before and after water treatment.

FIG. 13.

Pore-size distributions before and after water treatment.

Close modal
FIG. 14.

Pore-size distributions before and after CO2-water treatment.

FIG. 14.

Pore-size distributions before and after CO2-water treatment.

Close modal

The pore structure parameters including specific surface area (SSA), pore volume (PV), and average pore radius (Rave)was summarized in Table III and Figs. 15–18. Observable changes in pore structure become evident following water treatment. To elucidate the distribution of pores with varying sizes, the pores were classified according to IUPAC standards into macropores (>50 nm), mesopores (2–50 nm), and micropores (<2 nm).43 In scenarios where CO2 is absent in shale hydration, both the total pore volume (TPV) and total specific surface area (TSSA) of the rock samples decrease with increasing exposure time. The swelling caused by shale hydration results in the contraction of the macropores, forming new micropores, and simultaneously causing the disappearance of some pre-existing micropores. The volumes of macropores, mesopores, and micropores all decrease progressively, with the most substantial reduction observed within the initial three days of the reaction (Fig. 15). Additionally, in comparison to macropores, the reduction in volumes of micropores and mesopores is more pronounced, indicating that shale hydration-induced swelling has a greater impact on small and medium-sized pores.

TABLE III.

Pore structure parameters obtained by LT-NA.

Treatment methods SSA (m2/g) PV (10−3 cm3/g) Rave (nm)
SSAmic SSAmes SSAmac TSSA PVmic PVmes PVmac TPV
Untreated  3.417  0.63  0.082  4.13  0.51  15.61  2.75  18.87  18.3 
SC-CO2 + water 3 days  3.269  0.455  0.086  3.81  0.68  15.82  3.04  19.54  20.5 
SC-CO2 + water 5 days  1.9  0.269  0.053  2.222  0.21  10.68  1.81  12.7  22.9 
SC-CO2 + water 7 days  1.684  0.011  0.055  1.752  0.01  10.42  1.87  12.3  28.1 
Water 3 days  2.396  0.3351  0.0628  2.7939  0.37  10.23  2.41  13.01  18.6 
Water 5 days  1.8541  0.2797  0.047  2.1808  0.2  8.25  2.27  10.72  19.7 
Water 7 days  1.5029  0.1585  0.0318  1.6932  0.09  7.04  1.81  8.94  21.1 
Treatment methods SSA (m2/g) PV (10−3 cm3/g) Rave (nm)
SSAmic SSAmes SSAmac TSSA PVmic PVmes PVmac TPV
Untreated  3.417  0.63  0.082  4.13  0.51  15.61  2.75  18.87  18.3 
SC-CO2 + water 3 days  3.269  0.455  0.086  3.81  0.68  15.82  3.04  19.54  20.5 
SC-CO2 + water 5 days  1.9  0.269  0.053  2.222  0.21  10.68  1.81  12.7  22.9 
SC-CO2 + water 7 days  1.684  0.011  0.055  1.752  0.01  10.42  1.87  12.3  28.1 
Water 3 days  2.396  0.3351  0.0628  2.7939  0.37  10.23  2.41  13.01  18.6 
Water 5 days  1.8541  0.2797  0.047  2.1808  0.2  8.25  2.27  10.72  19.7 
Water 7 days  1.5029  0.1585  0.0318  1.6932  0.09  7.04  1.81  8.94  21.1 
FIG. 15.

SSA variations of shale samples under water treatment.

FIG. 15.

SSA variations of shale samples under water treatment.

Close modal
FIG. 16.

PV variations of shale samples under water treatment.

FIG. 16.

PV variations of shale samples under water treatment.

Close modal
FIG. 17.

SSA variations of shale samples under CO2-water treatment.

FIG. 17.

SSA variations of shale samples under CO2-water treatment.

Close modal
FIG. 18.

PV variations of shale samples under CO2-water treatment.

FIG. 18.

PV variations of shale samples under CO2-water treatment.

Close modal

Regarding TSSA, it exhibits a trend similar to TPV, with the most noticeable decrease occurring after the first three days of hydration, followed by a relatively steady decline (Fig. 16). Micropores have been proven to be the primary contributors to the specific surface area of shale,35,51 experience a notable decrease, driving the reduction in TSSA. These results suggest that shale hydration-induced swelling leads to a decrease in pore volume, causing the contraction of mesopores and macropores. The disappearance of micropores is more prominent than the formation of new micropores, leading to a decrease in micropore volume. This process further results in a gradual decrease in the proportion of micropore volume in TPV, while the proportion of mesopores and macropores demonstrates an increasing trend. Meanwhile, the average pore radius, calculated as 4×TPV/TSSA, exhibits an increasing trend due to the greater impact of hydration on mesopores and micropores. This impact leads to a rise in the proportion of macropores, consequently increasing the average pore radius.52 

As shown in Figs. 17 and 18, both the TPV and TSSA exhibit a decrease after CO2–water treatment, while the average pore radius tends to increase. The TPV of the untreated shale sample is 18.87 cm3/g, which increased to 19.22 cm3/g after 3 days of CO2–water treatment. Subsequently, the TPV tends to decrease, measuring 12.7 and 12.3 cm3/g after 5 and 7 days of treatment. The initial increase in pore volume after 3 days is associated with the dissolution of carbonate minerals, primarily calcite. As indicated in Sec. III B, the calcite mineral shows a decreasing trend during the reaction, with most of the calcite being consumed in the early stages. Carbonate minerals typically occupy larger pore spaces, and their dissolution can lead to the formation of larger-sized pores.53 Consequently, the volumes of macropores and mesopores exhibit an increasing trend. For PVmic, the impact of shale hydration swelling on micropores is higher than on mesopores and macropores, resulting in a decrease in micropore volume. However, when the changes in pore volume at different sizes are cumulative to the TPV, it shows an upward trend. This suggests that in the first three days of the reaction, the pore-generation effect dominated by chemical dissolution is the main factor driving changes in rock pore structure. The dissolution effect leads to the enlargement of pores, which counteracts the decrease in pore volume, resulting in a transient increase in pore volume after three days of reaction.

As the hydration process extends to 5 days, both TSSA and TPV experience a significant decrease, and further extension of exposure time shows minimal changes in TPV. XRD results indicate that, the dissolution of carbonate minerals almost ceases, while the content of clay minerals continues to decrease. As clay minerals are the main contributors to adsorption-induced swelling in shale, the reduction in clay mineral content appears to be the primary reason for the limited changes in pore volume. However, a detailed analysis of clay mineral composition reveals that the majority consists of illite-smectite and chlorite-smectite, with chlorite and illite closely following. Among these, illite and chlorite are non-expansive clay minerals, whereas illite-smectite and chlorite-smectite are classified as expansive clay minerals.54 In fact, the content of expansible clay minerals does not show a significant decrease. At the end of 7 days, the total content of illite-smectite and chlorite-smectite remains above 60%. Hence, the observed reduction in shale expansion cannot be solely attributed to the decrease in clay mineral content, and further analysis is required to identify the actual underlying cause. As indicated in Eqs. (4) and (5), the dissolution of illite and K-feldspar releases K+, which can effectively inhibit the water absorption and swelling of montmorillonite. Compared to other cations such as Mg2+ and Ca2+, K+ has stronger electrostatic attraction after lattice substitution, leading to a more tightly bound interlayer structure, effectively inhibiting clay swelling. Thus, in a weakly acidic environment, the K+ released from the dissolution of illite and K-feldspar can directly enter the interlayer of montmorillonite, reducing its swelling capacity. For illite-smectite minerals, which are formed by the alternating stacking of illite layers and montmorillonite layers, the K+ released from illite can more directly enter the interlayer of montmorillonite, better inhibiting the effect of clay swelling.

The fractal dimension (D) is a mathematical concept used to characterize the complexity and irregularity of geometric shapes or structures.55 The higher the value of D, the rougher the surface, and the more complex the pore structure.56 In this study, the Frenkel–Halsey–Hill (FHH) model was applied to evaluate the fractal dimension of the shale sample, which can be described as:57 
l n ( V ) = A ln [ ln P 0 / P ] + C , D = A + 3 ,
where V is the volume of adsorbed gas at P, cm3/g; A is a linear correlation coefficient; C is the intercept of the fitting curve.

The linear fitting curve of ln(V) and ln[ln(P0/P)] is presented in Figs. 19 and 20. Employing P/P0 = 0.45 as a threshold, the linear fitting curve is segmented into two distinctive regions. In the region where P/P0 < 0.45, characterized by low equilibrium pressure, gas molecules predominantly adsorb to the pore surface through multilayer or monolayer adsorption. This dataset is instrumental in evaluating the roughness of the pore surface. In the second region with 0.45 < P/P0 < 1, marked by higher equilibrium pressure, gas primarily fills the pore structure through capillary condensation. This set of data are utilized to characterize the heterogeneity of the pore structure. The fitting equation and calculated fractal dimensions are summarized in Table IV. For all the test samples, the correlation coefficients (R2) between ln(V) and ln[ln(P0/P)] are greater than 0.99, underscoring the robust fractal characteristics of the shale samples. Fractal dimension D1, representing the roughness of the pore surface, spans from 2.49 to 2.7066, while D2, representing the heterogeneity of the pore structure, ranges from 2.56 to 2.66. In the scenario without CO2, D2 exhibits a gradual increase with the progression of the reaction. The hydration swelling induces a reduction in all pore sizes, with no new pores generated by chemical reactions, thereby intensifying the heterogeneity of permeable pores. In CO2-involved scenarios, both D1 and D2 exhibit a decreasing trend, signifying a reduction in the roughness of the pore surface and the heterogeneity of the pore structure. Additionally, D2 is consistently smaller than D1 after CO2–water treatment, suggesting weaker heterogeneity of the permeable pores, which is more favorable for oil and gas production.

FIG. 19.

Fitting curve of water treated shale samples.

FIG. 19.

Fitting curve of water treated shale samples.

Close modal
FIG. 20.

Fitting curve of CO2-water treated shale samples.

FIG. 20.

Fitting curve of CO2-water treated shale samples.

Close modal
TABLE IV.

Fractal dimension of shale samples under different treatment methods.

Treatment methods P/P0 < 0.45 P/P0 > 0.45
Fitting equation D1 R2 Fitting equation D2 R2
Untreated  y1 = −0.2934x + 5.4766  2.7066  0.9955  y2 = −0.4116x + 3.9069  2.588 4  0.998 1 
SC-CO2 + water 3 days  y1 = −0.3101x + 4.9484  2.6899  0.9969  y2 = −0.4348x + 3.5186  2.565 2  0.998 5 
SC-CO2 + water 5 days  y1 = −0.3271x + 3.8836  2.6729  0.9981  y2 = −0.4381x + 5.5050  2.561 9  0.997 8 
SC-CO2 + water 7 days  y1 = −0.3364x + 4.2210  2.6636  0.9981  y2 = −0.4358x + 3.2462  2.564 2  0.999 64 
Water 3 days  y1 = −0.485x + 1.2538  2.515  0.9959  y2 = −0.3899x + 1.2834  2.610 1  0.998 8 
Water 5 days  y1 = −0.5032x + 1.042  2.4968  0.9998  y2 = −0.3689x + 1.09  2.631 1  0.995 2 
Water 7 days  y1 = −0.51x + 1.0437  2.49  0.9996  y2 = −0.372x + 1.0847  2.628  0.995 4 
Treatment methods P/P0 < 0.45 P/P0 > 0.45
Fitting equation D1 R2 Fitting equation D2 R2
Untreated  y1 = −0.2934x + 5.4766  2.7066  0.9955  y2 = −0.4116x + 3.9069  2.588 4  0.998 1 
SC-CO2 + water 3 days  y1 = −0.3101x + 4.9484  2.6899  0.9969  y2 = −0.4348x + 3.5186  2.565 2  0.998 5 
SC-CO2 + water 5 days  y1 = −0.3271x + 3.8836  2.6729  0.9981  y2 = −0.4381x + 5.5050  2.561 9  0.997 8 
SC-CO2 + water 7 days  y1 = −0.3364x + 4.2210  2.6636  0.9981  y2 = −0.4358x + 3.2462  2.564 2  0.999 64 
Water 3 days  y1 = −0.485x + 1.2538  2.515  0.9959  y2 = −0.3899x + 1.2834  2.610 1  0.998 8 
Water 5 days  y1 = −0.5032x + 1.042  2.4968  0.9998  y2 = −0.3689x + 1.09  2.631 1  0.995 2 
Water 7 days  y1 = −0.51x + 1.0437  2.49  0.9996  y2 = −0.372x + 1.0847  2.628  0.995 4 

Chemical-swelling coupling effects are the primary factors causing changes in shale pore structure. The schematic diagram of CO2 mitigation of shale swelling is proposed to illustrate the mechanism of permeability and pore-structure variations, as shown in Fig. 21. In the CO2–rock–water environment, a dynamic interplay of mineral dissolution, including carbonates and feldspar, and the deposition of secondary minerals occurs.51,58,59 In the initial 3 days of the reaction, the dissolution of carbonates enlarges existing pores and forms new ones, which can effectively compensate for the pore volume reduction caused by shale swelling, contributing to increased pore volume. The size of the generated pores is relatively large, contributing less to the SSA, while small pores are significantly affected by swelling and contraction, leading to a decrease in SSA. Simultaneously, despite significant mineral dissolution, the formation of secondary minerals occludes existing flow channels, resulting in a decrease in permeability. In the middle to later stages, water absorption-induced shale swelling become the main factor causing changes in pore structure and permeability, while the chemical dissolution effect acts to mitigation the impact of shale swelling. During this stage, the K+ released by chemical dissolution can effectively weaken the swelling of montmorillonite, thereby reducing the impact of shale swelling. Given that illite-smectite is the primary component of the clay mineral, it is also the main contributor to shale swelling. The K+ released by illite dissolution can enter the montmorillonite crystal layer more directly, thus better-mitigating shale swelling. However, the contraction of pores caused by shale swelling still exists, resulting in a decrease in permeability, PV, and SSA. The shrinkage of pore size increases the seepage resistance and results in a decrease in permeability. Both the permeability (Fig. 9) and PSD (Figs. 13 and 14) results indicate that the changes caused by the shale hydration reaction mainly occur within the first three days. However, the involvement of CO2 has been shown to effectively delay this process, with the most significant reduction observed on the fifth day of the reaction.

FIG. 21.

Schematic diagram of CO2 mitigates clay mineral swelling.

FIG. 21.

Schematic diagram of CO2 mitigates clay mineral swelling.

Close modal

Therefore, although pre-injected CO2 can effectively mitigate the clay swelling caused by subsequent water-based fracturing fluids, it cannot completely offset the impact of clay swelling. Currently, a multi-stage fracturing technique is often employed to generate a large SRV. This implies a significant time span between the first and last stages of fracturing, the decline in shale matrix permeability within the SRV due to the coupled effects of swelling-chemical effects is inevitable. Following the completion of the fracturing stages, shut-in operations are typically implemented to maximize the production of shale oil and gas. Hence, it is advisable to minimize the shut-in time after fracturing to expedite the flowback of water-based fracturing fluids and mitigate their impact on the shale permeability within the SRV.

Simultaneously, the impact of swelling-chemical coupling effects on CO2 sequestration is complex. On the one hand, the reduction in shale matrix permeability within the SRV can mitigate the risk of CO2 leakage. On the other hand, the decrease in TSSA within the affected zone leads to a reduction in the CO2 adsorption capacity of the shale matrix within this range. However, compared to the bulk shale formation, the contact area between the water-based fracturing fluid and the shale matrix is limited. Therefore, the region where the adsorption capacity is initially affected by swelling-chemical coupling effects is also limited. Consequently, it is believed that the alterations in shale properties induced by swelling-chemical coupling effects favor the geo-sequestration of CO2 in the shale.

In this study, the permeability and pore structure evaluation were conducted on a CO2-saturated shale. The time dependence of permeability and pore structure affected by chemical-swelling coupling effects was studied, and the mitigation effect of CO2 on shale hydration induce-swelling was confirmed. The following conclusions can be drawn:

  1. The intrusion of water-based fracturing fluid can significantly reduce the permeability of the Jimsar shale, with the permeability after seven days being 4% of the initial permeability. During the CO2 pre-fracturing process, the pre-injected CO2 can effectively mitigate the permeability reduction. The impact of CO2 becomes increasingly evident with the extension of reaction time. At a reaction time of 7 days, the permeability in the presence of CO2 is 5 times higher compared to the hydration scenario without CO2.

  2. The Jimusar shale has moderate carbonate and clay mineral content. The acidic environment formed by CO2 dissolution can effectively reduce the content of carbonates and clay minerals while causing an increase in quartz content. The expansive minerals like illite-smectite and chlorite-smectite are the main components of clay minerals, have a direct impact on shale permeability and pore structure. Under the effect of CO2–water, the self-released K+ caused by illite-smectite dissolution can inhibit the swelling of clay minerals.

  3. According to the LT-NA results, the adsorption capacity of untreated shale is greater than that of shale samples treated with CO2–water, while the water treated sample shows the lowest gas adsorption. Shale hydration swelling causes the contraction of the original pore structure, resulting in a decrease in TPV. At the same time, its impact on micropores and mesopores is greater, resulting in a decrease in TSSA and an increase in Rave. With the intervention of CO2, TPV shows a temporary upward trend and continues to decrease as the reaction progresses. Since the dissolution effect has a smaller impact on small pores, TSSA still shows a decreasing trend. Therefore, CO2 was proven to effectively mitigate the pore contraction but cannot prevent the swelling caused by shale hydration.

  4. During the CO2 pre-fracturing process, the swelling-chemical coupling effect is the main mechanism affecting shale physical properties. In the initial stage, the dissolution effect formed by chemical dissolution dominates, during which hydration swelling also occurs. However, the volume of the newly generated pores is higher than the pore contraction caused by swelling, resulting in an increase in pore volume. At the same time, the minerals shed and produced by the dissolution will block the original seepage channels, leading to a reduction in permeability. As the reaction continues, the role of hydration swelling gradually becomes dominant, with chemical dissolution playing a supporting role. The K+ formed by the acidification and hydrolysis of illite-smectite and illite weakens the swelling of clay minerals, thereby mitigating the subsequent shale matrix damage caused by water-based fracturing fluids.

However, it is important to note that some limitations existed in this study. Due to experimental constraints, the sample sizes used in this research are relatively small. While offer the advantage of ensuring similar reaction processes for samples required for different characterization experiments, it also limits a more comprehensive understanding. The combination of permeability test results, changes in mineral composition, and pore structure characteristics provides a mechanistic explanation for shale property changes during CO2 pre-fracturing. However, the small sample size restricts a more complete understanding. The physical changes in shale properties induced by CO2 pre-fracturing primarily result from the actions of CO2 and water-based fracturing fluids. The extent of their influence is mainly determined by the range of action of CO2 and water-based fracturing fluids. Therefore, a simulation approach should be employed to further investigate the impact of permeability changes on oil and gas production, and future research should comprehensively consider the action ranges of CO2 and water to further refine the understanding of the impact of CO2 pre-fracturing on shale properties.

The authors would like to acknowledge the financial support from the National Natural Science Foundation of China (Grant No. 52174045).

The authors have no conflicts to disclose.

Weiyu Tang: Writing – original draft (equal). Xiaoyu Zheng: Data curation (equal). Cheng Liu: Formal analysis (equal). Fujian Zhou: Supervision (equal); Validation (equal); Visualization (equal). Xiongfei Liu: Investigation (equal). Hang Zhou: Investigation (equal). Bo Wang: Resources (equal). Xiukun Wang: Visualization (equal). Xiaoyu Hou: Visualization (equal).

The data that support the findings of this study are available from the corresponding author upon reasonable request.

1.
L.
Li
,
B.
Wang
,
H.
Ge
,
H.
Zhou
,
Y.
Wang
,
L.
Zhang
et al, “
The intelligent optimization of perforation cluster locations incorporating the fiber optics monitoring results
,”
Phys. Fluids
35
(
12
),
126605
(
2023
).
2.
Q.
Wang
,
F.
Zhou
,
D.
Yang
,
S.
Yu
,
H.
Fei
,
E.
Yao
et al, “
Improving oil and gas flowability in tight carbonates with a novel solid delayed acid
,”
Phys. Fluids
36
(
1
),
012134
(
2024
).
3.
X.
Wang
,
S.
Wang
,
W.
Wu
,
Y.
Liang
, and
F.
Wang
, “
Coupled pressure-driven flow and spontaneous imbibition in shale oil reservoirs
,”
Phys. Fluids
35
(
4
),
042104
(
2023
).
4.
S.
Wang
,
Y.
Liang
,
Q.
Feng
, and
F.
Javadpour
, “
Sticky layers affect oil transport through the nanopores of realistic shale kerogen
,”
Fuel
310
,
122480
(
2022
).
5.
C.
Fu
and
N.
Liu
, “
Waterless fluids in hydraulic fracturing—A review
,”
J. Nat. Gas Sci. Eng.
67
,
214
224
(
2019
).
6.
Y.
Yang
,
S.
Wang
,
Q.
Feng
,
X.
Cao
,
Y.
Qin
,
C.
Shu
et al, “
Imbibition mechanisms of fracturing fluid in shale oil formation: A review from the multiscale perspective
,”
Energy Fuels
37
(
14
),
9822
9840
(
2023
).
7.
M.
Asadi
,
W.
Scharmach
,
T.
Jones
,
A.
Sampayo
,
E.
Chesney
, and
D.
Unruh
, “
Water-Free Fracturing: A Case History
,”
SPE/CSUR Unconventional Resources Conference
(
OnePetro
,
2015
).
8.
R.
Tudor
,
C.
Vozniak
,
W.
Peters
, and
M.-L.
Banks
, “
Technical advances in liquid CO2 fracturing
,” in
Annual Technical Meeting
(
OnePetro
,
1994
).
9.
L.
Li
,
Y.
Su
,
Y.
Hao
,
S.
Zhan
,
Y.
Lv
,
Q.
Zhao
et al, “
A comparative study of CO2 and N2 huff-n-puff EOR performance in shale oil production
,”
J. Pet. Sci. Eng.
181
,
106174
(
2019
).
10.
J. J.
Sheng
, “
Enhanced oil recovery in shale reservoirs by gas injection
,”
J. Nat. Gas Sci. Eng.
22
,
252
259
(
2015
).
11.
Y.
Xia
,
L.
Li
, and
Z.
Wang
, “
Experimental and numerical study on influencing factors of replacement capacity and slickwater flowback efficiency using pre-CO2 fracturing in tight oil reservoirs
,”
J. Pet. Sci. Eng.
215
,
110697
(
2022
).
12.
X.
Wang
, “
Practical application of liquid-CO2/slick-water hybrid fracturing technology in the lacustrine shale gas reservoir in Ordos Basin, China
,”
Interpretation
3
(
2
),
SJ75
SJ80
(
2015
).
13.
A.
Busch
,
S.
Alles
,
Y.
Gensterblum
,
D.
Prinz
,
D. N.
Dewhurst
,
M. D.
Raven
et al, “
Carbon dioxide storage potential of shales
,”
Int. J. Greenhouse Gas Control
2
(
3
),
297
308
(
2008
).
14.
A.
Abedini
and
F.
Torabi
, “
On the CO2 storage potential of cyclic CO2 injection process for enhanced oil recovery
,”
Fuel
124
,
14
27
(
2014
).
15.
Z.
Liu
,
Y.
Yang
,
Q.
Zhang
,
G.
Imani
,
L.
Zhang
,
H.
Sun
et al, “
Pore-scale flow simulation of CO2 sequestration in deep shale based on thermal-hydro-mechanical coupled model
,”
Phys. Fluids
36
(
2
),
022005
(
2024
).
16.
X.
Ao
,
Y.
Lu
,
J.
Tang
,
Y.
Chen
, and
H.
Li
, “
Investigation on the physics structure and chemical properties of the shale treated by supercritical CO2
,”
J. CO2 Util.
20
,
274
281
(
2017
).
17.
J.
Zhou
,
K.
Yang
,
S.
Tian
,
L.
Zhou
,
X.
Xian
,
Y.
Jiang
et al, “
CO2-water-shale interaction induced shale microstructural alteration
,”
Fuel
263
,
116642
(
2020
).
18.
Y.
Lu
,
J.
Liu
,
J.
Tang
,
X.
Ao
,
H.
Li
,
J.
Zhou
et al, “
Pore changes of slickwater-containing shale under supercritical CO2 treatment
,”
Fuel
312
,
122775
(
2022
).
19.
Y.
Lu
,
X.
Ao
,
J.
Tang
,
Y.
Jia
,
X.
Zhang
, and
Y.
Chen
, “
Swelling of shale in supercritical carbon dioxide
,”
J. Nat. Gas Sci. Eng.
30
,
268
275
(
2016
).
20.
Q.
Lyu
,
X.
Long
,
P.
Ranjith
,
J.
Tan
,
Y.
Kang
, and
Z.
Wang
, “
Experimental investigation on the mechanical properties of a low-clay shale with different adsorption times in sub-/super-critical CO2
,”
Energy
147
,
1288
1298
(
2018
).
21.
N. S.
Muhammed
,
T.
Olayiwola
, and
S.
Elkatatny
, “
A review on clay chemistry, characterization and shale inhibitors for water-based drilling fluids
,”
J. Pet. Sci. Eng.
206
,
109043
(
2021
).
22.
S.
Tian
,
J.
Zhou
,
X.
Xian
,
Q.
Gan
,
K.
Yang
,
Y.
Zheng
et al, “
Impact of supercritical CO2 exposure time on the porosity and permeability of dry and wet shale: The influence of chemo-mechanical coupling effects
,”
Energy
270
,
126905
(
2023
).
23.
Y.
Zou
,
S.
Li
,
X.
Ma
,
S.
Zhang
,
N.
Li
, and
M.
Chen
, “
Effects of CO2–brine–rock interaction on porosity/permeability and mechanical properties during supercritical-CO2 fracturing in shale reservoirs
,”
J. Nat. Gas Sci. Eng.
49
,
157
168
(
2018
).
24.
J.
Zhou
,
S.
Tian
,
L.
Zhou
,
X.
Xian
,
K.
Yang
,
Y.
Jiang
et al, “
Experimental investigation on the influence of sub-and super-critical CO2 saturation time on the permeability of fractured shale
,”
Energy
191
,
116574
(
2020
).
25.
Q.
Lyu
,
J.
Shi
, and
R. P.
Gamage
, “
Effects of testing method, lithology and fluid-rock interactions on shale permeability: A review of laboratory measurements
,”
J. Nat. Gas Sci. Eng.
78
,
103302
(
2020
).
26.
Z.
Lufeng
,
Z.
Fujian
,
Z.
Shicheng
,
W.
Jin
, and
W.
Yuechun
, “
Evaluation of permeability damage caused by drilling and fracturing fluids in tight low permeability sandstone reservoirs
,”
J. Pet. Sci. Eng.
175
,
1122
1135
(
2019
).
27.
T.
Liang
,
F.
Gu
,
E.
Yao
,
L.
Zhang
,
K.
Yang
,
G.
Liu
et al, “
Formation damage due to drilling and fracturing fluids and its solution for tight naturally fractured sandstone reservoirs
,”
Geofluids
2017
,
9350967
.
28.
A.
Fatah
,
H. B.
Mahmud
,
Z.
Bennour
,
M.
Hossain
, and
R.
Gholami
, “
Effect of supercritical CO2 treatment on physical properties and functional groups of shales
,”
Fuel
303
,
121310
(
2021
).
29.
S.
Li
,
S.
Zhang
,
H.
Xing
, and
Y.
Zou
, “
CO2–brine–rock interactions altering the mineralogical, physical, and mechanical properties of carbonate-rich shale oil reservoirs
,”
Energy
256
,
124608
(
2022
).
30.
C.
Qin
,
Y.
Jiang
,
J.
Zhou
,
S.
Zuo
,
S.
Chen
,
Z.
Liu
et al, “
Influence of supercritical CO2 exposure on water wettability of shale: Implications for CO2 sequestration and shale gas recovery
,”
Energy
242
,
122551
(
2022
).
31.
M.
Masri
,
M.
Sibai
,
J.-F.
Shao
, and
M.
Mainguy
, “
Experimental investigation of the effect of temperature on the mechanical behavior of Tournemire shale
,”
Int. J. Rock Mech. Min. Sci.
70
,
185
191
(
2014
).
32.
D. H.
Johnston
, “
Physical properties of shale at temperature and pressure
,”
Geophysics
52
(
10
),
1391
1401
(
1987
).
33.
A.
Chauhan
and
P.
Chauhan
, “
Powder XRD technique and its applications in science and technology
,”
J. Anal. Bioanal. Tech.
5
(
5
),
1000212
(
2014
).
34.
T. M.
Al-Bazali
,
J.
Zhang
,
M. E.
Chenevert
, and
M. M.
Sharma
, “
Factors controlling the membrane efficiency of shales when interacting with water-based and oil-based muds
,” in
SPE International Oil and Gas Conference and Exhibition in China
(
SPE
,
2006
), p.
SPE-100735-MS
.
35.
T. M.
AL-Bazali
,
Experimental Study of the Membrane Behavior of Shale during Interaction with Water-Based and Oil-Based Muds
(
The University of Texas at Austin
,
2005
).
36.
Y.
Zhi
,
Z.
Caineng
,
W.
Songtao
,
L.
Senhu
,
P.
Songqi
,
N.
Xiaobing
et al, “
Formation, distribution and resource potential of the “sweet areas (sections)” of continental shale oil in China
,”
Mar. Pet. Geol.
102
,
48
60
(
2019
).
37.
Y.
Cheng
,
M.
Zeng
,
Z.
Lu
,
X.
Du
,
H.
Yin
, and
L.
Yang
, “
Effects of supercritical CO2 treatment temperatures on mineral composition, pore structure and functional groups of shale: Implications for CO2 sequestration
,”
Sustainability
12
(
9
),
3927
(
2020
).
38.
A. A.
Hamouda
, “
Interaction between slug fracturing fluids energized with scCO2 and shale minerals
,”
J. Natural Gas Sci. Eng.
44
,
150
161
(
2017
).
39.
Q.
Lyu
,
P. G.
Ranjith
,
X.
Long
, and
B.
Ji
, “
Experimental investigation of mechanical properties of black shales after CO2-water-rock interaction
,”
Materials
9
(
8
),
663
(
2016
).
40.
B. L.
Alemu
,
P.
Aagaard
,
I. A.
Munz
, and
E.
Skurtveit
, “
Caprock interaction with CO2: A laboratory study of reactivity of shale with supercritical CO2 and brine
,”
Appl. Geochem.
26
(
12
),
1975
1989
(
2011
).
41.
K.
Wang
,
T.
Xu
,
F.
Wang
, and
H.
Tian
, “
Experimental study of CO2–brine–rock interaction during CO2 sequestration in deep coal seams
,”
Int. J. Coal Geol.
154–155
,
265
274
(
2016
).
42.
Y.
Tang
,
C.
Lv
,
R.
Wang
, and
M.
Cui
, “
Mineral dissolution and mobilization during CO2 injection into the water-flooded layer of the Pucheng Oilfield, China
,”
J. Nat. Gas Sci. Eng.
33
,
1364
1373
(
2016
).
43.
S.
Brunauer
,
L. S.
Deming
,
W. E.
Deming
, and
E.
Teller
, “
On a theory of the van der Waals adsorption of gases
,”
J. Am. Chem. Soc.
62
(
7
),
1723
1732
(
1940
).
44.
G.
Ni
,
S.
Li
,
S.
Rahman
,
M.
Xun
,
H.
Wang
,
Y.
Xu
et al, “
Effect of nitric acid on the pore structure and fractal characteristics of coal based on the low-temperature nitrogen adsorption method
,”
Powder Technol.
367
,
506
516
(
2020
).
45.
S.
Lowell
,
J. E.
Shields
,
M. A.
Thomas
, and
M.
Thommes
,
Characterization of Porous Solids and Powders: Surface Area, Pore Size and Density
(
Springer Science & Business Media
,
2006
).
46.
T.
Chen
,
X.-T.
Feng
, and
Z.
Pan
, “
Experimental study on kinetic swelling of organic-rich shale in CO2, CH4 and N2
,”
J. Nat. Gas Sci. Eng.
55
,
406
417
(
2018
).
47.
H.
Yin
,
J.
Zhou
,
Y.
Jiang
,
X.
Xian
, and
Q.
Liu
, “
Physical and structural changes in shale associated with supercritical CO2 exposure
,”
Fuel
184
,
289
303
(
2016
).
48.
P. J.
Jarboe
,
P. A.
Candela
,
W.
Zhu
, and
A. J.
Kaufman
, “
Extraction of hydrocarbons from high-maturity Marcellus shale using supercritical carbon dioxide
,”
Energy Fuels
29
(
12
),
7897
7909
(
2015
).
49.
X.
Luo
,
X.
Ren
, and
S.
Wang
, “
Supercritical CO2-water-shale interactions and their effects on element mobilization and shale pore structure during stimulation
,”
Int. J. Coal Geol.
202
,
109
127
(
2019
).
50.
A.
Goodman
,
S.
Sanguinito
,
M.
Tkach
,
S.
Natesakhawat
,
B.
Kutchko
,
J.
Fazio
et al, “
Investigating the role of water on CO2-Utica Shale interactions for carbon storage and shale gas extraction activities—Evidence for pore scale alterations
,”
Fuel
242
,
744
755
(
2019
).
51.
F.
Liu
,
P.
Lu
,
C.
Griffith
,
S. W.
Hedges
,
Y.
Soong
,
H.
Hellevang
et al, “
CO2–brine–caprock interaction: Reactivity experiments on Eau Claire shale and a review of relevant literature
,”
Int. J. Greenhouse Gas Control
7
,
153
167
(
2012
).
52.
Z.
Sun
,
H.
Zhang
,
Z.
Wei
,
Y.
Wang
,
B.
Wu
,
S.
Zhuo
et al, “
Effects of slick water fracturing fluid on pore structure and adsorption characteristics of shale reservoir rocks
,”
J. Nat. Gas Sci. Eng.
51
,
27
36
(
2018
).
53.
A.
Alalli
,
Q.
Li
,
A.
Jew
,
A.
Kohli
,
J.
Bargar
,
M.
Zoback
et al, “
Effects of hydraulic fracturing fluid chemistry on shale matrix permeability
,” in
SPE/AAPG/SEG Unconventional Resources Technology Conference, URTEC
,
2018
.
54.
C. E.
Weaver
and
L. D.
Pollard
,
The Chemistry of Clay Minerals
(
Elsevier
,
2011
).
55.
J.
Theiler
, “
Estimating fractal dimension
,”
J. Opt. Soc. Am. A
7
(
6
),
1055
1073
(
1990
).
56.
F.
Yang
,
Z.
Ning
, and
H.
Liu
, “
Fractal characteristics of shales from a shale gas reservoir in the Sichuan Basin, China
,”
Fuel
115
,
378
384
(
2014
).
57.
P.
Pfeifer
,
Y.
Wu
,
M. W.
Cole
, and
J.
Krim
, “
Multilayer adsorption on a fractally rough surface
,”
Phys. Rev. Lett.
62
(
17
),
1997
(
1989
).
58.
J.
Qiao
,
J.
Zeng
,
Y.
Ma
,
S.
Jiang
,
S.
Feng
, and
H.
Hu
, “
Effects of mineralogy on pore structure and fluid flow capacity of deeply buried sandstone reservoirs with a case study in the Junggar Basin
,”
J. Pet. Sci. Eng.
189
,
106986
(
2020
).
59.
N.
Liu
,
L.
Liu
,
X.
Qu
,
H.
Yang
,
L.
Wang
, and
S.
Zhao
, “
Genesis of authigene carbonate minerals in the Upper Cretaceous reservoir, Honggang Anticline, Songliao Basin: A natural analog for mineral trapping of natural CO2 storage
,”
Sediment. Geol.
237
(
3–4
),
166
178
(
2011
).