Introduction
The successful development of shale reservoirs in America alters the global petroleum supply and geopolitical pattern, further triggering a boom in exploration and development for shale reservoirs in the world [1⇓-3]. With its large abundance and great potential, shale oil is a vital strategic replacement resource in China [4⇓-6]. Compared with the marine shale oil in the U.S., however, the continental shale oil in China has a medium and low maturity mostly, featured with high wax content, high viscosity, and poor mobility [6-7]. Consequently, a fracture system with higher requirements of conductivity and effectiveness is claimed to achieve long-term stable production. To date, water-based fracturing fluid is most commonly used in shale oil reservoir fracturing. But supercritical CO2 fracturing is becoming a promising technology of non-aqueous fracturing due to the numerous advantages of CO2 in increasing the fracture complexity, replenishing the formation energy, increasing the mobility of crude oil, achieving carbon neutrality, and so on [8⇓-10].
When massive fracturing fluid is injected into reservoirs and retained in the fracture system, the fluid will interact with the reservoir rocks and further impact the physical and mechanical properties of shale. The main mechanism of water-shale interaction is the hydration expansion of water-sensitive clay minerals like montmorillonite and illite/smectite mixed layer. Sui et al. [11] and Xue et al. [12] investigated the micro-pore structure, hydration-induced fractures, and porosity and permeability of shale after hydration. Zeng et al. [13-14] analyzed the influence of various clay minerals and inorganic cations on the hydration characteristics of shale. Liu et al.[15] and Yang et al. [16] studied the deterioration behavior of mechanical parameters of different types of shales induced by hydration through triaxial mechanical experiments. The interaction mechanism between CO2 and shale includes adsorption expansion, chemical dissolution, extraction of dissolved organic matter, etc. Supercritical CO2 with strong adsorption capacity tends to adsorb on the surface of mineral particles such as organic matter and tiny clay, in turn reducing the surface free energy and leading to the expansion and deformation of shale [8,17]. Ao et al. [18] tested the deformation characteristics of shale induced by CO2 adsorption under different pressures. Jiang et al. [19] found that the specific surface area and porosity of shale were increased after the CO2-soaking, which is affected by soaking time, temperature, and pressure. Ozotta et al. [20] quantitatively characterized the changes in pore structure and connectivity of organic-rich shale after supercritical CO2-soaking. Tang et al. [17] and Ni et al. [21] studied the changes in the mechanical strength of shale under different CO2-soaking conditions through rock mechanics experiments. In addition, Zhang et al. [22] and Zhou et al. [23-24] respectively compared the softening degree of shale soaked by different types of fluids.
Most of the research mentioned above merely focused on the mechanical properties in a macroscale, but in the fields of reservoir fracturing and CO2 storage, lots of mechanical behaviors are closely related to micro-mechanics of rock. Taking sand fracturing in shale reservoirs as an example, the embedment and crushing of proppants related to hydraulic fracture closure are microscopic phenomena on the fracture surface. Therefore, it is more reasonable to use the micro-mechanical properties of shale to explain the closure behaviors of hydraulic fracture and analyze the fracture failure mechanism [25⇓-27]. At present, micron/nanoindentation technology has been widely used to measure the micro-mechanical parameters of shale [28⇓⇓-31], but the micro-mechanical properties of shale under the water/supercritical CO2-rock interaction still need to be further studied.
Taking the calcareous lamellar shale of the lower submember of the third Member of the Shahejie Formation of Paleogene (lower Es3 for short) in the Jiyang Depression of the Bohai Bay Basin as the research object, this paper investigated microstructure changes on the surface of shale soaked in water/supercritical CO2 under different soaking time, pressure and temperature through a series of high-temperature and high-pressure soaking experiments. Meanwhile, nanoindentation testing was performed to quantitatively characterize micro-mechanical properties of shale under water/supercritical CO2-rock interaction, which provides a basis for understanding the micro-mechanical response of shale in the process of CO2 storage and non-aqueous fracturing.
1. Materiel and methodology
1.1. Sample preparation
The samples used in this study were down-hole cores collected from the interval of 3080.75 m to 3126.95 m in the lower Es3 of the Jiyang Depression in the Shengli Oilfield. The lithology is calcareous lamellar shale (Fig. 1a ). According to the mineral composition test, the shale consists of 71.4% calcite, 9.0% quartz, 6.7% dolomite, 6.6% clay minerals, 3.9% pyrite, and a small amount of plagioclase as well as iron-dolomite. Among them, the clay minerals are mainly illite/smectite mixed layer and illite, with relative contents of 60.8% and 34.8%, respectively.
Fig. 1. The photos of core samples. |
To begin with, the core was processed into cubic samples with a side length of 1.5 cm and a height of 1.0 cm, along the directions parallel and perpendicular to the beddings by wire-cutting equipment (Fig. 1b ). Alcohol was used as a coolant during cutting in order to avoid water-core contact. One of the end faces perpendicular to the beddings was mechanically polished several times, and then polished with a wide-band argon ion device to guarantee the roughness of the testing surface meets the requirements of the nanoindentation experiment.
1.2. Soaking treatment
To simulate the fluid-rock interaction process under reservoir conditions, soaking experiments were performed using a high-temperature and high-pressure soaking device. This device was mainly composed of a CO2 source, a constant-rate and constant-pressure syringe pump, intermediate container, CO2 heating pipelines, reaction kettle, high-temperature oven, temperature, and pressure probes [32] (Fig. 2 ). This study focused on the influences of fluid type (water, supercritical CO2), soaking time (2, 4, 6 d), soaking pressure (10, 20, 30 MPa) and soaking temperature (60, 80, 100 °C) on the micro-mechanical properties of calcareous lamellar shale.
Fig. 2. High-temperature and high-pressure soaking device (modified according to Reference [32]). |
The procedures of the soaking test can be summarized as follows: (1) Before the soaking test, the shale sample was displaced by water to remove the oil inside and dried; (2) The equipment was tested to avoid any leakage during the experiment; (3) The sample was placed with the polished surface upward in the reaction kettle, and then the reaction kettle was heated to target temperature and maintained constant; (4) The reaction kettle was filled with soaking fluid through the constant-rate and constant-pressure syringe pump till the target pressure was reached. When using supercritical CO2 to soak the sample, the CO2 was preheated to the target temperature by heating pipeline to ensure the phase in the reaction kettle was supercritical; (5) Two probes were used to monitor temperature and pressure respectively in the reaction kettle during the soaking; (6) When the designed soaking time was reached, the fluid pressure in reaction kettle was slowly released to the atmospheric pressure. After the reaction kettle was cooled to room temperature naturally, the sample was taken out for surface cleaning and drying treatment. The soaked sample was not polished again to avoid any mechanical damage to the shale surface, which may interfere with the experimental results of shale damage observation and micro-mechanical parameters.
1.3. Nanoindentation test
Keysight G200 nanoindentation instrument was used to quantitatively characterize the micro-mechanical properties of shale before and after soaking [25-26]. Continuous stiffness measurement was used to study the continuous change of micro-mechanical parameters with the indentation depth. Berkovich indenter was used in the test, whose constant strain rate in the loading stage was 0.05 s−1, harmonic displacement was 2 nm, the frequency was 45 Hz, and the target penetration depth was 4000 nm. Dot matrix indentation was used to statistically characterize the changes in elastic modulus and hardness on the shale surface after soaking with different fluids. To avoid disturbance between indentation points, the interval between adjacent indentation points was 100 μm.
2. Experimental results and analysis
2.1. Shale damage characteristics under water/supercritical CO2-rock interaction
To avoid the impact of the initial micro-fractures on the surface of the original shale sample on the shale damage characteristics under the interaction of water/supercritical CO2-rock, the surfaces of all shale samples were examined by field emission environmental scanning electron microscopy (FE-SEM) before the soaking experiment, and an area about 2 mm×2 mm without initial fractures was selected for observation. After the soaking experiment, the microstructure changes on the surface of the shale in the observation area were inspected (Fig. 3 ). After the shale sample was soaked in water/supercritical CO2, its surface damage was mainly induced fractures in the clay-rich laminae.
Fig. 3. Microstructure characteristics of shale after liquid soaking. |
Under the hydration effect, the illite/smectite mixed layer is prone to occur hydration expansion, and illite is prone to collapse [12]. Therefore, significant particle falling off (Fig. 3a , 3b), clay mineral crystal layer stripping (Fig. 3c ), and induced fracture bifurcation (Fig. 3d ) can be seen within the clay-rich laminae. The adsorption of supercritical CO2 can lead to shale expansion and deformation [17,33]. During the soaking experiment, CO2 entered into the shale mainly along the laminae with strong permeability under differential pressure and was adsorbed on the surface of clay mineral particles. Therefore, the expansion deformation caused by adsorption mainly occurred in the lamellar structure. When the expansion stress was higher than the mechanical strength of the laminae, induced fractures were formed. According to the FE-SEM observation, the induced fracture caused by CO2 adsorption was narrower than that caused by hydration (Fig. 3e , 3f). In addition, the pores on the shale surface were enlarged or new micro-pores were formed due to the dissolution caused by CO2.
The number and width of induced fractures inside the sample were statistically analyzed to quantitatively evaluate the damage degree of shale surface under different soaking conditions (Fig. 4 ). Although the total content of clay minerals in shale samples only accounts for 6.6%, the influence of hydration cannot be ignored because the illite/smectite mixed layer is the main clay minerals, which is easy to hydrate and expand. Under varying conditions of soaking time, temperature, and pressure, the hydration-induced fracture width generally surpasses supercritical CO2-induced fracture width.
Fig. 4. Statistical results of induced fractures in shale under different soaking conditions. |
The sensitivity of induced fracture width to different variables significantly varies under the two types of fluid-soaking. When the soaking pressure was 20 MPa and the temperature was 80 °C, the number and the average width of induced fractures gradually increased after water-soaking for 2, 4, and 6 d, respectively, which were 1.52, 2.87, 4.73 μm, and the maximum increase was 211.2% (Fig. 4a ). This indicated that the degree of hydration damage on the shale surface aggravated with the soaking time, which was consistent with the results of Ma et al. [34]. After supercritical CO2-soaking, the average width of the fracture was 1.11, 1.10, 1.46 μm respectively, and the maximum increase was 31.5% (Fig. 4a ). Due to the characteristics of ultra-low interfacial tension and ultra-low viscosity of supercritical CO2, it is easy to enter the micro-pores and laminae in shale, and quickly reach adsorption equilibrium in the vicinity of the fracture, so the soaking time has a limited effect on the fracture width.
Increasing the soaking pressure can increase the number of water/supercritical CO2-induced fractures and significantly increase the fracture width. With the increase of soaking pressure, the depth of water imbibition into the rock sample along the laminae increases and the imbibition volume increases, so the hydration is enhanced. However, when the shale is soaked with supercritical CO2, the increase in soaking pressure promotes the adsorption of CO2 in shale laminae and pores and enhances the dissolution ability of CO2. In addition, in the process of pressure unloading, the pressure in the container gradually decreases, and the change of CO2 pressure inside the shale has a hysteretic effect. Under the effect of pressure difference inside and outside the core, CO2 phase change and expansion occur to some extent [35], resulting in the generation and further expansion of induced fractures. When the fluid temperature was 80 °C and the soaking time was 6 d, with the soaking pressure increasing from 10 MPa to 30 MPa, the statistical mean value of induced fracture width of shale surface after soaking with water and supercritical CO2 increased from 1.95 μm, 0.49 μm to 3.99 μm, 1.65 μm, respectively. The increases were 104.6% and 236.7%, respectively (Fig. 4b ).
When the soaking pressure was 20 MPa and the soaking time was 6 d, the influence of soaking temperatures on induced fracture width was more complex in the shale soaked with water. As the soaking temperature increased from 60 °C to 100 °C, the statistical average value of induced fracture width first decreased and then increased, but the maximum width of induced fracture gradually increased. This is related to the difference in thermal expansibility among shale samples and the difference in thermal deformation during the heating process caused by the heterogeneity of thermal expansion of shale samples. For the shale soaked in supercritical CO2, two competing mechanisms existed for the influence of temperature: on the one hand, the increase of temperature leads to thermal expansion of shale, which promotes the propagation of induced fractures; On the other hand, rising temperature inhibits the adsorption of CO2 and weakens the adsorption expansion and deformation of shale [17,21]. In this experiment, the promotion effect of thermal expansion on fractures was dominant. With the increase in the soaking temperature of supercritical CO2, the number of induced fractures increased, and the statistical average and maximum fracture widths increased gradually (Fig. 4c ).
2.2. Failure patterns and characteristics of load-indentation depth curves
After the nanoindentation experiment, the micro-mechanical parameters before and after soaking were obtained by the load-indentation depth curve, and the failure characteristics near the indentation points were observed by FE-SEM. The corresponding relation between micro-mechanical parameters and indentation points was determined by the position of indentation points in the matrix. Based on the failure patterns and characteristics of the load-indentation depth curve of different indentation points on the shale surface, indentation points can be roughly divided into three categories.
(1) The indentation in the matrix area far away from the laminae: The indentation pattern was mostly standard triangular pyramid shape, and there was no obvious induced fracture near the indentation point or only a few induced microfractures were developed at the indentation edge (Fig. 5a ). As the indentation depth increased, the load gradually increased, while the elastic modulus and hardness decreased and gradually stabilized. The values at the maximum indentation depth were 65.60 GPa and 2.10 GPa, respectively (Fig. 5b ).
Fig. 5. Failure morphology and micro-mechanical response curves of indentation points at different positions. |
(2) The indentation near the induced fracture: Affected by the laminae and the fracture induced by soaking in water/supercritical CO2, the failure characteristics and micro-mechanical response characteristics of the indentation point were more complex. With the increase of indentation depth, the contact area between the indenter and shale surface increased, and the influence of laminae was gradually obvious. The indentation points close to the laminae were morphologically incomplete, and the deformation of surface minerals resulted in local compaction and closure of induced fractures generated during the soaking (Fig. 5c ). In addition, under the action of drag force, a number of stress-induced tension fractures parallel to the lamina were formed within the indentation range. The emergence of new fractures on the shale surface or the influence of clay-rich laminae led to an obvious "accelerated advance" phenomenon on the load curve [28,36], indicating that the indentation depth increased rapidly, but the load increased slowly (Fig. 5d ). During the whole process, the micro-mechanical parameters fluctuated in a decreasing trend with the increase of the indentation depth, and gradually stabilized after reaching a relatively large indentation depth (about 2500 nm). The elastic modulus and hardness at the maximum depth were 48.30 GPa and 1.00 GPa, respectively, which were lower than the indentation point in the matrix area.
(3) The indentation point at the induced fracture: "No-load" phenomenon appeared in the load-indentation depth curve of the loading stage [29], which means that the load approaches 0 as the indentation depth increases rapidly in the early stage (Fig. 5f ). When the indentation depth reached a certain degree (about 1000 nm), the load gradually increased with the increasing of indentation depth. In the middle and late loading stage, the elastic modulus and hardness of the indentation point at the induced fracture increased and tended to be stable. The elastic modulus and hardness at the maximum indentation depth were 23.90 GPa and 0.40 GPa, respectively, which were only 36.4% and 19.0% of the micro-mechanical parameters of the matrix. Although the maximum load was less than 50 mN, there was an obvious compress-crushed zone at the fracture wall of the indentation point (Fig. 5e ). When mineral debris is fell into the fracture, it can lead to blockage of the pore throat in the proppant pack, increasing resistance of oil flow.
2.3. Comparison of micro-mechanical characteristics of shale
Statistical analysis of nanoindentation test results of shale samples soaked in water/supercritical CO2 shows that there is an obvious positive correlation between elastic modulus and hardness, but the test results of the same sample show great discreteness. In this study, the maximum, minimum, and average values of mechanical parameters under different conditions were summarized, and the parameters with the frequency of 25%-75% in each group of sorted data were statistically analyzed. In addition, for the original shale sample, the statistical average value of elastic modulus at the indentation point is 66.30 GPa, and the value distributed in 25%-75% is 61.10-75.30 GPa. The statistical average value of hardness is 2.10 GPa, and the values distributed in 25%-75% are 1.68-2.53 GPa (Fig. 6 ).
Fig. 6. Statistical results of micro-mechanical parameters of original shale. |
2.3.1. Effect of soaking time
The observation results of shale surface damage after water-soaking show that the longer the soaking time, the more fractures observed on the sample surface and the larger the width of hydration-induced fractures (Fig. 4a ). The microstructure damage of the shale surface aggravates the deterioration of shale micro-mechanical strength. The elastic modulus-hardness diagram shows that the main distribution of micro-mechanical parameters shifts from the region of high elastic modulus and high hardness to the region of low elastic modulus and low hardness (Fig. 7a -7c). After being soaked in water for 2, 4, 6 d, the statistical average values of elastic modulus of shale were 57.32, 52.00, 43.34 GPa, respectively, which decreased by 13.5%, 21.6%, and 34.6%, respectively, compared with the original samples. The statistical average values of hardness were 1.52, 1.49, 0.89 GPa, respectively, which were 27.6%, 29.0%, and 57.6% lower than those of the original samples respectively (Table 1 ). The distribution of mechanical parameters of shale soaked in supercritical CO2 does not change significantly with soaking time (Fig. 7d -7f), which is consistent with the observation of microscopic damage on the shale surface from a scanning electron microscope (Fig. 4a ). After being soaked in supercritical CO2 for 2, 4, 6 d, the average values of elastic modulus were 58.85, 56.50, 56.66 GPa respectively, with the maximum decrease compared with the original sample of 14.8%. The damage induced by CO2 was more obvious in the early stage of soaking. The average hardness was 1.71, 1.65, and 1.47 GPa, with a maximum decrease of 30.0% compared with the original sample, suggesting that the influence of supercritical CO2 on shale micro-mechanical properties is lower than that of hydration (Table 1 ).
Fig. 7. Distribution of micro-mechanical parameters of shale after different soaking times (pressure of 20 MPa, temperature of 80 °C). |
Table 1. Mechanical parameters of nanoindentation after water/supercritical CO2-soaking for different times (20 MPa and 80 °C) |
| Soaking fluid | Soaking time/d | Modulus of elasticity/GPa | Hardness/GPa | ||||||
|---|---|---|---|---|---|---|---|---|---|
| Minimum value | Maximum value | 25%-75% interval value * | Average | Minimum value | Maximum value | 25%-75% interval value * | Average | ||
| Water | 2 | 28.50 | 75.34 | 52.95-66.00 | 57.32 | 0.45 | 2.59 | 1.10-1.83 | 1.52 |
| 4 | 36.90 | 67.10 | 46.15-57.90 | 52.00 | 0.65 | 2.47 | 1.17-1.81 | 1.49 | |
| 6 | 29.90 | 64.00 | 39.50-47.00 | 43.34 | 0.41 | 1.67 | 0.73-0.98 | 0.89 | |
| Supercritical CO2 | 2 | 28.18 | 71.21 | 54.40-64.10 | 58.85 | 0.44 | 2.55 | 1.42-2.07 | 1.71 |
| 4 | 42.40 | 68.50 | 51.25-61.95 | 56.50 | 0.96 | 2.42 | 1.33-2.06 | 1.65 | |
| 6 | 32.12 | 70.28 | 51.80-62.30 | 56.66 | 0.53 | 2.19 | 1.21-1.71 | 1.47 | |
Note: * Parameter distribution between 25% and 75% (i.e. between upper and lower quartiles) after data sorting |
The softening of the fracture surface (that is, the decrease of micro-mechanical strength) is one of the mechanisms affecting proppant embedding and fragmentation, which is of great significance for understanding the failure of the induced fracture system. The micro-mechanical test results show that even in shale with low clay mineral content, the influence of water-rock interaction on the microscopic strength of the fracture surface cannot be ignored. At present, after shale reservoir fracturing, shale gas well is often shut in for a certain period to further expand the stimulation volume and take maximum advantage of the fracturing fluids to energize the formation and enhance the imbibition and oil displacement effects, which means there is a physical process of fluid-rock interaction [37-38]. Without considering sand carrying capacity and treatment difficulty, supercritical CO2 has certain advantages in reducing the softening degree of the fracture surface and maintaining the effectiveness of induced fractures.
2.3.2. Influence of soaking pressure
The statistical results of micro-mechanical parameters under different soaking pressures show that the micro-mechanical parameters of shale soaked in water and supercritical CO2 decrease with the increase of soaking fluid pressure (Table 2 and Fig. 8 ). However, compared with water, the softening degree of shale surface caused by supercritical CO2 at the same pressure is relatively low. When the soaking pressure was 10 MPa and 30 MPa, respectively, the surface elastic modulus of shale soaked in water for 4 d was 59.08 GPa and 43.49 GPa, respectively, 10.9% and 34.4% lower than that of the original shale, respectively. The hardness was 1.74 GPa and 1.06 GPa, respectively, 17.1% and 49.5% lower than that of the original shale, respectively. The surface elastic modulus of shale soaked in supercritical CO2 for 4 d was 62.34 GPa and 51.91 GPa, respectively, 6.0% and 21.7% lower than that of the original shale, respectively. The hardness was 1.80 GPa and 1.28 GPa respectively, 14.3% and 39.0% lower than that of the original shale respectively.
Table 2. Mechanical parameters of nanoindentation of rock samples after water/supercritical CO2-soaking at different pressures (soaking for 4 d at 80 °C) |
| Soaking fluid | Soaking pressure/ MPa | Modulus of elasticity/GPa | Hardness/GPa | ||||||
|---|---|---|---|---|---|---|---|---|---|
| Minimum value | Maximum value | 25%-75% interval value* | Average | Minimum value | Maximum value | 25%-75% interval value* | Average | ||
| Water | 10 | 23.80 | 91.81 | 47.00-70.10 | 59.08 | 0.60 | 3.39 | 1.27-2.15 | 1.74 |
| 30 | 32.60 | 54.90 | 39.00-48.35 | 43.49 | 0.70 | 1.75 | 0.85-1.23 | 1.06 | |
| Supercritical CO2 | 10 | 39.60 | 88.60 | 57.10-67.90 | 62.34 | 0.64 | 2.73 | 1.43-2.28 | 1.80 |
| 30 | 23.90 | 87.50 | 43.20-63.35 | 51.91 | 0.40 | 2.52 | 0.73-1.68 | 1.28 | |
Note: * Parameter distribution between 25% and 75% (i.e. between upper and lower quartiles) after data sorting |
Fig. 8. Distribution of micro-mechanical parameters of shale under different soaking pressures (soaking for 4 d at 80 °C). |
The statistical results of induced fractures on the shale surface after being soaked in two fluids (Fig. 4b ) show that the softening degree of the fracture surface increases with the increase of soaking pressure, which is related to the increasing number and width of induced fractures after water/supercritical CO2-shale interaction, leading to the expansion of the area affected by induced fractures. Therefore, the number of indentation points located near and at induced fractures increases. In addition, hydration and CO2 chemical dissolution lead to the formation of new pores in the matrix, which is also one of the reasons for the overall deterioration of shale micro-mechanical strength.
2.3.3. Effect of soaking temperature
Espinoza et al. [31] found that the elastic modulus of siliceous shale (with the quartz content of 66.3%) of the Silurian Longmaxi Formation in the Sichuan Basin only decreased by 10%, and the hardness did not change significantly when the temperature rose from 20 °C to 300 °C. Lu et al. [39-40] found that temperature can facilitate shale hydration and reduce hydration characteristic time. In this study, the statistical average values of elastic modulus and hardness of shale soaked in water for 4 d at 60 °C were 51.57 GPa and 1.50 GPa, respectively, which were 22.2% and 28.6% lower than those of the original shale, respectively, and had no obvious difference with those of shale soaked in water at 80 °C (Table 1 and Table 3 ). When the temperature rose to 100 °C, the deterioration of micro-mechanical strength was further aggravated, and the number of indentation points with low elastic modulus and low hardness was obviously increased, which means that the distribution of indentation data points moved to the lower left corner (Fig. 9a , 9b). The statistical average values of elastic modulus and hardness decreased to 40.84 GPa and 1.24 GPa, respectively, with a decrease of 38.4% and 41.0%, respectively (Table 3 ). Generally, the deeper the reservoir is buried, the higher the temperature is. Because the hydration is promoted by temperature, the softening degree of the fracture surface in deep fracturing is higher. However, the increase in burial depth is accompanied by the illitization of montmorillonite [41], and the transformation from strong water-sensitive clay minerals to weak water-sensitive clay minerals will weaken the damage degree of hydration to shale. Therefore, the influence of temperature on hydration still needs to be studied more systematically and deeply.
Table 3. Mechanical parameters of nanoindentation after water/supercritical CO2-soaking at different temperatures (soaking for 4 d at 20 MPa) |
| Soaking fluid | Soaking temperature/ °C | Modulus of elasticity/GPa | Hardness/GPa | ||||||
|---|---|---|---|---|---|---|---|---|---|
| Minimum value | Maximum value | 25%-75% interval value* | Average | Minimum value | Maximum value | 25%-75% interval value* | Average | ||
| Water | 60 | 31.90 | 72.10 | 45.75-58.05 | 51.57 | 0.48 | 2.43 | 1.13-1.89 | 1.50 |
| 100 | 25.40 | 59.32 | 36.50-44.28 | 40.84 | 0.47 | 2.48 | 0.93-1.43 | 1.24 | |
| Supercritical CO2 | 60 | 44.10 | 73.18 | 56.77-64.60 | 60.55 | 0.88 | 2.46 | 1.60-2.05 | 1.81 |
| 100 | 24.80 | 73.30 | 39.20-57.72 | 49.57 | 0.41 | 2.44 | 1.18-2.00 | 1.55 | |
Note: * Parameter distribution between 25% and 75% (i.e. between upper and lower quartiles) after data sorting |
Fig. 9. Distribution of micro-mechanical parameters of shale at different soaking temperatures (soaking for 4 d at 20 MPa). |
The observation results of shale microscopic damage characteristics show that soaking temperature promotes the formation of induced fractures in calcareous lamellar shale during supercritical CO2-soaking (Fig. 4c ). In addition, the higher the temperature, the stronger the dissolution ability of CO2 to carbonate minerals such as calcite, and the more significant the alteration of microscopic pore structure on shale surface [17]. Therefore, the higher the temperature, the more severe the softening phenomenon of the matrix is (Fig. 9c , 9d). The statistical results of micro-mechanical parameters show that the elastic modulus of shale decreased from 60.55 GPa to 49.57 GPa when the soaking temperature of supercritical CO2 increased from 60 °C to 100 °C, which was 8.7% and 25.2% lower than that of original shale respectively. The hardness decreased from 1.81 GPa to 1.55 GPa, which was 13.8% and 26.2% lower than that of the original shale, respectively.
3. Discussion
In this paper, the micro-mechanical properties of shale under the interaction of water/supercritical CO2-rock were experimentally studied. As the basic research of non-aqueous fracturing technology, this study still needs to be further investigated in the following three directions. Firstly, this paper only carried out experimental research on the micro-mechanical properties of shale under the action of water/supercritical CO2 alone. CO2 pre-injection and composite fracturing, as the injection method commonly used in the field at present, is often accompanied by more complex coupled physical and chemical processes of CO2-water-shale. Zou et al. [42], Li et al. [30] proved that the coupled physical and chemical process of CO2-water-rock has a significant impact on the macro-physical and mechanical properties of shale, but its influence on the micro-mechanical characteristics remains to be investigated. Secondly, although this study discussed the influence of soaking pressure on the experimental results, there are still differences from the actual underground conditions. Shale laminae are affected by compaction under high confining pressure, which inhibits the hydration expansion and supercritical CO2 adsorption expansion, and consequently reduces the shale damage degree under the interaction of water/supercritical CO2-rock. The comparison between the laboratory results and those in the reservoir conditions needs to be further investigated. Thirdly, the upscaling study of rock mechanical parameters using homogenization theory has become an important means to understand the rock mechanical characteristics of reservoirs [28]. The degree of fluid-rock interaction is closely related to the core size effect, and has different influence degrees on micro and macro mechanical properties. The influence of different fluid-shale interactions should be discussed to improve the accuracy of the upscaling model in future studies.
4. Conclusions
After being soaked in water/supercritical CO2, the damage of calcareous lamellar shale is mainly caused by the formation of induced fractures in the clay-rich laminae. Compared with supercritical CO2-soaking, the induced fracture width caused by water-soaking is larger, and increases with the increase of soaking time and soaking pressure. The induced fracture width caused by supercritical CO2-soaking increases with the increase of soaking pressure and temperature, but is not sensitive to soaking time.
The characteristics of the micro-mechanical response of shale vary at different positions after soaking. The matrix dominated by calcite is characterized by high elastic modulus, high hardness, and intact indentation morphology; the elastic modulus and hardness near the induced fractures decrease, and compaction closure of fractures and stress-induced tension fractures parallel to the direction of the laminae can be seen near the laminae; the elastic modulus and hardness at the induced fractures are the lowest, where the abnormal phenomenon that the micro-mechanical parameters increase with the increase of the indentation depth occurs during the loading process.
Generally, the declining degree of the micro-mechanical properties of shale surface caused by water/supercritical CO2-soaking intensifies with the increasing of soaking time, soaking pressure, and soaking temperature. Under the soaking pressure of 20 MPa and temperature of 80 °C, with the soaking time increasing to 6 d, the elastic modulus of shale soaked in water and supercritical CO2 decreases by 34.6% and 14.8%, respectively, compared with those of the original shale, and the hardness decreases by 57.6% and 30.0%, respectively. When the soaking temperature is 80 °C and the soaking time is 4 d, with the soaking pressure increasing to 30 MPa, the elastic modulus of shale after soaking in water and supercritical CO2 decreases by 34.4% and 21.7%, and the hardness by 49.5% and 39.0%. When the soaking pressure is 20 MPa and the soaking time is 4 d, with the soaking temperature increasing to 100 °C, the elastic modulus of the shale surface after soaking in water and supercritical CO2 decreases by 38.4% and 25.2%, and the hardness by 41.0% and 26.2%. The mechanical damage degree of shale surface under the interaction of supercritical CO2 and shale is lower than that under the interaction of water and shale, so supercritical CO2 can act as a fracturing fluid capable of preventing fracture surfaces of shale reservoir from softening.