Keywords

1 Introduction

Supercritical CO2 has gained significant attention over recent years as a novel form of fracturing fluid. This is attributed to its advantageous properties, including high density, low viscosity, high diffusion coefficient, low surface tension, and superb heat and mass transfer performance [1,2,3]. For tight sandstone reservoirs, experimental research on fracture propagation through supercritical CO2 fracturing has found that The commencement force for fracturing using supercritical CO2 is approximately 28.2% less than that of utilizing slick water. Furthermore, there is an increased complexity of fractures in conditions with low horizontal stress differences [4,5,6]. Supercritical CO2 does not cause damage to the reservoir and is prone to backflow, effectively avoiding near wellbore formation blockage, protecting oil and gas reservoirs, and improving reservoir permeability [7, 8]. In tight sandstone, CO2 generates calcite precipitation through physical and chemical interaction with reservoir minerals, which can also realize the local storage of CO2, thus reducing environmental pressure [9, 10]. Therefore, supercritical CO2 has the potential for reservoir transformation, carbon utilization, and carbon sequestration. Guided by the “dual carbon” goal, the technology of fracturing using supercritical CO2 exhibits vast potential for application in various fields.

With the proposal and development of supercritical CO2 fracturing technology, the study of fracture propagation law in such technology is a crucial aspect. Zhang X et al. [11] used CT scanning technology to compare fracture propagation in water, liquid CO2 and supercritical CO2, and reflected changes in fracture pressure through AE signals; JIANG Y [12] plotted injection pressure and temperature curves and proposed the mechanism of expansion and thermal stress promoting fracture propagation using active acoustic waves P and S waves. Zhou D [13, 14] conducted visualization experiments on fracture propagation in supercritical CO2 and believed that CO2 phase transition leads to dynamic fracture propagation and the formation of complex fractures. Chen H [15] used cement specimens to test the influence of supercritical CO2 on the complexity of fractures, and believed that the stress state has a significant impact on supercritical CO2 fracturing. Due to the lack of systematic research on the initiation and fracturing of tight sandstone in the Sichuan Basin, fracture propagation patterns, and post fracturing fracture morphology caused by supercritical CO2 and slickwater, it is not possible to effectively guide the formulation of supercritical CO2 fracturing plans. In combination with the geological characteristics and engineering needs of the block, this paper uses the Shaximiao formation outcrop in Sichuan Basin and uses a large true triaxial fracturing experimental device and CT scanner to carry out true triaxial supercritical CO2 and slick water fracturing experimental study in tight sandstone, and focus on quantifying the fracture pressure, tortuosity and fractal dimension of fractures in tight sandstone in Sichuan Basin while investigating the impact of various fluids, flow rates, and stress disparities on tight sandstone fracture expansion. This study aims to address the significant challenges associated with supercritical CO2 fracturing in tight sandstone.

2 Experimental Apparatus

The experimental apparatus employed for the true triaxial fracturing experiment consists of several components, including a triaxial model system, confining pressure system, heating control system, preheating system, pumping system, measurement system, and data acquisition system. The procedure for the experiment is illustrated in Fig. 1. To simulate hydraulic fracturing, similarity criteria were followed [16, 17]. Samples with lower strength were used or the viscosity of the fracturing fluid was increased to reduce the impact of fracture toughness on fracture growth, avoid dynamic fracture growth during the fracturing experiment and affect the experimental results.

Fig. 1.
figure 1

Flow Chart of True Triaxial Fracturing Experimental Device.

Full size image

CT scanners measure the object being measured by emitting an X-ray beam, and then establish a 3D image of the object based on the differences in X-ray absorption and transmittance of different components within the object using a computer processing system. The CT scanning system allows for real-time observation and imaging of the internal structure evolution of the specimen during the experimental process, accurately reflecting the internal structure of rock bedding, fractures, and other structures. It is a completely non-destructive testing technology for rock samples.

In order to get the coordinate parameters of the fracture surface and further analyze the parameters such as tortuosity and roughness of the fracture surface [18], some rock specimens after fracturing were cut open and Trimble FX 3D laser scanner was used. Scan the fracture surface of the rock specimen to obtain the scanning data point coordinates x, y, and z, and the data collection is based on the laser triangle principle.

3 Experimental Samples and Methods

To guarantee the selected rock samples’ representativeness, a systematic field survey was performed on the Shaximiao Formation outcrop in the Sichuan Basin. The survey aimed to identify the reservoir’s outcrop location, followed by a comparison of the mineral composition, microstructure, and other characteristics of the outcrop and formation core. Finally, the Shaximiao Formation tight sandstone outcrop samples in the Jiangyou area, Mianyang, Sichuan were selected. After retrieving the samples from the field to the indoor area, the sandstone was processed into a cube specimen with a side length of 20 cm, and a simulated wellbore with a diameter of 1.2 cm and a height of 11 cm was drilled in the middle of the specimen. Special chemical adhesive (AB adhesive) is used to fix the steel injection pipe into the simulated wellbore of the specimen, which is used to simulate the fracturing wellbore. To obtain the required samples for the true triaxial physical simulation experiment, we have designed the injection pipe with a sealing length of 9 cm, which includes a 2 cm open hole section located at the center of the specimen. Our laboratory study focuses on the true triaxial supercritical CO2 and slick water fracturing of tight sandstone. This study aims to investigate the impact of various factors such as different fluids, displacement, and stress differences on the propagation of tight sandstone fractures. To conduct the experiments, we have developed a specific experimental plan detailed in Table 1, and the experimental temperature has been set at 40 °C.

Table 1. Functional parameters of the proppant transport characteristic complex fracture experimental device.
Full size table

Among them, in order to simulate the crustal stress state of vertical wells in tight sandstone reservoirs [19, 20], triaxial hydraulic loading is used to achieve triaxial stress loading, and the minimum horizontal principal stress is applied respectively σ H and maximum horizontal principal stress σ H perpendicular to the wellbore axis, vertical stress σ V is parallel to the wellbore axis, as shown in Fig. 2.

Fig. 2.
figure 2

Schematic diagram of stress loading on a true triaxial large physical model.

Full size image

Based on the experiment’s objectives and plan, the procedure for conducting the experiment entails the following steps:

  1. (1)

    To clarify the internal bedding and natural fracture distribution of tight sandstone, CT scans were performed on the prepared samples;

  2. (2)

    Seal the test piece into the rubber sleeve and let it air dry for 3 days. After checking the sealing of the equipment, place the test piece into the loading triaxial pressure chamber;

  3. (3)

    Connect the fracturing fluid injection pump, injection pipeline, and simulated wellbore;

  4. (4)

    After inspecting the entire pipeline and data collection system, according to the experimental conditions in Table 1, preheat the loading chamber in a water bath for 30 min to subject the sample to three-dimensional stress loading;

  5. (5)

    Turn on the heater for heating, start the booster pump and data acquisition software, inject fluid, and start the fracturing experiment;

  6. (6)

    After the fracturing experiment is completed, stop the booster pump and data acquisition software, and slowly unload the three-dimensional stress to normal pressure;

  7. (7)

    Remove the loading device from the water bath, retrieve the sample, take photos and record, use CT to scan the fracturing sample, and take photos to record the fracture morphology.

4 Experimental Results and Analysis

4.1 The Influence of Fluid Type on the Propagation Law of Fractures in Tight Sandstone

In terms of fracture pressure, due to the small size of the sample, when the ultimate strength is reached around the wellbore, the initiation and propagation of fractures occur almost simultaneously, so the peak value of the curve is considered as the initiation pressure. The fracturing curve of supercritical CO2 and slick water fracturing is shown in Fig. 3.

Fig. 3.
figure 3

Fracturing curve of supercritical CO2 and slickwater.

Full size image

From the figure, it illustrates that under the conditions of a stress difference of 8MPa and a displacement of 30 mL/min, the fracture pressure of supercritical CO2 is 13.9 MPa, rock fracture takes 303 s, the fracture pressure of slickwater is 23.4 MPa, rock fracture takes 81 s. This indicates that compared to slickwater, the fracture pressure of supercritical CO2 has been observed to decline by 9.5 MPa, with a mean reduction of 40.6%, and the rock fracture time has increased by 274.1%. This is because the surface tension of supercritical CO2 is extremely low, so supercritical CO2 fracturing only requires a small amount of internal pressure to extend, while smooth water has a higher surface tension and can only enter large-scale fractures. The pressure-acting distance inside the fractures is short, which is not conducive to activating micro fractures and leading to fracture pressure. The pressure change curve of sandstone during supercritical CO2 fracturing can be clearly categorized into three distinct stages: compression stage, rapid pressurization stage, and fracture initiation and failure stage. In the compression stage, the initial CO2 quickly fills the pipeline and holes, and the high compression characteristics of supercritical CO2 make it exhibit a gentle pressurization phenomenon, and enter the rapid pressurization stage. The pressure curve shows a sharp rise, showing a linear growth similar to the water fracturing curve. The pressure change curve of smooth hydraulic fracturing sandstone during the initial injection stage is slow due to the discharge of air from the pipeline and borehole. Once filled, the pressure within the specimen increases rapidly, leading to its eventual damage and a subsequent notable extension of the time taken for rock fracture.

In terms of fracture morphology, the fracture morphology formed by supercritical CO2 and slick water fracturing under the same injection displacement and stress difference conditions is shown in Fig. 4.

Fig. 4.
figure 4

Fracture morphology diagram of supercritical CO2 (left) and slick water (right) fracturing.

Full size image

From the figure, it illustrates that the hydraulic fracture HF formed by supercritical CO2 fracturing extends along the direction of the maximum horizontal principal stress. During the fracture process, it is common for branch fractures, such as F1 and F2, to form and slightly deviate from their original propagation direction. This leads to the creation of an intricate network of fractures where the major and minor fractures intersect and connect with each other. The hydraulic fracture HF formed by smooth water fracturing starts at the open hole section and extends along the direction of the maximum horizontal principal stress to form a simple transverse fracture with a relatively straight and single fracture morphology.

As the main seepage channel in the sandstone after reservoir reconstruction, the artificial fracturing fracture directly determines the quality of the fracturing effect. The fracture characteristics mainly include: fracture space morphology, complexity, tortuosity and roughness of the fracture surface and other parameters. To further study the fracture morphology after fracturing with different fluids, CT scans were performed on the rock core to obtain the distribution and type of fracturing fractures, as shown in Fig. 5.

Fig. 5.
figure 5

CT scan images of supercritical CO2 (left) and slick water (right) after fracturing.

Full size image

The figure reveals that supercritical CO2 fracturing yields a more intricate and uneven fracture surface in contrast to hydraulic fracturing. Tortuosity, the proportion of the actual fracture length to the straight-line length, reflects the fluid flow’s curvature within the fracture. To determine fractal tortuosity, the rock specimen’s fracture surface underwent 3D laser scanning. The resulting data indicates that the tortuosity of supercritical CO2 fracturing is 2.23 times higher than that of hydraulic fracturing. The surface roughness of the fracture is larger and the spatial morphology of the fracture is more complex, which helps to provide a more effective seepage channel for the formation fluid.

4.2 The Influence of Injection Displacement Rate on the Fracture Propagation Law in Tight Sandstone

The injection displacement rate is another important engineering factor that affects the morphology of hydraulic fractures. To study the effect of different injection rates on fracture pressure and fracture propagation, the fracture curves of supercritical CO2 under different injection rates under the condition of 8 MPa horizontal stress difference are shown in Fig. 6.

Fig. 6.
figure 6

Fracturing curves of supercritical CO2 at different injection rates

Full size image

From the figure, it illustrates that given similar horizontal stress differential conditions, an increase in supercritical CO2 displacement from 30 mL/min to 40 mL/min caused a 11.1% rise in fracture pressure from 13.6 MPa to 15.1 MPa, while the fracture time declined by 13.3% from 300 s to 260 s. However, when the displacement was further increased to 60 mL/min, the fracture pressure rose by 26.5% to 17.2 MPa, and the fracture time decreased by 43.3% to 170 s. It should be noted that an increase in injection displacement results in a more rapid CO2 pressurization rate during the pore pressurization phase, which reduces the time required to inject the identical volume of CO2. Owing to the restricted permeability of the tight sandstone matrix, the infiltration rate of CO2 into the pores of the matrix shows no significant enhancement. Consequently, an elevated injection displacement causes an increment in initiation pressure and a reduction in fracture time.

In terms of fracture morphology, under the condition of 8 MPa horizontal stress difference, the fracture morphology of supercritical CO2 under different injection displacement rates is shown in Fig. 7.

Fig. 7.
figure 7

Fracture morphology diagram of supercritical CO2 fracturing under different injection rates

Full size image

From the figure, it illustrates that under the 8 MPa horizontal stress difference, the hydraulic fracture HF formed by supercritical CO2 fracturing extends along the direction of the maximum horizontal principal stress at an injection flow rate of 30 mL/min. Furthermore, branching fractures F1 and F2 are produced. At 40 mL/min, the hydraulic fracture HF extends along the direction of the maximum horizontal principal stress angle of 6.8°, generating additional branching fractures F1, F2, and F3. Finally, at 60 mL/min, supercritical CO2 fracturing forms the hydraulic fracture HF, which extends along the direction of the maximum horizontal principal stress while creating branch fractures F1, F2, F3, and F4. The angle between the branch crack and the main crack is 13–24°. Overall, as the displacement increases, the number of branch fractures grows, and the angle between the branch fractures and the main fracture increases. Additionally, the length of the branch fractures will increase, leading to the formation of a complex fracture network where the main and branch fractures communicate with each other. This is due to the increase of CO2 discharge, the continuous accumulation of energy in the fracture, the increase of pore pressure gradient, and the accumulation of Strain energy stress, which lead to the increase of normal and tangential flow velocity in the fracture, the obvious increase of pressure, the shortening of fracture initiation time, the shortening of fracture propagation time, and the increasing of propagation speed.

4.3 The Influence of Horizontal Stress Difference on the Fracture Propagation Law in Tight Sandstone

The horizontal stress difference is an important geological factor that determines the propagation form of hydraulic fractures. To study the effect of different stress differences on fracture pressure and fracture propagation law, the fracture curves of supercritical CO2 under different stress differences at a displacement of 30 mL/min are shown in Fig. 8.

Fig. 8.
figure 8

Fracturing curves of supercritical CO2 under different stress differences.

Full size image

From the figure, it illustrates that at a displacement rate of 30 mL/min, the fracture pressure and fracture time differ when the horizontal stress variation changes. With a horizontal stress difference of 6 MPa, rock fracture pressure is 11.8 MPa, and the fracture time is 305 s. However, the horizontal stress difference is 8 MPa, rock fracture pressure is 13.6 MPa, and the fracture time is 300 s. This represents a coefficient increase of 15.3% in fracture pressure and a decrease of 1.6% in fracture time. At a higher horizontal stress difference of 10 MPa, the fracture pressure and fracture time are 15.5 MPa and 282 s, respectively, signifying a 31.4% increase in fracture pressure and a 7.5% decrease in fracture time. The hydraulic fracture is influenced by internal pressure, in-situ stress, and other factors leading to increased pressure required for fracture initiation when the horizontal stress difference is large. As the fracture extends a certain distance, the hydraulic fracture experiences reduced resistance when subjected to horizontal traction, primarily towards the direction of maximum horizontal principal stress, resulting in decreased expansion pressure.

In terms of fracture morphology, the fracture morphology of supercritical CO2 under different stress differences at a displacement of 30 mL/min is shown in Fig. 9.

Fig. 9.
figure 9

Fracture morphology diagram formed by supercritical CO2 fracturing under different stress differences

Full size image

Based on the figure presented, when the horizontal stress difference is 6MPa and the displacement is 30 mL/min, a hydraulic fracture with a direction of 26° to the maximum horizontal principal stress is generated after compression. During its expansion, the main fracture demonstrates a clear deviation, leading to the formation of two branch fractures with an included angle of 13°. Furthermore, branch fractures F3 and F4 are generated on both wings of the main fracture, inclined at angles of 86° and 34° with respect to the main fracture, respectively. Upon increasing the horizontal stress difference to 8 MPa, hydraulic fractures HF propagate in the direction of the maximum horizontal principal stress, and branch fractures F1 and F2 appear at angles of 23° and 9°, respectively, with respect to the main fracture. Similarly, at a stress difference of 10 MPa, a hydraulic fracture HF extends along the direction of the maximum horizontal principal stress and triggers the formation of a branch fracture F1, inclined at an angle of 13° with the main fracture. Therefore, the smaller the horizontal stress difference, the more conducive it is to the multi-point initiation and turning of hydraulic fractures, thereby increasing the complexity of reservoir fractures and improving the effectiveness of oil and gas flow.

5 Conclusions

Supercritical CO2 has the potential for reservoir transformation, carbon utilization, and carbon sequestration. Through the large-scale true triaxial fracturing experimental device, the impact of different fluids, displacement, and stress differences on the expansion of tight sandstone fractures is clarified. It is believed that under the guidance of the “dual carbon” goal, supercritical CO2 fracturing technology has broad application prospects.

The fracture pressure of supercritical CO2 fracturing is 13.9 MPa, which is 40.6% lower than that of slick water fracturing. The rock fracture time has increased by 274.1% compared to slick water fracturing. The fracture tortuosity formed by supercritical CO2 fracturing is 2.23 times that of slick water fracturing. The spatial morphology of fractures formed by supercritical CO2 is more complex, and the surface roughness is large.

In supercritical CO2 fracturing, larger displacement and lower in-situ stress difference are more conducive to multi-branch initiation and fracture turning of hydraulic fractures, thereby increasing the complexity of the reservoir fractures and improving the flow capacity.