1 Introduction

With the increasing demand for unconventional energy sources, the development of shale oil and gas holds a highly significant position (Wang et al. 2016; Guo et al. 2008). During the oil and gas extraction process, drilling fluid is indispensable, and the selection and usage of drilling fluid have a direct impact on wellbore stability, reservoir integrity, and stability (Huang et al. 1995; Wang et al. 2007; Zhang et al. 2002). By gaining a deeper understanding of the interactions between drilling fluids and shale, we can better predict the fracture development in shale reservoirs, the path of crack propagation, and changes in rock mechanical properties, providing a scientific basis for oil and gas exploration and development (Zhao et al. 2007; Deng et al. 2006; Okere et al. 2023). Simultaneously understanding the complex fracture failure characteristics of shale is crucial for predicting reservoir fracture expansion, optimizing hydraulic fracturing plan design, and implementing a rational oil and gas production strategy (Chen et al. 2017; Long et al. 2011; Li et al. 2012; He et al. 2021; Okere and Sheng 2023; Zhang et al. 2021).

Extensive research has been conducted on the impact of drilling fluids on the mechanical properties of rocks. Studies have found that with the immersion of drilling fluid, the rock's strength, elastic modulus, cohesion, and hardness all experience a certain degree of reduction (**ang et al. 2018; Fan et al. 2020; Wang et al. 2020a, b; Lu et al. 2012; Wang et al. 2018; ** et al. 2017). However, there are diverse opinions on the fundamental reasons for its strength reduction. Among these, scholars have considered mechanical damage caused by the hydration of clay minerals (Luo et al. 2022; Kang et al. 2016) and hydraulic fracturing (Kang et al. 2016; Wang et al. 2018; Lu et al. 2012) as the main factors contributing to rock strength reduction. Some researchers also believe that chemical damage induced by the dissolution of minerals by alkaline components in drilling fluids (**ang et al. 2018; Kang et al. 2016) is the primary cause of rock strength reduction. Some scholars have also conducted research on the frictional forces on rock structural surfaces after immersion in drilling fluid (Wang et al. 2018) and changes in rock surface morphology (Liu et al. 2014). Additionally, investigations have been made into the changes in the fracture patterns of water-immersed rocks (** et al. 2017). While previous researchers have extensively studied the reasons for damage to rocks after immersion in drilling fluid and the resulting alterations, there has been limited research on the effects of drilling fluid on the fracture process zone of rocks and its impact on fracture patterns. Building upon prior work, this study investigates the influence of drilling fluid immersion on rock fracture toughness, the fracture process zone, and changes in damage patterns.

In this study, for the determination of fracture toughness, one of the four standard test methods recommended by the International Society for Rock Mechanics (ISRM), the Semi-Circular Bend (SCB) three-point bending test, was chosen to measure the composite fracture toughness of rocks (Suo et al. 2023, 2021, 2020). Due to its simplicity in specimen preparation and accuracy in testing, the SCB method has been extensively researched by scholars both domestically and internationally. Studies have revealed that the initial load, peak load, critical crack length, and initial fracture toughness significantly depend on the notch width and pre-crack angle (Angle between prefabricated crack and longitudinal direction of SCB sample) (Xu et al. 2023). When the loading mode shifts from Type I to Type II, crack propagation speed and crack acceleration increase (Alneasan et al. 2022). And we also need to investigate the variations in the stress intensity factors KI and KII. KI represents the Mode I stress intensity factor, quantifying the strength of tensile stress distribution near the tip of a pre-crack. On the other hand, KII represents the Mode II stress intensity factor, quantifying the strength of shear stress distribution near the tip of a pre-existing crack. Some researchers have observed that with an increase in the pre-crack angle β, KI decreases while KII increases, with KI consistently greater than KII, indicating the dominance of Type I fracture (Huang 2022). As the pre-crack angle β increases, the proportion of Type I fractures gradually decreases, and stress intensity factor KI decreases, while the proportion of Type II fractures gradually increases, along with an increase in stress intensity factor KII and T stress (Zuo et al. 2019; Feng et al. 2016). Some researchers have also investigated the deterioration characteristics of I + II composite fracture toughness of sandstone under the influence of chemical solutions and dry–wet cycles. They found that with an increasing number of dry–wet cycles, the fracture toughness of sandstone gradually deteriorates, with a decreasing rate of deterioration over time. The impact of chemical solutions on fracture toughness follows the order: acidic > alkaline > neutral. The overall deterioration of Type II fracture toughness is less than that of Type I fracture toughness (Liu et al. 2023).

In the process of three-point bending testing, real-time monitoring of energy release during the fracture process is achieved through Acoustic Emission (AE), while the Digital Image Correlation (DIC) technique is employed for tracking the microcrack region at the crack tip, known as the fracture process zone (FPZ). Some researchers employed DIC to monitor displacement jumps and identify stress-free crack tips, considering the distance between the crack tip and the stress-free crack tip as the fully developed FPZ length (Pan et al. 2023). Others used DIC to monitor maximum tensile strain and defined the FPZ tip as the location where maximum tensile strain is reached (Wu et al. 2011). Some scholars regarded areas with high acoustic emission event density and event counts as corresponding to the fracture process zone, using acoustic emission events as criteria for evaluating FPZ length and width (Boukais et al. 2023). Additionally, certain researchers utilized CT scanning technology, combined with post-processing image techniques, to compare fracture patterns within the FPZ and obtain the shape and size of the FPZ (Ghamgosar et al. 2021). There were also instances where the FPZ length was directly calculated using tensile modulus and Linear Elastic Fracture Mechanics (LEFM) theory (Miarka et al. 2023). Scholars have drawn inspiration from various definitions of FPZ length, so in this study, a new method for determining FPZ length using DIC monitoring was defined based on previous research. Regarding fracture patterns, many researchers have investigated the fracture modes of different types of specimens, such as Cracked Straight-Through Brazilian Disc (CSTBD) (Ma et al. 2023), Semi-Circular Bend (SCB) (Suo et al. 2023), and Three-Point Bending Beams with a Straight Offset Notch (TPB-SO) (Gong et al. 2020). The research methods for distinguishing between tensile and shear failure in these different specimen types were based on the combination of rock acoustic emission and stress–strain analysis, using the Amplitude-Frequency Ratio Analysis (AF-RA) method. This provides a reliable basis for rock mass stability assessment and engineering design (Li et al. 2023; Teng et al. 2023; Dong et al. 2023).

In summary, existing research on the immersion damage of rocks by drilling fluids has primarily focused on the influence of drilling fluids on the mechanical properties of rocks. However, there has been relatively limited research on the fracture process zone, fracture toughness, and failure modes. Building upon this gap, the present study conducted three-point bending tests on shale specimens from the Songliao Basin G Area. Combining Digital Image Correlation (DIC) and Acoustic Emission (AE) monitoring techniques, the loading process was monitored. DIC analysis was employed to assess changes in the Fracture Process Zone (FPZ) length, strain fields, and displacement fields in SCB specimens. Additionally, acoustic emission technology was used to analyze changes in AE energy and b-values during the fracture process of the specimens. The study established schematic diagrams depicting the proportion of different fracture modes under different crack angles after immersion in various drilling fluids. This study quantifies the damage characteristics of complex fractures in shale after drilling fluid treatment, which is of paramount importance for simulating reservoir fracture propagation and optimizing hydraulic fracturing schemes in real field scenarios. The research provides a theoretical foundation and technical support for construction design in layered rock formations.

2 Experiment methods

2.1 Sample preparation and mineral content text

The specimens were obtained from the shale formation in Block G of the Songliao Basin. As shown in Fig. 1. The specimen preparation process is as follows: Initially, cylindrical rock cores with a diameter of 76 mm were drilled from a large shale block. Subsequently, a wire cutting machine was utilized to slice the cylindrical rock cores into circular discs with a thickness B of 30 mm. These discs were further divided into semicircular discs with a radius R of 36 mm. Simultaneously, a wire cutter was used to create an incision with a length of 11 mm at the center of the semicircular disk, where the crack is oriented at a specific angle β with the vertical direction. This created a pre-crack. The sample obtained was meticulously smoothed and polished from all aspects to fall within the acceptable error range, especially the surface intended for measuring crack propagation velocity. The pre-crack angles β were set at 15°, 30°, and 45°, with reference to their deviation from the vertical direction. For each pre-crack angle, 12 SCB specimens were prepared, resulting in a total of 36 experimental specimens. These specimens were subjected to three separate experiments, including one without drilling fluid immersion and three with immersion in different drilling fluids, resulting in a total of four test scenarios for each pre-crack angle. These samples were processed into SCB samples and the bedding orientation was crack-divide (Lv 2018). All the geometric parameters were in accordance with the recommendations of the International Society of Rock Mechanics (ISRM) (Wei et al. 2016). The specimens were subjected to a three-point bending loading experiment to test I + II mixed-mode fracture toughness, the support spacing at the bottom two points is 60 mm for 2S.

Fig. 1
figure 1

Sample preparation and sample geometry

Full size image

A certain amount of shale was taken from the same batch of shale samples and ground into fine powder. X-ray diffraction analysis was conducted to determine the mineral composition of the shale in the Songliao Basin. The analysis revealed that the shale in the Songliao Basin is primarily composed of brittle minerals such as quartz, accounting for an average of 49.2% of the composition. Feldspar minerals, including potassium feldspar and sodium feldspar, make up 11.6% of the composition. Clay minerals constitute an average of 28.4% of the composition, with an average of 77.8% being illite and 22.2% being eumong interbed. As shown in Fig. 2.

Fig. 2
figure 2

The average mineral composition of the shale samples

Full size image

2.2 Sample treatment

In the Songliao Basin's G region, three types of drilling fluids are commonly used: water-based drilling fluids(4% Bentonite + 0.25% Caustic Soda + 0.02%KOH + 1–2.5% Resin-Based Filtration Control Agent + 1.5–2.5% Ammonium Salt-Based Filtration Control Agent + 1.5–2.5% Bridging and Anti-Collapse Agent + 0.3–0.5% Polyamine + 0.2–0.4% Encapsulating Agent), alcohol-based drilling fluids(30% Propylene Glycol + 5%KCL + 0.05%NaOH + 3% Sodium Bentonite + 0.2% Anionic Cellulose + 0.2% Heat-Resistant Modified Starch + 2% Heat-Resistant Blocking Agent + 0.5% Na2SO3 + Barite) and oil-based drilling fluids(2–3% Primary Emulsifier + 1–1.5% Secondary Emulsifier + 3% Organic Clay + 2%CaO2 + 2.5–3% Filtration Control Agent + 2–3% Micro-sized Plugging Agent + 1–2% Nano-sized Plugging Agent + Calcium Chloride (20% Aqueous Solution), Oil–Water Ratio 80:20) to treat rock samples in this study. The first step involved placing rock samples with different pre-crack angles into a beaker and slowly pouring the prepared drilling fluid over them to ensure complete coverage. Each type of drilling fluid was used to soak the rock samples with three different pre-crack angles, and the beaker was sealed with tape to ensure a reliable seal. The second step is to take the soaked rock sample out every 12 h and wipe the outer drilling fluid with a clean paper towel. Measure the rock quality with a precision balance record the initial mass m0, and then put the rock sample into the dryer to dry for 2 h. The final mass m of the rock is measured again. This process was repeated until the mass change was small enough or close to zero. The third step involved using the drilling fluid to soak saturated samples for three-point bending loading experiment. The study utilized Digital Image Correlation (DIC) and Acoustic Emission (AE) techniques for monitoring. The formal experimental scheme was divided into four groups, each group of samples being tested three times.

2.3 Experimental instrument

As shown in Fig. 3. The test machine GCTS RTX-3000 was utilized to conduct a three-point bending loading experiment at room temperature. The experiment employed a displacement control loading approach with a loading speed of 0.05 mm/min. The actual load–displacement curve was obtained during the experiment. The digital image acquisition equipment is a high-speed camera Revealer M220, which has a resolution of 1920*1080 and an acquisition speed of 3003 frames/s. It records the whole process from the beginning of loading to the fracture of the specimen, and can effectively capture the moment of cracking. Additionally, it accurately observed the displacement field and strain field of the prefabricated crack growth. The selected acoustic emission instrument is the DS5-8A acoustic emission sensor. In the experiment, a threshold of 40 dB is set to exclude noise, ensuring that only sufficiently strong emission signals are recorded. This is advantageous for reducing the burden of data storage and processing, as well as enhancing the quality and accuracy of acoustic emission events.

Fig. 3
figure 3

Experimental equipment

Full size image

3 Results and discussion

3.1 Fracture toughness

The calculation formula for fracture toughness of SCB sample is represented by Formula (1) and (2). The SCB sample contains a straight notch and is subjected to three-point bending loading during the fracture toughness test. The maximum load P during the SCB test is recorded. KI and KII are determined by the following formula (Suo et al. 2022a, b).

$${K}_{I}=\frac{P}{2RB}\sqrt{\pi a}{Y}_{I}$$
(1)
$${K}_{II}=\frac{P}{2RB}\sqrt{\pi a}{Y}_{II}$$
(2)

In the given formula, R and B represent the radius and thickness of SCB sample. P denotes the radial loading load, and the dimensionless stress intensity factors YI and YII are dimensionless parameters that rely on the specimen’s geometry and loading conditions. The purpose of dimensionless stress intensity factors is to enable comparisons of crack problems with different sizes and geometric shapes, independent of specific material parameters. These parameters are typically calculated by finite element analysis. KI and KII were Mode I and Mode II fracture toughness (MPa m1/2), respectively. Table 1 shows the geometric parameters of the sample when S/R = 0.8 and a/R = 0.3 in the Mixed-mode fracture (Ayatollahi et al. 2007).

Table 1 Geometric parameters of different pre-crack angles in I + II mixed-mode fracture mode
Full size table

In fracture mechanics, in order to further reveal the fracture mechanical characteristics of rocks under the I + II mixed-mode, the effective fracture toughness value Keff is defined for the I + II fracture condition.

$${\mathrm{K}}_{\mathrm{eff}}=\sqrt{{K}_{\mathrm{I}}^{2}+{K}_{II}^{2}}$$
(3)

Moreover, the mixing degree is an important parameter of the dominant component in mixed-mode fracture, as it determines the dominant component and is often used to analyze the impact of various fracture toughness components on the fracture behavior of samples. In order to qualitatively determine the effect of chemical coupling on I + II mixed-mode fracture, the mixing parameter Me was defined:

$${M}^{e}=\frac{2}{\pi }\mathrm{arctan}\left(\frac{{K}_{I}}{{K}_{II}}\right)$$
(4)

In Formula (4), the value of Me varies from 0 to 1. Under the condition of pure Mode I fracture, the value of Me is 1, and under the condition of pure Mode II fracture, the value of Me is 0. Me represents the loading from pure Mode II to pure Mode I.

Figure 4 illustrates the fracture toughness changes of SCB shale samples with different pre-crack angles that were soaked in different drilling fluids. The x-axis represents the angle of pre-crack, the y-axis represents the types of drilling fluid treatment, and the z-axis represents the corresponding fracture toughness values. It can be seen from the figure that as the pre-crack angle increases, the fracture toughness of Mode I and the effective fracture toughness continuously decrease, while the fracture toughness of Mode II increases. This is due to the increase in stress concentration at the fracture tip increases when the pre-crack angle increases. It alters the distribution of the stress field and subsequently affects the fracture toughness. Furthermore, the study found that water-based drilling fluids had the greatest weakening effect on the fracture toughness of shale samples and the trending is water-based drilling fluids > Alcohol-base drilling fluids > oil-based drilling fluids. The main reason for this is that drilling fluids penetrate shale fractures and bedding planes under capillary action (Wang et al. 2020a, b), causing the expansion of clay minerals in shale, with illite being a major constituent of clay minerals. Moreover, a recent study (Yang et al. 2022) has demonstrated that illite exhibits macroscopic characteristics of high stress and low strain during hydration, and higher stress levels can lead to the initiation, extension, and propagation of fractures within the shale. The intrusion of drilling fluid leads to the connectivity of pore throat structure between skeleton grains in rock materials, including the connectivity between small pores and medium pores. The continuous increase of various pores makes the separation of rock skeleton grains produce micro-cracks, which constantly blend to form new cracks and destroy the fabric integrity of pores between rock skeleton grains.

Fig. 4
figure 4

Influence of drilling fluid immersion on fracture toughness at different crack angles

Full size image

3.2 DIC analysis of fracture process

3.2.1 Length of fracture process zone

The digital image correlation method was utilized to match the images before deformation. The two sets of digital images, one taken before deformation and the other after deformation, are used to extract deformation information in the region of interest through a correlation-based calculation. In the image taken before deformation, the region of interest is subdivided into grids, treating each subregion as a rigid body. Subsequently, for each subregion, a predefined correlation function is employed, along with a specific search method, to perform a correlation calculation in the image taken after deformation. This calculation aims to find the area in the post-deformation image that exhibits the highest correlation coefficient with the respective subregion, indicating the final position of the subregion after deformation. This process allows for the determination of displacement for each subregion, which, when applied to all subregions, yields a comprehensive dataset of deformation information for the entire field. The deformation characteristics of the specimen were analyzed using a pre-crack angle β = 45° as an example. As depicted in Fig. 5, there is an obvious dividing line between tension and pressure in the horizontal displacement field. The upper part of the sample is subject to pressure while the lower part is subject to tension. The vertical direction of the pre-crack tip shows irregular symmetry in the horizontal displacement field. To further illustrate this issue, a small observation window was placed at the prefabricated crack tip. The horizontal displacement figure of the observation window is shown in Fig. 6. The contour lines of the horizontal displacement field converge to form a line above the pre-crack, and the length taken from the intersection point that coincides with the highest point of the vertical displacement field to the pre-crack tip represents the length of the fracture process area.

Fig. 5
figure 5

Schematic diagram of fracture length in horizontal and vertical displacement fields

Full size image
Fig. 6
figure 6

Horizontal displacement difference of sampling line in displacement field

Full size image

Five lines with a length of 4.78 mm were selected inside and outside the intersection point, with a spacing of 1.88 mm between them. Sampling line 4 was the line outside the intersection point, sampling line 3 passed through the intersection point, and sampling line 1 and 2 were located within the intersection point. Sampling line 0 was located close to the prefabricated crack tip. The displacement distance difference of the sampling line was then determined by calculating the change in horizontal displacement of the sampling line under the loading of the specimen, as shown in Fig. 6.

As can be seen from Fig. 6, it is evident that the distances between sampling lines 0 and 1 at the intersection point are relatively far apart. Consequently, when there is a significant displacement change along the sampling lines, lines 0 and 1 experience substantial variations in displacement. However, as the sampling lines approach the intersection point, the magnitude of displacement changes progressively diminishes, as illustrated by sampling line 2. When the sampling lines pass through the intersection point, their displacement alterations become exceedingly minimal, as demonstrated by sampling line 3. Sampling line 4 is oriented vertically upward from the intersection point, and its displacement remains largely unaffected when other sampling lines experience noticeable changes. At the peak load, sampling lines in the vertical downward region of sampling line 3 exhibit relatively substantial displacement variations. Concurrently, as stress levels remain at their peak without diminishing, it indicates that at this point, the fracture can still transmit stress. The discontinuity in displacement and stress transmission characteristics serves as two critical indicators for identifying the fracture process zone. Consequently, it can be inferred that the region below sampling line 3 in the vertical downward direction represents the fracture process zone of the shale sample, with sampling line 3 marking the tip of the fracture process zone.

To summarize, the tip of the fracture process area can be determined by finding the intersection of the convergence line of the horizontal displacement contour line and the highest coincidence point in the vertical direction of the maximum vertical displacement contour line at the peak load. The length of the fracture process area is then defined as the distance from the end point of the pre-crack to the intersection point. The same method can be used to analyze the length of fracture process zone at the maximum load of shale samples soaked with different pre-crack angles and drilling fluids. Different samples may have different lengths of fracture process zones. Five sampling lines were selected to ensure the same degree of precision of analysis for different samples. These include 1 sampling line close to the prefabricated crack tip, 1 sampling line past the intersection point, 2 sampling lines between the intersection point and the crack tip, and 1 sampling line above the intersection point. A unified analysis standards and accuracy of analysis were adopted to ensure consistency in the results.

As shown in Fig. 7. The FPZ length corresponding to peak load decreases with an increase in the pre-crack angle. Additionally, the FPZ length increases to varying degrees with immersion in different drilling fluids, with the water-based drilling fluid having the most significant effect, followed by the alcohol-based drilling fluid, and then the oil-based drilling fluid. Therefore, it can be inferred that the length of the I + II mixed-mode fracture process zone is related to the sample pre-crack angle and drilling fluid immersion. As the angle of pre-cracks in SCB specimens increases, several factors contribute to a reduction in the fracture process zone: Firstly, it causes stress to concentrate more at the crack tip, promoting faster crack propagation and potentially reducing the zone of plastic deformation during fracture, thus leading to a decrease in the fracture process zone. Secondly, cracks do not propagate strictly along the direction of the pre-crack; they may change direction and rapidly extend to the other side within a shorter distance, resulting in a shorter fracture process zone. The strength properties of shale also influence the length of the fracture process zone. With the immersion of drilling fluid, several phenomena occur: Firstly, the hydration of clay minerals induces expansion stress, and when the stress intensity factor at the tip of microcracks exceeds the critical stress intensity factor, microcracks in shale develop, resulting in a decrease in shale strength. Among the three types of drilling fluids, water-based drilling fluid primarily consists of water, and the extent of hydration reactions between the infiltrating free water within the shale and the clay minerals in the shale is the most pronounced (Liu et al. 2018). Consequently, the application of water-based drilling fluid leads to the most significant reduction in shale mechanical strength. Secondly, both oil-based and water-based drilling fluids are characterized as alkaline systems, with pH values typically ranging from 8 to 10 (Lu et al. 2012; Huang et al. 1995). The drilling fluid creates an alkaline environment, and under such conditions, certain shale minerals undergo chemical reactions, which erode the minerals at the crack tip and induce structural damage, thereby lowering the critical stress intensity factor at the crack tip and resulting in a decrease in shale fracture toughness. Therefore, the fracture process zone elongates.

Fig. 7
figure 7

illustrates the fracture process zone length of shale samples with varying pre-crack angles under peak load after immersion with different drilling fluids

Full size image

3.2.2 I + II mixed-mode fracture strain field and displacement field

As shown in Fig. 8, taking the load–displacement curve of the SCB specimen with a pre-crack angle of 45° without soaking can be divided into four stages: initial loading and compaction stage, elastic stage, plastic stage, and approaching fracture stage. The labeled points in the figure represent the characteristic points of each stage. The descriptions of the four stage points are as follows:

Fig. 8
figure 8

SCB load displacement curve of unsoaked crack with a pre-crack angle of 45°

Full size image

The OA phase corresponds to the compaction stage, where the initial loading induces slight stress on the specimen's surface. During this phase, particles within the rock may undergo reconfiguration, and small microcracks may form. The duration of this stage is relatively brief.

The AC phase represents the elastic stage, characterized by the elastic deformation of the specimen as loading continues. The relationship between load and strain is linear in this stage, which extends over a considerable duration. Within the AC phase, it can be further divided into the AB and BC segments. In the AB segment, external energy input accumulates within the rock, with only a small portion of this energy being released. Deformation energy increases at a slow rate and at a low level, and this stage persists for an extended period. The BC segment is marked by a gradual acceleration in upward trends. As loading continues, noticeable crack propagation occurs within the horizontal strain field. The energy stored within the rock gradually dissipates, and this stage also endures for a relatively lengthy period.

The point C signifies the yield point, marking the transition from the elastic phase to the plastic phase. The CD phase corresponds to the plastic stage, where, with the continued increase in load, the relationship between load and displacement is no longer linear. This phase continues until point D, which represents the point of failure. At this juncture, the specimen begins to experience macroscopic fracture, leading to a sharp decline in the load–displacement curve, indicating imminent complete failure.

The contour maps in Table 2 depict the horizontal strain field at various pre-crack angles for each stage point. The table shows that the horizontal strain field exhibits a relatively uniform change during the initial loading stage. As the peak load is approached, the horizontal strain concentration area at the crack tip reaches its maximum. When approaching the peak load, the horizontal strain concentration area at the crack tip reaches its maximum, followed by sample fracture and load unloading. Upon comparing SCB samples with different pre-crack angles, it was observed that the horizontal strain increases continuously with the increase of the pre-crack angle. In other words, the larger the pre-crack angle, the greater the horizontal strain generated before fracture of the rock sample. Additionally, the strain width at the crack tip also increases with the increase of the pre-crack angle.

Table 2 Horizontal strain contour maps with different pre-crack angles
Full size table

Figure 9 displays the horizontal strain contour maps of the SCB sample at the peak moment with a pre-crack angle of 45° under the influence of three drilling fluids. The figures reveal that there is an increase in strain magnitude in the fracture process zone after immersion in drilling fluids, with the length variation being the most prominent.

Fig. 9
figure 9

Horizontal strain contour maps of the SCB samples soaked by different drilling fluids when the pre-crack angle is 45°

Full size image

The horizontal displacement contour maps for each stage under varying pre-crack angles are presented in Table 3. From the table, it can be observed that in the initial loading stage the horizontal displacement distribution is relatively uniform. At point B, a noticeable difference in horizontal displacement can be observed on both sides of the crack. At point C, the discrepancy is even more pronounced and the tendency of crack extension before the pre-crack can be slightly discerned. At point D, there is a clear partition in horizontal strain on both sides of the crack propagation path, indicating significant displacement under the applied load. The visualization of the crack propagation path is evident, and macroscopic crack formation occurs subsequently, continuously extending until the rock sample experiences instability and failure.

Table 3 Horizontal displacement contour maps with different pre-crack angles
Full size table

Furthermore, the results indicate that as the angle of the pre-crack increases, the tendency of crack extension becomes more pronounced in the horizontal displacement field. This is due to the increased stress concentration effect resulting from an increased pre-crack angle, which leads to higher stress intensity factors at the crack tip and consequently larger displacement changes.

The horizontal displacement contour maps of the SCB samples at the peak moment, with a pre-crack angle of 45°, is presented in Fig. 10 under the influence of three drilling fluids. It can be observed that there are notable differences in the level of organization of the contour lines in the horizontal displacement field of the SCB sample after immersion in drilling fluids. Specifically, the SCB sample immersed in water-based drilling fluid exhibits a relatively more organized horizontal displacement field. This consistency can be attributed to the reaction between the drilling fluid and the rock sample, which leads to a decrease in the mechanical properties of the rock, resulting in a softening loading process where displacement is more likely to occur.

Fig. 10
figure 10

Horizontal displacement contour maps of the SCB sample by different drilling fluids when the pre-crack angle is 45°

Full size image

3.3 The results and analysis of acoustic emission

3.3.1 Acoustic emission energy and b-value

Acoustic emission energy refers to the energy released per unit of time during the rock damage process. It provides an indication of the intensity of the signal energy and serves as an important parameter for assessing the severity of rock damage. Acoustic emission is a phenomenon in which strain energy is released in the form of elastic waves during rock deformation and failure, similar to seismic events. The b-value is a parameter used in seismic research that relates to the magnitude and frequency of seismic events. Therefore, the analysis of b-value in acoustic emission can be used to study the characteristics of rock deformation and failure. In natural seismic events, the G-R formula is used to represent the relationship between seismic events magnitude and frequency. It was prominently established by Gutenberg and Richter in 1954 (Gutenberg et al. 1954).

$$lgN=a-b*M$$
(5)

The equation includes M represents the seismic events magnitude, N represents the cumulative number of seismic events with magnitudes greater than or equal to M, and a and b as positive constants representing the seismic activity, primarily determined by the maximum seismic events magnitude in the sequence. The b-value represents the slope of the line and is typically a number close to 1, reflecting the proportional relationship between the number of larger seismic events. In this study, the seismic events magnitude in the G-R relationship is represented by the amplitude as follows:

$$M=lgA$$
(6)

In the equation, A represents the maximum amplitude of the acoustic emission events. In this study, calculations are performed using the least squares method to investigate the impact of different pre-crack angles and immersion in different drilling fluids on the b-value.

The load–displacement curves, acoustic emission energy, and b-values of shale SCB samples under different pre-crack angles in the I + II mixed-mode are presented in Fig. 11. The results indicate that the peak strength of the shale increases with the increase of the pre-crack angle, which suggests an increase in compressive strength. During the initial loading stage, the load–displacement curves of the shale samples with three pre-crack angles exhibit a concave-down pattern. This is because the samples experience compressive forces, causing microcracks within the samples to close. This resulted in almost no acoustic emission energy generated, indicating the presence of fewer microcracks inside the rock. In the elastic consolidation stage, there is a clear linear relationship between load and displacement. At the peak load, the acoustic emission energy remains relatively stable, but the b-value fluctuates within a large range. After reaching the peak load, the post-peak curve of the shale exhibits a sharp drop with increasing displacement. At this stage, energy is continuously released during the process of crack propagation and agglomeration, resulting in accelerated release of acoustic emission energy and higher energy levels. Meanwhile, the b-value shows a downward trend, and the SCB sample experiences brittle fracture, resulting in a rapid loss of bearing capacity. As the pre-crack angle increases from 15° to 45°, the time period in which acoustic emission signals are generated becomes larger, but the maximum acoustic emission energy becomes smaller. This is due to the increased bending and stress concentration at the edges of the sample, which increases the internal stress and strain. It promoting more crack propagation and fracture events in this stage. This leads to more acoustic emission events, but the energy of each event may decrease.

Fig. 11
figure 11

Influence of different pre-crack angles on AE energy and b value

Full size image

The shale samples were saturated with three types of drilling fluids and subjected to loading experiments under the I + II mixed-mode. The load–displacement curves of the shale samples were obtained, and the results for the sample with a pre-crack angle of 45° are shown in Fig. 12. The figure shows that the strength of shale sample is affected by the immersion of drilling fluids. The peak load of the shale samples after immersion in different drilling fluids decreases to varying degrees, with the largest reduction observed in water-based drilling fluid, followed by alcohol-based drilling fluid, while oil-based drilling fluid has the least effect on the strength of the shale samples. Furthermore, the displacement at the peak load of shale samples increases as their strength decreases, indicating a varying degree of plastic enhancement and decreased brittleness in the shale samples immersed in different drilling fluids. During the loading process, the initiation and propagation of microcracks in the specimen are accompanied by the release of elastic waves. The initial damage stage corresponds to the nonlinear stage of the specimen, where the damage represented by AE energy is negligible.

Fig. 12
figure 12

Influence of different drilling fluid immersion on AE energy and b value when the pre-crack angle is 45°

Full size image

The stages of damage progression correspond to approximately linear phases in the specimens, indicating that there are virtually no new cracks appearing. During this stage, cracks begin to initiate and propagate within the specimen, acoustic emission (AE) activity gradually intensifies, and the b-value shows both increasing and decreasing trends, suggesting relatively stable variations. This indicates a consistent occurrence of both large and small AE events, signifying a progressive and stable growth of internal cracks within the rock. As observed in Fig. 12, the magnitude of variation in b-values follows the order of water-based drilling fluid > alcohol-based drilling fluid > oil-based drilling fluid. The internal fractures of rocks immersed in water-based drilling fluids are more susceptible to propagation and expansion. The destructive stage corresponds to the point where the specimen undergoes failure. AE energy suddenly spikes, and the b-value experiences a significant decrease, indicating an increased proportion of large-scale events. Microcracks within the material primarily contribute to large-scale damage, and the evolution of crack development becomes more dramatic. Notably, after immersion in water-based drilling fluid, the decrease in b-values is the most pronounced, followed by the specimen's failure. Consequently, it can be inferred that the presence of pores and cracks alters the characteristics of AE energy release during the loading process of the specimen. For shale specimens immersed in drilling fluids, the intrusion of drilling fluid into shale significantly impacts the evolution of AE energy release characteristics. Load displacement, AE energy, and b-value all correspondingly demonstrate the weakening effect of drilling fluids on shale specimens, with the degree of weakening following the order: water-based drilling fluid > alcohol-based drilling fluid > oil-based drilling fluid.

3.3.2 Tensile shear failure changes

The AF-RA method is a useful tool for analyzing the fracture mechanism of rocks (Lv et al. 2022), particularly in understanding the formation of I + II mixed-mode fracture cracks. The method involves analyzing the acoustic emission parameters of shale failure, specifically the ratio of rise time to maximum amplitude (RA) and the average frequency (AF). By defining the parameter k = AF/RA, it is possible to distinguish between tensile and shear failure mechanisms. When the ratio of AF to RA is greater than a certain value of k, it is considered to be tensile failure, while when it is less than a certain value of k, it is considered to be shear failure. (reference experience, the dividing line for fracture mechanism is typically taken as k = 50). In the Fig. 13b, the failure mechanisms are categorized into tensile and shear modes by the diagonal line. The selection of the k value is determined by the properties and structure of the rock. The average frequency (AF) is defined as the ring count divided by the duration, while the RA value is the ratio of rise time to maximum amplitude. Figure 13 illustrates the fracture modes of the unsoaked SCB specimen with a pre-crack angle of 45° using the AF-RA method.

Fig. 13
figure 13

Schematic diagram of AF-RA method dividing shear and tensile fracture modes. a Acoustic emission waveform parameters schematic diagram b AF-RA schematic diagram of unsoaked SCB specimen at a pre-crack angle of 45°

Full size image

Table 4 displays the distribution of fracture modes in SCB specimens analyzed using the AF-RA method after immersion in different drilling fluids at varying pre-crack angles. From the table, it can be observed that as the pre-crack angle of the SCB specimen increases, the proportion of shear-type fractures gradually increases, while the proportion of tensile-type fractures relatively decreases. After immersion in drilling fluids, the SCB specimens exhibit an increase in the proportion of shear-type fractures and a decrease in the proportion of tensile-type fractures regardless of the pre-crack angle. Among them, the SCB specimens immersed in water-based drilling fluid show the largest increase in the proportion of shear-type fractures, followed by those immersed in alcohol-based drilling fluid. In contrast, the increase in the proportion of shear-type fractures is relatively small after immersion in oil-based drilling fluid. Additionally, for the unsoaked SCB specimens, the proportion of shear fractures increases by 12% as the pre-crack angle varies from 15° to 45°. Comparing the unsoaked SCB specimens with those soaked in oil-based drilling fluid, alcohol-based drilling fluid, and water-based drilling fluid, the proportion of shear fractures increases by 5.9%, 10.1%, and 17.2%, respectively.

Table 4 The proportion of fracture modes in SCB specimens after immersion in different drilling fluids at various crack angles
Full size table

The angle of pre-cracks in SCB specimens significantly influences the fracture propagation mode. A smaller pre-crack angle leads to increased stress concentration near the crack, reinforcing the tensile stress effect and resulting in a higher occurrence of tensile fractures. Conversely, a larger pre-crack angle causes more shear stress concentration near the crack, making shear-type fractures more likely. As the angle of pre-cracks in SCB specimens increases, the proportion of tensile fractures (Mode I fractures) gradually decreases, while the proportion of shear fractures (Mode II fractures) steadily increases. Therefore, with an increasing pre-crack angle, tensile fractures decrease, and shear fractures become more prevalent.

When the degree of damage within the specimen is substantial, several factors contribute to an increased likelihood of shear failure. Firstly, microcracks within the shale matrix tend to develop further, resulting in the involvement of a larger area during specimen failure and an increased probability of shear failure (Suo et al. 2023). Secondly, cracks usually propagate along paths offering less resistance, often found near damaged regions or areas around the crack tip that are more prone to extension (Lv et al. 2022). This contributes to a higher frequency of shear-type fractures. Thirdly, numerous micro-damaged and deformed zones within the shale matrix can lead to localized shear stress concentration (Huang et al. 2023), serving as initiation points for shear failure. Therefore, with greater damage induced by drilling fluids, the likelihood of shear failure also increases.

4 Conclusion

The study investigated the impact of different pre-crack angles and immersion in different drilling fluids on the fracture behavior of shale SCB specimens. Three-point bending tests were conducted to measure the changes in Mode I, Mode II, and effective fracture toughness. Detailed information for the full displacement measurements using DCI and the characteristics of AE specimens were recorded and analyzed during the experiments. Based on the results obtained from the experiments, the following conclusions can be drawn:

  1. 1.

    As the pre-crack angle of the SCB specimens increased, the fracture toughness of Mode I and effective fracture toughness gradually decreases, while the fracture toughness of Mode II increases. The drilling fluid had a significant impact on the fracture behavior of the specimens. The fracture toughness of the SCB specimens is observed to decrease to varying degrees after immersion in drilling fluids, with the weakening effect on fracture toughness in the order of water-based drilling fluid > alcohol-based drilling fluid > oil-based drilling fluid.

  2. 2.

    The FPZ (Fracture Process Zone) of the SCB specimens was analyzed using DIC (Digital Image Correlation) and an improved method for FPZ length analysis was implemented. The study found that the FPZ length decreases as the pre-crack angle of the SCB specimens increases. However, after immersion in drilling fluids, the FPZ length of the SCB specimens tends to increase to some extent. In addition, the study also analyzed the horizontal strain field and displacement field. The results showed that the horizontal strain increases with the increase in the pre-crack angle. The region where crack propagation occurs becomes more pronounced in the horizontal displacement field. After immersion in drilling fluids, the strain magnitudes in the FPZ show varying degrees of increase and the horizontal displacement field appears more uniform.

  3. 3.

    By analyzing the energy and b-value of AE (Acoustic Emission) and found that as the pre-crack angle of the SCB specimens increases, the relative magnitude of acoustic emission energy decreases, but the frequency of events increases. The magnitude of b-value reduction also increases. After immersion in drilling fluids, the acoustic emission events become more frequent and the b-value becomes more disordered with a greater reduction. Furthermore, the study revealed that the relationship among load–displacement, AE energy, and b-value can effectively characterize the fracture behavior of the rock. The weakening effect of drilling fluids on shale specimens indicates that the degree of weakening is in the order of water-based drilling fluid > alcohol-based drilling fluid > oil-based drilling fluid. Finally, the AF-RA method in AE was used to analyze the fracture modes of SCB specimens with different pre-crack angles and immersed in different drilling fluids, and relevant fracture mode diagrams were established. The research results provide theoretical basis and technical support for the construction design of rock mass engineering.