The ground stress, water pressure and temperature are higher, and disturbance is stronger (Zhang et al. 2022; Pretorius and Mathews 2019) with the continuous transfer of engineering excavation activities to the deep. Moreover, due to geological structure, the engineering rock mass develops macro-fractures and micro-fractures of different scales (Zhang 2010; Yang and Shan 2017; Li and Liu 2013). Under continuous excavation disturbance, the rock mass is soaked and softened to varying degrees caused by groundwater migration. Its mechanical structure, deformation and failure characteristics will undergo unconventional evolution when subjected to external loads (Zhang and Wang 1994; Huang and Xu 2004; Li and Zhang 2021). At the same time, the shear stress field at the excavation surface is formed by the formation of overlying rock hanging surface and the downward transmission of load. This condition will lead to the instability and failure of rock mass and the development of fracture network in mining, tunnel excavation, underground chamber excavation and other engineering construction. It is easy to induce geological disasters (Zhang and Li 2020).

Many valuable achievements in fractured rock mass and water–rock coupling were formed by relevant experts and scholars at home and abroad. It is found that prefabricated fractures with different occurrences (different inclination angles, lengths, spacings, numbers, fillings, etc.) have a common effect on the macro–micro physical and mechanical properties of rock mass under different stress loading paths of uniaxial (Gupta 2013; Wang and Liu 2018; Chen and Jing 2017; Gutierrez and Melentijevic 2020; Su and Qiu 2020; Tian and Xu 2020; Krzaczek and Nitka 2021) or triaxial (Rukhaiyar and Samadhiya 2016; You and Dai 2021; Huang and Gu 2021; Li and Wang 2022) tests. The properties of fractured rock mass will show different regular changes in the shear stress field compared with uniaxial or triaxial tests. In the existing research, a shear mechanical model of prefabricated fractured granite was established and proposed a calibration process for PFC3D numerical simulation of prefabricated fractured granite and intact granite(Uxía and Leandro 2020), and someone established a settlement model by studying the variation of pore water pressure in clay by multi-directional cyclic shear test (Tran and Hiroshi 2017). At the same time, Some scholars have studied the energy evolution and mechanical properties of rock by conducting shear tests. They analyzed the energy evolution law of fractured stone (Hu and Xia 2022; Xu and Wu 2012; Liang and Tang 2021) and studied the shear mechanical properties and fractures propagation mode of marl with parallel fractures (Yin and Liu 2021; Chen and Jian 2021). Also, a new method for evaluating the strength parameters of rock mass is proposed by shear test, and the reliability of friction coefficient is analyzed and verified (Choo and Ong 2015). In addition, in order to reveal the water–rock coupling mechanism, Scholars have carried out a lot of research on laboratory tests (Tang and Yao 2019; Wang and Liu 2015; Zhao and Taheri 2020; Yao and Wang 2021; Wei and Apel Derek 2018; Chen and Jian 2021). They found that the micro-structural characteristics and creeping mechanism of rock (soil) under different freeze–thaw cycles and different water content had significant differences.

It can be seen that the current research on fractured rocks focuses on the stress field effects of different loading paths primarily. But most of the excavated rock mass is subjected to long-term immersion and erosion of groundwater. It leads to the weakening of the physical and mechanical properties of rock mass, and the unconventional failure will occur when the rock mass is subjected to external force. As a result, the hidden disaster-causing factors increase and the construction process is slowed down. However, most of the existing research results are about the influence of water on rock properties under conventional stress loading paths (uniaxial, triaxial, etc.). Few studies have been carried out on the superposition of water content and fracture occurrence as the main factors.

Therefore, In this paper, the direct shear tests of fractured rocks under different water contents were carried out, and the shear mechanical properties and fracture morphology evolution of rocks under the influence of water environment and fracture morphology were analyzed. The research conditions are more suitable for engineering practice in this paper.

2.1 Making rock samples

In this experiment, red sandstone with good homogeneity was selected, and the rock samples were taken from the same complete red sandstone rock block. Square rock samples with geometric dimensions of 50 mm × 50 mm × 50 mm were prepared by wire-cut technology and according to relevant standards. The surface flatness deviation range is from −0.2 mm to + 0.2 mm, and the maximum deviation of the right angle between adjacent surfaces is less than 0.3°. The rock samples were tested by JSR-DPR300 ultrasonic detector to reduce the influence of discreteness on the test results, and decrease the material discrete influence. A size of 1 mm × 25 mm × 50 mm and a dip angle of 0°, 20°, 40°, and 60° single penetrating fracture was processed on the rock sample (Gupta 2013; You and Dai 2021; Dai et al. 2020). Four samples were processed at each dip angle. The processed red sandstone sample is shown in Fig. 1.

The red sandstone samples with prefabricated fractures of different angles

2.2 Testing equipment

This test was carried out on the MTS 816.01 rock mechanics shear test instrument. The equipment's structure is shown in Fig. 2, where Fig. 2a is the physical map and Fig. 2b is the software drawing. The structure diagram of each subsystem is drawn as shown in Fig. 3 to display the specific structure of the equipment visually.

MTS 816.01 rock mechanics shear test instrument

MTS 816.01 flow chart of rock mechanics shear tester

The MTS 816.01 rock mechanics shear tester includes a pressure supply system, a loading system and a control system. The rock samples with different geometric scale can be tested by embedding different geometric size molds in shear box, and the side of shear box is 200 mm × 200 mm × 340 mm. The system of uniaxial and shear loading included in the control system can be designed on demand.

The test machine's maximum shear load and displacement are + 250 kN and 100 mm. Four normal displacement sensors and two shear displacement sensors, not less than 0.5% of the actual reading accuracy, were built in the experimental cabin. The testing machine can carry out uniaxial loading, direct shear, cyclic loading and unloading, and other tests.

2.3 Test scheme and steps

The test is divided into 6 groups of 20 rock samples to carry out uniaxial compression, immersion, and direct shear tests respectively. The number of rock samples in each group and the constant parameters in the test are shown in Table 1.

The test steps are as follows:

1. (1)

   Apply normal stress to the first group of rock samples to destruction at a rate of 0.06 kN/s;

2. (2)

   Apply normal stress to 20%, 40% and 60% of the peak value at a rate of 0.06 kN/s, and then sheared it to destroy at a rate of 0.02 kN/s.

3. (3)

   First, dry the 3–6 groups of rock samples. Then calculate the water content at different time intervals by soaking them in water. The step can be ended after the dynamic stability of the water content with the change of soaking time;

4. (4)

   Apply 30% of the normal stress maximum value to the third group of rock samples at a rate of 0.06 kN/s, and then sheared it to failure at a rate of 0.02 kN/s;

5. (5)

   The 4–6 groups of rock samples were taken to repeat the test steps (4) until all rock samples were tested.

3.1 Analysis of water absorption law of rock sample

According to the calculation results of the immersion test, the water content-immersion time evolution curve is drawn (Fig. 4).

Evolution law of water content and immersion duration

The significant regular change of water content with the soaking time can be divided into similar to the linear growth, growth reduction and dynamic stability stages. Two key points are taken from the stage where water content increases rapidly similar to the linear growth. The change in water content slowed down after the growth reduction stage, and the demarcation point of each stage was selected as the key point. The water content of 0% (dry), 1.103%, 1.521% and 1.606% (saturation) were chosen as the key points based on the analysis of the changing trend of water content and combined with the previous research ideas(Fang et al. 2018; Yao and Wang 2021). These are the constant parameters of water content for subsequent experiments (as shown in Table 1).

3.2 Mechanical properties of rock samples

The uniaxial compression test was carried out on the complete rock sample to show the stress–strain curve (Fig. 5).

Stress–strain curve of complete rock sample

The rock sample has a relatively long duration in the micro-pore pressure sealing stage. A slight downward trend after the end of this stage indicates that the rock sample has slight damage. Therefore, the rock samples are similar to the conventional material stress–strain curve trend and segmentation. The rock sample's uniaxial compressive strength and elastic modulus are 60.2 MPa and 1.2 GPa. The normal stress value of the fractured sample during the test can be determined by synthesizing other test requirements.

Direct shear tests were carried out on the intactsamples in the second group by applying different normal stresses. It can study the basic shear mechanical properties of the intact rock samples. The test results show that the shear displacement-shear stress curve is drawn as shown in Fig. 6.

Shear displacement-shear stress evolution trend diagram

The maximum shear stress of the rock sample increases successively as the constant normal stress increases. But the corresponding shear displacement decreases when the shear stress reaches the maximum. At the same time, the maximum shear stress is proportional to the shear displacement corresponding to the maximum point. Combined with the Mohr-Coulomb strength criterion, the curve of the shear stress of the rock sample with the normal stress is fitted and calculated ( the fitting curve is shown in Fig. 7). The cohesion and the internal friction angle of the rock sample used in this test are about 10.65 MPa and 45.71°.

Normal stress -maximum shear stress fitting curve

4.1 Effect of fracture Angle on shear properties of water-bearing rock samples

Figure 8 is drawn to describe the evolution law of maximum shear stress with fracture dip angle, and analyzes the effect of fracture dip angle on the mechanical properties of water-bearing rock samples.

Angle of primary fractures -maximum shear stress evolution law and zoning characterization

The evolution law of the shear resistance of rock samples affected by primary fractures has significant zoning characteristics. The shear stress reaches the maximum when the dip angle of the primary fractures is 20° and the minimum is 0° in the test results. Taking 20° as the demarcation point, to the left of the demarcation point, the dip angle of the fracture is correlated with the maximum shear stress positively; to the right of the demarcation point, dip angle is correlated with the maximum shear stress negatively.

The possible reasons for the evolution of the curve in Fig. 8 are analyzed. Due to the different fractures dip angles, the relative position of the concentrated stress direction at the fracture's tip is also different from the dominant fracture surface direction. It also leads to the difference in the concentrated stress direction at the fracture's tip. When the dip angle of the prefabricated fractures is 20°, the direction of the concentrated stress at the fracture's tip is the most inconsistent with the direction of the theoretical dominant fracture surface, and the condition for the stress component in the direction of the dominant fracture surface to reach the threshold is the highest. Therefore, the overall strength of it is maximum relatively after the test.

4.2 Effect of water content on shear properties of rock samples with fissure

Water content is one of the key factors inducing the weakening of shear strength of rock mass. According to the test results, the water content-peak shear stress evolution curve is drawn (Fig. 9).

Evolution law of water content-maximum shear stress

In the results of this experiment, taking 1.521% as the demarcation point, to the left of the demarcation point, the maximum shear stress de-creases with the increase of water content of red sand-stone samples. To the right of the demarcation point, the maximum shear stress of the rock sample rises. At the same time, it can be seen that the shear performance of unsaturated water-bearing rock samples is weaker than that of intact and saturated rock samples. It indicates that the strength of fractured rock samples is weakened after water intervention.

The possible reasons for the weakening damage degree of saturated water content to the shear performance of rock samples are analyzed. The rock samples are composed of microscopic particles with different geometric scales, irregular arrangements and different bonding degrees, and there are certain pores between particles (Jiang and Wen 2012; Jin and Wang 2022). When the external load compresses the pore structure of the rock sample, the microscopic particles are squeezed and connected to form a support structure, which constitutes the overall performance of the rock sample. When the water flows in the gap of the rock sample, the erosion of the water flow will cause the microscopic particles to be worn, the geometric size to decrease, and the cohesive force to be damaged. Finally, the particles with a smaller geometric scale are washed out after the constraint is removed, which weakens the microscopic particles' supporting effect and reduces the rock sample's strength.

When the rock sample is saturated, the internal voids are filled with water medium. Under conventional mechanical compression, the volume change of the water body is minimal, which can be regarded as incompressible. During the test, because the rock sample is separated from the water-rich immersion environment, the internal water body will lose and near the surface loss mostly. The inward loss gradually decreases, forming a moisture content distribution law that decreases from the centre of the rock sample to the surface gradually (Fig. 10). The water body in the central area is squeezed when the rock sample is sheared under the external load to form a trend of diffusion to the weak water area. But limited to the internal pore structure scale of the rock sample, the seepage rate of the squeezed water body is low. The rock sample is destroyed before the squeezed water body flows out of the central area. The formed of water body in the slow flow area supports the microscopic particles and improves the overall strength of the rock sample.

Macroscopic and mesoscopic saturated water state diagram of rock sample

After the test, the rock sample's macroscopic fracture development and expansion characteristics were reconstructed. The original rock sample and the reconstruction diagram are shown in Tables 2 and 3.

5.1 Influence of fracture Angle on fracture morphology of water-bearing rock sample

According to Tables 2 and 3, the development-expansion law of macroscopic fracture surface rock sample was analyzed: at 0° and 20°, The fracture surface initiates at the tip of the primary fracture and terminates on the surface of the rock sample close to the fracture point, and there is no longitudinal fracture surface. At 40° and 60°, the transverse fracture surface initiates at the tip of the primary fractures. It terminates at the discontinuous stress point on the longitudinal surface far from the fracture initiation point. The two sets of transverse fracture surfaces are parallel to each other approximately. The two ends of longitudinal fracture surface of the two groups are the tip of the primary fracture and the transverse fracture surface that cracks at the tip of another primary fracture, and the two groups are also parallel approximately. Analysis of macro-fracture development-expansion law: macro-fractures generally initiate at the tip of primary fractures and extend to both sides of primary fractures and fracture surfaces. The fracture density increases and then decreases with the increase of the dip angle of the primary fractures specifically. The relative maximum fracture density is at a 20° dip angle, and the fractures run through the primary fractures and the edge of the rock sample only at 20°.

As shown in Fig. 11, the changing trend of rock sample debris area increases first and then decreases with the increase of 20° as the boundary. The specific performance is as follows: in the range of 0° to 20°, it increases significantly with the increase of primary fractures dip angle; in the range of 40° to 60°, it decreases approximately linear with the growth of fracture dip angle. It can be seen from the concentric circle of the distribution area of the rock sample debris area in Table 3 that the development position of the rock sample debris area is the most extensive when the dip angle of the primary fractures is 20°. The distribution area gradually shrinks and concentrates to the corner and the areas on both sides of the fracture surface decrease progressively or disappear.

The macroscopic area of the rock sample debris evolves with the dip angle of the primary fracture

Other types of failure zones also have corresponding development rules: the shedding area is mainly distributed at the tip of the fracture surface, of the dry state and large dip angle primary fractures rock samples. The flake spalling area only exists on the rock sample with 0% water content. It extends along the edge line of the rock sample, and the variation law with the dip angle of the primary fractures is the same as that of the debris area. The area evolution law of the original rock sample area approximately decreases first and then increases.

5.2 Effect of water content on fracture morphology of rock samples containing fractures

It can be seen that the transverse fracture surface of rock samples does not change significantly with the increase of water content when the dip angle of primary fracture is 0° and 20°, combined with Tables 2 and 3. At 40° and 60°, the longitudinal fracture surface is developed in other groups of rock samples except for the drying state. The quantity and opening of its fracture surface increase successively. For the macroscopic fracture development characteristics, taking the 20° dip angle rock sample as an example, the fracture density drops sharply during the water content change from 0% to 1.103%. The fracture density is relatively uniform before water saturation.

It can be seen from Fig. 12 that the macroscopic fracture morphology of rock samples change significantly after water absorption: compared with the surface morphology of dry rock samples, the area of debris area increases for the first time after the rock sample absorbs water; the area of the debris area begins to decrease to varying degrees when the water content increases to 1.103%, and the curve rises after the water content is 1.606%. From the perspective of distribution characteristics, the rock sample debris area is concentrated on the corner and both sides of the fracture surface when the water content is 0%; after the rock sample is water-bearing, the area of the debris area at the corner gradually decreases and retracts to the edge of the rock sample. The distribution area of the debris area on both sides of the fracture surface gradually decreases or completely disappears, and it is again spread to both sides of the fracture surface after the rock sample is saturated.

The evolution trend of a macro area of rock sample debris area with water content

5.3 Fracture development and expansion morphology and type

According to the characteristics of the macroscopic fracture development and expansion on the surface of the rock sample, the new fractures on the surface of rock samples with the same dip angle were extracted and superimposed (Fig. 13) and then summarized and classified qualitatively.

Prefabricated macro fracture superposition diagram of fractured rock samples with different dip angles

When the dip angle of the prefabricated fracture is 0°, the principal fracture direction has a certain angle with the horizontal direction. It initiates at the tip of the prefabricated fracture and terminates on the surface of the rock sample close to the fracture point. The direction of the fracture surface is approximately parallel to the horizontal direction when the dip angle is 20°, its initiation and termination points are the same as at 0° inclination. AC, BD and AE are the fracture direction when the dip angle is 40°. The fracture initiation points of AC and BD do not change. The termination point becomes the discontinuous stress point on the longitudinal surface of the rock sample far away from the fracture initiation point. AE takes the prefabricated fracture tip A as the fracture initiation point and terminates on the fracture surface BD. AE and BD are approximately vertical. When the prefabricated fractures dip angle is 60°, the fracture direction increases to 4 because the BF direction becomes one of the fracture surfaces.

Based on the law of the development and expansion path of the fracture surface of rock samples, the development direction and spatial position generalization diagram (Fig. 14) are drawn, and the classification and qualitative expression are carried out according to the spatial position parameters of each fracture surface.

Fracture propagation path generalization diagram

It can be seen from Fig. 14 that the crack initiation points of different fracture surfaces are different, and the fracture tendency and number are also different. Therefore, the fracture surfaces can be divided into shortest path single through type, longest path single through type and cross path through type.

The angle of the shortest path single through type is ≥ 0° (the clockwise rotation is positive, and the primary fracture tip is the origin). The fracture initiation point is the primary fracture tip. It terminates at the discontinuous stress point on the longitudinal surface of the rock sample near the fracture's initiation point, such as the fracture surfaces \(\text{AD}\) and \(\text{BC}\) in Fig. 14a. The angle of the longest path single through type is less than 0°. Its initiation point is the tip of the primary fracture and its termination point is on the surface of the rock sample far away from to the fracture point, such as the fracture surfaces \(\text{AC}\) and \(\text{BD}\) in Fig. 14b. The cross path through types of fracture surfaces include positive tendency and negative tendency, and the positive tendency fracture surface belongs to the longest path single through type. The angle of the negative tendency fracture surface is less than the longest path single through type. It initiates at the tip of the primary fracture and terminates at the fracture surfaces of the longest path single through type initiated at the tip of another primary fracture. They are approximately vertical, such as the fracture surfaces \(\text{AE}\) and \(\text{FB}^{\prime}\) in Fig. 14c.

In addition, according to the distribution characteristics of the development and expansion path of the fracture surface, the rock sample is divided into single through type and cross through type. Figures 14a and b are single through types, and Fig. 14c is cross through type.

1. (1)

   The shear strength of rock samples is the largest when the dip angle of primary fractures is 20°, and the influence of fractures on it is the lowest. With the increase in water content, the weakening degree of shear resistance of rock samples decreases gradually. The weakening degree of shear resistance decreases from increasing to decreasing after 1.521% of water content, and the overall mechanical structure increases after saturated water content.

2. (2)

   The rock sample only develops transverse rupture surface when the dip angle of the primary fracture is 0° and 20°; there are approximately vertical transverse rupture surface and longitudinal rupture surface when the dip angle is 40° and 60°. With the dip angle of 20° as the boundary, the macroscopic fracture density increases first and then decreases. The change law of the area of the debris area is the same as that of the macroscopic fracture density, and the spatial distribution range is the most extensive when the dip angle is 20°.

3. (3)

   An increasing trend is shown in the number and opening degree of fracture surface with the increased water content. The density of macro fractures decreases sharply with the water absorption of rock samples and then increases to a certain extent with the saturation of rock samples. There is a cyclic fluctuation between the area of the debris area and the water content, with double rises and double falls. The distribution area of the debris area shrinks to the edge with the water absorption of the rock sample, and the areas on both sides of the fracture surface gradually decrease or disappear. The distribution range returns to both sides of the fracture surface again after the rock sample is saturated.

4. (4)

   A model is established to classify and quantitatively characterize the spatial position and geometric relationship of the fracture surface of rock samples under the influence of different primary fracture dip angles. The fracture surface is divided into shortest path single through type, longest path single through type and cross path through type according to the initiation point and tendency. The rock samples after failure are divided into single through type and cross through type. There is only one horizontal fracture surface of the shortest path single through type. But the type of fracture surfaces of longest path single through and cross path through include both horizontal and vertical, and they are perpendicular to each other approximately.