Shear Strength and Re-Failure Characteristics of Intact Red Sandstone and Grouting-Reinforced Body of Fractured Red Sandstone under Different Shear Angles

Abstract

To reveal the strengthening mechanism and re-failure mechanism of grouting on fractured rock, the variable-angle shear tests, real-time acoustic emission (AE) tests and VIC-3D (non-contact full field strain measurement system) tests were carried out on intact red sandstone (IRS) and grouting-reinforced body of fractured red sandstone (GRBFRS). The results show that the peak shear strength of IRS and GRBFRS decreases with the increase of shear angle. Grouting reinforcement mainly increases the cohesion of GRBFRS to improve its shear strength, but its reconstructability decreases with the increase of shear angle. In the whole process of shear deformation, the shear micro damage and shear fracture of GRBFRS are more than those of IRS. Although the shear re-failure before and after the peak stress stage is the most notable, its intensity or degree is reduced. When the shear angle is 45°, both IRS and GRBFRS undergo shear-splitting failure. However, when the shear angle is large, the IRS and GRBFRS only occurs shear failure. Moreover, the larger the shear angle, the more likely IRS and GRBFRS is to produce secondary shear cracks. The low strength of the rock–grout interface in GRBFRS is the main inducer of shear re-failure.

1. Introduction

The natural rock mass is easy to deform and destroy into a fractured rock mass under the action of complex geological structure evolution or late production disturbance (such as roadway or tunnel excavation and mineral resource exploitation), as shown in Figure 1. It loses structural integrity and reduces its bearing capacity, resulting in great instability possibilities of rock mass engineering [1,2]. In order to prevent the occurrence of catastrophic accidents of rock engineering, quantities of engineering practices have proven that grouting reinforcement can effectively improve the integrity and strength of fractured rock mass, and thus maintain its stability [3,4,5,6]. However, due to the inevitable existence of plenty of weak rock–grout interfaces (Figure 1c) in the fractured rock mass after grouting [7,8], it is easy to produce re-failure under repeated external disturbance, especially shear re-failure due to low shear strength, which will affect the safety and long-term stability of the grouting reinforcement project of the fractured rock mass. Therefore, studying the shear strength and shear re-failure characteristics of a grouting-reinforced body of fractured rock will help evaluate the grouting reinforcement effect on fractured rock mass engineering and predict its stability.

To this point, Evdokimov et al. [9] pointed out by shear test as early as 1970 that grouting reinforcement can improve the shear strength of a fractured rock mass, and the more broken the rock, the better the grouting reinforcement effect, which was confirmed and improved by Li et al. [10] and Zhang et al. [11]. Han et al. [12] concluded that grouting reinforcement can improve the shear strength and structural stiffness of fractured rock mass, and the number of saw teeth (approximate roughness of saw teeth) has a significant impact on its shear strength. This lays a foundation for studying the shear strength of the grouting-reinforced body of fractured rock with rock–grout interfaces. Salimian et al.’s [13] further study shows that grouting reinforcement has a positive effect on the shear strength of a grouting-reinforced body of fractured rock. The maximum cohesion increases with the decrease of the water–cement ratio, but the maximum internal friction angle decreases. Liu et al. [14,15,16] acquired test results from the shear test on the grouting-reinforced body of mudstone and sandstone, which indicated that the peak shear strength and residual shear strength of grouted mudstone and sandstone increased by about 68–117%, 17–21%, 108–170% and 54–72%, respectively. Wang et al. [17] obtained the relationship between the shear strength of the grouting-reinforced body and normal stress, and the strength of grouting material and rock matrix under pressure grouting. Stavropoulou et al. [18] acquired the adhesion, shear performance and failure mode of wet weak carbonate with rock–cement interfaces under cyclic loading. Lan et al. [19] studied the grouting reinforcement mechanism of heterogeneous fractured rock and soil mass under uniaxial compression, and pointed out that the rock/soil-slurry interface is a weak structural plane for a reinforced body, which is not conducive to improving the strength of a reinforced body. The above research results have greatly promoted the cognition of the shear strength effect of grouting reinforcement. However, since the research mainly focuses on the grouted body containing the artificial regular fracture interface of concrete pouring, the randomness and irregularity of the fractures formed by the disturbance stress are ignored. Moreover, most of the above studies do not discuss the shear re-failure characteristics of a grouted body, which leads to the inability to effectively evaluate and predict the reinforcement effect and stability of the grouted rock mass.

For the purpose of studying the shear failure characteristics of a grouting-reinforced body, Zheng et al. [20] found through testing that the grouted body of a rock-like substance first suffered bond failure, and then sliding failure. Wang et al. [21] obtained that the shear strength of a grouting-reinforced body increased with the increase of the number of saw teeth, and the sawtooth interface of the grouted body under different normal stress will appear with three basic failure modes: adhesion failure, climbing failure and shear failure. Lu et al.’s further study [22] shows that there are four failure modes of the grouting-reinforced body: climbing first and then shearing, saw tooth directly gnawing, cement slurry separation and gnawing, and slurry shearing. He also pointed out that the shear strength of the grouting-reinforced body increases first and then decreases until the equilibrium with the increase of the grouting filling ratio. Lu and Zhang et al. [23] obtained the influence laws of grouting on the shear strength and deformation of soft rock joints, and pointed out that the typical shear failure occurred at the rock–grout interface. Li et al. [24] used direct shear tests and nanoindentation tests to study the macro- and micro-mechanical properties of the grout–rock interface and its influence on the shear mechanical behavior of grout-infilled joint rock. These studies will promote the understanding of the shear re-failure mechanism of the fractured rock mass after grouting. However, most of them do not discuss the influence of damage to intact rock during shear fracture on the shear re-failure of a grouting-reinforced body, and ignore the connection and difference of shear failure between intact rock and grouting-reinforced body. In other words, they may lack the research on the whole process of rock form integrity, to shear failure, to grouting reinforcement to shear re-failure.

Therefore, in view of the shortcomings and flaws in the existing research, this paper further focuses on the study of the shear strength and failure characteristics of a grouting-reinforced body with random rough surface formed by shear failure. The variable-angle shear tests, AE (Acoustic emission) detecting test and VIC-3D (non-contact full field strain measurement system based on Digital Image Correlation) capturing test on IRS and GRBFRS under the same conditions were performed. The strengthening law of shear strength and parameters, and the evolution characteristics of shear failure and re-failure of red sandstone before and after grouting reinforcement were compared and analyzed. Through these studies, it is expected to reveal the shear strength strengthening mechanism and shear re-failure mechanism of a grouting-reinforced body of fractured rock, improve the theory of grouting reinforcement and lay the foundation for the study of stability of grouting engineering of a fractured rock mass.

2. Test Materials and Test Schemes

2.1. Test System

The test system mainly includes an electronic universal material testing machine DDL600 (Changchun Institute of Mechanical Science Co., LTD, Changchun, China), a shear instrument with variable angle, an AE instrument, a VIC-3D (Correlated Solutions Inc., Irmo, SC, USA) and two data acquisition computers. The overall arrangement and installation connection are shown in Figure 2. The DDL600 is used to apply axial pressure, and the loading rate is 0.5 mm/min. The variable-angle shear instrument fixes the red sandstone sample and conducts the shear test under different shear angles. The contact surface between the rock sample and the shear fixture is a normal constraint, while the other surfaces have free boundaries, as shown in Figure 2a. The AE instrument (Figure 2b) has the functions of full-waveform acquisition and processing and real-time acoustic emission positioning. These are used to collect the acoustic emission characteristic parameters such as amplitude, ringing count, duration and energy in the process of the variable-angle shear test, to identify the generation, development and evolution of micro-cracks in red sandstone samples [25]. To obtain the AE signal during loading, an AE probe was placed at the center of the back of the sample, as shown in Figure 2b. The VIC-3D system mainly includes a CCD (charge-coupled device) camera and a data acquisition system, as shown in Figure 2b,c, which is used for real-time capture of the macroscopic surface deformation and failure evolution images of samples, and calculating the full-field strain in the failure process. The specific testing methods of VIC-3D can be seen in references [26,27,28]. The layout of the VIC-3D system is shown in Figure 2b, and its test schematic is shown in Figure 2c.

2.2. Sample Preparation

In order to minimize the dispersion and test errors caused by the differences of the samples themselves in the test process, the standard square sample was prepared from a large piece of homogeneous and complete red sandstone, and its size was 50 mm × 50 mm × 50 mm. The variable-angle shear tests were carried out on standard red sandstone samples (i.e., IRS) to obtain the shear fracture samples with random rough surfaces. Marked according to the shear angle, the shear fracture samples were grouting-reinforced using grout with a water–cement (ordinary 325# portland cement) ratio of 0.5:1 in self-made grouting equipment. The thickness of the in-filled grout after the upper and lower fractured rock blocks are compressed under constant pressure is usually about 1.0–1.5 mm. After consolidation into a whole, the samples were demolded and maintained for 30 days to obtain the GRBFRS samples for the test. The longer maintenance time is to ensure that the fractured rock blocks and the grout are fully bonded. The specific preparation process of the GRBFRS sample is shown in Figure 3.

2.3. Test Scheme

In order to compare and analyze the shear strength and failure characteristics of red sandstone before and after grouting reinforcement, a variable-angle shear test scheme is designed, and its specific test steps are as follows: (1) Red sandstone blocks were taken from the engineering site and processed into standard cube samples of 50 mm × 50 mm × 50 mm by cutting and grinding, as shown in Figure 3a. (2) The shear tests at the shear angles of 45°, 60° and 75° were carried out on the standard cube samples (i.e., IRS), to obtain the shear stress-shear strain curve, the peak and residual shear strength of IRS and the fractured red sandstone samples, as shown in Figure 3b. At the same time, the AE monitoring test and the VIC-3D capturing test were carried out during shear loading. (3) Ordinary cement grout was used to bond the fractured red sandstone into a grouted body sample, i.e., GRBFRS, as shown in Figure 3c. (4) The shear test, the AE monitoring test and the VIC-3D capturing test were carried out on the GRBFRS samples prepared by the above method under the same shear angle. In order to eliminate the dispersion in the test as much as possible, three groups of tests were carried out under each shear angle, and then the average value is taken as the approximate test result. The specific test scheme and process are shown in Figure 4.

3. Results and Discussion

3.1. Shear Deformation and Strength Behavior of IRS and GRBFRS

3.1.1. Shear Deformation Characteristics of IRS and GRBFRS

For the variable-angle shear test, the normal stress σ and shear stress τ (Figure 2a) on the shear plane of the rock sample can be calculated as follows:

$$\sigma =\frac{P}{A}\cos \alpha$$

(1)

$$\tau =\frac{P}{A}\sin \alpha$$

(2)

The shear strain of rock sample is:

$${\epsilon }_{\tau }=\frac{S}{L}\sin \alpha$$

(3)

where σ is normal stress on shear plane of rock sample (MPa), τ is shear stress on shear plane of rock sample (MPa), P is the axial load applied to the rock sample by the test machine DDL600 (Figure 2a) (N), A is the shear area of rock sample (mm²), α is shear angle (Figure 2a) (°), ε_τ is shear strain of rock sample, S is the axial displacement of the test machine DDL600 (mm) and L is the length of the rock sample (mm).

During the variable-angle shear test, after obtaining the axial load P and the axial displacement S of the test machine DDL600, the shear stress–strain (τ–ε_τ) curve of rock samples in the whole loading process can be acquired by using the Equations (2) and (3). Considering that the variation of the τ–ε_τ curve under the same shear angle is similar, a set of test data (sample numbers 4–2, 6–2 and 7–2, respectively) from the IRS and GRBFRS under each shear angle are selected to draw the shear stress–strain (τ–ε_τ) curves as shown in Figure 5. All the test results are listed in

Table 1. Shear strength and parameters of IRS and GRBFRS under different shear angles.

α /°	Sample Number	IRS										GRBFRS
		σp /MPa	τp /MPa	τpa /MPa	φp /°	cp /MPa	σr /MPa	τr /MPa	τra /MPa	φr /°	cr /MPa	σpg /MPa	τpg /MPa	τpag /MPa	φpg /°	cpg /MPa
45°	4-1	23.56	23.56	20.82	35.2	6.84	15.5	15.5	13.93	37.6	3.88	18.38	18.38	17.21	37.9	4.59
	4-2	20.39	20.39				12.8	12.8				17.76	17.76
	4-3	18.50	18.50				13.6	13.6				15.50	15.50
60°	6-1	6.48	11.22	14.21			3.50	6.10	9.79			4.56	7.90	11.72
	6-2	8.27	14.32				6.06	10.50				7.57	13.12
	6-3	10.00	17.10				6.93	12.01				8.17	14.16
75°	7-1	1.92	7.15	7.39			1.09	4.10	3.93			1.23	4.60	4.50
	7-2	2.43	9.06				1.15	4.30				1.37	5.10
	7-3	1.60	5.97				0.91	3.40				1.01	3.80

.

It can be seen from Figure 5 that during the shear loading process, the IRS and GRBFRS have roughly gone through the typical compaction stage (section OA in Figure 5), elastic deformation stage (section AB in Figure 5), plastic deformation stage (section BC in Figure 5) and post-failure stage (section CD in Figure 5). These are similar to the deformation characteristics obtained from the variable-angle shear test of coal samples in the reference [29]. However, due to the differences in the properties and integrity of red sandstone and coal samples, there are still certain differences in the specific details of each stage of the loading process.

With the increase of shear angle, the shear compaction stage and elastic stage of IRS and GRBFRS are shortened. The main reason is that the larger the shear angle, the lower the normal stress of the sample, resulting in the reduction of the friction between various components in the sample. Under the same shear angle, the compaction stage of GRBFRS is obviously longer than that of IRS. This is mainly because when the IRS is sheared and fractured, many micro pores and fissures with different sizes and distributions will be formed in the sample due to loading. Moreover, due to insufficient filling and cementation during cement grouting consolidation, the rough fracture surface formed by shear failure will leave more pores and cracks in the rock–grout interface. These significantly increase the number of initial pores and fissures in the GRBFRS, which then increase the time of compaction stage.

3.1.2. Shear Strength of IRS and GRBFRS

In addition to the differences in various stages of shear deformation and failure, the shear stress and shear strain of IRS and GRBFRS samples will also change with different shear angles and sample integrity.

Combined with Figure 5 and Table 1, it can be seen that both peak shear stress τ_p and residual shear stress τ_r of IRS (i.e., before grouting) decrease with the increase of shear angle, which is consistent with the direct shear test results in references [30,31]. Their shear strain also decreases with the increase of shear angle. After taking the average value of the peak and residual shear stress (i.e., shear strength) obtained in each group under the same shear angle as shown in Table 1, the peak shear strength τ_pa of IRS decreases from 20.82 MPa when the shear angle α is 45° to 7.39 MPa when the shear angle α is 75°, decreasing by about 64.51%. The residual shear strength τ_ra decreases from 13.97 MPa to 3.93 MPa, a decrease of about 71.87%. This may be because the normal stress σ on the shear plane of the red sandstone samples will decrease with the increase of shear angle, so that the friction on the shear plane decreases. At the same time, the shear stress τ on the shear plane will increase, leading to shear failure of the samples under smaller shear stress and shear strain. However, under the same shear angle, the peak shear strength τ_pag of the GRBFRS (i.e., after grouting) is less than the peak shear strength τ_pa of the IRS, but greater than the residual shear strength τ_ra of the IRS, as shown in Table 1. It indicates that the grouting reinforcement can fill the rough surface and cracks of the red sandstone after shear failure, and make the fractured rough surfaces bond with each other to form a whole, so as to improve the integrity of the structure and limit the deformation ability, and then improve the shear strength.

In order to evaluate the strengthening effect of grouting reinforcement on the shear strength, the peak shear strength τ_pag of GRBFRS is divided by the residual shear strength τ_ra of IRS, and the increase ratio of grouting reinforcement on the shear strength of fractured red sandstone can be obtained. According to the calculation results of the data obtained in Table 1, when the shear angle α is 45°, 60° and 75°, the shear strength ratios of GRBFRS are 123.19%, 119.71% and 115.38%, respectively. This indicates that as the shear angle increases, the improvement ability of grouting reinforcement on the shear strength of fractured red sandstone will decrease. This may be due to the reduction of the normal stress σ on the shear plane of the sample, which not only reduces the friction of the rough surface of the fractured rock, but also reduces the bonding force at the grouting cementation. Meanwhile, the shear stress on the shear plane also increases.

3.1.3. Shear Strength Parameters Based on the Mohr–Coulomb Criterion

For the purpose of further exploring the strengthening mechanism of grouting reinforcement on the shear strength of fractured red sandstone, the cohesion and internal friction angle, two important parameters characterizing the shear strength of rock are studied. Considering that the shear stress τ and normal stress σ on the shear plane of the red sandstone sample should meet the Mohr–Coulomb strength criterion during the shear failure process [29]:

$$\tau =\sigma \tan \phi +c$$

(4)

where φ is the internal friction angle of rock (°) and c is the cohesion of rock (MPa).

The peak shear stress τ_p, residual shear stress τ_r and normal stress σ_p of IRS, as well as the peak shear stress τ_pg and normal stress σ_pg of GRBFRS, obtained by the variable-angle shear tests are listed in Table 1. They are plotted as scatter plots, as shown in Figure 6. By linear fitting of shear stress and normal stress according to Equation (4), the Mohr–Coulomb strength criterion of peak shear strength and residual shear strength of IRS, as well as peak shear strength of GRBFRS can be expressed as follows, respectively:

$${\tau }_p=0.7056{\sigma }_p+6.84$$

(5)

$${\tau }_r=0.776{\sigma }_r+3.88$$

(6)

$${\tau }_{pg}=0.7805{\sigma }_{pg}+4.59$$

(7)

The specific fitting results are shown in Figure 6.

Compared the fitting results with Equation (4), the shear strength parameters (i.e., cohesion and internal friction angle) of IRS and GRBFRS can be obtained, which are shown in Table 1. It can be seen from Table 1 that the peak internal friction angle φ_p (35.2°) of IRS is smaller than the residual internal friction angle φ_r (37.6°). This is probably because the IRS will form a rough surface during the process of shear failure, accompanied by the formation of clastic particles, which will increase the friction of the shear plane and then increase the residual internal friction angle. However, the peak cohesion c_p (6.84 MPa) of IRS is much larger than the residual cohesion c_r (3.88 MPa) and the peak cohesion c_pg (4.59 MPa) of GRBFRS, so the peak shear strength of IRS is larger than the residual shear strength of IRS and the peak shear strength of GRBFRS. This shows that the shear strength of red sandstone is jointly determined by the internal friction angle and cohesion. Moreover, the grouting reinforcement cannot make the peak shear strength of GRBFRS exceed that of IRS, but can only make it higher than the residual shear strength of IRS. Therefore, the cohesion of the rock should be an important evaluation index to measure the effect of grouting reinforcement.

As can be seen from Table 1, the peak cohesion c_pg and the peak internal friction angle φ_pg of GRBFRS are 1.18 times and 1.01 times of the residual cohesion c_r and the residual internal friction angle φ_r of IRS, respectively. This maybe because the injected cement grout can not only fill defects such as pores and cracks formed in the shear failure, but also bond and cement the shear fracture surface. This greatly improves the structural integrity and limits the deformation ability of the grouting-reinforced body, and then significantly increases its peak cohesion c_pg. However, due to the properties of cement grout and the randomness of cement particle settlement, consolidation and distribution, the peak internal friction angle φ_pg of GRBFRS will be increased to a lesser extent.

The above shows that grouting reinforcement can greatly enhance the cohesion of the fractured red sandstone, but has little effect on the internal friction angle, which is consistent with the theoretical analysis conclusion of a large amount of experimental data in reference [32]. This further indicates that grouting reinforcement mainly improves the shear strength of fractured red sandstone by increasing the cohesion, which is consistent with the conclusion in reference [33].

3.1.4. Strengthening Evaluation and Prediction of Shear Strength Parameters

In order to evaluate and predict the strengthening ability (i.e., grouting reconstructability) of grouting reinforcement on the shear strength of fractured rock masses, Xu et al. [32,34,35] proposed to use cohesive growth rate ξ_coh and internal friction angle growth rate ξ_f to characterize it, respectively:

$${\xi }_{coh}=\frac{c_{ag}-c_{bg}}{c_{bg}}$$

(8)

$${\xi }_{\phi }=\frac{{\phi }_{ag}-{\phi }_{bg}}{{\phi }_{bg}}$$

(9)

where c_ag is peak cohesion of a grouting-reinforced body of fractured rock (MPa), c_bg is residual cohesion of fractured rock (MPa), φ_ag is peak internal friction angle of grouting-reinforced body of fractured rock (°) and φ_bg is residual internal friction angle of fractured rock (°). Moreover, Xu et al. [32,34,35] also deduced and established the empirical equation between the cohesive growth rate ξ_coh and the internal friction angle growth rate ξ_f and the compressive strength growth rate ξ_c of a grouting-reinforced body:

$${\xi }_{coh}=0.95{\xi }_c$$

(10)

$${\xi }_{\phi }=0.05{\xi }_c$$

(11)

$${\mathrm{\xi }}_c=\frac{\mathrm{\Delta }{S}_c}{S_{cr}}=\frac{S_{cg}-S_{cr}}{S_{cr}}=\frac{\frac{2c_{pg}\cos {\phi }_{pg}}{1-\sin {\phi }_{pg}}-\frac{2c_r\cos {\phi }_r}{1-\sin {\phi }_r}}{\frac{2c_r\cos {\phi }_r}{1-\sin {\phi }_r}}$$

(12)

where ∆S_c is the increment of uniaxial compressive strength of fractured rock before and after grouting (MPa), S_cg is peak uniaxial compressive strength of a grouting-reinforced body of fractured rock (MPa) and S_cr is residual uniaxial compressive strength of intact rock (MPa). Both of them can be calculated from the corresponding cohesion and internal friction angle [32,34,35].

According to the test results listed in Table 1, the test values of cohesion growth rate ξ_coh and internal friction angle growth rate ξ_f of GRBFRS can be obtained according to Equations (8) and (9), and the results are listed in

Table 2. Comparative analysis of growth rate of shear strength parameters of GRBFRS.

Growth Rate of Shear Strength Parameters	Test Value/%	Theoretical Value/%		Relative Error/%
		Wang et al. [36]	Present Study	Wang et al. [36]	Present Study
Growth rate of cohesion	18.30	19.86	18.38	7.85	0.44
Growth rate of internal friction angle	0.96	1.05	0.97	8.57	1.04

. With the help of the test results of reference [36], it can be seen that the compressive strength growth rate of GRBFRS is 0.209 [32]. Using Equation (12) and the results listed in Table 1, it can be obtained that the compressive strength growth rate of the fractured red sandstone in this paper after grouting is 0.194. According to the Equations (10) and (11), the theoretical values of the cohesive growth rate ξ_coh and the internal friction angle growth rate ξ_f of GRBFRS can be obtained, and the results are listed in Table 2.

Table 2 shows that the relative errors between the experimental and theoretical values of the cohesion growth rate and the internal friction angle growth rate of GRBFRS are 0.44% and 1.04%, respectively. Even if using the compressive strength growth rate obtained in reference [36], their relative errors are 7.85% and 8.75%, respectively, which are less than 10%. The result indicates that the shear strength parameters of GRBFRS obtained in this experiment are reliable. It is also proved that the theoretical equation for calculating the growth rate of shear strength parameters of a grouting-reinforced body of fractured rock proposed by Xu et al. [32,34,35] is correct and reasonable. This can be used to evaluate and predict the strengthening ability of grouting reinforcement on the shear strength of a grouting-reinforced body of fractured rock.

3.2. Shear Failure and Re-failure Characteristics of IRS and GRBFRS

3.2.1. Microscopic Characteristics of Shear Fracture Based on AE Signals

According to the study of the correlation between crack evolution and AE parameters of coal and rock during loading process, it is shown that the AE hits and AE energy can better reflect the deformation and crack evolution process of coal and rock [29,37]. Therefore, in order to study the micro-damage evolution characteristics of shear failure and re-failure of IRS and GRBFRS under different normal stress and shear stress, the AE is used to obtain the variation characteristics of AE hits and AE energy with time under different shear angles. The results are shown in Figure 7.

It can be seen from Figure 7 that the change characteristics of AE hits and AE energy in IRS and GRBFRS samples during the process of shear deformation and fracture are correlated and consistent, but the shear angle and sample integrity have a significant impact on them. Combined with the shear stress–strain (τ–ε_τ) curve shown in Figure 5, it can be divided into three typical stages: pre-peak stage I, peak stage II and post-peak stage III, as shown in Figure 7. In the pre-peak stage I (roughly corresponding to the compaction stage and elastic deformation stage OB in Figure 5), the AE hits and AE energy of IRS are low, and their increase is small during the process of continuous loading. This shows that under the action of compression and shear load, although some of the primary micropores and microcracks inside the sample are closed, there is still the expansion and evolution of primary micropores and microcracks, and the initiation and expansion of new micropores and microcracks. However, in general, the initiation, evolution and expansion of pores and cracks is not much, so only less vibration occurs and less energy is released. This result is quite different from the acoustic emission test results of coal samples obtained under different shear angles in reference [29]. The main possible reason for this is that the number and size of pores and cracks in IRS are far less than those in coal samples, and their integrity and uniformity are also significantly better than those in coal samples. However, the number of AE hits and AE energy of GRBFRS in the pre-peak stage I is significantly higher than that of IRS. It may be due to the large number of pores and cracks as well as the rock–grout interface in GRBFRS which makes the structural integrity and uniformity of GRBFRS worse than that of IRS.

When entering peak stage II (roughly corresponding to the plastic failure stage BC in Figure 5), the pores and cracks in IRS and GRBFRS continued to propagate, evolve and penetrate to form larger pores and cracks, which emitted strong AE signals. Therefore, the number of AE hits and AE energy increased significantly, and rapidly increased to the maximum in a very short time before and after reaching the peak shear stress, indicating that the IRS and GRBFRS samples formed macro cracks near the shear plane and were broken into two parts. By contrast, the duration of the AE hits in GRBFRS is longer than that of IRS, but the maximum AE energy is lower than that of IRS, as shown in Figure 7a,d. This shows that the shear re-failure of GRBFRS before and after peak stress is more likely to occur and the number of fractures is more, but the energy released due to the fracture is lower.

In post-peak stage III (which can roughly correspond to the post-failure stage CD in Figure 5), the AE hits and AE energy of IRS decrease rapidly and tend to be stable. This shows that when macroscopic shear failure occurs in IRS, the rough shear fracture surface formed will still occur, causing friction or even shear failure of the micro-convex during the loading process, so the AE hits and AE energy can still be emitted in the later stage of the failure. However, because its bearing capacity has been lost and the only residual bearing capacity is generated by friction, the AE signals are weak. However, the AE hits and AE energy of GRBFRS are significantly higher than that of IRS in the later stage of failure, indicating that it has more shear failure.

In addition, it can be seen from Figure 7 that with the increase of shear angle, the AE hits and AE energy of IRS and GRBFRS show a decreasing trend. This shows that the smaller the ratio of shear stress to normal stress of the sample is, the higher the shear fracture degree of the sample.

3.2.2. Macroscopic Characteristics of Shear Fracture Based on VIC-3D

In order to study the evolution characteristics of shear fracture under different shear angles, the VIC-3D capturing test is used to obtain the shear strain of samples during loading, and the results are shown in Figure 8.

It can be seen from Figure 8 that when the shear angle is 45°, the shear strain of IRS in the direction near the shear angle increases significantly, and converges to a band to form the primary shear crack, as shown in Figure 8(a₂). With the loading progress, the main shear crack develops and expands continuously, and the crack width increases gradually. At the same time, a new secondary crack is initiated at the lower end of the primary shear crack, where the propagation direction of secondary crack is along the vertical direction (i.e., the loading direction), and it has a certain angle with the propagation direction of the main shear crack, which is a split crack, as shown in Figure 8(a₃). That is to say, in addition to shear failure, the sample also has a longitudinal splitting failure. Furthermore, outside the boundary of the primary shear crack and secondary crack, there are many areas where the shear strain increases locally on the surface of the specimen, which are the microcrack initiation and development areas of the sample, that is, the local damage area, as shown in Figure 8(a₃). These areas eventually aggregate and expand into macroscopic secondary cracks with the increase of loading force, and the development and expansion direction of them is along the vertical direction, and they are parallel to each other; they are the split cracks as shown in Figure 8(a₄).

Compared with IRS, there are a large number of rock–grout interfaces in GRBFRS under the shear angle of 45°. The strength of these rock–grout interfaces formed is obviously different due to the difference in the quality of cement-grout filling and consolidation, resulting in shear re-failure and compression failure of GRBFRS at the original primary shear crack and weak secondary cracks, as shown in Figure 8(b₂). With the increase of loading, the rock–grout interface of the grouting-reinforced body at the original primary shear crack is continuously peeled off and destroyed (Figure 8(b₃)). Moreover, some new splitting and compression re-failures occur near the weak rock–grout interfaces or micro-damage of original primary and secondary cracks, as shown in Figure 8(b₃) and (b₄).

Comparing Figure 8(a₁–a₄) with Figure 8(b₁–b₄), it is also found that under the shear angle of 45°, although the rock–grout interface and the adjacent area in the grouting-reinforced body are the main occurrence areas of shear re-failure, some splitting and compression re-failures are not initiated from the rock–grout interface. This may be because, on the one hand, some partial micro-damage and micro-cracks in the sample are generated during shear failure, and cannot be effectively filled and cemented during grouting reinforcement, leading to its low strength; on the other hand, the strength of a part of the macroscopic cracks produced in sample is significantly restored after grouting reinforcement, and they exceed the strength of micro-damage and micro-cracks that are not filled and cemented.

When the shear angle is 60°, the concentration area of shear strain in IRS appears in the middle of the sample, and develops to both sides along the direction of shear angle to form a through macroscopic shear fracture, as shown in Figure 8(c₂) to (c₄). However, there is no macroscopically visible secondary crack near the shear crack, which is different from the sample with a shear angle of 45°. The main reason for this difference is that the shear stress of the sample with a shear angle of 60° is greater than the normal stress (up to 1.73 times), that is, the shear stress is the main control stress that causes the shear failure of the sample. In contrast, the increased area of shear strain in GRBFRS appears at the rock–grout interface formed from the original shear crack. With the increase of loading, the grouting-reinforced body peels off at the rock–grout interface, causing microcracks to develop and expand into a macroscopic shear crack, and eventually undergoes macroscopic shear re-failure, as shown in Figure 8(d₂) to (d₄). In addition, during the whole loading process, although there is a local increasing area of shear strain on the surface of GRBFRS, it is not able to develop and expand into macroscopic cracks, as shown in Figure 8(d₂) to (d₄).

When the shear angle is 75°, the shear strain of IRS at one end of shear increases (Figure 8(e₂)) and develops and expands along the direction of near shear angle to the other side to form a through macroscopic shear crack, as shown in Figure 8(e₃). Moreover, a local increasing area of shear strain in IRS is generated (Figure 8(e₃)). As the loading continues, the local increasing area of shear strain develops and aggregates into macroscopic secondary shear cracks, and the direction of the cracks is approximately parallel to the primary shear crack, as shown in Figure 8(e₄). This is different from the failure of samples under the shear angles of 45° and 60°. The possible main reason is that the shear stress of the sample under the shear angle of 75° is much greater than the normal stress (up to 3.73 times), which makes it easy for the sample to form a secondary shear crack in the direction of the shear angle. This also leads to shear re-failure of GRBFRS only at the rock–grout interface formed by one of the shear cracks, and obvious local shear re-failure occurs at the upper shear point, as shown in Figure 8(f₂–f₄). In addition to the larger shear stress of the sample, the reason for the shear failure may be that the strength of the rock–grout interface in GRBFRS is different due to the different filling cementation. This makes the one with the lowest strength among all rock–grout interfaces become the main site of inducing shear re-failure. However, there are many micropore cracks and rock–grout interfaces formed by grouting at the local shear fracture of IRS, which leads to low strength and easy to produce local shear re-failure.

From the above analysis, it can be seen that when the shear angle is 45°, the shear stress and normal stress on the sample are both the main control stress of rock failure, which causes the sample to undergo the shear-split failure with shear failure as the main and split failure as the supplementary. With the increase of shear angle, the shear stress of the sample is the main control stress to control the failure, which makes the sample mainly shear failure. The strength of shear-fractured red sandstone can be improved by grouting and filling cementation, but the rock–grout interface formed becomes the main inducer of shear re-failure due to its low bonding strength. Therefore, when the fractured rock mass is reinforced by grouting, improving the strength of the rock–grout interfaces will help to improve the overall strength of the grouting-reinforced body, and then improve the grouting reinforcement effect.

4. Conclusions

In this paper, the variable-angle shear test, AE detecting test and VIC-3D capturing test on the IRS and GRBFRS under the same conditions were performed. Meanwhile, the shear strength and parameters, and the shear failure and re-failure characteristics of red sandstone before and after grouting reinforcement were compared and analyzed. The main conclusions are as follows:

(1) The shear compaction stage and elastic stage of IRS and GRBFRS are shortened with the increase of shear angle. At the same shear angle, the compaction stage of GRBFRS is longer than that of IRS. The integrity of GRBFRS is worse than that of IRS, and there are many micro-pores or cracks and rock-grout structural planes in GRBFRS, which will affect the deformation of GRBFRS.

(2) With the increase of shear angle, the peak shear strength of IRS and GRBFRS decreased. Compared with the residual shear strength of IRS, grouting reinforcement mainly increases the cohesion of GRBFRS by reconstructing the structure of fractured red sandstone to effectively improve the shear strength up to 123.19%. However, the reconstructability of grouting reinforcement on the shear strength of fractured red sandstone decreases with the increase of shear angle. The normal stress on GRBFRS is an important factor affecting the reconstructability of grouting.

(3) Differences from the IRS only occurred with the obvious shear failure in a very short time before and after the peak stress stage, where the shear micro-damage and shear failure of GRBFRS was more over the whole shear deformation process. However, the intensity or degree of shear re-failure is reduced. When the shear angle is 45°, the shear stress and normal stress are both the main control stress of shear-splitting failure. When the shear angle is 60° and 75°, the shear stress is the main control stress of the failure, so that only shear failure occurs. The larger the shear angle, the more likely the sample to produce secondary shear cracks.

(4) The low strength of the rock–grout interface in GRBFRS is the main cause of shear re-failure. The key to improve the strength of a grouting-reinforced body is to improve the strength of rock–grout interfaces.