J. Cent. South Univ. (2021) 28: 2472-2484

DOI: https://doi.org/10.1007/s11771-021-4780-0

Evolution of anisotropy during sandstone rockburst process under double-faces unloading

LIU Dong-qiao(刘冬桥)¹, LING Kai(凌凯)^{1, 2}, LI Dong(李东)¹, HE Man-chao(何满潮)¹,

LI Jie-yu(李杰宇)^{1, 2}, HAN Zi-jie(韩子杰)^{1, 2}, ZHANG Shu-dong(张树东)^{1, 2}

1. State Key Laboratory for GeoMechanics and Deep Underground Engineering, China University of Mining and Technology, Beijing 100083, China;

2. School of Mechanics and Civil Engineering, China University of Mining and Technology,Beijing 100083, China

Central South University Press and Springer-Verlag GmbH Germany, part of Springer Nature 2021

Abstract: Rockburst is one of the major disasters in deep underground rock mechanics and engineering. The precursors of rockbursts play important roles in rockburst prediction. Strainburst experiments were performed under double-face unloading on sandstone with horizontal bedding planes using an independently designed rockburst testing facility. P-wave propagation time during the tests was automatically recorded by the acoustic emission apparatus. The P-wave velocities were calculated in both two directions to analyze their patterns. To find a characteristic precursor for rockburst, the dynamic evolution of rock anisotropy during the rockburst test is quantified by the anisotropic coefficient k, defined as the ratio of the two P-wave velocities in the directions vertical to and parallel to the bedding planes. The results show that rockburst occurs on the two free surfaces asynchronously. The rockburst failure occurs in the following order: crack generation, rock peeling, particle ejection, and rock fracture. In the process of rockburst under double-face unloading, the potential evolution characteristics of anisotropy can be generalized as anisotropy-isotropy-anisotropy. The suddenly unloading induces damage in the rock and presents anisotropic coefficient k steeply increasing departing from one, i.e., isotropy. The rocks with horizontal bedding planes will reach the isotropic state before rockburst, which could be considered as a characteristic precursor of this kind of rockburst.

Key words: strainburst; bedding plane; P-wave velocity; anisotropy; rockburst prediction

Cite this article as: LIU Dong-qiao, LING Kai, LI Dong, HE Man-chao, LI Jie-yu, HAN Zi-jie, ZHANG Shu-dong. Evolution of anisotropy during sandstone rockburst process under double-faces unloading [J]. Journal of Central South University, 2021, 28(8): 2472-2484. DOI: https://doi.org/10.1007/s11771-021-4780-0.

1 Introduction

Rockburst is frequently encountered in the deep mining excavations given a high geo-stress condition. As one kind of drastic rock mass failure, it is featured by the ejection of rock fragment with high kinetic energy [1, 2]. Such a dynamic failure not only poses a significant threat to workers, but also causes great economic losses. Therefore, rockbursts have become an obstacle in the deep underground engineering operations. According to COOK [1], rockbursts can be categorized into the following three types, i.e., fault slip burst, pillar burst and strainburst. The most common form of rockburst is strainburst [3]. In recent years, many researchers have spent a lot of efforts to investigate strainburst in terms of mechanism, numerical simulation and physical modeling [4-7].

The strain energy amount accumulated in the rock mass has been regarded as the most important factor affecting the strainburst. COOK [8] initially proposed a method to predict rockburst based on the excessive energy. SALAMON [9] proposed an idea of energy transition during the mining operation. KAISER et al [2] developed the rockburst support countermeasure based on energy release approaches. Many researchers studied the strainburst mechanism from the view of energy based on numerical and experimental methods [10-18]. Although abundant research achievements have demonstrated that enough energy is the basic mechanism of rockburst, the characteristic precursor of rockburst is still not clear which is the key for successful prediction.

The rock mass forms a mutual layered structure under diagenesis. In the underground engineering with complex stress environment, the rock masses often exhibit physical and mechanical anisotropy. Nowadays, numerous attempts have been made to investigate anisotropy and find that it can affect rock properties evidently. For example, NIANDOU et al [19] carried out triaxial compression tests to investigate the mechanical behavior of shale and found that anisotropy had an influence on plastic deformation and failure behavior. LI et al [20] conducted triaxial compression tests to investigate anisotropic characteristics of crack initiation and crack damage thresholds for shale. However, these studies are mainly focusing on the effect of anisotropy on the mechanical properties and static brittle failure processes of rocks rather than rockburst, i.e., dynamic failure or ejection failure of rocks [21]. Few studies have addressed how anisotropy features evolve during the rockburst process.

Physical modeling in the laboratory provides an alternative method to understand the rockburst process in a better way. The advancements in experimental techniques change the past way of testing work, from uniaxial tests [8, 22-24], biaxial tests [25, 26] to true triaxial tests [27-34]. The in-depth study of rockburst demonstrated that the true triaxial compression tests disobeyed the in-situ stress path during excavation. Therefore, to simulate real rockburst process as much as possible through the true triaxial unloading experiments prevails in recent years. The true triaxial experimental apparatus designed by HE et al [4] was the earliest attempt towards this direction. This independent designed apparatus is capable of unloading from one direction in a short time to simulate the excavation process in site. A large number of rockburst experiments were carried out using this system on acoustic emission characteristics, fragmentation characteristics, unloading rate, thermal damage, etc [4, 35-45]. Later, SU et al [46-50] developed a true triaxial rockburst experimental system with higher loading capacity and the capability of rapid unloading in one direction. Moreover, many researchers have studied the influence of unloading on the rock mass stability. WU et al [51] conducted a series of laboratory experiments to investigate the effects of unloading rate, initial shear stress, and initial normal stress on the frictional instability of the fault. A numerical study was also carried out to interpret the stress variation and particle evolution during the unloading process. SI et al [52] conducted the triaxial unloading compression tests on the cubic fine-grained granite specimens to investigate the effects of three-dimensional stress state and unloading rate on the rockburst process. LI et al [53] used a mathematical physical model to explain the unloading mechanisms of the brittle rock under different stress paths at two dimensional scale. The proposed model agreed well with the theoretical solution and provided a basis for understanding the evolution of the unloading response around the tunnel.

Due to the equipment limitations, almost all true triaxial unloading experiments used a modified true triaxial apparatus that could suddenly unload the stress only in one surface of the specimen. The experimental method can only simulate the rockburst occurred at the tunnel face. With the rapid development in the underground engineering, there are many complex intersection tunnels, such as “U-type” tunnels which could be regarded as unloading from double-faces.

In this study, the strainburst experiments of sandstone were carried out by an independent designed new true triaxial experimental apparatus which is capable of unloading from two directions. The anisotropic property is analyzed to capture the characteristic precursor of rockburst using the P-wave velocity acquisition unit. In order to observe the rockburst process phenomenon, a high-speed camera was also employed in the experiments.

2 Rockburst experimental procedures

2.1 Specimen preparation

In many underground engineering, such as the N-J hydropower station diversion tunnel in Pakistan and Ruhr coal in Germany, strainbursts occurred in sandstone. Sandstone is a typical sedimentary rock, whose anisotropic behaviors are caused by the bedding planes. Hence, sandstone with horizontal bedding planes was used to simulate the strainburst phenomenon.

Two groups of specimens (YA and YB) were cut from massive sandstone blocks, from Shandong Province and Jiangxi Province, China. The sandstone is red, medium-to-fine-grained 150 mm× 150 mm×150 mm cube. X-ray diffraction (XRD) analysis reveals the major minerals in the rock are quartz (59.64%), albite (31.96%), microcline (4.52%), clinochlore (2.25%) and clay minerals (1.63%, illite dominant). The basic physical and mechanical parameters are listed in Table 1. The surface of the specimen was carefully ground to satisfy the orthogonality requirements. The sandstones used in the experiment were shown in Figure 1. The two horizontal bedding planes with thickness varying from 7 to 10 mm are obviously observed.

2.2 Experimental system

The 5000 kN hydraulic servo triaxial rockburst experimental system available at the State Key Laboratory for GeoMechanics and Deep Underground Engineering was used in the experiment. The loading capacities of the experimental system are 5000 and 2000 kN in the vertical and horizontal directions, respectively. The system is capable of loading independently from three directions and different control modes (load control or displacement control) can be chosen for each direction. The loading system adopts spherical seats to increase the tolerance to the possible unparallel defects from the sample preparation. During the loading on six surfaces along with the three directions, rapid unloading can be realized in one surface, two surfaces, three surfaces or four surfaces, which are used to simulate rockbursts at a roadway, roadway intersection, a pillar with three faces, and a pillar with four faces, respectively. To investigate the P-wave velocity and failure characteristics, both the P-wave velocity acquisition and high-speed photography units were used during the experiments. The experimental setup is shown in Figure 2.

2.3 P-wave velocity acquisition system and testing scheme

The automatic sensor test (AST) function of the PCI-2 acoustic emission (AE) monitoring system was used to record the travelling time differences of P-wave to different sensors. The P-wave velocities can be thusly calculated. The system consists of sensors, preamplifiers, computer mainframe, and transmission cables. In order to ensure the time consistency of the experimental data, the time of all equipments are synchronized before the test. Moreover, the gain value of the preamplifier and threshold value are set at 40 dB to filter external noise effectively. A small amount of butter was applied to the interfaces between the sensor and specimen to reduce signal attenuation. Consequently, the change of P-wave velocity during the whole process can be effectively captured.

Using the AST function, four sensors are used to measure the P-wave velocities both parallel or perpendicular to the bedding planes. In each test, the sensor transmits 10 pulses. The duration of a pulse is 5 ms. Time interval between each pulse is 100 ms. In the convenience of attaching the sensors, a hole was dug in the loading plate to place sensor and a spring was put in the hole to ensure that sensors are tightly attached to the sample surfaces. Loading plate and sensor installation are shown in Figure 3.

Table 1 Basic physical and mechanical parameters of granite rock material

Figure 1 Experimental samples

Figure 2 True triaxial rockburst experimental setup

Figure 3 Loading plate and sensor installation:

The positions of the sensors on the rock sample are shown in Figure 4. On the left side and right side of the specimen, No. 4 and No. 3 sensors are arranged on the center of the specimen, another two sensors (No. 1 and No. 2) were placed on the top and bottom faces of the specimen. The distance from the boundary to No. 1 and No. 2 sensor is 35 mm and 75 mm, respectively. The formula used for measuring P-wave velocity is as follows:

                              (1)

where △d is the distance between the sensors, the distance between the Nos. 1 and 2 sensors is 170 mm, and the distance between the No.3 and No.4 sensors is 150 mm; △t is time difference of P-wave propagation between two sensors.

2.4 Designed loading path

In this paper, the main experimental method is to simulate strainburst on sandstone. Strainburst is the most common rockburst, which is caused by strain energy release of the rock mass near the free boundary after excavation.

Since the elastic wave generated by continuous damage in the rock will be received by the acoustic emission sensor, it will significantly interfere with the pulse information emitted by the acoustic emission system. Therefore, a multi-stage loading method was adopted. The loadings increment for every stage is 25 kN, and the loading rate is 2 kN/s. When the preset stress was reached, it was maintained for 0.5 min (simplified as “L-M” in the following text). Then the acoustic emission monitoring system was used to test the P-wave velocity during the load holding period. The initial stress is determined using the geo-stress regression formula at the sampling area and the calculation formula of geo-stress are shown below [54]:

σ_v=0.017D;

σ_H=0.0233D+4.4665;

σ_h=0.0162D+2.1                          (2)

where σ_v, σ_h and σ_H are two horizontal and one vertical in suit stress, respectively, and D denotes the depth. The sampling depth of sandstones used in the experiment is 1000 m, therefore initial stress is σ_h=18.3 MPa, σ_H=28.0 MPa, σ_v=27.0 MPa, respectively.

The test process was designed to simulate the in situ stress state of rockburst. The detailed test procedures were as follows:

1) Applied loads gradually using the L-M method until the stress on all six surfaces reached the minor principal stress.

2) Maintained the loads on two horizontal surfaces with the minor principal stress and increased the stress on the other surfaces using L-M method until they reached the intermediate principal stress.

Figure 4 Layout of AE sensors:(Unit: mm)

3) After stress on the top surface reached intermediate principal stress, increased the stress on the other horizontal surfaces to the expected major principal stress using L-M method while keeping constant the loads on the other surfaces.

4) After stress in three directions increased to preset value, the loads were kept unchanged for several minutes to ensure a uniform stress distribution.

5) Suddenly remove the minor principal stress while increasing the major principal stress using the L-M method until rockburst occurred.

The stress transformation during the process of rockburst is shown in Figure 5. The loading path is shown in Figure 6, and the relationship between the bedding orientation and loading direction is shown in Figure 7.

Figure 5 Stress transformation process of strainburst under double-faces unloading

3 Experimental results

3.1 Rockburst phenomena

The failure characteristics under double-faces unloading are similar for the two groups of rock samples. Limited by article length, only one specimen of each group is selected for detailed discussion here. Figures 8 and 9 show high-speed photographic screenshots of the unloading surface of YA-1 and YB-1, respectively.

Figure 6 Loading path for strainburst

Figure 7 Relationship between bedding orientation and loading direction

On the left side of sample YA-1, a crack appeared on the right edge of the rock at 6050.310 s of the experiment corresponding to a stress state of σ_v=155.6 MPa, σ_H= 28.0 MPa and σ_h=0 MPa. However, no audible cracking sound could be heard. 1.225 s later, the small cracks in the upper right corner penetrated each other. Meanwhile, rock fragments were formed, accompanied by a quiet spalling sound. After 11.018 s, the upper half of the rock sample can be seen to have split and bent. This evolution can be seen clearly until it finally fell away from the rock. After 13.538 s, rockburst occurred on the entire surface. The rockburst fragments were ejected violently at high speed, producing a large amount of rock dust and accompanying by a loud blasting sound.

On the right side of YA-1, small particles began to be ejected constantly from the lower part of the sample at 6052.313 s of the experiment (σ_v=155.6 MPa). 8.39 s later, a slight cracking sound could be heard. Fragments began to spall off at the lower-left corner of the rock sample, though only slowly. After 9.01 s, most areas on the right side of the sample started to bulge outward, accompanied by a clear cracking sound. After 10.023 s, a violent rockburst occurred in the lower-left corner of the rock, with small particles ejected violently, accompanied by a loud sound.

Figure 8 Macroscopic phenomena associated with rockburst of YA-1:

Figure 9 Macroscopic phenomena associated with rockburst of YB-1:

Visible failure phenomena began to appear on the left side of rock sample YB-1 at 4351.821 s corresponding to a stress state of σ_v=85 MPa, σ_H= 28.0 MPa and σ_h=0 MPa. Three cracks appeared at the bottom of the rock and gradually expanded to the upper left corner, accompanied by a slight cracking sound. After 7.102 s, the cracks at the lower right corner penetrated each other. Rock fragments began to peel off, accompanied by a clear sound. After 8.21 s, rock fragments began to be ejected outward which was accompanied by the extrusion and fall of small particles from the bottom edge. After 8.84 s, rockburst occurred on the left side of the rock, accompanied by a loud cracking sound.

The right side of rock sample YB-1 also began to show damage phenomena at 4352.931 s (σ_v=85 MPa). A large crack was observed in the lower-left corner. The sound of minor cracking could be heard. No ejection of particles could be observed. After 6.082 s, the crack had penetrated the bottom of the rock. Fine particles began to be ejected. An associated squeaking sound was heard. After 8.342 s, the rock plate had folded on the right side, and the entire right part had peeled off; this was accompanied by a dull sound. Rock fragments were continuously ejected outward. After 9.357 s, a violent rockburst occurred. The rock fragments were ejected outward at high speed, and a loud explosion sound could be heard.

Figures 8 and 9 clearly show that rockburst can occur in both unloading faces at different elapsed time. The obvious rockburst pits are formed after the tests.

3.2 Variation characteristic of P-wave velocity

P-wave velocity is a comprehensive index that reflects the mechanical properties of rocks. The P-wave velocity in different directions provides useful information about the characteristics of anisotropic rock masses. Thus, research on P-wave velocity variation is an excellent method to describe the anisotropy characteristics of rockburst [55, 56].

The AST data recorded during the experiment were processed and analyzed. The variation of the P-wave velocities either parallel or perpendicular to the bedding planes were obtained. The results are shown in Figure 10. The v_a is the P-wave velocity parallel to the beddings, and v_b is the P-wave velocity perpendicular to the beddings. It is important to notice that when the stress in the direction of σ_h is unloaded, the stress in the direction of σ_H will fluctuate slightly. The reason that causes stress fluctuation is that when unloading minimum principal stress σ_h, the rock will deform significantly along the horizontal direction because of the vertical loading. The interaction between the rock and platen induced the slight fluctuation of the stress σ_H.

The following characteristic can be summarized from Figure 10.

Before unloading: At the beginning of the experiment, v_b is lower than v_a which can be attributed to the effect of the sandstone bedding planes. The sandstone exhibits significant anisotropy. When the initial three-dimensional loading reaches the minimum principal stress value (σ_h=σ_v=σ_H=18.3 MPa), the P-waves velocity in both directions increases continuously with the increase in stress. The bedding structures and voids are prone to compaction under the impact of stress. The sandstone becomes denser. Therefore, the P-wave velocity in both directions increases. In this stage, the sandstone becomes more homogeneous due to the compaction of the original cracks. While σ_h remains constant, σ_v and σ_H simultaneously increase to the intermediate principal stress value (σ_v=σ_H=27.0 MPa). The voids in the sandstone continue to shrink under stress. The sandstone becomes more compact, so v_a and v_b increase continuously. However, because the stress in the σ_h direction remains unchanged, the rates of increase in v_b and v_a slow down. In the third stage, σ_v remains constant. σ_H reaches the maximum principal stress value (σ_H=28.0 MPa). v_a and v_b slightly increase.

Figure 10 Relationships between P-wave velocity and stress:

Unloading point: The sandstone expands along the direction of unloading. The previously compacted cracks and voids are expanded instantaneously, affecting v_a and v_b. So there is a change in P-wave velocity in both directions or a single direction. It can be seen from Figure 10 that v_b will decrease at the unloading point due to damage in the rock when unloading minimum principal stress. But v_a has a significant increase (YB-1, YB-2, YB-3), decrease (YA-3) or no significant change (YA-1, YA-2) at unloading point. By observing the sample photos, some inconspicuous vertical bedding planes are found in YB-1, YB-2, YB-3 and YA-3. The rock will deform significantly in the horizontal direction when unloading. The vertical bedding planes will be compacted due to the dilation in the horizontal direction when the vertical bedding planes are perpendicular to free face. Therefore, v_a has a significant increase in YB-1, YB-2 and YB-3. When the vertical beddings are parallel to free face, the propagation of the P-wave is hindered. So v_a will decrease in YA-3.

After unloading: σ_h and σ_H remain constant, and σ_v continues to increase independently. v_a remains almost unchanged, but v_b continues to increase. Because of σ_H remaining unchanged, there is little compression of voids and cracks in the horizontal direction. So v_a along the horizontal direction tends to remain unchanged. σ_v continues to increase. Bedding structures and cracks in the sandstone continue to be compressed in the vertical direction, and rock becomes denser. Therefore, the v_b continues to increase.

Just before rockburst: Due to the continuous increase of σ_v, the number of cracks in the rock gradually increases, which hinders the propagation of P-waves. Therefore, it is not possible to measure P-wave propagation times, and P-wave velocity is no longer available.

During the rockburst experiment, the P-wave velocity in both directions increases firstly, then an abrupt change during unloading, finally a slower increase. v_a>v_b in the early stage before unloading, and there was a development trend toward v_a=v_b with the σ_v increase after unloading, and then v_ab.

4 Discussion

At present, the researchers have paid more attention to the anisotropy coefficient k because it is a favorable indicator, which can reflect the degree of anisotropy. The anisotropy coefficient has been widely used in many fields, such as rock mechanics, shale gas extraction. Because rock is an anisotropic brittle material, it is of great significance to study rock failure, especially rockburst, by using anisotropic coefficient k, so as to find the precursor information of rockburst. In this paper, the anisotropy coefficient k is used to characterize the evolution of rock anisotropy:

k=v_a/v_b                                 (3)

The closer the k value is to one, the lower the anisotropy of the rock. The greater the difference between the k value and one, the stronger anisotropy exhibited by the rock. When k=1, the rock behaves as an isotropic medium.

The anisotropy coefficient k during the process of rockburst was obtained from the P-wave velocity, and evolution was shown in Figure 11. The following evolution characteristic can be generalized from Figure 11:

1) Before unloading: Anisotropy coefficient k is greater than one and it fluctuates slightly but generally shows a decreasing trend. Because of the initial cracks closure, the anisotropy coefficient evolves from greater than one to one which indicated that the rock gradually becomes isotropic.

2) Unloading point: It can be seen from Figure 11, the anisotropy coefficient increases sharply, and the rock shows stronger anisotropy again. It demonstrates that unloading in the double-faces induces damage in the rock.

3) Stress concentration after unloading: The anisotropy coefficient shows a downward trend. At a certain time, k=1 (v_a=v_b), and the rock is isotropic. With the increase in stress, k<1, and the rock changes from isotropy to another kind of anisotropy that is different from the initial anisotropy.

4) Just before rockburst, P-wave velocity cannot be collected due to damage in the rock, which results in failing to calculate the anisotropy coefficient.

5 Conclusions

A new true triaxial rockburst experimental system is used to simulate the rockburst phenomenon at intersection tunnels. Changing pattern of the P-wave velocity during the rockburst caused by double-faces unloading are revealed, and the anisotropy characteristics of the rock mass are further analyzed. The result revealed that the anisotropy changes significantly during the whole experiment. For the sandstones with horizontal bedding planes, the evolution of anisotropy will transform from strong anisotropy to isotropy. The main conclusions are as follows:

Figure 11 Relationships between stress and anisotropy coefficient k:

1) In the double-faces unloading rockburst experiments on sandstone, rockburst can occur at two free surfaces. Their failure processes are similar. The failure process comprises several stages, including crack generation, rock spalling, particle ejection, and rock fracture. Rockburst occurs on the two free surfaces after different time delay.

2) During the rockburst experiments, the P-wave velocity in two directions experienced a process of increase, sudden change, and then slowly increase. In the early stage of loading, both of v_a and v_b increase; when unloading intermediate principal stress, v_a and v_b will fluctuate slightly; after unloading and increasing vertical stress, v_b keeps increasing and v_a remains unchanged.

3) The anisotropy coefficient in the rockburst experiments went through a process of first decreasing, then suddenly increasing when unloading and then decreasing to below one. Before unloading, the anisotropy coefficient decreased from a higher value to one. At the moment of unloading, the anisotropy coefficient increased sharply. After unloading, the anisotropy coefficient presented a decreasing trend. For the rocks with horizontal bedding planes, the isotropy state could be considered as a characteristic precursor of rockburst.

Contributors

The overarching research goals were developed by LIU Dong-qiao and HE Man-chao. LI Dong, LING Kai, HAN Zi-jie, and ZHANG Shu-dong conducted the experiments. LI Jie-yu analyzed the measured data.The initial draft of the manuscript was written by LING Kai and LI Dong and LIU Dong-qiao.All authors replied to reviewers’ comments and revised the final version.

Conflict of interest

LIU Dong-qiao, LING Kai, LI Dong, HE Man-chao, LI Jie-yu, HAN Zi-jie, and ZHANG Shu-dong declare that they have no conflict of interest.

References

[1] COOK N G W. A note on rockbursts considered as a problem of stability [J]. Journal of the South African Institute of Mining and Metallurgy, 1965, 65(10): 551-554. https://www.researchgate.net/publication/285089529_A_note_on_rockbursts_considered_as_a_problem_of_stability.

[2] KAISER P K, MCCREATH D R, TANNANT D D. Canadian rockburst support handbook [M]. Sudbury, Canada: Geomechanics Research Centre, 1996. https://www.researchgate.net/ publication/285771770.

[3] HE Man-chao, XIA Hong-man, JIA Xue-na, GONG Wei-li, ZHAO Fei, LIANG Kang-yuan. Studies on classification, criteria and control of rockbursts [J]. Journal of Rock Mechanics and Geotechnical Engineering, 2012, 4(2): 97-114. DOI: 10.3724/SP.J.1235.2012.00097.

[4] HE Man-chao, MIAO Jin-li, FENG Ji-lin. Rock burst process of limestone and its acoustic emission characteristics under true-triaxial unloading conditions [J]. International Journal of Rock Mechanics and Mining Sciences, 2010, 47(2): 286-298. DOI: 10.1016/j.ijrmms.2009.09.003.

[5] FENG Guang-liang, FENG Xia-ting, CHEN Bing-rui, XIAO Ya-xun, YU Yang. A microseismic method for dynamic warning of rockburst development processes in tunnels [J]. Rock Mechanics and Rock Engineering, 2015, 48(5): 2061-2076. DOI: 10.1007/s00603-014-0689-3.

[6] MANOUCHEHRIAN A, CAI Ming. Simulation of unstable rock failure under unloading conditions [J]. Canadian Geotechnical Journal, 2016, 53(1): 22-34. DOI: 10.1139/cgj-2015-0126.

[7] HE Jiang, DOU Lin-ming, MU Zong-long, CAO An-ye, GONG Si-yuan. Numerical simulation study on hard-thick roof inducing rock burst in coal mine [J]. Journal of Central South University, 2016, 23(9): 2314-2320. DOI: 10.1007/ s11771-016-3289-4.

[8] COOK N G W. The basic mechanics of rockbursts [J]. Journal of the South African Institute of Mining and Metallurgy, 1963, 64(3): 71–78. https://journals.co.za/doi/10.10520/AJA00382 23X_3752

[9] SALAMON M D G. Energy considerations in rock mechanics: Fundamental results [J]. Journal of the South African Institute of Mining and Metallurgy, 1984, 84(8): 233-246.

[10] WANG J A, PARK H D. Comprehensive prediction of rockburst based on analysis of strain energy in rocks [J]. Tunnelling and Underground Space Technology, 2001, 16(1): 49-57. DOI: 10.1016/S0886-7798(01)00030-X.

[11] MANSUROV V A. Prediction of rockbursts by analysis of induced seismicity data [J]. International Journal of Rock Mechanics and Mining Sciences, 2001, 38(6): 893-901. DOI: 10.1016/S1365-1609(01)00055-7.

[12] HUA An-zeng, YOU Ming-qing. Rock failure due to energy release during unloading and application to underground rock burst control [J]. Tunnelling and Underground Space Technology, 2001, 16(3): 241-246. DOI: 10.1016/S0886-7798(01)00046-3.

[13] SUN Jin-shan, ZHU Qi-hu, LU Wen-bo. Numerical simulation of rock burst in circular tunnels under unloading conditions [J]. Journal of China University of Mining and Technology, 2007, 17(4): 552-556. DOI: 10.1016/S1006-1266(07)60144-8.

[14] LI Zhi-hua, DOU Lin-ming, LU Cai-ping, MU Zong-long, CAO An-ye. Study on fault induced rock bursts [J]. Journal of China University of Mining and Technology, 2008, 18(3): 321-326. DOI: 10.1016/S1006-1266(08)60068-1.

[15] JIANG Quan, FENG Xia-ting, XIANG Tian-bing, SU Guo-shao. Rockburst characteristics and numerical simulation based on a new energy index: A case study of a tunnel at 2500 m depth [J]. Bulletin of Engineering Geology and the Environment, 2010, 69(3): 381-388. DOI: 10.1007/s10064-010-0275-1.

[16] HE Man-chao, REN Fu-qiang, LIU Dong-qiao, ZHANG Shu-dong. Experimental study on strain burst characteristics of sandstone under true triaxial loading and double faces unloading in one direction [J]. Rock Mechanics and Rock Engineering, 2021, 54(1): 149-171. DOI: 10.1007/s00603-020-02272-3.

[17] HE Man-chao, LI Jie-yu, LIU Dong-qiao, LING Kai, REN Fu-qiang. A novel true triaxial apparatus for simulating strain bursts under high stress [J]. Rock Mechanics and Rock Engineering, 2021, 54(2): 759-775. DOI: 10.1007/s00603-020-02285-y.

[18] CHEN Zhi-yong, SU Guo-shao, JU J W, JIANG Jian-qing. Experimental study on energy dissipation of fragments during rockburst [J]. Bulletin of Engineering Geology and the Environment, 2019, 78(7): 5369-5386. DOI: 10.1007/ s10064-019-01463-9.

[19] NIANDOU H, SHAO J F, HENRY J P, FOURMAINTRAUX D. Laboratory investigation of the mechanical behaviour of Tournemire shale [J]. International Journal of Rock Mechanics and Mining Sciences, 1997, 34(1): 3-16. DOI: 10.1016/S1365-1609(97)80029-9.

[20] LI Cun-bao, XIE He-ping, WANG Jun. Anisotropic characteristics of crack initiation and crack damage thresholds for shale [J]. International Journal of Rock Mechanics and Mining Sciences, 2020, 126: 104178. DOI: 10.1016/ j.ijrmms.2019.104178.

[21] KAISER P K, CAI Ming. Design of rock support system under rockburst condition [J]. Journal of Rock Mechanics and Geotechnical Engineering, 2012, 4(3): 215-227. DOI: 10.3724/SP.J.1235.2012.00215.

[22] NEMAT-NASSER S, HORII H. Compression-induced nonplanar crack extension with application to splitting, exfoliation, and rockburst [J]. Journal of Geophysical Research: Solid Earth, 1982, 87(B8): 6805-6821. DOI: 10.1029/JB087iB08p06805.

[23] NEMAT-NASSER S, HORII H. Rock failure in compression [J]. International Journal of Engineering Science, 1984, 22(8-10): 999-1011. DOI: 10.1016/0020-7225(84)90101-0.

[24] BAGDE M N, PETROS V. Fatigue properties of intact sandstone samples subjected to dynamic uniaxial cyclical loading [J]. International Journal of Rock Mechanics and Mining Sciences, 2005, 42(2): 237-250. DOI: 10.1016/ j.ijrmms.2004.08.008.

[25] ZHANG Xiao-jun. Experimental research on splitting rock burst of surrounding rocks in deep roadways (tunnels) [J]. Journal of Mining & Safety Engineering, 2011, 28(1): 66-71.

[26] WANG Zhao-feng, FENG Xia-ting, YANG Cheng-xiang, ZHOU Yang-yi, XU Hong, HAN Qiang, GAO Yao-hui. Experimental investigation on fracturing process of marble under biaxial compression [J]. Journal of Rock Mechanics and Geotechnical Engineering, 2020, 12(5): 943-959. DOI: 10.1016/j.jrmge.2020.05.002.

[27] TAKAHASHI M, PARK H, TAKAHASHI N, FUJII Y. True triaxial tests–using permeability and extensional stress parameters to simulate geological history in rocks [J]. Geosystem Engineering, 2013, 16(1): 75-82. DOI: 10.1080/ 12269328.2013.780760.

[28] HU Xiao-chuan, SU Guo-shao, CHEN Kang, LI Tian-bin, JIANG Quan. Strainburst characteristics under bolt support conditions: An experimental study [J]. Natural Hazards, 2019, 97(2): 913-933. DOI: 10.1007/s11069-019-03682-5.

[29] FENG Xia-ting, ZHANG Xi-wei, KONG R, WANG G. A novel mogi type true triaxial testing apparatus and its use to obtain complete stress-strain curves of hard rocks [J]. Rock Mechanics and Rock Engineering, 2016, 49(5): 1649-1662. DOI: 10.1007/s00603-015-0875-y.

[30] HOFER K H, THOMA K. Triaxial tests on salt rocks [J]. International Journal of Rock Mechanics and Mining Sciences & Geomechanics Abstracts, 1968, 5(2): 195-196. DOI: 10.1016/0148-9062(68)90034-X.

[31] JIANG Bang-you, GU Shi-tan, WANG Lian-guo, ZHANG Guang-chao, LI Wen-shuai. Strainburst process of marble in tunnel-excavation-induced stress path considering intermediate principal stress [J]. Journal of Central South University, 2019, 26(4): 984-999. DOI: 10.1007/s11771-019-4065-z.

[32] GONG Feng-qiang, LUO Yong, LI Xi-bing, SI Xue-feng, TAO Ming. Experimental simulation investigation on rockburst induced by spalling failure in deep circular tunnels [J]. Tunnelling and Underground Space Technology, 2018, 81: 413-427. DOI: 10.1016/j.tust.2018.07.035.

[33] GONG Feng-qiang, SI Xue-feng, LI Xi-bing, WANG Shan-yong. Experimental investigation of strain rockburst in circular Caverns under deep three-dimensional high-stress conditions [J]. Rock Mechanics and Rock Engineering, 2019, 52(5): 1459-1474. DOI: 10.1007/s00603-018-1660-5.

[34] LUO Yong, GONG Feng-qiang, LI Xi-bing, WANG Shan-yong. Experimental simulation investigation of influence of depth on spalling characteristics in circular hard rock tunnel [J]. Journal of Central South University, 2020, 27(3): 891-910. DOI: 10.1007/s11771-020-4339-5.

[35] HE Man-chao, JIA Xue-na, COLI M, LIVI E, SOUSA L. Experimental study of rockbursts in underground quarrying of Carrara marble [J]. International Journal of Rock Mechanics and Mining Sciences, 2012, 52: 1-8. DOI: 10.1016/ j.ijrmms.2012.02.006.

[36] HE Man-chao, NIE Wen, ZHAO Zhi-ye, GUO Wei-hua. Experimental investigation of bedding plane orientation on the rockburst behavior of sandstone [J]. Rock Mechanics and Rock Engineering, 2012, 45(3): 311-326. DOI:10.1007/s00603-011-0213-y.

[37] ZHAO X G, WANG J, CAI M, CHENG C, MA L K, SU R, ZHAO F, LI D J. Influence of unloading rate on the strainburst characteristics of Beishan granite under true-triaxial unloading conditions [J]. Rock Mechanics and Rock Engineering, 2014, 47(2): 467-483. DOI: 10.1007/s00603-013-0443-2.

[38] HE M C, ZHAO F, CAI M, DU S. A novel experimental technique to simulate pillar burst in laboratory [J]. Rock Mechanics and Rock Engineering, 2015, 48(5): 1833-1848. DOI: 10.1007/s00603-014-0687-5.

[39] HE Man-chao, SOUSA L, MIRANDA T, ZHU Gua-long. Rockburst laboratory tests database—Application of data mining techniques [J]. Engineering Geology, 2015, 185: 116-130. DOI: 10.1016/j.enggeo.2014.12.008.

[40] GONG Yu-xin, SONG Zhan-jie, HE Man-chao, GONG Wei-li, REN Fu-qiang. Precursory waves and eigenfrequencies identified from acoustic emission data based on singular spectrum analysis and laboratory rock-burst experiments [J]. International Journal of Rock Mechanics and Mining Sciences, 2017, 91: 155-169. DOI: 10.1016/j.ijrmms.2016.11.020.

[41] HE Man-chao, REN Fu-qiang, LIU Dong-qiao. Rockburst mechanism research and its control [J]. International Journal of Mining Science and Technology, 2018, 28(5): 829-837. DOI: 10.1016/j.ijmst.2018.09.002.

[42] AKDAG S, KARAKUS M, TAHERI A, NGUYEN G, HE Man-chao. Effects of thermal damage on strain burst mechanism for brittle rocks under true-triaxial loading conditions [J]. Rock Mechanics and Rock Engineering, 2018, 51(6): 1657-1682. DOI: 10.1007/s00603-018-1415-3.

[43] LIU Dong-qiao, LI De-jian, ZHAO Fei, WANG Cheng-chao. Fragmentation characteristics analysis of sandstone fragments based on impact rockburst test [J]. Journal of Rock Mechanics and Geotechnical Engineering, 2014, 6(3): 251-256. DOI: 10.1016/j.jrmge.2014.04.001.

[44] REN Fu-qiang, ZHU Chun, HE Man-chao. Moment tensor analysis of acoustic emissions for cracking mechanisms during schist strain burst [J]. Rock Mechanics and Rock Engineering, 2020, 53(1): 153-170. DOI: 10.1007/s00603-019-01897-3.

[45] WANG Yang, HE Man-chao, REN Fu-qiang, ZHU Chun, FARAMARZI L. Source analysis of acoustic emissions during granite strain burst [J]. Geomatics, Natural Hazards and Risk, 2019, 10(1): 1542-1562. DOI: 10.1080/ 19475705.2019.1593888.

[46] SU Guo-shao, JIANG Jian-qing, ZHAI Shao-bin, ZHANG Gang-liang. Influence of tunnel axis stress on strainburst: An experimental study [J]. Rock Mechanics and Rock Engineering, 2017, 50(6): 1551-1567. DOI:10.1007/s00603-017-1181-7.

[47] SU Guo-shao, ZHAI Shao-bin, JIANG Jian-qing, ZHANG Gang-liang, YAN Liu-bin. Influence of radial stress gradient on strainbursts: An experimental study [J]. Rock Mechanics and Rock Engineering, 2017, 50(10): 2659-2676. DOI: 10.1007/s00603-017-1266-3.

[48] SU Guo-shao, CHEN Zhi-yong, JU J W, JIANG Jian-qing. Influence of temperature on the strainburst characteristics of granite under true triaxial loading conditions [J]. Engineering Geology, 2017, 222: 38-52. DOI: 10.1016/j.enggeo. 2017.03.021.

[49] SU Guo-shao, HU Li-hua, FENG Xia-ting, YAN Liu-bin, ZHANG Gang-liang, YAN Si-zhou, ZHAO Bin, YAN Zhao-fu. True triaxial experimental study of rockbursts induced by ramp and cyclic dynamic disturbances [J]. Rock Mechanics and Rock Engineering, 2018, 51(4): 1027-1045. DOI: 10.1007/s00603-017-1384-y.

[50] SU Guo-shao, SHI Yan-jiong, FENG Xia-ting, JIANG Jian-qing, ZHANG Jie, JIANG Quan. True-triaxial experimental study of the evolutionary features of the acoustic emissions and sounds of rockburst processes [J]. Rock Mechanics and Rock Engineering, 2018, 51(2): 375-389. DOI: 10.1007/ s00603-017-1344-6.

[51] WU Wei, ZHAO Zhi-hong, DUAN Kang. Unloading-induced instability of a simulated granular fault and implications for excavation-induced seismicity [J]. Tunnelling and Underground Space Technology, 2017, 63: 154-161. DOI: 10.1016/j.tust.2017.01.002.

[52] SI Xue-feng, GONG Feng-qiang. Strength-weakening effect and shear-tension failure mode transformation mechanism of rockburst for fine-grained granite under triaxial unloading compression [J]. International Journal of Rock Mechanics and Mining Sciences, 2020, 131: 104347. DOI: 10.1016/ j.ijrmms.2020.104347.

[53] LI Xi-bing, CAO Wen-zhuo, ZHOU Zi-long, ZOU Yang. Influence of stress path on excavation unloading response [J]. Tunnelling and Underground Space Technology, 2014, 42: 237-246. DOI: 10.1016/j.tust.2014.03.002.

[54] HUANG Lu-yuan, YANG Shu-xin, CUI Xiao-feng, CHEN Qun-ce, YAO Rui. Analysis of characteristics of measured stress and stability of faults in North China [J]. Rock and Soil Mechanics, 2013, 34(S1): 204-213. DOI: 10.16285/ j.rsm.2013.s1.012.

[55] HORNBY B E, HOWIE J M, INCE D W. Anisotropy correction for deviated-well sonic logs: Application to seismic well tie [J]. Geophysics, 2003, 68(2): 464-471. DOI: 10.1190/1.1567212.

[56] STOEP D M V. Velocity anisotropy measurements in wells [J]. Geophysics, 1966, 31(5): 900-916. DOI: 10.1190/1.1439822.

(Edited by HE Yun-bin)

中文导读

砂岩双面卸载岩爆过程各向异性特征实验研究

摘要：岩爆是深部地下岩石工程中主要的灾害之一，岩爆前兆信息在岩爆预测中至关重要。利用自主研发的岩爆设备和具有水平层理的红砂岩开展了双面卸载应变岩爆实验。利用声发射设备自动采集了实验过程中P波的传播时间，并且计算分析了P波在水平层理方向和垂直层理方向的传播速度。为了找到岩爆的特征前兆信息，将垂直于层理的波速与平行于层理的波速的比值定义为各向异性系数k，定量分析了岩爆过程中岩石各向异性的演化规律。实验结果表明，两个卸载面的岩爆并非同时发生，岩爆主要包括裂纹萌生，岩片剥落，颗粒弹射，岩板断裂等过程。双面卸载岩爆实验的各向异性演化特征可以总结为由各向异性向各向同性转化，而后再次向各向异性转化，并且快速卸载会对岩石产生损伤，各向异性系数会突增。具有水平层理的岩石在岩爆之前会达到各向同性状态，可以将此视为岩爆的前兆信息。

关键词：应变岩爆；层理；P波速度；各向异性；岩爆预测

Foundation item: Projects(41941018, 51704298) supported by the National Natural Science Foundation of China; Project(2021JCCXSB03) supported by the Fundamental Research Funds for the Central Universities, China

Received date: 2020-08-28; Accepted date: 2021-03-08

Corresponding author: LIU Dong-qiao, PhD, Associate Professor; Tel: +86-13811861688; E-mail: liudongqiao@yeah.net; ORCID: https://orcid.org/0000-0001-9946-7505