Introduction

The chemical environment of mining areas is very complicated and various chemical substances will react with the minerals in rocks. These reactions will change the mineral compositions of the rocks, and damage their internal structure and integrity. This damage results in deterioration of the mechanical parameters of the rock, thus posing a great threat to the safety of rock engineering (Chen et al. 2002; Feng and Ding 2011; Ding et al. 2015). Mine wastewater is acidic or weakly acidic and pH value is as low as 1.9 (Baker and Banfield 2003; Gaikwad and Gupta 2008; Mohajane et al. 2014; Fernando et al. 2018; Zhang et al. 2020); acid dissolution by such wastewater has a deleterious effect on the mechanical strength of rock (Yao et al. 2009; Liao et al. 2020).

Acid dissolution is the most commonly used experimental method to study the mechanical behavior of rock materials affected by acid mine drainage (Hutchinson et al. 1993; Putnis et al. 2014; Ryou J et al. 2015; Zhang 2017). Uniaxial compression testing has been widely applied to study the mechanical properties and fracture mechanisms of different rocks. Tang et al. (2002a, b) performed uniaxial compressive strength tests and fracture effect tests on granite, limestone, and sandstone that had been treated with chemical solutions of different ion concentrations. Jiang and Wen (2011) performed uniaxial compression tests on sandstone eroded by acid mine drainage solution. Miao et al. (2018) studied the constitutive model of fractured rock subjected to uniaxial compression and chemical dissolution, and deduced that the rock damage resulted from pores generated by hydro–physical–chemical interactions. Guo et al. (2014) conducted experiments on natural and saturated karst limestone under uniaxial and triaxial loading and studied the energy features and energy mechanism during failure. These studies demonstrated that acid-treated rocks are more easily broken than untreated rocks.

During uniaxial compression tests, rock material failure occurs within a very short time. Acoustic emission (AE) has been widely used to record the failure progress of rock specimens under loading (Scholz et al. 1968; Sammonds et al. 1992; Li et al. 2019; Gu et al. 2019; Zhu et al. 2020). Multiple techniques such as imaging, computed tomography, ultrasonic pulse velocity, Hg intrusion porosimetry, polarized light, and fluorescent optical microscopy have also been applied to explain the deformation and fracture mechanism of rock materials (Ehsan et al. 2017; Zhang et al. 2017; Li et al. 2018; Li et al. 2020; Murru et al. 2018).

The mechanical strength and failure mode of rock are sensitive to the growth of cracks. Though AE has been widely used to investigate uniaxial and triaxial compression of rocks, most experiments have not considered structural planes or the effect of acid dissolution (Edward et al. 2008; Li et al. 2009; Zha et al. 2020). Although there has been some limited work applying AE to rocks containing structural planes (Wang et al. 2018; Zhang et al. 2019; Wang et al. 2020), there is little information on the response of limestone containing structural planes to acid treatment.

After the mine is shut down, the underground water level in the original goaf changes, so that the overlying strata or coal and rock columns that have been in a state of stress equilibrium and stability are soaked and dissolved in water. Due to the decrease of strength, geological disasters such as ground collapse, mountain collapse, and landslide are induced. The aim of this study was to study the mechanical properties of layered limestone (both untreated and after exposure to acid solution) by means of uniaxial compression tests and AE monitoring. The failure mode, stress–strain curves, mechanical parameters, and AE evolution of specimens with dip angles of 0 to 90° were analyzed. The mechanism of degradation by dissolution was also assessed, by considering the differences in the mineral composition of the structural plane and the rock matrix. The results of this study will provide useful reference information for the stability of rock engineering in areas affected by acid mine drainage.

Materials and methods

Specimen preparation

The limestone specimens were obtained from the middle-weathered gray thin-layer to middle-thick limestone of the lower Triassic Yongningzhen Formation (300–875) in Anshun City, Guizhou Province, China (105° 54′ 30′′, 26° 24′ 24′′) (Fig. 1) (Lehrmann et al. 2020; Enos et al. 2006; Wang et al. 2017). Undeveloped joint fissures, with parallel bedding, the spacing between bedding was about 0.5–2cm. In practical engineering, the pH value of the acidic wastewater produced after the stoppage of the mine is as low as 1.9, resulting in strong karst phenomenon (Fig. 2). Therefore, the experimental scheme is designed to carry out the corrosion test. According to the results of polarizing microscope, the limestone matrix is mainly crystalline limestone, and the bedding plane is black argillaceous limestone (Fig. 3a). The matrix structure of the rock is dense, and there were small pores between local grains (Fig. 3b). The bedding plane of limestone was lamellar structure (Fig. 3c). According to the X-ray diffraction results of limestone, the matrix minerals of limestone mainly include calcite, partial quartz, and dolomite (Fig. 3d, Table 1), and the bedding includes clay minerals (Fig. 3e). The quarry sampling method is to make horizontal bedding rectangular blocks by blasting and cutting (Fig. 4a). Following the ISRM recommendations (Fairhurst and Hudson 1999), standard limestone cylinders (diameter 50 mm, length 100 mm) were drilled using a rock coring machine to yield specimens with five different dip angles (0°, 30°, 45°, 60°, and 90°; Fig. 4b, Table 2).

Fig. 1
figure 1

Limestone sampling location and regional geology (Guizhou Academy of Geological survey)

Full size image
Fig. 2
figure 2

Corrosion of rock mass caused by mining (Yudong river coal mining district, Guizhou Province)

Full size image
Fig. 3
figure 3

Microstructure and mineral composition of limestone. (a) Polarizing microscope for limestone specimen. (b) Scanning electron microscopy of crystalline limestone (matrix). (c) Scanning electron microscopy of muddy limestone (bedding). (d) X-ray diffraction results of limestone. (e) X-ray diffraction results of limestone bedding plane

Full size image
Table 1 Rock mineral and chemical composition
Full size table
Fig. 4
figure 4

Typical specimens with different bedding dip angles (0°, 30°, 45°, 60°, and 90°, respectively (measured from the horizontal)). (a) Limestone specimens production site. (b) Schematic diagram of drilling and coring from block bedding rock. (c) Photo of typical specimens

Full size image
Table 2 Basic physical indexes of limestone specimens
Full size table

Specimens were labeled with their dissolution status (US, untreated or DS, dissolved), bedding angle (0 to 90°), and specimen number; for example, US-0-1 is the first untreated specimen with a bedding angle of zero, and DS-0-1 is the first acid-treated specimen with a bedding angle of zero.

Experimental system

The system used for the experiments comprises a dissolution device, an axial loading system, and an AE data-acquisition system. The dissolution device included a solution box, a pH meter, and an acid solution. The axial loading system, a servo-controlled WAW-1000KN compression testing machine (Shanghai Hualong Test Instruments Corporation, Shanghai, China), has closed loop controls and can perform constant stress control and stress retention tests. It has a maximum load capacity of 1000 kN and the press has two control modes: displacement and force.

Experimental scheme

An aqueous solution of hydrochloric acid at pH 2 was prepared. Five groups, each of three limestone specimens with similar density and P-wave velocity, were selected, and the prepared rock specimens were weighed and immersed in the acidic solution. The solution remained at pH 2 throughout the experiment (Fig. 5). After immersion for 22 days, the specimens were removed from the acid solution, washed in water, and subjected to uniaxial compression testing. Before each experimental run, two AE sensors were attached to the specimen (Fig. 6c). The AE probe was smeared with Vaseline for better recording of AE events. Finally, the load was applied by displacement control mode at a loading rate of 0.2 mm/min. The AE monitoring systems and the loading system were turned on simultaneously at the moment that the experiment began, to acquire AE data and mechanical parameters. In this study, four aspects of the deformation and failure of the limestone were analyzed (AE counts, cumulative AE counts, cumulative AE energy, and b-value), to reveal the evolution of cracks inside the rocks under uniaxial loading.

Fig. 5
figure 5

Dissolved test for limestone specimens under different bedding angles

Full size image
Fig. 6
figure 6

Schematic diagram and photos of the experimental system. (a) Loading control system. (b) Loading and AE signal collecting system. (c) AE data storage and processing system. (d) Photo of a limestone sample and bedding inclination angle. (1) Servo hydraulic press. (2) Limestone sample. (3) Preamplifier. (4) Computer of data display. (5) AE monitoring device. (6) Manual lifting switch. (7) Computer of loading control and data storage. (8) Pressure gauge

Full size image

Results and discussion

Mechanical properties of the acid-immersed limestone

The stress–strain behavior of rock specimens under uniaxial compression has been extensively investigated to understand the cracking process (Wawersik and Brace 1971). The stress–strain curve of the uniaxial loading process can be divided into five stages, illustrated for specimens US-0-1 and DS-0-1 in Fig. 7a. During the initial compression stage (O–A), existing micro-fractures in the limestone specimen are compressed continuously, and the stress–strain curve is nonlinear and slightly concave. During the succeeding elastic deformation stage (A–B), the stress values of the acid-treated and untreated specimens are 0.03σp and 0.04σp, respectively (σp represents the peak stress of the stress–strain curves). The micro-cracks and weak joints are further compressed, most micro-cracks have closed, and the stress–strain curve is linear. The stable crack growth stage (B–C) is reached at stress values of 0.21σp (acid-treated specimens) and 0.51σp (untreated specimens). In this stage, the formation of new micro-cracks induced by the increasing stress continuously deteriorates the mechanical performance of the limestone specimen. As the deformation of limestone approaches the plastic deformation stage, the stress–strain curve deviates from a straight line. The stress value of the acid-treated specimens is markedly less than that of the untreated specimens; therefore, the acid solution can reduce the strength of the limestone bedding planes and the parent rock. Subsequently, at stress values of 0.34σp (acid-treated) and 0.93σp (untreated), the specimens enter the unstable crack growth stage (C–D). During this stage, micro-cracks expand, and the concentration of stress at the crack tip and continuous fracture growth causes release of both stress and energy. Finally, when the σp is reached, the limestone specimen enters the failure stage (D–E). During this stage, macroscopic rupture surfaces form, the dilatancy is strong, and the strength falls rapidly until the rock breaks completely and loses its carrying capacity. Compared with the test curve results before and after limestone bedding was dissolved at 0° (Fig. 7a), 30° (Fig. 7b), 45° (Fig. 7c), 60° (Fig. 7d), and 90° (Fig. 7e), when the bedding angle is 60°, the dissolved and untreated limestone enters the stable crack growth stage, the stress value is lower than that of other bedding dip angles, and the stress value after dissolution is lower than that of untreated limestone. For the stress value when unstable crack growth stage, according to the bedding angles of 0°, 30°, 45°, 60°, and 90°, the stress values of untreated limestone samples are 136.42 MPa, 79.58 MPa, 37.86 MPa, 53.95 MPa, and 110.10 MPa, while the stress values of limestone samples after dissolution are 42.39 MPa, 34.41 MPa, 33.66 MPa, 27.31 MPa, and 50.82 MPa, It can be concluded that the stress values of dissolved bedding limestone are lower than those of untreated limestone, and the stress values remain low when the bedding angle is 45° and 60°.

Fig. 7
figure 7figure 7

Relationship between number of AE counts, cumulative energy, b-values, and stress–stain curve. (a–e) Specimens UN-0-1 and DS-0-1, UN-30-1 and DS-30-1, UN-45-1 and DS-45-1, UN-60-1 and DS-60-1, UN-90-1 and DS-90-1, respectively

Full size image

The uniaxial compressive strength (UCS) and Young’s modulus were different in limestone specimens with different bedding dip angles. The UCS and Young’s modulus of untreated and acid-treated layered limestone markedly decreased as the bedding angle (α) increased from 0 to 45°, then rose as the bedding angle increased from 45 to 90° (Fig. 8). In particular, the UCS and Young’s modulus exhibited minimum values for α = 45°, because under axial compression, the shear stress of the 45°-bedding-angle decomposition was large, and its deformation and failure resistance were low. In addition, relative to the untreated limestone, the UCS of acid-treated samples was reduced by 29%, 29%, 20%, 9%, and 41%, and Young’s modulus by 22%, 33%, 2%, 4%, and 25%, for α values of 0°, 30°, 45°, 60°, and 90°, respectively. The UCS was markedly lower for α = 90° in acid-treated specimens, indicating that the strength of the bedding plane and the matrix of limestone specimens were weakened by acid solution, the bedding plane was easier to tension, and the matrix was damaged.

Fig. 8
figure 8

UCS and Young’s modulus of versus bedding orientation with untreated and dissolved limestone specimens. (a) Uniaxial compressive strength. (b) Young’s modulus

Full size image

The failure characteristics of limestone in uniaxial compression tests are illustrated in Fig. 9. The observed failure patterns for specimens with different bedding angles were categorized into three different mechanisms: splitting, sliding, and mixed (i.e., a combination of both splitting and sliding; Wasantha et al. 2014). Specimens with shallow bedding dip angles (0°, 30°, and 45°) showed “splitting” failure, which occurred suddenly and quite violently through the material. In contrast, acid-treated specimens with bedding angles of 30° and 45° showed “mixed” failure. In this failure style, the local cracks in the acid-treated limestone specimen exhibited both vertical splits and damage along the bedding surfaces, resulting in formation of a zigzag crack during crack propagation through the bedding (Fig. 10) (Zuo et al. 2020). “Sliding” along a weaker bedding layer is another major failure mechanism, and was observed for specimens with bedding angles of 60°. In this style, failure occurred along a single weaker plane. In contrast, there were several large shear cracks and few tensile cracks in the acid-treated specimens. Finally, “splitting” along a weaker bedding layer was observed for specimens with bedding angles of 90°.

Fig. 9
figure 9

Failure characteristics of untreated and dissolved limestone specimens under different bedding dip angles. (a–j) Specimens UN-0-1, UN-30-1, UN-45-1, UN-60-1, UN-90-1, DS-0-1, DS-30-1, DS-45-1, DS-60-1, DS-90-1, respectively. T and S represents tensile and shear, and the red and blue lines represent crack paths and bedding planes, respectively

Full size image
Fig. 10
figure 10

Fracture diagram of bedding limestone. Red line represents fractures, and blue line represents bedding plane, respectively

Full size image

AE responses during the loading process

To facilitate the analysis, we defined the strain value at the peak stress of a limestone specimen as the peak strain (εp). When α = 0°, the stress–strain curve and AE counts of limestone specimens were very similar, with maximum instantaneous AE counts appearing at strain values of approximately 0.6εp (Figs. 7a and 11a). However, after the maximum instantaneous AE count, the AE events of specimen UN-0-1 exhibited slow growth and a plateau prior to the accelerated crack growth stage. In contrast, AE events occurred after specimen DS-0-1 experienced the maximum instantaneous AE counts, because the strength of limestone was reduced after acid treatment, and cracks formed more easily. Specimens UN-30-1, UN-45-1, and UN-90-1 exhibited maximum instantaneous AE counts at strain values of 0.99εp, 1.0εp, and 0.84εp, respectively, whereas specimens DS-30-1, DS-45-1, and DS-90-1 displayed maximum instantaneous AE counts at strain values of 0.59εp, 0.7εp, and 0.56εp, respectively. Similarly, after maximum instantaneous AE counts, AE events occurred. However, the periods of AE activity for DS-30-1, DS-45-1, and DS-90-1 were relatively longer than for untreated specimens. When α = 60°, the maximum instantaneous AE counts of UN-60-1 and DS-60-1 appeared at the peak strain (εp), because shear slip failure of limestone specimens along a single bedding plane results in the maximum instantaneous AE counts.

Fig. 11
figure 11

AE signal characteristics of limestone with different bedding. (a) Variation of instantaneous AE counts and strain against bedding orientation for limestone specimens. (b) Variation of accumulated AE counts and accumulated AE energy for limestone specimens

Full size image

Although the AE characteristics of limestone with different bedding angles exhibited similar trends, there were some differences between untreated and acid-treated specimens. From the cumulative AE energy, we infer that the obvious regular was related to the failure mode. With increasing bedding dip angle, the cumulative AE counts increased. Specimen DS-45-1 exhibited maximum AE counts, because internal failure of limestone specimens occurred along bedding and cracks propagated along the limestone matrix (Fig. 11b). In addition, the cumulative AE count for DS-45-1 was greater than that of UN-45-1, indicating that the bedding plane and matrix strength of limestone were reduced after acid treatment, and new cracks were more likely to occur in the specimen after loading. This result was confirmed in specimens DS-0-1 and DS-90-1. In contrast, the cumulative AE counts of DS-60-1 were noticeably lower than other specimens. The failure surface of DS-60-1 was relatively simple: the crack mainly propagated along a single weak surface, so the number of AE events was lower.

The variations in instantaneous AE counts and strain for different bedding orientations were studied to understand the AE events for different dip angles. Plots of instantaneous AE counts and strain against bedding dip angle for limestone specimens before and after acid treatment are shown in Fig. 11a. The instantaneous AE counts and strain value were clearly higher for untreated limestone than for acid-treated limestone, demonstrating that high instantaneous AE counts do not readily occur in untreated specimens under loading, and that micro-cracks form more readily in acid-treated limestone.

b-value characteristics of AE

Larger b-values indicate a higher proportion of small-scale cracks, which are mainly micro-cracks (Lockner et al. 1991; Zeng et al. 1995). Constant b-values imply no change in the size-scale crack distribution, and the proportions of micro-cracks on different scales are relatively constant. Reduced b-values indicate a proportional increase in large-scale cracks. If b-values are confined to a small range, the number of AE events is stable, i.e., the cracks in the rock are undergoing gradual, consistent expansion. A substantial fall in b-value means that crack development has changed sharply and the rock has failed.

Application of b-values is not limited to the field of seismology. AE events in rock failure can be regarded as seismicity (microseismicity), and the change in b-values of AE in the process of rock deformation and failure under different conditions can be calculated using Eq. (1):

$$\lg N = a - b\lg M$$
(1)

where M is the magnitude of the AE event during loading, N is the total number of AE events greater than (and including) amplitude M during loading, a is the statistical constant related to acoustic emission, and b is the b-value.

The b-values are equal to the absolute value of the slope of the regression line between magnitude M and lgN within a certain range on the right side of the peak magnitude. In the calculation of b-values, the AE amplitude can be divided by 20 (ML) to replace the earthquake magnitude M (Rao et al. 2005).

$$M_{L} = \frac{A}{20}$$
(2)

where ML is the AE magnitude and A is the AE amplitude (dB).

The b-values of each specimen were obtained using Eqs. (1) and (2) and the data plotted in Fig. 7. When the stress–strain curve was in the compaction stage, the b-values were stable. During the elastic deformation stage, the b-values of UN-0-1, UN-30-1, UN-45-1, UN-60-1, and DS-45-1 began to increase; subsequently, in the stage of stable crack growth, the b-values rose. Although the cumulative AE energy and counts rose sharply with increased stress, the b-values decreased rapidly. This pattern demonstrates that more cracks were formed in the limestone and energy was released rapidly, and so AE signals were generated. Finally, with the release of fracture energy, the specimen lost its bearing capacity. Every macro-cracking event led to a rapid decrease in b-values, indicating that b-values can be applied to predict large-scale events.

The b-values of UN-0-1 and DS-0-1 were higher than those of specimens with other dip angles; in addition, the b-values fluctuated greatly when the crack propagation was accelerated. Furthermore, there was a similar b-value trend for specimens UN-30-1 and UN-45-1, for which strain values close to εc yielded the minimum b-values. However, the b-values of DS-30-1 fluctuated before failure and decreased rapidly when macro-cracks occurred. In contrast, the b-values of DS-45-1 rose when approaching the failure strength, indicative of the formation of the next macroscopic crack. In addition, the strain values of UN-60-1 and DS-60-1 were close to 0.65εp at lowest b-value, much lower than for other specimens. At the minimum b-value, the stress values were 0.35σp (UN-60-1) and 0.46σp (DS-60-1); after this point, b-values continued to rise until failure of the specimens occurred. When α = 90°, the AE signal in the rock was weak in the initial stage, and the b-values increased until the peak strength was reached. In contrast, the b-values of UN-90-1 were smallest at the peak strength, and the b-values of DS-90-1 were smallest at a strain value of 0.66εp, characteristics similar to those of DS-60-1.

The b-values of limestone with different bedding dip angles were regularity under the uniaxial test conditions of acid-treated and untreated (Fig. 12). The trend is mainly manifested in the decrease of b-values first and then increase, the minimum b-values occurred when the specimen forms large-scale cracks, the acid-treated specimens were generally smaller than that of the untreated limestone specimens. When the bedding dip angles of 45° and 60°, the uniaxial compression of the acid-treated specimen has a small b-value. Combined with the rock deformation and failure modes, the strength of the rock matrix and the bedding plane decreases under acid-treated, and the failure modes of the 45° and 60° bedding dip angles are controlled by the bedding plane, and crack propagation through the bedding. In addition, compared with the untreated specimen at different dip angles, the b-values of acid-treated specimens are concentrated in the lower strain value. The main reason was that the acid-treated limestone specimens were prone to deformation and failure under small strain, so b-values were mainly concentrated in the smaller strain value.

Fig. 12
figure 12

The b-values of untreated and dissolved specimens varied with strain. Black and red lines were the untreated and dissolved limestone specimens, respectively

Full size image

The b-values of limestone specimens with different bedding dip angles were similar, the b-values of the acid-treated specimens were concentrated in the lower strain values, and b-values were low in the compaction and elastic deformation stages (Fig. 12). During elastic deformation and accelerated crack growth, the b-values of untreated limestone specimens decreased overall. After reaching the failure stage, the b-values rose.

Mechanical degradation mechanism of limestone

The bedding planes within the limestone were mainly argillaceous limestone containing a small amount of clay minerals (Fig. 3a). Treatment with HCl and H2O alters the surface, internal structure, mineral particles, and pores of limestone specimens. Macroscopic changes include loss of limestone mass and unevenness of the surface of rock specimen.

Clay minerals in rock swell and soften after exposure to water, which deteriorates the mechanical properties of the rock (Ruedrich et al. 2011; Berthonneau et al. 2016; Cherblanc et al. 2016). When limestone specimens were put into hydrochloric acid solution, the calcite began to react with HCl, and CO2 bubbles were visible on the surface of the specimen. The reaction solution became turbid and insoluble material accumulated at the bottom of the container. Yellowish material consisting of unreactive clay minerals was visible on bedding surfaces. After the specimens had been removed from acid solution, differences in corrosion between the bedding plane and the matrix could be clearly seen, and the bedding was slightly more affected than the matrix. Moreover, dissolution pores in the matrix were visible by the eye, and large pores were obvious in the bedding plane (Fig. 13a, b). After dissolution, the microstructure of limestone was obviously different. The limestone matrix was mainly dissolved along the calcite cleavage, and the dissolution pores are formed between the cleavages. The bedding plane formed a bulge and showed honeycomb dissolution holes, soluble substances were dissolved away, insoluble substances remained, forming a dissolution skeleton (Fig. 14a–c). According to the energy spectrum analysis, the main residual silicon and aluminum minerals in the bedding are speculated to be residual quartz and clay minerals, and the dissolved matter is mainly calcite minerals. The rock matrix was mainly calcium, and it is speculated to be calcite mineral composition (Fig. 14d).

Fig. 13
figure 13

Dissolution process of layered limestone. (a) Reaction mechanism of calcite, dolomite, and quartz with acid solution (Li et al. 2018, She et al. 2016). (b) Crystalline swelling results in the widening of interlayers of swellable clay minerals by hydration of cations (Madsen et al. 1988)

Full size image
Fig. 14
figure 14

Results of scanning electron microscopy and energy spectrum analysis after limestone dissolution. (a) Limestone bedding and matrix dissolution microstructure. (b) Matrix dissolution microstructure. (c) Bedding dissolution microstructure. (d) Energy spectrum analysis of interface between limestone bedding and matrix (the arrow indicates the starting direction of the energy spectrum)

Full size image

By considering the macroscopic differences in the limestone specimens before and after acid treatment, the mechanism responsible for the degradation of the mechanical strength of the specimens can be inferred. The structure of the limestone specimens was altered and the mechanical strength of the matrix and bedding plane decreased as a result of acid treatment (Fig. 14b, c). Differences in the composition of the matrix and the bedding planes mean that the mechanical strength of the bedding planes is easily affected by acid solution. This conclusion can be drawn from the failure mode of limestone specimens before and after acid treatment (Fig. 9). When α = 30° and 45°, crack propagation occurred along the matrix and the bedding plane, but this phenomenon was not obvious in untreated specimens.

Conclusions

Through the dissolution test, mechanical test, and AE study of bedding limestone, we drawled the following conclusions. The loading process was divided into five stages on the basis of the stress–strain curve for untreated and acid-treated limestone specimens: compaction; linear elastic deformation; stable crack propagation; accelerating crack propagation; and failure. Acid-treated specimens reached the accelerating crack growth stage at markedly lower stress values than untreated specimens. The uniaxial compressive strength and Young’s modulus of untreated and acid-treated limestone specimens exhibited obvious anisotropy. The mechanical properties decreased with increasing bedding dip angle from 0 to 45°, then rose as the bedding dip angle increased from 45 to 90°. Acid treatment obviously deteriorates the mechanical strength of bedded limestone. Mixed failure was more likely to occur in acid-treated specimens with bedding dip angles of 30 to 60°, this was mainly caused by the decrease of mechanical strength of limestone matrix after dissolution. AE activity was closely related to rock deformation and fracture. AE events and cumulative AE energy increased markedly as a result of closure, fracture, and sliding of cracks. Furthermore, the b-values varied with the bedding angle, and minimum b-values were obtained at lower strain values for the acid-treated limestone than for untreated specimens. The main reason was that after dissolution, limestone was more likely to rupture to form AE events, so as to form macroscopic rupture. Through the analysis of electron microscopy and energy spectrum, the mechanical deterioration of limestone was that the minerals in limestone react with HCl, resulting in increased internal porosity and softening of bedding planes.