Effects of Thermal and Stress Cycles on the Mechanical Properties and Acoustic Emission Characteristics of Rocks

An underground energy storage chamber is a subterranean facility designed for energy storage. Its design and construction must consider rock mechanics, geological conditions, and engineering techniques to ensure safety and long-term stability [1-3]. The stress conditions of the surrounding rock significantly impact the storage effectiveness of the chamber. The surrounding rock is influenced not only by static loads from initial stress and disturbances such as earthquakes and construction activities but also by alternating hot and cold environments. Consequently, the mechanical properties of the surrounding rock vary greatly [4]. Existing research predominantly focuses on the effects of cyclic loads and freeze–thaw cycles on the mechanical properties and failure modes of rocks, yielding various findings.

Cyclic loading significantly impacts the mechanical performance and damage mechanisms of rocks. After multiple cyclic loadings, the compressive strength, elastic modulus, and fatigue life of rocks change [5, 6]. Song et al. [7] found through experiments that stress patterns with higher loading rates and lower unloading rates correspond to lower fatigue strength and shorter fatigue life. Miao et al. [8] discovered that under triaxial stress, the angle between the main fracture surface of granite and the longitudinal axis increases with confining pressure, further revealing the evolution characteristics of plastic energy consumption and damage energy consumption. He et al. [9] used experimental-theoretical methods to study the fracture and energy evolution of rock specimens with circular holes under multistage cyclic loading. Xu et al. [10] compared the peak strength and AE characteristics during the failure process of sandstone discs with different prefabricated fractures, summarizing the impact of crack angles on the AE cumulative count and energy.

Temperature, an unavoidable environmental factor in the field, also significantly influences the mechanical and failure properties of rocks [11, 12]. Li et al. [13] found a negative correlation between the number of freeze–thaw cycles and both fatigue life and failure stress. Increased freeze–thaw damage and fatigue damage lead to increased ductility and decreased stiffness in specimens. Zhao et al. [14] subjected red sandstone to up to 300 freeze–thaw cycles, discovering that with more cycles, the porosity increased exponentially while strength, residual strength, elastic modulus, and Poisson’s ratio decreased linearly. Song et al. [7] noted that more freeze–thaw cycles resulted in more pronounced damage in sandstone, characterized by lower fatigue life and higher strain rates. Zhu et al. [15] analyzed the stress–strain, peak stress evolution, and macroscopic failure characteristics of sandstone under uniaxial cyclic loading at four constant loading rates. Zheng et al. [16] conducted graded cyclic loading–unloading tests on cement mortar, showing that under the same cyclic stress, specimens hardened with more cycles, and with increased cyclic stress, the dissipated energy density in each stage gradually surpassed the elastic energy density, becoming dominant and rapidly increasing near failure.

Some researchers have studied the combined effects of temperature and load on rocks. Qiu et al. [17] investigated the mechanical properties of coal gangue from macroperspective and microperspective, finding that with more cyclic loading and freeze–thaw cycles, the relative dynamic elastic modulus and compressive strength gradually decreased, while the pore structure deteriorated. Zhou et al. [18] used digital image correlation technology to study the effects of combined freeze–thaw cycles and cyclic loading on defective rocks, identifying the types of cracks formed during rock fracture in detail. Wang et al. [19] examined the fatigue damage evolution of rocks caused by coupled cyclic temperature and load-induced volume deformation, establishing a coupled damage model that effectively describes the accumulation of rock damage. Fei et al. [20] quantitatively studied the impact of combined cyclic mechanical loading and freeze–thaw cycles (i.e. stress history effects) on the microstructural evolution and mechanical degradation of three different porosity porous sandstones. Xiao et al. [21] found that with increasing temperature, the stress–strain curves of heat-treated rock samples exhibited obvious hysteresis loops, showing a three-stage evolution characteristic of “sparse→dense→sparse.”

AE is a nondestructive monitoring tool that provides real-time monitoring and early warning of internal damage in materials and structures. It is often used in structural crack monitoring, pipeline leakage monitoring, material fracture behavior research, and tunnel rock mass deformation studies [22-25]. In the field of rock mechanics, AE technology has also been widely applied [26]. During laboratory tests, AE technology can effectively monitor internal damage in rocks [27], and analyzing AE waveform parameters can quantify the proportion of tensile-shear cracks during rock failure [28]. Studying the time-frequency characteristics of AE can reveal the crack propagation process inside the rock, and some researchers have studied the spatiotemporal characteristics of AE time [29-31], reproducing the process from microcrack generation to macrocrack formation within the rock.

Previous research on the effects of cyclic load and temperature on the mechanical properties of rocks either focused solely on cyclic load or temperature, with few studies exploring the combined effects of load and temperature on rocks. The surrounding rock of underground chambers is often subjected to both temperature and load. Additionally, due to the influence of groundwater, rocks are often saturated with water. Laboratory studies on rocks under single environmental factors deviate from engineering reality, making the findings less applicable. Moreover, most research has concentrated on rock failure during cyclic loading–unloading, with few scholars considering cyclic loading–unloading as a preconditioning method. Furthermore, few studies have combined AE technology to monitor the failure characteristics of rocks subjected to cyclic temperature and load pretreatment. Therefore, this article selects basalt and sandstone (for comparison with basalt) to process test samples and applies different preconditioning methods, such as water saturation, cyclic load after water saturation, cyclic temperature pretreatment after water saturation, and combined cyclic load and temperature pretreatment after water saturation. A series of uniaxial compression tests are conducted using AE equipment to study the mechanical properties and failure mechanisms of the surrounding rock of chamber from macroperspective and microperspective.

This study focuses on the Zhangjiakou compressed air energy storage rock cavern as the engineering background, selecting its typical surrounding rock, basalt, as the research subject. Sandstone, as a very typical sedimentary rock, exists in many underground chambers as a surrounding rock, many scholars have also chosen sandstone to study the stability of the chamber, and its mechanical properties are relatively stable while relatively clear. Two types of cylindrical specimens, φ50 × 100 mm and φ50 × 50 mm, were prepared, as shown in Figure 1 In addition, all the rock specimens used were carefully polished to ensure that the degree of nonparallelism and nonperpendicularity was controlled less than 0.02 mm.

Considering that underground rock chambers are often in humid environments and that the water content of rocks significantly affects their properties, all rock specimens were soaked in water for 48 hours before the experiments to fully simulate the underground environment. The rock specimens were divided into four groups: the first group consisted of saturated rock samples (SR), the second group included rock samples subjected to cyclic load pretreatment (SR-CLT), the third group comprised rock samples subjected to cyclic temperature pretreatment (SR-CTT), and the fourth group contained rock samples subjected to both cyclic load and cyclic temperature pretreatments (SR-CLT-CTT). The processes of the three pretreatment methods are shown in Figure 2.

Based on the monitored variations in cyclic stress and cyclic temperature experienced by the surrounding rock in underground chambers, the cyclic load pretreatment protocol was designed as follows: the SR samples underwent a total of 40,000 cyclic stress cycles using the MTS Landmark high-frequency fatigue testing system, which has a maximum load capacity of ±100 kN and high-precision load measurement accuracy of ±0.5%, with a dynamic frequency range of 0–80 Hz, as shown in Figure 2(a). The minimum load in each cycle was set to 4 MPa, the maximum load to 10 MPa, and the loading frequency to 30 Hz.

The cyclic temperature pretreatment was conducted using a TMS9012-260 high–low temperature test chamber, capable of controlling temperatures from −40°C to 210°C with minimal fluctuations. Each SR sample underwent 100 cyclic temperature pretreatments with temperature cycling between 0.1°C and 0.2°C. During the first hour, the temperature rapidly increased from 10°C to 50°C; in the following hour, it gradually decreased from 50°C to 30°C; subsequently, the temperature dropped from 30°C to 0°C within an hour; and in the final hour, the temperature rose again from 0°C to 10°C. A complete temperature cycle is illustrated in Figure 2.

The Brazilian splitting tests were conducted using the INSTRON 1346 electro-hydraulic servo testing system, which has a maximum load capacity of 250 kN and a load accuracy of ±0.5%. The loading mode was displacement control, with a control rate of 0.21 mm/min, as shown in Figure 3(a). The uniaxial compression tests and variable angle shear tests were also performed on the INSTRON 1346 system, with a maximum vertical load capacity of 2000 kN and the same load accuracy of ±0.5%. Both tests were conducted in displacement control mode but with different loading rates: 0.18 mm/min for the uniaxial compression tests and 0.3 mm/min for the variable angle shear tests, as shown in Figures 3(b) and 3(c).

Additionally, for all rock specimen tests, a PCI-2 acoustic emission (AE) system was employed to monitor the failure processes. The AE system’s threshold was set to 45 dB, the sampling rate to 10 MSPS, and the sampling length to 2 k. The peak definition time, hit definition time, and hit lockout time were set to 50 μs, 200 μs, and 300 μs, respectively. The AE sensors used were Nano 30 models with a response frequency of 140 kHz, as shown in Figure 3(d), and the preamplifier’s main amplification frequency was set to 40 dB, as shown in Figure 3(e).

The uniaxial compressive strength (σ_ucs), indirect tensile strength (σ_its), and shear strength parameters (cohesion, c, and internal friction angle, ϕ) of basalt and white sandstone obtained from UCT, BITT, and VAST are presented in Table 1 and Figure 4.

The mechanical properties of basalt and white sandstone, including σ_ucs, σ_its, c, and ϕ, exhibit significant variations under different processing methods (SR, SR-CLT, SR-CTT, and SR-CTT-CLT). For the SR method, basalt and white sandstone show σ_ucs of 82.62 MPa and 21.39 MPa, σ_its of 1.23 MPa and 0.64 MPa, c of 7.80 MPa and 2.66 MPa, and ϕ of 43.96° and 44.64°, respectively. Under SR-CLT, σ_ucs for basalt decreases to 76.69 MPa while σ_its increases to 2.56 MPa, with c increasing to 8.69 MPa and ϕ decreasing to 40.12°; for white sandstone, σ_ucs and σ_its increase to 22.87 MPa and 0.98 MPa, with c rising to 2.92 MPa and ϕ slightly decreasing to 44.43°. The SR-CTT method results in increased σ_ucs of 89.97 MPa and σ_its of 3.44 MPa for basalt, with c significantly rising to 15.42 MPa and ϕ to 44.87°; white sandstone shows increases in σ_ucs and σ_its to 25.96 MPa and 0.79 MPa, with c at 2.91 MPa and ϕ at 44.78°. The combined SR-CTT-CLT pretreatment yields the highest σ_ucs and σ_its for both lithologies, with basalt at 93.27 MPa and 3.83 MPa, but a lower c of 4.59 MPa and ϕ of 38.97°; white sandstone shows σ_ucs of 27.93 MPa and σ_its of 1.30 MPa, with c increasing to 3.21 MPa and ϕ slightly decreasing to 44.35°. Comparing these pretreatments to the baseline SR method, SR-CLT and SR-CTT show distinct increases in tensile strength and cohesion, indicating enhanced rock stability under cyclic loading and thermal conditions, while SR-CTT-CLT provides the highest compressive and tensile strengths but with variable effects on cohesion and friction angle. This analysis highlights the complex interplay of stress and temperature cycles on rock mechanics, providing valuable insights for engineering applications.

The failure modes of saturated basalt and sandstone under uniaxial compressive stress, both before and after cyclic loading, are illustrated in Figure 5. In the uniaxial compression tests, different pretreatment methods lead to distinct failure modes of the rocks. The primary failure mode of saturated basalt is the formation of elongated slab-shaped blocks on the side, with all block lengths smaller than that of rock specimen. Additionally, a small amount of fragmented blocks is generated at the upper and lower ends of the rock due to end effects. The failure of saturated basalt can be defined as inclined shear. Moreover, the main failure mode of saturated sandstone is the formation of conical shapes at the upper and lower ends, with flaking occurring on the side of the rock, resulting in large blocky fragments. The lengths of the main blocks are close to the length of the rock, indicating pronounced X-shaped conjugate shear failure in sandstone, suggesting the occurrence of fractures due to shear stress. Comparing the failure modes of these two types of rocks reveals that rocks with lower strength are more susceptible to end effects.

After undergoing cyclic loading, damage accumulates within the rocks, leading to a significant change in the failure mode. Basalt transitions from shear failure to predominantly X-shaped conjugate shear failure, with the length of elongated slab-shaped blocks formed on the side equaling the length of the rock specimen. The degree of rock failure is more thorough, and the residual bearing capacity after reaching peak strength is lower. Sandstone, on the other hand, transitions from X-shaped conjugate shear failure to inclined shear and splitting, with more pronounced conical shapes at the upper and lower ends and plate-like blocks penetrating through the rock on the side, with all block lengths equal to the specimen’s length.

The comparison of the failure modes of saturated basalt and sandstone with and without temperature cycle pretreatment under uniaxial compressive stress is illustrated in Figure 6. After undergoing cyclic temperature pretreatment, the main failure mode of saturated basalt is the formation of significantly pronounced conical shapes at the ends and with shearing along the edges of the cones into the middle of the rock. Additionally, at one-quarter of the length of rock specimen, it splits vertically along the direction of the perpendicular end face to the lower end plane, ultimately resulting in only one cone and elongated slab-shaped block covering 75% of the rock length. Like sandstone after cyclic loading pretreatment, the primary failure mode of sandstone after cyclic temperature pretreatment is also the production of conical failure, with two cones formed at the upper and lower ends. The only difference lies in the higher degree of columnar splitting on the side and a more perpendicular angle to the end plane.

The failure modes of saturated basalt and sandstone with and without coupled loading and temperature cycle pretreatment under uniaxial compressive stress are depicted in Figure 7. After undergoing coupled cyclic loading and temperature pretreatment, the main failure mode of saturated basalt under uniaxial compression is the formation of various fragments with multiple fracture directions. Cracks are distributed in a mesh pattern on the side of the rock. However, the fracture that causes the loss of load-bearing capacity at its side bottom originates from the near-edge position of the upper end face of the rock, extending vertically to the middle of the rock and then downward, eventually penetrating through the entire rock.

Sandstone subjected to temperature and loading cycle coupling pretreatment exhibits flaking in the form of sheet-like and slab-like fragments under uniaxial compression, with the fragment lengths equal to the specimen length. Unlike sandstone subjected only to temperature or loading cycle pretreatment, the sandstone appears to have undergone surface stripping, leaving behind a more cylindrical central portion. Fractures in the central portion indicate the final failure of the rock.

In the above content, the effects of CLT, CTT and combined temperature and load on the failure mode of the surrounding rock of underground chamber were studied, respectively. However, due to the complex conditions of underground environment, there are two other states in the surrounding rock of the chamber, one is to experience the CLT and then suffer the CTT action, and the other is to experience the CTT action and then suffer the CLT action. In these two different conditions, what is the difference between the failure modes of rocks? It is of great reference significance for us to investigate and analyze the stability analysis and disaster prevention of rocks at different periods.

By comparing the failure modes of Figures 6(b) and 7(b), it can be found that both the basalts treated by CTT and coupled treatment are damaged in the form of spalling on the side of the rock sample, and both are in strip shape. However, the strip fragments of the former are longer and their length is close to the height of the rock sample, while the size of the fragments of the latter is only half of the length of the rock sample. This shows that CLT can reduce the hardness of rock. As can be seen from Figures 6(d) and 7(d), the failure mode of sandstone has no such change, mainly because the cohesion of sandstone is lower.

By comparing the failure modes of Figures 5(b) and 7(b), it can be found that the complete failure of basalt after CLT is significantly different from that after combined action. The former is mainly shear failure through the entire rock mass, and the spalling of rock blocks on the side of rock is not obvious. This is mainly since the additional CTT reduces the cementation strength of the basalt matrix, which makes it easy to peel off the rock body under the Poisson effect. The failure mode of sandstone under these two treatment methods is consistent with that of basalt, which also proves that sandstone as a stratified rock is much more affected by temperature than load.

The development of microcracks and the overall damage of rock with different pretreatments can also be studied by AE system. According to the previous studies [26, 27], the variation trend of AE counts is closely related to the development of microcracks inside rocks, the AE energy can reflect the intensity of microfailure inside rocks, and the scatter map of RA-AF can help to qualitatively determine the type of crack that causes rock failure.

The variations in AE counts and cumulative AE counts over loading time for basalt and sandstone subjected to SR, SR-CLT, SR-CTT, and SR-CTT-CLT are illustrated in Figure 8. The results indicate that the AE counts for basalt are significantly lower than those for sandstone. Both basalt and sandstone with SR-CTT-CLT exhibit higher AE counts compared to other pretreatment conditions. The cumulative AE count curve for basalt shows a stepwise increase, with the majority of AE counts appearing in the later stages for untreated samples, while those treated with SR-CTT exhibit substantial counts during the midloading stage. This suggests that temperature promotes the propagation of microcracks within basalt. Conversely, the cumulative AE count curve for sandstone is smoother. For sandstone with SR-CTT, the AE counts predominantly appear in the later loading stages, whereas untreated sandstone exhibits AE counts throughout the entire loading phase. This indicates that SR-CTT inhibits the propagation of microcracks within sandstone under uniaxial loading.

The variations in AE energy and cumulative AE energy over loading time for basalt and sandstone subjected to SR, SR-CLT, SR-CTT, and SR-CTT-CLT are illustrated in Figure 9. For basalt, AE energy is predominantly released at the moment of failure, with very low energy emissions during the early and midloading stages. In contrast, sandstone with SR and sandstone with SR-CLT exhibit AE energy emissions throughout the entire loading phase. Sandstone subjected to other prepretreatment methods shows increased AE energy emissions in the mid- to late-loading stages. Basalt with SR-CLT and sandstone with SR-CTT generate the highest AE energy under uniaxial loading compared to other pretreatment methods. This indicates that loading pretreatment and temperature pretreatment, respectively, enable basalt and sandstone to accumulate more internal energy.

The scatter distribution of AE RA-AF for basalt and sandstone under uniaxial compression is illustrated in Figure 10. The data volume for basalt is significantly smaller than that for sandstone. For basalt with SR-CTT, the AF values primarily range from 0 to 150 kHz, while for other basalt samples, the AF values are mainly within 0–100 kHz. The RA values for all basalt samples mainly range from 0 to 400 ms/V. In contrast, for sandstone with SR and SR-CTT, the AF values are primarily distributed within 0–100 kHz, whereas for sandstone subjected to other prepretreatment methods, the AF values mainly range from 0 to 150 kHz. The RA values for sandstone under all prepretreatment methods primarily range from 0 to 1400 ms/V. A scatter distribution along the RA axis indicates predominant shear failure, while distribution along the AF axis indicates tensile failure. Therefore, under uniaxial loading, both basalt and sandstone primarily exhibit shear failure, with sandstone exhibiting a higher degree of shear compared to basalt. Basalt and sandstone with SR-CTT exhibit a higher degree of tensile failure compared to other pretreatment methods. This indicates that SR-CTT promotes splitting failure in rocks under uniaxial loading.

The natural logarithmic scatter plot and fitting line of AE peak (lnA) and AE count (lnN) for basalt and sandstone under uniaxial compression are illustrated in Figure 11. The results show that basalt with SR-CTT exhibits poor linear correlation between lnN and lnA, with a fitting coefficient of only 0.64. In contrast, basalt samples subjected to other prepretreatment methods demonstrate a better linear correlation between lnN and lnA. For sandstone with SR-CTT-CLT, the linear correlation between lnN and lnA is significantly higher than that for sandstone subjected to the other two pretreatment methods. The scale of crack propagation within the rock influences the acquisition of AE signals, thereby affecting the correlation between lnN and lnA. A larger scale of propagation results in poorer correlation. Therefore, it can be inferred that basalt with SR and sandstone with SR-CTT, as well as sandstone with SR-CLT, experience more intense large-scale microcrack propagation during uniaxial loading. Basalt and sandstone subjected to other prepretreatment methods exhibit similar scales of microcrack propagation. For the phenomenon of low lnA-lnN fitting coefficient of sandstone in Figures 11(b) and 11(d), the reason is that sandstone, as a sedimentary rock, has high water absorption and becomes looser after the action of water inside. So, there is a lot of noise caused by the interaction of water in the process of applying loads to the samples.

When the rock sample is under the action of load, various types of energy will be generated, such as elastic energy E_e, loss energy E_d, kinetic energy E_k, and friction energy Δ_E, and so forth [32, 33]. The relationship between the total input energy E_t and these energies is as follows:

The uniaxial compression experiment in this article is a static experiment, and friction reduction measures are used, so compared with elastic energy and loss energy, kinetic energy and friction energy can be ignored. Therefore, formula (1) can be changed into formula (2):

where

Thus:

Figure 12 illustrates the evolution of total input energy, elastic energy, and dissipated energy with strain for basalt and sandstone under uniaxial compression. Due to the calculation method, the elastic energy curve resembles the stress–strain curve, with a small accumulation of elastic energy during the initial crack compaction stage, steady energy storage during the elastic deformation stage, rapid accumulation during the unstable development stage, and significant energy release upon complete rock failure. A sharp increase in dissipated energy indicates localized failure within the rock. The total energy in basalt is greater than in sandstone, and basalt consistently shows higher elastic energy than dissipated energy. In contrast, for sandstone, dissipated energy exceeds elastic energy before 0.4% strain for untreated samples, before 0.25% strain for cyclic temperature-treated samples, and before 0.2% strain for cyclic temperature and load-treated samples. For basalt, dissipated energy begins to rapidly increase at 80f of peak strength, while for sandstone, this sharp increase generally occurs at peak strength. The strain at which dissipated energy begins to rapidly increase for basalt progressively grows from SR to SR-CLT, to SR-CTT, and finally to SR-CLT-CTT.

The total energy (E_pt), elastic energy (E_pd), and dissipated energy (E_pe) at peak strength for rock samples are critical parameters. Figure 13 illustrates E_pt, E_pd, and E_pe for basalt and sandstone under various prepretreatment conditions. The results indicate that cyclic loading pretreatment reduces E_pt, E_pd, and E_pe in basalt, as well as E_pt and E_pe in sandstone. Conversely, cyclic temperature pretreatment increases E_pt, E_pd, and E_pe in both basalt and sandstone. Combined pretreatment increases the peak energies in basalt but decreases them in sandstone, suggesting that temperature plays a decisive role in the response of basalt, whereas load plays a decisive role in the response of sandstone during combined pretreatment. From an energy perspective, basalt is more significantly affected by cyclic temperature and cyclic loading than sandstone.

During the compression process, rocks typically store energy primarily in the form of elastic energy. Therefore, the ratio of elastic energy (E_e) to dissipated energy (E_d) at peak strength can be used as the rock damage intensity factor λ_max, which provides a direct and quantitative measure of the rock’s damage severity [21]. Figure 14 illustrates the variation characteristics of λ_max for basalt and sandstone under different prepretreatment methods. The results indicate that basalt experiences the highest damage under cyclic loading, whereas sandstone experiences the highest damage under cyclic temperature pretreatment. Both basalt and sandstone exhibit the least damage under combined pretreatment. This analysis suggests that cyclic loading has a more significant impact on basalt, and cyclic temperature has a more significant impact on sandstone. Conversely, the combined pretreatment method has the most substantial overall impact. Except for sandstone under cyclic loading pretreatment, the damage severity of sandstone is generally higher than that of basalt under other prepretreatment methods.

This study selected basalt and sandstone (as the reference) and conducted a series of indoor experiments combined with an AE

monitoring system on rock sample with different pretreatments, including water saturation, water saturation followed by cyclic temperature pretreatment, water saturation followed by cyclic loading pretreatment, and water saturation followed by combined cyclic loading and temperature pretreatment. The main conclusions are as follows:

1. Different processing methods have significant impacts on the mechanical properties of basalt and white sandstone, including σ_ucs, σ_its, c, and φ. Among these methods, the SR-CTT and SR-CTT-CLT methods are particularly effective in enhancing compressive and tensile strengths, although their effects on c and φ vary.

2. Basalt with SR shows splitting failure, while sandstone with SR displays X-shaped conjugate shear failure. SR-CLT makes the failure mode of basalt transform into inclined shear failure and sandstone into conical failure. SR-CTT causes basalt to form pronounced conical shapes and sandstone to exhibit increased columnar splitting. With combined loading and temperature pretreatments, basalt fractures into various fragments with mesh-pattern cracks, whereas sandstone undergoes surface stripping, leaving a cylindrical central portion.

3. Basalt shows significantly lower AE counts than sandstone, with both rocks exhibiting the highest AE counts under SR-CTT-CLT. SR-CTT leads to the inhibition of microcrack propagation. Basalt with SR-CLT and sandstone with SR-CTT have the highest AE energy, indicating these pretreatments enable greater internal energy accumulation. RA-AF scatter distribution reveals both rocks mainly exhibit shear failure under uniaxial loading, with sandstone showing a higher degree of shear. SR-CTT could increase tensile failure in both rocks, promoting splitting failure.

4. SR-CTT and SR-CLT influence the energy dynamics, with basalt being more affected by temperature and sandstone by load. At peak strength, SR-CLT makes these energy parameters decrease for both rocks, while SR-CTT increases them. Combined pretreatments enhance these three energy parameters in basalt but reduce them in sandstone. Based on λ_max, SR-CLT causes the most damage to basalt, while SR-CLT affects sandstone the most. Combined pretreatments result in the least damage for both rocks, though sandstone generally experiences higher damage than basalt under most conditions.

Data of this work will be made available on request.

The authors declare that they have no conflicts of interest.

This work was supported by the National Natural Science Foundation of China (grant no. 52374150) and the Science and Technology Innovation Program of Hunan Province (grant no. 2021RC3007).

We thank all those who contributed to this work, especially the reviewers and editorial office who helped to greatly enhanced the quality of this submissions, and the publisher staff in organizing and editing the manuscript.