1 Introduction

With the continuous increase in population growth and the fast development of the economy, fossil fuels such as coal and oil, with the resulting release of acid rain and greenhouse gases, are becoming increasingly depleted. Clean and efficient energy is the direction of future energy development, for which deep geothermal energy offers substantial advantages in terms of cost, reliability and environmental friendliness [26]. Thus, deep geothermal energy exploitation is now recommended and identified as a renewable and alternative energy source [20]. Aiming to extract deep geothermal energy, the enhanced geothermal system (EGS), which recovers heat from a sufficiently hot thermal reservoir at shallow depth by creating an artificial circulation system, was designed and implemented by Los Alamos National Laboratory in the 1970s [32]. EGS uses hydraulic fracturing and simulation to create highly conductive zones that interconnect an injection well to a production well to form a doublet system [29]. Thermal energy is extracted by circulating water, or another suitable fluid, into the hot fractured rock and pumped to a power plant (binary or flash plant) on the surface to generate electricity [3]. During the exploration of geothermal energy, cold water may be injected and circulated through the fractures in the geothermal reservoirs, which may change the properties of geothermal reservoir rocks (rocks at elevated temperatures) and further influence the stability and safety of the geothermal wellbore wall. Thus, studying the physical and mechanical properties of cyclic heat-treated granite via water cooling is of substantial importance.

Numerous laboratory studies on various types of rocks, such as granite [6, 7, 36, 46, 54], marble [27, 28, 33], sandstone [24, 25, 38, 47, 48], limestone [14, 27, 49, 51] and claystone [39], that have been exposed to high temperatures have been conducted to investigate the effects of high temperature on the physical and mechanical properties of rocks and their the microstructures. In the terms of rock physical properties, the volume, porosity, permeability and electrical conductivity of rocks increase with the thermal temperature, while the density, wave velocity and thermal conductivity typically decrease. The mechanical properties mainly focus on strength and deformation characteristics, such as compressive stress, tensile stress, elastic modulus and post-peak characteristics. The tensile strength, compressive strength and elastic modulus gradually decrease with the temperature and exhibit various levels of deterioration. The rocks after high-temperature treatments are less brittle and exhibit much stronger ductile characteristics compared to those without treatment. Scanning electron microscopy (SEM) [4, 7, 47], optical microscopy [28, 34] and X-ray microcomputed tomography (CT) [11, 22, 53] analyses were conducted to identify the change mechanisms of rock physical and mechanical properties after exposure to high temperature.

Deep granite reservoirs have adequate temperatures for serving as geothermal reservoirs, according to exploratory geothermal well tests [12]. Although the influences of high temperature on granite physical and mechanical properties have been investigated systematically as discussed above, the granites in most of the previous studies were cooled naturally to room temperature inside or outside of the furnace; hence, those studies are not applicable to geothermal energy extraction/EGS. Meanwhile, only a few experiments have captured the effect of the cyclic natural cooling treatment on granite physical and mechanical properties [22, 34] and limited research has been conducted on heat-treated granite via water cooling [18, 21, 36]. Recently, Xu and Sun [43] have investigated effects of quenching cycle on tensile strength of granite with water cooling and found that the static tensile strength decreases with increasing temperature and increasing number of quenching cycles. Ge and Sun [13] and Zhu et al. [56] studied the damage of granite after cyclic heating and cooling with circulating water by acoustic emission (AE) and found that the AE curve agrees well with the stress–time curve at each test temperature point and can reflect the mechanical damage increasing with the increase in water cooling at high-temperature cycles. However, to the authors’ knowledge, the microscopic change mechanisms of granite mechanical properties under cyclic heating and water cooling have not been published. In the presented study, granite material is used to study the physical (weight, volume, density and P-wave velocity) and mechanical properties (strength, deformation and failure behaviours) of rock after heating and water-cooling treatment for 1 and 30 cycles. According to a comprehensive review of international literature, the effects of cyclic water cooling on mechanical properties of granite are analysed, and the relations between the physical and mechanical properties of water-cooled granite are also discussed. Then, the microscopic mechanism of heating and water cooling on the physical and mechanical behaviours of the tested granite was revealed based on SEM analysis. Therefore, the objectives of this paper are to investigate the effects of the cyclic heating and water-cooling processes on both the physical and mechanical properties of granite and to identify the microscopic change mechanisms based on SEM observations. This study is expected to support analytical calculations and numerical simulations in geothermal energy extraction.

2 Experimental design

2.1 Description of rock specimens

Gray and fine-grained granite blocks were collected from a mine in Suizhou city, Hubei province, China. According to the method that is suggested by the International Society for Rock Mechanics (ISRM) [10], cylindrical specimens with a diameter of 50 mm and a height of 100 mm were machined from the same block (Fig. 1). The natural density of the specimens is 2.603 g/cm3, and the average longitudinal wave velocity is 3770 m/s. Petrophysical analysis with X-ray diffraction demonstrates that the main mineral components of the granite are 77.68% feldspar, 10.58% quartz, 6.28% biotite, 3.24% chlorite and 2.22% amphibole.

Fig. 1
figure 1

Untreated granite specimens

Full size image

2.2 Experimental procedure

The procedures of the present experiment are illustrated in Fig. 2. All the specimens were initially subjected to dry processing, which is performed by placing the specimens in a drying oven that is maintained at 105 °C for 24 h to remove all moisture content to eliminate the effect of the natural water content on the experimental results. The quality, size and longitudinal wave velocity of each specimen were measured with a Vernier calliper, an electronic balance and an RSM-SY5 (T) ultrasonic concrete tester, respectively. Any specimen with an abnormal density or a wave velocity was abandoned to ensure the reliability of the test results. The remaining specimens were divided into nine groups, with 4 specimens in each group (three for mechanical testing and one for making thin slices for SEM observation), including a control group, which was not subjected to thermal treatment. The specimens were heated in the SG-XL1200 high-temperature box furnace to the predetermined testing temperatures (150, 300, 400 and 500 °C) using a modest heating rate of 5 °C/min, as in the previous experiments on rocks [4, 27, 38, 46, 49]. Then, specimens were maintained at the predetermined temperatures for 2 h to ensure that all the specimens had been adequately heated. Once the constant temperature stage was over, the tested specimens were cooled via water immersion in a water bath with a sufficient volume of distilled water. After the specimens had cooled to room temperature, they were subjected to dry processing again, which is regarded as one thermal cycle. Then, the diameter, height, mass and wave velocity of each specimen after the thermal cycle were measured once again. To investigate the effects of thermal cycles on granite physical and mechanical properties, half of the specimens were subjected to 30 thermal cycles and the quality, size and longitudinal wave velocity of each specimen were measured after each thermal cycle.

Fig. 2
figure 2

Test flow chart

Full size image

The microstructural characteristics of the specimens after heating and water-cooling treatment were observed using a Quanta250 SEM, and uniaxial compression tests were conducted using a TAW-2000 electro-hydraulic servo-controlled rock mechanics testing system in displacement-controlled conditions at a rate of 0.3 mm/min. A separate group of natural granite specimens without heat treatment were also tested for reference. The test results of the granite physical and mechanical properties before and after high-temperature treatment are listed in Table 1.

Table 1 Physical and mechanical parameters of water-cooling-treated granite after exposure to various temperatures
Full size table

3 Experimental results

3.1 Volume, mass and bulk density

To more accurately characterize the change laws of the mass, volume and bulk density with the temperature of the granite after heating and water-cooling treatment, the volume increase rate (ηv), mass decrease rate (ηm) and density decrease rate (ηρ) are introduced, which are, respectively, defined as follows:

$$ \eta_{v} = \frac{{V_{a} - V_{0} }}{{V_{0} }} \times 100\% $$
(1)
$$ \eta_{m} = \frac{{m_{0} - m_{a} }}{{m_{0} }} \times 100\% $$
(2)
$$ \eta_{\rho } = \frac{{\rho_{0} - \rho_{a} }}{{\rho_{0} }} \times 100\% $$
(3)

where V0, m0 and ρ0 denote the volume, mass and density, respectively, of the specimen prior to thermal treatment and Va, ma and ρa denote the volume, mass and density of the specimen after heating and water-cooling treatment.

As shown in Fig. 3, the average ηv of 1 cycle increases linearly with the temperature and it increases to 1.73% at 500 °C. The average value of ηv over 30 cycles increases exponentially with the temperature and accelerates substantially above 300 °C. The average value of ηv over 30 cycles is 1.16% at 300 °C, and it increases to 1.94% and 2.55% at 400 °C and 500 °C, respectively. The average value of ηm also increases with the thermal temperature; however, the average values of ηm over 1 and 30 cycles only increase to 0.16% and 0.31% at 500 °C. Compared to the volume increase rates, the mass decrease rates are smaller; thus, the trends of the density decrease rate and the volume increase rate with temperature are highly similar below 500 °C. The average value of ηρ over 30 cycles increases with temperature; at 150 °C, ηρ is only 0.72%, while it reaches 1.30%, 2.12% and 2.78% at 300 °C, 400 °C and 500 °C, respectively.

Fig. 3
figure 3

Relationships between temperature and the average change rates of volume, mass and bulk density

Full size image

3.2 P-wave velocity

The P-wave velocity of each specimen for each desired thermal cycle was measured to investigate the effect of thermal cycling on granite specimens after water-cooling treatment. The variations of the P-wave velocity with the number of thermal cycles are presented in Fig. 4. The P-wave velocities of granite after treatment at high temperatures exhibit a similar trend. Most of the decrease in the P-wave velocity occurred within 5 thermal cycles, especially after the first cycle. Afterwards, the P-wave velocities changed little with the number of thermal cycles. In addition, the P-wave velocity decreases with the thermal temperature. The values of the P-wave velocity of 30 cycles decrease by 74.4%, 54.0%, 37.6% and 23.4% at 150 °C, 300 °C, 400 °C and 500 °C, respectively.

Fig. 4
figure 4

Relationship between the number of thermal cycles and the P-wave velocity of granite after heating and water-cooling treatment

Full size image

3.3 Stress–strain curves

As shown in Fig. 5, the stress–strain curves for the specimens after heating and water-cooling treatment for 1 and 30 cycles exhibit four stages: compaction, elastic deformation, yield and failure. According to Fig. 5, the strength and deformation behaviour of granite depend strongly not only on the heating and water-cooling treatment but also on the number of cycles. In the compaction stage, all the curves display a concave-up shape from the start of the test and the nonlinearity in this stage increases with the thermal temperature for both 1 and 30 cycles, which is likely the result of more thermal cracks being induced by higher-temperature treatments. The slope of the curve in the elastic deformation stage decreases with the increase in the temperature of the heating treatment, especially for 30 cycles. Meanwhile, a yield platform can be observed in the yield stage above 300 °C for both 1 and 30 cycles; hence, the brittleness of the granite gradually decreases and the ductility gradually increases as the temperature rises. When exposed to a higher thermal temperature after water-cooling treatment, the granite specimens fail more slowly after the peak strength with increasing axial deformation.

Fig. 5
figure 5

Stress–strain curves for high-temperature granite with the water-cooling treatment: a one thermal cycle and b 30 thermal cycles

Full size image

3.4 Rock strength and deformation behaviour

The relationships between the temperature and the mechanical parameters of granite after heating and water-cooling treatment for 1 and 30 cycles are plotted in Fig. 6. Both the uniaxial compressive strength (UCS) and the elastic modulus (E) decrease with the heating temperature. Prior to the thermal treatments, the average values of UCS and E were 118.55 MPa and 18.39 GPa, respectively, which decreased to 105.11 MPa and 16.86 GPa at 150 °C after the first thermal cycle. However, the average UCS and E values decreased to 63.78 MPa and 5.55 GPa even at 150 °C when the number of thermal cycles increased to 30. The thermal cycling after water-cooling treatment had a substantial effect on the granite mechanical properties. The values of UCS decreased by 33.37% and 78.91% at 500 °C after 1 and 30 thermal cycles, respectively, and the values of E decreased by 38.82% and 88.42% at 500 °C after 1 and 30 thermal cycles, respectively. According to Fig. 6, the average peak strain (ε) increases with the thermal temperature and changes dramatically after 30 thermal cycles, which accords with the trends of UCS and E versus the heating temperature.

Fig. 6
figure 6

Relationships between temperature and granite mechanical parameters

Full size image

3.5 Microscopic observation

Thin sections of the water-cooling-treated granite specimens before and after thermal treatments of 1 and 30 cycles were enlarged by a factor of 40 and observed via SEM. As shown in Fig. 7a, mineral grains are arrayed closely and few initial pores and fissures can be observed in the specimen without heating. After the first thermal cycle, when the temperature is increased to 150 °C, we observe a few microstructural alterations compared with the result at normal temperature in Fig. 7b. Increasingly many thermal microcracks are induced with the gradual increase in the temperature to 500 °C (Fig. 7c–e). After 30 thermal cycles, many microcracks are observed, even at 150 °C (Fig. 7f), and the crack density of granite at 30 thermal cycles is substantially higher than that at 1 thermal cycle. Many microcracks in the 300 °C and 400 °C specimens have begun to interact and coalesce with each other (Fig. 7g–h), which leads to the further increase in the crack density compared to the specimens at 1 thermal cycle. Finally, a microcrack network is formed in the thin section of the specimen at 500 °C (Fig. 7i).

Fig. 7
figure 7

SEM images of water-cooling-treated granite after heating to various temperatures: 1 cycle: b–e; 30 cycles: f–i

Full size image

Chen et al. [4] and Yang et al. [46] observed that no microcracks can be found inside the granite specimen after thermal heating to 200 °C with natural cooling. However, in the current study, a few microcracks are observed after heating to 150 °C with water-cooling treatment. In addition, according to Chen et al. [4], a microcrack network is formed in thin sections of the specimen above 573 °C and for the heat-treated granite specimens that are subjected to water cooling, a microcrack network is formed in the thin section of each specimen at 500 °C after 30 thermal cycles. Thus, it is concluded that cyclical water-cooling treatment can induce microcracks more easily and induces more severe thermal damages to granite compared with natural cooling treatment.

To quantify the microcracks inside the granite specimens after heating and water-cooling treatment for 1 and 30 cycles, the microcrack density (ρf) is defined as follows:

$$ \rho_{f} = \frac{L}{S} $$
(4)

where ρf is the microcrack density that is measured in the thin section of the specimen after thermal treatment; L represents the total length of the microcrack that is observed in the thin section of the specimen after thermal treatment; and S represents the area of the thin section of the specimen.

The use of methods of damage mechanics to study rock thermodynamics is a new strategy in rock mechanics [24]. Thermal damage variables DVT and DET are introduced for quantifying the damage degree of granite specimens after heating and water-cooling treatment for 1 and 30 cycles, which are defined as follows:

$$ D_{VT} = 1 - \left( {\frac{{V_{T} }}{{V_{0} }}} \right)^{2} $$
(5)
$$ D_{ET} = 1 - \frac{{E_{T} }}{{E_{0} }} $$
(6)

where DVT and DET are the thermal damage variables; VT and ET represent the longitudinal wave velocity and the elastic modulus, respectively, of granite after heating and water-cooling treatment; and V0 and E0 represent the longitudinal wave velocity and the elastic modulus, respectively, of granite without thermal treatment.

The values of the microcrack density and the thermal damage variable are listed in Table 2. Figure 8 displays the relationships between the temperature and the microcrack density and thermal damage variable. Thermal damage accords with the microcrack density after 1 and 30 cycles. The values of microcrack density after 30 cycles (15.253, 18.491, 22.260 and 30.204 mm/mm2) are higher than those after 1 cycle (6.799, 10.310, 13.637 and 18.021 mm/mm2). The values of DVT and DET exhibit a similar trend to the thermal temperature after 1 and 30 cycles. The values of DET after 30 cycles are larger than the values of DVT after 30 cycles, while the values of DET after 1 cycle are smaller than the values of DVT after 1 cycle; hence, the heating and water-cooling treatment may have little effect on the physical and mechanical properties.

Table 2 Microcrack density and the thermal damage variable of granite after heating and water-cooling treatment
Full size table
Fig. 8
figure 8

Relationships between the temperature and the microcrack density and thermal damage variable of granite after heating and water-cooling treatment for 1 and 30 cycles

Full size image

3.6 Macroscopic failure modes

The macroscopic failure modes of granite after heating and water-cooling treatment for 1 and 30 cycles are presented in Fig. 9. When there is no thermal damage, the granite is a typically brittle rock material and shows a multiple axial splitting tensile failure mode. As the applied temperatures increase to 150 and 300 °C for 1 cycle, there is a tendency to transition from axial splitting to shear failure. From 400 °C onward, a through-going shearing plane emerges in the granite specimens and the granite exhibits a macroscopic shear failure mode, which is consistent with the previous observations [40]. After 30 thermal cycles, the integrity of specimen is lower than after treatment for 1 thermal cycle. According to Rong et al. [34], this is because more thermally induced microcracks developed in specimens that were subjected to more thermal cycles. After heating at 150 and 300 °C, the specimens show two parallel shear planes, which represent double parallel shear plane failure. Above 400 °C, a single shear failure plane occurs in the specimens and the angle of the failure plane is higher than that after 1 thermal cycle. In summary, the heating and water-cooling treatment and the number of thermal cycles both substantially influence the failure mode of granite.

Fig. 9
figure 9

Macrofractures of granite after heating and water-cooling treatment for 1 and 30 cycles

Full size image

4 Discussion

4.1 Influence mechanism of high temperature

The increase in the volume, the decrease in the wave velocity and the deterioration of mechanical properties are closely related to the generation and development of microcracks. Granite mineral grains will expand after heating treatment, and the heating temperature may have a substantial effect on the structure of the granite specimens. Differences in the thermal-expansion characteristics of among the minerals in the assemblage of mineral grains can cause structural damage upon heating the granite. In addition, differences in thermal expansion along the crystallographic axes of the same mineral can also cause structural damage upon heating [8]. Thus, microdefects will be produced between mineral grain boundaries or inside mineral grain bodies, and the higher the thermal temperature is, the more microdefects are produced (Fig. 7). Meanwhile, the states of water form water vapour and escape from the microcracks, thereby causing high air pressure which intensifies the formation and expansion of microcracks and micropores [50]. Due to severe deformation and microcracks, the original mineral structures will be destroyed and irrecoverable deformation will be produced, even after cooling to room temperature [39]. In addition, the open microcracks in granites that are caused by thermal treatment will rapidly close under a minor load; as a result, the brittleness of granite gradually decreases and the ductility gradually increases as the temperature rises [46].

The mass decrease of granite after temperature change is mainly caused by the loss of various types of water in the temperature range that is studied in this research (20–500 °C). The temperature ranges of vaporization of attached water, bound water and constitution water are room temperature to 100 °C, 100–300 °C and 300–500 °C, respectively [25]. The average mass decrease rate increases gradually with the thermal temperature (Fig. 3); hence, the higher the temperature the granite experiences, the greater the loss of granite. According to the above analysis, the granite mass decreases with temperature, while the granite volume increases with temperature. Consequently, the granite density decreases with temperature and the change amplitude increases with temperature.

It is concluded that the values of P-wave, UCS and E of the granite specimens after heating and water-cooling treatment all decrease with the temperature, which is attributed to the production and development of microcracks. Therefore, the changes of the physical and mechanical properties after high-temperature treatments must be related. The relationships between the normalized VpT/Vp0 values and the normalized UCST/UCS0 and ET/E0 values of the heat-treated granite by water cooling for 1 and 30 cycles are plotted in Figs. 10 and 11, respectively. Via fitting analysis of the above experimental data, normalized VpT/Vp0 is linear with respect to normalized UCST/UCS0 and ET/E0 after the first thermal cycle and exhibits exponential relationships with normalized UCST/UCS0 and ET/E0 after 30 thermal cycles. Moreover, the experimental properties of granite at 30 thermal cycles change more intensely than those at 1 thermal cycle. The fitting results are presented in Table 3, and the correlation coefficients of the fitting curves all exceed 0.9008; hence, there are strong links between the changes of the wave velocity and UCS and E of the granite after heating and water-cooling treatment for 1 and 30 thermal cycles.

Fig. 10
figure 10

Relationships between the normalized P-wave velocity and the normalized uniaxial compressive stress of water-cooled granite

Full size image
Fig. 11
figure 11

Relationships between the normalized P-wave velocity and the normalized elastic modulus of water-cooled granite

Full size image
Table 3 Relationships between the normalized wave velocity and the normalized uniaxial compressive stress and the elastic modulus of water-cooled granite
Full size table

4.2 Influence mechanism of cyclic water cooling

Normalized values of the uniaxial compressive stress (UCST/UCS0) and the elastic modulus (ET/E0) of granites at various temperatures are collected from a comprehensive review of the international literature, which includes Chinese publications that are not yet available for the English-speaking scientific community. The values of UCST/UCS0 and ET/E0 that were obtained for granites after heating and water-cooling or air-cooling treatment are presented in Figs. 12 and 13, respectively. The experimental data of Kumari et al. [21] and Xi et al. [42] were collected from heat-treated granite by water cooling, and “Average” represents the average values of UCST/UCS0 and ET/E0 at each temperature for all heat-treated granites via natural cooling. Below 500 °C, the changes in UCST/UCS0 and ET/E0 of the heat-treated granites by water cooling are larger than those of the heat-treated granites by natural cooling. At 400 °C, the average UCST/UCS0 and ET/E0 values of the heat-treated granites by natural cooling are 0.94 and 0.91, while the UCST/UCS0 and ET/E0 values by water cooling after the first thermal cycle in this paper decrease to 0.76 and 0.74, respectively, and the UCST/UCS0 and ET/E0 values at 30 thermal cycles are only 0.36 and 0.23. When the temperature is increased to 500 °C, the UCST/UCS0 and ET/E0 values at 1 thermal cycle that are obtained in this paper decrease by 33% and 0.61, respectively, and the UCST/UCS0 and ET/E0 values at 30 thermal cycles decrease by 79% and 88%.

Fig. 12
figure 12

Relationships between the temperature and the normalized uniaxial compressive stress of water-cooling- and natural-cooling-treated granites after being heated to various temperatures (“1 cycle” and “30 cycles” represent the values of UCST/UCS0 that were obtained in this study; “[21, 42]” represent the values of UCST/UCS0 of granite specimens after heating and water-cooling that were obtained from references; “[2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55]” represent the values of UCST/UCS0 of granite specimens after heating and air-cooling treatment that were obtained from references; and “Average” represents the average value of UCST/UCS0)

Full size image
Fig. 13
figure 13

Relationships between the temperature and the normalized elastic modulus of water-cooling- and natural-cooling-treated granites after being heated to various temperatures (“1 cycle” and “30 cycles” represent the values of ET/E0 obtained in this study; “[21, 42]” represent the values of ET/E0 of granite specimens after heating and water-cooling that were obtained from references; “[2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54]” represent the values of ET/E0 of granite specimens after heating and air-cooling treatment that were obtained from references; and “Average” represents the average value of ET/E0)

Full size image

Therefore, both the water-cooling and thermal cycling treatments substantially influence the granite physical and mechanical properties. The temperature decrease rate of heat-treated granites by water cooling is much more rapid than that by air cooling and causes more intense and sudden thermal shock [35]. This thermal shock further widens the cracks that have developed in the sample; thus, high cooling rates or more rapid and larger temperature reductions enhance thermally induced pre-fractures, thereby resulting in visible macroscale cracks in the rock mass [31]. In addition, for heat-treated granite specimens by water cooling, water will intrude into the granite bodies through the microcracks and micropores that are produced by heating and the strong contract among mineral grains will be further jeopardized, which will intensify the propagation and development of microdefects [21]. As a result, the mechanical properties of granites after heating and water-cooling treatment will be further deteriorated and weakened. Moreover, as shown in Fig. 7, thermal cycling treatment will cause a further change in the microstructure of the granite specimens [22, 34] and will lead to huge degradations in the physical and mechanical properties of granites after 30 thermal cycles.

4.3 Implications for geothermal energy exploitation

During long-term water circulation for exploiting deep geothermal energy, the hot geothermal reservoir rocks will be inevitably subjected to cyclic rapid temperature changes [3]. The above analysis suggests that further deteriorations of the mechanical properties were induced by rapid temperature reductions during cyclic water cooling. As a result, the deterioration of the mechanical properties tends to cause instability of the wall rock. Cooling may induce a pervasive tensile microcracking process prior to macroscopic failure location [1], which ultimately results in wall rock breakouts [37]. The experimental results demonstrated the substantial effect of the cyclic rapid cooling treatment and provided reliable parameter values for the accurate simulation of wellbore stability. In addition, the exploitation of geothermal energy entails the use of hydraulic fracturing and simulation to create highly conductive zones that interconnect an injection well to a production well [29] and hydraulic stimulation of a geothermal reservoir is a coupled process of thermal microcracking and hydraulic fracturing [18]. Considering the deterioration of the mechanical properties of geothermal reservoir rocks that is caused by cyclic rapid water cooling, it is beneficial for well drilling and producing fracture systems in geothermal reservoirs.

5 Conclusions

The physical and mechanical properties of granite after heating and water-cooling treatment for 1 and 30 cycles were studied experimentally, and the change mechanisms were identify via the SEM image analyses. Based on an extensive review of the mechanical properties of granites after high-temperature treatment, the following conclusions are drawn:

After heating and water-cooling treatment, the volume of granite increases with the temperature, the mass and the density of granite decrease with the temperature, and the changes in these quantities increase with the temperature. At 500 °C, the volume of granite increases by 1.73% and 2.55%, the mass of granite decreases by 0.16% and 0.31%, and the density of granite decreases by 1.86% and 2.78% over 1 and 30 thermal cycles, respectively.

The average values of UCS and E of granite after heating and water-cooling treatment for 1 and 30 cycles decrease as the temperature increases, while the peak strains exhibit the reverse trend. The decrease extents of UCS and E of the heat-treated granites by water cooling are larger than those of the heat-treated granites by natural cooling due to considerable thermal damage that is caused by intense thermal shock and water intrusion. A yield platform appears in the yield stage of the stress–strain curve above 300 °C, and the ductility of granite gradually increases as the temperature rises.

The average P-wave velocities of the heat-treated granite by water cooling for 1 and 30 cycles decrease with temperature and they decrease by 58.6% and 76.6%, respectively, at 500 °C. The normalized P-wave is linear with respect to the normalized UCS and E after first thermal cycle and shows exponential relationships with the normalized UCS, E after 30 thermal cycles.

According to microscopic observation of the thin sections of the water-cooling-treated granite specimens after thermal treatments of 1 and 30 cycles, the deterioration of the physical and mechanical properties of granite is mainly due to the generation and development of microcracks inside the specimen. Based on the observation of more microcracks at 30 thermal cycles than at 1 cycle, the thermal cycling treatment can further deteriorate and weaken the granite physical and mechanical properties.