Experimental Investigation of Shear Behavior and Pore Structure Evolution in Heat-Treated Granite Subjected to Liquid Nitrogen and Water Cooling

Abstract

It is imperative to understand the shear mechanical properties and pore evolution of granite under thermal shock to assess the fracturing of hot dry rock reservoirs. In this study, variable-angle shear tests were performed on coarse- and fine-grained granite samples following liquid nitrogen (LN₂) cooling under different high-temperature conditions. The effect of thermal treatment temperature, particle type, and cooling method on the shear strength, cohesion, and angle of internal friction of granite was then analyzed. To this end, low field nuclear magnetic resonance (NMR) and scanning electron microscopy (SEM) were used to investigate the pore size distribution and microstructural evolution of granite. The experimental results indicate that both the shear strength and cohesion of granite initially increase and then decrease with the rise in thermal treatment temperature. The maximum increases in shear strength and cohesion are 38.0% and 36.7%, respectively, while the maximum decreases reach 43.7% and 42.4%. Notably, the most pronounced thermal hardening effect is observed at 200 °C. In contrast, the internal friction angle exhibits a decreasing-then-increasing trend as the temperature rises, with a maximum reduction of 5.4% and a maximum increase of 14.5%. In addition, fine-grained granite exhibits superior shear strength and a more pronounced thermal hardening effect compared to coarse-grained granite. Furthermore, the damage effect caused by thermal shock increases with increasing heat treatment temperature, with the damage effect induced by liquid nitrogen cooling being particularly significant compared to water cooling. Furthermore, for both types of granite at the same shear angle, an increase in the heat treatment temperature results in a corresponding increase in the total fracture area, with the fracture area after liquid nitrogen cooling being more significant. The macroscopic failure mode changes from a mixed compression–shear failure mode to a direct shear failure mode with increasing shear angle. NMR testing shows that liquid nitrogen cooling can effectively increase the proportion of medium pores and large pores in the granite and increase the connectivity of internal pores; specifically, in coarse-grained granite, medium pores and large pores collectively increased by 10.5%, while in fine-grained granite, the total increase in medium pores reached 51%. As the heat treatment temperature increases, the type of crack that develops in granite changes from intragranular to transgranular. In addition, the fracture surface of granite is more prone to form micropores and small pores when cooled with liquid nitrogen, increasing the connectivity of the crack network. The results of this research will be useful for fracturing hot dry rock reservoirs.

1. Introduction

Geothermal energy has been identified as a highly efficient and reliable form of renewable energy, often referred to as ‘green energy’, due to its ability to maintain a consistent output regardless of seasonal or diurnal variation [1,2]. The term ‘hot dry rock’ (HDR) refers to rock formations found at depths of 3–10 km and characterized by high temperatures, which account for a significant proportion of geothermal resources [3,4]. The permeability of HDR reservoirs is primarily enhanced by hydraulic fracturing, creating enhanced geothermal systems (EGSs) that provide efficient heat extraction pathways [5]. However, fractures created by conventional hydraulic fracturing are typically large and singular, which may not result in the desired increase in production. In contrast, liquid nitrogen fracturing has been shown to improve the effectiveness of reservoir fracturing [6]. Hong et al. [7] observed via SEM scanning that, compared with conventional hydraulic fracturing fluids, nitrogen fracturing significantly increased the total crack length, overall fracture surface area, and flow conductivity of the rock, primarily due to the significant temperature difference between the low-temperature liquid nitrogen and the reservoir matrix. This induces numerous microcracks in the matrix under cold shock, resulting in the formation of a complex network of channels. Consequently, there is an imperative for comprehensive studies to be conducted on the evolution of the physical and mechanical properties of high-temperature granite after exposure to liquid nitrogen cold shock.

In recent years, a considerable body of research has been dedicated to investigating the alterations in the physical and mechanical properties of granite subjected to thermal treatment [8,9,10,11,12,13]. For instance, Zhang et al. [14] investigated the mechanical properties of granite under the combined influence of temperature and loading rate. The experimental findings demonstrated that compressive strength and elastic modulus initially increased and then decreased with temperature, reaching a maximum at 200 °C. When the temperature exceeded 600 °C, the effect of loading rate on compressive strength diminished, and the failure mode of the rock transitioned from brittle to semi-brittle. In a comparative study on the effects of two cold-shock methods (natural cooling and water-cooling shock) on the mechanical properties of granite after high-temperature heating, Srinivasan et al. [15] found that peak displacement significantly decreased under water-cooling shock, and the crack damage threshold was reached earlier. The study also found that water-cooling shock exacerbated strength degradation by inducing the expansion of intra-grain cracks. Dwivedi et al. [16] conducted a systematic investigation into the mechanical properties of Indian granite under elevated temperatures (30–160 °C), encompassing uniaxial compressive strength, tensile strength, Poisson’s ratio, and elastic modulus. Their research revealed that, under high-temperature conditions, granite exhibited significant anisotropy in its coefficient of thermal expansion. Furthermore, the use of scanning electron microscopy (SEM) images revealed that mineral particle expansion at 160 °C led to partial crack closure. Qin et al. [17] conducted physical and mechanical tests on granite following its natural cooling from 25 °C to 1000 °C. They identified 400 °C, 600 °C, and 800 °C as characteristic temperatures, which they suggest mark distinct stages in the evolution of the physical and mechanical properties of granite. In a related study, Zhang et al. [18] performed triaxial compression tests on granite subjected to high-temperature water-cooling shock. Their findings revealed that the Young’s modulus and cohesion of granite initially increased and then decreased with increasing temperature. Khalid Elwegaa et al. [19] investigated the effects of low-temperature nitrogen thermal shock on the porosity, permeability, and mechanical properties of shale, and they proposed a multi-cycle thermal shock experimental method. By subjecting the rock to three cycles of heating followed by nitrogen injection, they observed significant enhancements in porosity, permeability, brittleness index, and fracturing index.

Furthermore, a number of studies have investigated the physical and mechanical properties of granite following exposure to high-temperature liquid nitrogen cooling. For instance, Wu et al. [20] conducted a study on the properties of three types of rocks following exposure to high-temperature and liquid nitrogen cold-shock treatment. Through mechanical strength and permeability analysis, it was determined that liquid nitrogen cold shock significantly increased the permeability of the rocks. Sandstone, with its larger pores and mineral composition, exhibited the least thermal shock effect. Zhang et al. [21] conducted liquid nitrogen jet experiments on granite at temperatures ranging from 200 °C to 300 °C and analyzed the effects of liquid nitrogen fracturing on the fracture characteristics of dry hot rock reservoirs. The results showed that liquid nitrogen cold shock significantly increased the number and complexity of fractures as well as the permeability of the rock. In a related study, Wu et al. [22] investigated the effects of liquid nitrogen, water, and air cooling on the physical and mechanical properties and microstructural changes of high-temperature granite. Their findings indicated that liquid nitrogen cooling resulted in more substantial damage to granite compared to other cooling methods. In a related study, Cai et al. [23] investigated the effects of liquid nitrogen treatment on the damage characteristics of high-temperature granite. This included the effects of liquid nitrogen cold shock on P-wave velocity, tensile strength, energy evolution, and acoustic emission parameters. The study concluded that liquid nitrogen cold shock significantly increased the damage level of high-temperature granite. Longinos S.N. et al. [24] investigated the fracturing effects of liquid nitrogen (LN₂) on coal rock from the Karaganda Basin in Kazakhstan and proposed two novel treatment methods: freeze time (FT) and freeze–thaw cycles (FTCs). Their study demonstrated that liquid nitrogen fracturing can weaken the mechanical properties of coal by regulating freeze–thaw conditions, thereby enhancing permeability.

In the domain of practical rock engineering, the rock body is typically in a two- or three-dimensional stress state, with compressive shear damage being the most prevalent mechanical response of the rock body [25]. He et al. [26] investigated the impact of elevated temperatures on the shear characteristics through conducting variable-angle shear tests on granite samples subjected to different temperatures. The experimental findings demonstrated that both shear strength and cohesion decreased with increasing temperature, while the friction angle increased. Zhang et al. [27] investigated the mechanical characteristics of rock shear damage at high temperatures and determined that an increase in temperature led to a decrease in shear strength and modulus. The friction angle was found to have a significant effect on the behavior of shear damage. Du et al. [28] analyzed the shear strength and acoustic emission characteristics of rocks in the variable-angle shear test and the direct shear test. The authors found that, although the main types of microcracks generated by shear damage and compressive damage were the same, the shear damage generated by large-scale microcrack extension activity was more intense. Wael R. Abdellah et al. [29] conducted numerical simulations of rock mechanical behavior under triaxial compression and proposed a finite element method based on the Mohr–Coulomb criterion. The results demonstrated that an increase in confining pressure notably enhances both compressive strength and deviatoric stress; additionally, under small displacement increments, the rock more readily reaches its ultimate strength, thereby verifying the feasibility of substituting experimental tests with numerical simulations. However, the aforementioned studies have primarily focused on the shear characteristics of granite under natural cooling or ambient conditions, and have not thoroughly investigated the compressive–shear failure mechanisms induced by high-temperature/liquid nitrogen thermal shock and water thermal shock. The considerable difference in thermal stress gradients generated by liquid nitrogen thermal shock (–196 °C) versus water thermal shock (26 °C) could significantly alter the shear failure mode through crack anisotropy, yet no study has quantitatively characterized this phenomenon to date.

Despite the significant body of research conducted on the subject of liquid nitrogen cold shock and its effects on dry hot rock, the majority of studies have focused primarily on its impact on the physical properties and compressive and tensile strengths of high-temperature granite. Research on shear properties remains limited, and furthermore, the grain size of granite also influences its shear properties after liquid nitrogen cold shock. Therefore, in this study, variable-angle shear tests were carried out on high-temperature granite samples of different grain sizes subjected to liquid nitrogen and water-cooling shocks. This study analyzed the effects of treatment temperature, cooling method, and grain size on the shear strength, cohesion, and angle of internal friction of granite. In addition, nuclear magnetic resonance (NMR) and scanning electron microscopy (SEM) tests were used to investigate the internal pore evolution. The results of this study are expected to assist in the fracturing and modification of dry hot rock reservoirs.

2. Materials and Methods

2.1. Sample Preparation

In this study, two types of granite, distinguished by their coarse and fine particle sizes, were used in the experimental setup. The coarse granite was obtained from Huang-bai Town, Miluo City, Hunan Province. The fine granite was obtained from Chenxiang Town, Changtai District, Zhangzhou City, Fujian Province. Thin sections of coarse- and fine-grained granite samples were observed with a M330P_HK830 polarizing microscope at 5× magnification. Figure 1 presents the cross-polarized micrographs of the coarse- and fine-grained granite samples. Preliminary measurements from the imaging system indicated that in the coarse-grained granite, the particle size ranges for mica, feldspar, and quartz were 0.074–0.726 mm, 0.063–0.577 mm, and 0.085–0.635 mm, respectively, whereas in the fine-grained granite, the respective ranges were 0.048–0.415 mm, 0.074–0.225 mm, and 0.050–0.454 mm. In this study, the terms ‘coarse-grained’ and ‘fine-grained’ granite are employed to denote the relative differences in grain size between the two sample groups, as opposed to adhering strictly to the IUGS classification standards. This nomenclature is employed exclusively to differentiate the heterogeneity characteristics of the two groups for subsequent analysis. The polarized microscope images corresponding to each granite type are shown in Figure 1. X-ray diffraction (XRD) analyses showed that the main mineral compositions of both granite types include quartz, mica, and feldspar, with the coarse granite having a higher quartz content than the fine-grained granite. The mineralogical composition of the two granite types differs, with the coarse granite containing both sodium and potassium feldspar, whereas the fine granite is characterized by the presence of sodium feldspar alone (see

Table 1. Coarse and fine-grained granite: main mineral components and physical properties.

Granite	Mineralogical Composition (%)				Natural Density (g/cm3)	Porosity (%)
	Quartz	Micas	Feldspar-Sodium	Potassium Feldspar
Coarse	23.16	18.45	30.37	28.02	2.63	0.81
Fine	14.5	27.02	58.47		2.80	0.44

). All samples were obtained from a homogeneous granite source and carefully ground into cubic samples measuring 20 mm × 20 mm × 20 mm, ensuring that the surface was flat during the polishing process.

2.2. Experimental Equipment

The heating of the granite specimens was conducted within a muffle furnace (TNX1200-30, Shenzhen Wance Testing Equipment Co., Ltd., Shenzhen, China), which has a maximum heating temperature of 1200 °C, a chamber volume of 12 L, a maximum heating rate of 20 °C/min, and a temperature control system consisting of a 30-segment intelligent program controller, a PID controller, and a silicon-controlled rectifier master controller. The variable-angle shear tests on the granite specimens after high-temperature cold shock were carried out using a computer-controlled electronic universal testing machine (ETM305D, Luoyang Chunqing Furnace Industry Co., Ltd., Luoyang, China). The loading system of the machine consists mainly of a movable crosshead, a reaction frame, a force sensor, a manual controller, and shear fixtures. The reaction frame is made of high strength alloy steel and has a maximum load capacity of 300 kN. The force sensor system utilizes Transcell load cells, with an accuracy of up to 0.04% FS. The shear fixture angle (θ) is capable of ranging from 5° to 70° in 5° increments. To assess the pore size distribution of the granite samples after saturation treatment, a MicroM12-025VR nuclear magnetic resonance (NMR) core analyzer was employed. The microstructure of the granite was observed using a JSM-IT300 scanning electron microscope (SEM). The corresponding experimental equipment is shown in Figure 2.

2.3. Experimental Procedure

2.3.1. High-Temperature Cold-Shock Test

The granite samples were initially subjected to heating in a muffle furnace in order to simulate the ambient temperature of dry hot rock reservoirs. The heating temperatures were set at 200 °C, 300 °C, 400 °C, 500 °C, and 600 °C, and the heating rate was maintained at 5 °C/min [30] in order to avoid the occurrence of thermal shock effects during the heating process. Once the target temperature had been reached, the samples were held at the set temperature for a period of 4 h in order to ensure uniform heating. Thereafter, the samples were subjected to liquid nitrogen cold shock and water-cooling shock. Following this, the samples were removed from the liquid nitrogen and water, allowed to return to room temperature, and placed in an oven to dry at 105 °C for 24 h before further use. The detailed high-temperature cold-shock procedure is shown in Figure 3. The high-temperature treatment was conducted in the air environment of a muffle furnace (TNX1200-30) in order to simulate the actual conditions of dry hot rock reservoirs. While elevated temperatures can potentially induce mineral oxidation or phase transitions (e.g., the quartz α → β transition), all samples were subjected to identical thermal conditions to ensure reliable intergroup comparisons. Previous studies have demonstrated that within the temperature range of 200–600 °C, the effects of oxidation and phase transitions on the mechanical properties of granite are negligible [31]. Moreover, the present study focuses primarily on the interaction between thermal shock modes and particle size heterogeneity, with oxidation effects not constituting a core variable. The granite samples after high-temperature cold shock under different conditions are shown in Figure 4. The color changes observed in both coarse and fine granite were found to be predominantly influenced by the heating temperature, with minimal impact from the cold-shock method. Specifically, both granite types demonstrated a gradual color brightening with increasing heating temperature, which is attributed to the decline in mica content at higher temperatures [9]. The color change was more pronounced in the fine-grained granite, which underwent a transition from dark grey to grey-brown.

2.3.2. Shear Test

The high-temperature, cold-shock granite specimens were situated at the center of the shear fixture for the purpose of conducting variable-angle shear tests. Based on validation from previous experimental results, specimens are prone to interfacial slip failure when the shear angle is below 45°, and data anomalies may occur due to moment imbalance when the angle exceeds 70°. Therefore, the range of 45–70° is considered a valid testing interval [32]. Accordingly, shear angles of 45°, 55°, and 65° were selected for this experiment. The fixture setup is shown in Figure 5. During the course of the test, according to the recommendations for rock shear tests issued by the International Society for Rock Mechanics (ISRM) [33], displacement loading was applied at a rate of 0.1 mm/min until specimen failure occurred.

3. Results and Discussion

3.1. Shear Load–Displacement Curves

As demonstrated in Figure 6, Figure 7, Figure 8 and Figure 9, the shear load–displacement curves of coarse and fine granite under different high-temperature cold impact modes can be categorized into five stages [34].The first stage is the compression-density stage, during which the existence of pores and microcracks within the granite results in compression. When the specimen is subjected to compression–shear loading, the pores and microcracks will be gradually closed, resulting in the load curve showing a ‘concave’ shape in the initial stage of the test. The second stage is the linear elasticity stage, during which the load increases linearly with the increase in displacement, and the load–displacement curve is basically a straight line. This stage is characterized by the emergence and stable expansion of cracks within the specimen, resulting in a transition of the load–displacement curve from a linear to a non-linear trajectory. The third stage is marked by the non-stable expansion of these cracks, leading to their rapid propagation and penetration within the granite. The fourth stage, known as the post-peak stress softening stage, is characterized by a sudden drop in load following the peak, resulting in a sudden rupture of the specimen and the manifestation of typical brittle damage characteristics.

The load–displacement curves demonstrate that, for a given cold-shock method, both coarse-grained and fine-grained granite exhibit shortened compaction and linear elastic stages with increasing shear angle, with peak load and peak displacement decreasing continuously. This phenomenon is primarily attributable to the reduced normal stress on the granite, which reduces friction between internal particles and weakens its ability to resist shear loads [35]. Furthermore, as illustrated in Figure 10, at identical treatment temperatures and shear angles, high-temperature liquid nitrogen cold shock exhibited a greater effect on the peak displacement of fine-grained granite, with a smaller total displacement compared to high-temperature water-cooled shock. This phenomenon was less pronounced in coarse granite. The analysis suggests that fine-grained granite experiences greater thermal stresses under high-temperature liquid nitrogen cold shock, resulting in an increased number of thermally induced cracks. When shear loading is applied, these cracks propagate rapidly, significantly reducing the peak displacement. Furthermore, when the treatment temperature reaches 600 °C, both the peak load and peak displacement show a significant reduction, probably due to the α-β phase transition of quartz (≥573 °C), which causes crystal volume expansion and an increase in interparticle cracking, significantly degrading its mechanical properties [36].

3.2. Shear Mechanical Properties

The normal stresses and shear stresses exerted on the rock samples at differing shear angles were calculated using Formulae (1)–(3).

$$\tau ={\displaystyle \frac{P}{A}}\left(\sin \theta -f\cos \theta \right)$$

(1)

$$\sigma ={\displaystyle \frac{P}{A}}\left(\cos \theta +f\sin \theta \right)$$

(2)

$$f={\displaystyle \frac{1}{nd}}$$

(3)

where τ is the normal stress (MPa), σ is the shear stresses (MPa), P is the peak load of the specimen at the time of damage (N), A is the shear area (mm²), θ is the shear angle (°), f is the roller friction factor, n is the number of rollers and takes the value of 9, and d is the diameter of the roller and takes the value of 10 mm.

Furthermore, cohesion and angle of internal friction are essential parameters for the characterization of the shear strength of rocks, and their values were obtained based on the Mohr–Coulomb criterion [37]. A linear fit of the peak shear stress and the corresponding normal stress at varying shear angles was performed using the least squares method. The cohesion and angle of internal friction of the granite were derived from the fitted curve, as shown in the following formulae [38]:

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

(4)

3.2.1. Effect of Heat Treatment Temperature on Shear Mechanical Properties

As demonstrated in Figure 11, the variation in shear strength of different granite types with varying heat treatment temperatures is evident. It is observed that, for a constant shear angle, both coarse- and fine-grained granites demonstrate an initial increase followed by a subsequent decrease in shear strength as the heat treatment temperature is increased. In addressing the implications of high temperature on granite strength, the scientific community has put forward two predominant hypotheses [18,39,40,41]. Firstly, the thermal expansion of mineral grains has been suggested to fill the original pores in the rock, thereby enhancing its integrity. This is accompanied by the evaporation of internal moisture, thereby increasing the friction between the mineral grains. Conversely, the presence of different thermal expansion coefficients among minerals leads to the generation of thermal stresses within the rock, ultimately resulting in thermally induced cracking. The former process is designated as thermal hardening, while the latter is termed thermal damage. The net effect of temperature on the strength of granite is the result of a competition between these two processes. When thermal hardening is predominant, the shear strength of granite exhibits a substantial increase compared to its strength at room temperature (e.g., at 200 °C, where the shear strength of granite reaches its maximum). Conversely, when thermal damage is dominant, the shear strength of granite is significantly lower than its room temperature shear strength, as observed at 500 °C.

As demonstrated in Figure 12, the variation in cohesion and internal friction angle of coarse- and fine-grained granite with heat treatment temperature is observed. It can be deduced from the figure that the change in cohesion follows a trend consistent with that of shear strength, while the internal friction angle initially decreases and then increases with increasing heat treatment temperature. This is in contrast to the trends previously observed for shear strength and cohesion, and is primarily attributed to the thermal hardening effect, which enhances the cohesion of the granite and mitigates the influence of internal friction. Conversely, thermal damage generates a substantial number of thermal cracks in the granite, thereby weakening its cohesion and increasing the internal friction, resulting in an increase in the internal friction angle.

3.2.2. Effect of Cooling Method on Shear Mechanical Properties

As demonstrated in Figure 11 and Figure 12, the trends in the shear mechanical properties of coarse- and fine-grained granite under different cooling methods are, in general, consistent. However, in comparison with water cooling, liquid nitrogen cooling results in reduced shear strength and cohesion for granite. To illustrate this, let us consider fine-grained granite under a shear angle of 45°. As shown in

Table 2. Variation in shear characteristics of different types of coarse-grained granite with temperature.

	Temperature (°C)	Shear Strength (%)			Cohesion (%)	Internal Friction Angle (%)
		45°	55°	65°
High-Temperature Water-Cooled Impact	200	20.0	15.2	38.0	36.7	−5.2
	300	10.6	14.2	30.2	31.3	−5.4
	400	6.6	4.0	18.5	16.9	−2.9
	500	4.	−3.9	8.8	7.1	−1.6
	600	−10.	−21.9	−9.0	−11.8	−0.9
High-Temperature Liquid-Nitrogen-Cooled Impact	200	7.9	10.9	22.9	23.2	−3.6
	300	6.0	−0.9	12.4	10.9	−1.9
	400	2.9	−5.4	9.0	7.1	−2.0
	500	−2.3	−10.7	4.0	2.0	−2.3
	600	−14.2	−31.8	−13.3	−19.4	−0.3

and

Table 3. Variation in shear characteristics of different types of fine-grained granite with temperature.

	Temperature (°C)	Shear Strength (%)			Cohesion (%)	Internal Friction Angle (%)
		45°	55°	65°
High-TemperatureWater-CooledImpact	200	10.2	25.0	19.1	24.5	−4.3
	300	5.1	19.9	−5.6	−1.8	6.3
	400	0.9	6.4	−10.1	−8.1	5.9
	500	−7.0	−5.0	−22.3	−22.6	9.5
	600	−12.2	−32.5	−37.3	−42.4	14.5
High-TemperatureLiquid-Nitrogen-CooledImpact	200	3.3	14.0	6.0	9.1	−0.7
	300	2.1	5.9	−11.9	−10.4	7.4
	400	−2.3	2.5	−20.6	−18.6	10.0
	500	−14.3	−10.6	−27.5	−26.3	8.4
	600	−28.2	−34.0	−43.7	−44.8	10.7

, the shear strength of granite increased by 10.2% and 3.3%, respectively, after 200 °C water-cooling and liquid-nitrogen-cooling shocks, while the cohesion increased by 24.5% and 9.1%, respectively. Conversely, when subjected to water-cooling shock or liquid-nitrogen-cooling shock at 600 °C, the shear strength of granite decreased by 12.2% and 28.2%, respectively, while cohesion declined by 42.4% and 44.8%, respectively. These findings suggest that liquid-nitrogen-cooling shock diminishes the thermal hardening effect and amplifies the thermal damage effect in granite. This phenomenon can be attributed to the greater temperature difference between liquid nitrogen and high-temperature granite compared to that between water and high-temperature granite. When granite is subjected to liquid nitrogen, the thermal stresses generated within the material are greater than those generated by water, resulting in a greater number of thermally induced cracks [30]. This ultimately leads to a reduction in shear strength and cohesion in liquid nitrogen-shock-cooled granite compared to water-shock-cooled granite at the same heat treatment temperature.

3.2.3. Effect of Particle Size on Shear Mechanical Properties

As demonstrated in Figure 11, the shear strength of coarse granite is considerably lower than that of fine-grained granite when subjected to the same conditions. At a shear angle of 45°, within the range of 25 °C to 600 °C, the shear strength of coarse granite under water-cooling shock is measured at 119.28 MPa, 143.10 MPa, 131.89 MPa, 127.04 MPa, 124.23 MPa, and 106.27 MPa, respectively. Conversely, the shear strength of fine-grained granite under water-cooling shock is 158.33 MPa, 173.99 MPa, 165.88 MPa, 159.37 MPa, 135.28 MPa, and 113.39 MPa, respectively. This observation indicates that, in variable-angle shear tests, the shear strength of granite decreases as the particle size increases. This phenomenon can be attributed to the fact that coarser granite exhibits poorer internal homogeneity, which in turn affects its overall shear strength [40]. As demonstrated in Figure 12, within the range of 200 °C to 600 °C, the cohesion of coarse granite under liquid nitrogen shock is 123.2%, 110.9%, 107.1%, 102.0%, and 80.6%, respectively, and the cohesion of fine granite under liquid nitrogen shock is 109.1%, 89.6%, 81.4%, 73.7%, and 55.2%, respectively. This observation indicates that during the cohesion increase phase, the thermal hardening effect exerts a greater influence on coarse granite. This phenomenon can be attributed to the fact that following thermal treatment, the thermal expansion of the mineral grains in the granite leads to the filling of the original internal pores. Furthermore, coarse granite possesses a higher initial pore rate and a greater proportion of large pores (Figure 13), resulting in a more substantial enhancement in shear strength. Conversely, during the cohesion reduction phase, fine-grained granite exhibits a heightened susceptibility to thermal damage.

3.3. Macroscopic Failure

As illustrated in Figure 14, the macroscopic failure images of coarse- and fine-grained granite in variable-angle shear tests demonstrate that failure of the specimen primarily occurs along the shear plane. Under identical heat treatment conditions, when θ = 45°, a significant number of fragments are ejected during failure, and the surface of the specimen forms pronounced spall craters, with only a few cracks around the shear failure plane. Conversely, at an angle of 65°, the specimen predominantly fails along the shear plane, exhibiting a relatively flat shear failure surface, a hallmark of pure shear failure. This phenomenon can be attributed to the increasing shear stress, which, at higher angles, causes the variable-angle shear test to approach a direct shear test. Conversely, as the shear angle decreases, the proportion of compressive stress increases and the compressive stress inhibits shear failure, rendering the variable-angle shear test more analogous to a uniaxial compression test [25]. With increasing heat treatment temperature, the degree of fracture of coarse- and fine-grained granite at the same shear angle generally increases, and this phenomenon is more pronounced in the samples treated with liquid-nitrogen-cooling shock. This is related to the decrease in cohesion with increasing heat treatment temperature, as mentioned in Section 3.2.2. The high temperature reduces the cohesion within the specimen and weakens the cementation between the particles. When the specimen is loaded, the thermal cracks join and propagate rapidly, resulting in higher fracture levels at failure.

3.4. Pore Evolution

Nuclear magnetic resonance (NMR) is a widely utilized technique for the measurement of the size and distribution of pores in rocks [42,43]. The principle underpinning this method is the interaction between hydrogen nuclei in pore fluids (such as free and bound water) and an external magnetic field. The application of radio frequency pulses induces nuclear magnetic resonance, thereby enabling the acquisition of the relaxation time distribution [44]. The relaxation time distribution of fluids in porous media can be expressed by Equation (5), which demonstrates that the pore radius is proportional to the relaxation time [45]. In accordance with the extant research methodologies and in conjunction with the properties of the data of this specimen [45,46,47,48], the classification of rock pores is as follows: micropores (T₂ ≤ 1 ms), small pores (1 ms ≤ T₂ ≤ 100 ms), medium pores (100 ms ≤ T₂ ≤ 1000 ms), and large pores (T₂ ≥ 1000 ms).

$${\displaystyle \frac{1}{T_2}}\approx {\displaystyle \frac{1}{T_{2S}}}=\rho ({\displaystyle \frac{S}{V}})=\rho ({\displaystyle \frac{F_S}{r}})$$

(5)

where T₂ is the total relaxation time (ms), T_2s is the surface relaxation time (ms), ρ is the surface relaxation strength, S is the surface area of the pores (μm²), F_S is the shape factor of the rock, and r is the radius of the pores (μm).

3.4.1. Spectrum Curve Analysis

The T₂ spectral curves of granite after different heat treatments are shown in Figure 14, from which it can be seen that the peaks of the T₂ spectral curves are mainly bimodal, and most of the curves between the two peaks are smoothly and continuously connected, indicating that the pore connectivity of different dimensions within the granite is relatively good. Overall, the peaks and areas of the spectral curves gradually increase with increasing heat treatment temperature, indicating that the high-temperature cold shock can affect the pore structure of the granite and promote the development of pores [45]. Comparing the T₂ spectral curves under different cooling modes, it is found that the elevation and right shift of the second peak of coarse and fine granite under liquid nitrogen cold shock are more significant, indicating that the size and number of pores have been greatly increased, and liquid nitrogen cold shock has effectively changed the distribution of pores; furthermore, the peaks and valleys between the two peaks have been shifted upward more obviously with the increase in heat treatment temperature, which means that compared with water-cooled shock, liquid nitrogen cold shock can effectively improve the connectivity between the pores of granite. This indicates that compared with water-cooled shock, liquid nitrogen cold shock can effectively improve the connectivity between the pores of granite.

3.4.2. Pore Size Distribution Changes

The proportions of different types of internal pores in granite under various thermal shock conditions are shown in Figure 15. The distribution of different pore sizes in granite under high-temperature cold-shock conditions is observed to vary with heat treatment conditions. In particular, the pore structure of coarse granite is dominated by medium pores and large pores, whereas that of fine-grained granite is dominated by small pores and medium pores. It is evident that within the temperature range of 25 °C to 400 °C, the pore structure of both coarse- and fine-grained granite exhibits negligible changes in response to water-cooling shock. However, a pronounced alteration in pore structure is observed under liquid-nitrogen-cooling shock conditions. Specifically, the proportion of large pores in coarse granite increases from 15.64% to 38.02%, while the proportion of medium pores in fine granite rises from 12.88% to 28.4%. This observation indicates that liquid-nitrogen-cooling shock significantly increases the proportion of medium pores and large pores within the granite [44], which macroscopically weakens the shear mechanical properties of the granite. This finding is consistent with the results reported in Section 3.2.2. Within the range of 400 °C to 600 °C, the proportion of large pores and medium pores in both coarse- and fine-grained granite increases dramatically under both cooling methods, indicating that high temperature severely damages the internal pore structure of granite. It is noteworthy that under high-temperature cold shock, the proportion of large pores in coarse-grained granite increases with heat treatment temperature, while the proportion of medium pores in fine-grained granite increases with heat treatment temperature. This finding indicates that the deterioration of shear mechanical properties in coarse-grained granite is predominantly attributable to alterations in large pores induced by high-temperature cold shock, while the degradation observed in fine-grained granite is attributed to changes in medium pores.

3.5. Microscopic Failure Mechanism Analysis

As illustrated in Figure 16, which presents SEM images of coarse- and fine-grained granite under varying cooling conditions, it is evident that, at ambient temperature, the mineral crystals within both coarse- and fine-grained granite are predominantly intact, exhibiting smooth surfaces and firmly bound mineral grains. Some pores demonstrate a propensity to expand, and, in general, there is a paucity of intragranular cracks and fragments in the stressed regions. However, when the heat treatment temperature reaches 200 °C, the mineral crystals within both coarse and fine granite remain intact with smooth surfaces. The microcrack types are mainly transgranular and intragranular cracks, and the width of the cracks in the rock is relatively small. This is because at this stage the temperature is low and the thermal stress generated is minimal. The process of evaporation of free and bound water within the rock results in enhanced particle contacts, with thermal hardening dominating, leading to a significant increase in the shear strength and cohesion of the rock (Figure 17a). The application of a cooling shock, in the form of liquid nitrogen, results in the formation of numerous micro and small pores on the surface of the granite fracture. This phenomenon occurs due to the lower temperature of the liquid nitrogen, which, upon contact with the high-temperature rock, generates greater thermal stress [49].

As the temperature of the heat treatment increases from 200 °C to 400 °C, the integrity of the granite fracture surface gradually decreases, with both the number and width of cracks increasing. Transgranular cracking gradually becomes the dominant type of cracking in the rock, and at this stage the volume of the mineral grains inside the granite expands due to the high temperature, increasing the stress between the grains (Figure 17b). The formation of small cracks at the periphery of the grains, coupled with the evaporation of internal water, results in the deterioration of the mineral structure, with certain minerals undergoing decomposition and oxidation. The process of thermal damage escalates, leading to a progressive and continuous deterioration of the rock’s physical properties. At a temperature of 400 °C, the fracture surface of fine-grained granite transitions from smooth to rough [50], a consequence of the uncoordinated deformation of mineral particles, which serves to reduce the original cementation between these particles [45]. In contrast, the fracture surface of coarse-grained granite becomes rough at 500 °C, a phenomenon that may be attributed to the disparate initial levels of cementation between the two granite types. Based on the findings presented in Section 3.2.1 and Section 3.3 of this study, an inverse relationship was observed between fracture surface roughness and the rock brittleness index. Specifically, smoother fracture surfaces were associated with higher brittleness indices, whereas an increase in fracture surface roughness corresponded to a progressive reduction in rock brittleness [50].

Figure 16. SEM images of granite under different cooling methods (500×): (a) coarse-grained granite after high-temperature water-cooled impact, (b) coarse-grained granite after high-temperature liquid-nitrogen-cooled impact, (c) fine-grained granite after high-temperature water-cooled impact; (d) fine-grained granite after high-temperature liquid-nitrogen-cooled impact.

Figure 16. SEM images of granite under different cooling methods (500×): (a) coarse-grained granite after high-temperature water-cooled impact, (b) coarse-grained granite after high-temperature liquid-nitrogen-cooled impact, (c) fine-grained granite after high-temperature water-cooled impact; (d) fine-grained granite after high-temperature liquid-nitrogen-cooled impact.

Figure 17. Schematic diagram of particle crack structure in granite after high-temperature treatment: (a) 25 °C, (b) 400 °C, (c) 600 [51].

Figure 17. Schematic diagram of particle crack structure in granite after high-temperature treatment: (a) 25 °C, (b) 400 °C, (c) 600 [51].

Figure 17c provides a schematic illustration of the interparticle changes in granite when the heating temperature reaches 600 °C; when considered together with Figure 16 and Figure 17, it reveals when the heat treatment temperature reaches 600 °C, the degree of fracturing of the granite fracture surface increases significantly, with a marked increase in the number and width of cracks. Quartz undergoes an α-β phase transition at this stage (≥573 °C) [52], accompanied by significant volume expansion. The interparticle thermal stress increases, significantly reducing cementation between mineral particles. The number and width of transgranular cracks between particles increase significantly. In addition, the different coefficients of thermal expansion of minerals at high temperatures generate large thermal stresses during both the heating and cooling phases [53], which further promote the development and interconnection of microcracks, forming a crack network. At this stage, the thermal damage effect far outweighs the thermal hardening effect, leading to a sharp reduction in the shear strength, cohesion, and other strength indicators of the rock.

3.6. Damage Variable

Quantitative analysis of changes in the mechanical properties of granite under different hot–cold impacts can be undertaken using the generalized damage variable D:

$$D=1-{\displaystyle \frac{W_T}{W_0}}$$

(6)

where W_T and W₀ are the mechanical properties (shear strength and cohesion) of the granite before and after the heat–cold impact, respectively. Compared to conventional damage variables, if D is negative, it indicates that thermal hardening of the rock is occurring at this time.

Based on the experimental results, the damage to coarse- and fine-grained granite after various high-temperature and cold-shock treatments was calculated, as shown in Figure 18. The figure shows that, overall, the degree of damage to granite increases significantly with increasing heat treatment temperature, reaching a maximum at 600 °C. In addition, under liquid nitrogen cold shock, the damage characteristics of the rock are more pronounced. For example, at 600 °C, the cohesion damage of coarse-grained granite under high-temperature liquid nitrogen cold shock is 0.2, significantly higher than the 0.12 observed under high-temperature water cold shock. In addition, fine-grained granite exhibits more significant damage under high-temperature cold shock than coarse-grained granite. For example, at 400 °C, the maximum damage value of fine-grained granite under high-temperature water cold shock is 0.11, while under high-temperature liquid nitrogen cold shock it is 0.23. As mentioned in Section 3.2.1, the effect of high temperature on granite is the result of the competition between thermal hardening and thermal damage. From Figure 18, it can be seen that the transition from thermal hardening to thermal damage is closely related to temperature, cold-shock method, and particle size, with the transition temperature occurring earlier for liquid nitrogen cold shock than for water cold shock. The degree of damage to granite resulting from high-temperature cold shock varies with different mechanical parameters. Taking fine-grained granite as an example, if the damage level of granite is assessed on the basis of its cohesion change, it is significantly higher than the results obtained using other mechanical parameters. Therefore, relevant mechanical parameters should be reasonably selected to assess the fracture effect of dry hot rock reservoirs.

4. Conclusions

In this study, variable-angle shear tests were performed on coarse- and fine-grained granite samples after high-temperature heating, followed by liquid nitrogen cold shock and water-cooling shock. Combined with nuclear magnetic resonance (NMR) and scanning electron microscopy (SEM) tests, the following main conclusions were reached:

• As the shear angle is increased, a continuous decrease in peak load and peak displacement is exhibited by both coarse- and fine-grained granite samples. Furthermore, liquid nitrogen cold shock exhibited a more pronounced effect on the peak displacement of fine-grained granite when the heat treatment temperature and shear angle were constant, with the overall effect being less than that of water-cooling shock.
• Granite shear strength and cohesion generally exhibit a trend of initial increase followed by a decrease with rising thermal treatment temperatures, whereas the internal friction angle shows an inverse trend. Notably, the thermal hardening effect is most pronounced at 200 °C. Furthermore, at the same thermal treatment temperature, granite subjected to liquid nitrogen thermal shock has lower shear strength and cohesion compared to granite treated with water thermal shock.
• This study established that the application of liquid nitrogen cold shock resulted in a pronounced alteration in the internal pore distribution within granite, thereby enhancing the connectivity between pores. In comparison to the impact of water-cooling shock, liquid nitrogen cold shock exhibited a significant augmentation in the proportion of medium pores and large pores within the granite.
• At ambient temperature, intragranular cracking is the predominant form of internal cracking in granite; however, as the heat treatment temperature is increased, the cracks gradually transition to transgranular cracks and the integrity of the fracture surface is reduced, with the surface changing from smooth to rough. Furthermore, the application of liquid nitrogen cold shock has been demonstrated to result in the emergence of numerous micro- and small-sized pores on the fracture surface of granite. This effect underscores its superior efficacy in fostering the development of intricate fracture networks within dry hot rock reservoirs.