2.1 Using only water or water and air as a working fluid for geothermal energy

Studies have explored the physical mechanisms involved when water is used as a working fluid for the extraction of heat from EGS (as illustrated in Figure 1). Rong et al. (2018) conducted an experimental investigation of thermal cyclic effects (0, 1, 2, 4, 6, 8, and 16 thermal cycles) on dolomitic marble and granite. To this end, the porosity and P-wave velocity (Vp) of the marble and granite samples were recorded. To simulate thermal cyclic effects, the samples were heated to 600°C (to imitate high temperatures in deep geothermal reservoirs) and maintained at that temperature for 4 h (Rong et al., 2018). The samples were then cooled naturally to room temperature (25°C).

               

The test results show that Young's modulus (Figure 2a), wave velocity (Figure 2b), and uniaxial compressive strength (UCS) (Figure 3) of the samples decrease as the number of thermal cycling treatment increases (Rong et al., 2018). This might have occurred due to the thermal cycling treatment inducing microcracks in the rock samples, and with an increasing number of cycles, the microcracks grow into larger fractures (Rong et al., 2018). The early stages of the experiment show a huge decrease in Young's modulus and P-wave velocity, which gradually decreases with an increase in the number of thermal cycles. This could be attributed to the possibility that the cracks may have propagated sufficiently after a few cycles. The effect of heat extraction on the Young's modulus of hot rocks is further supported by a study conducted by Shu et al. (2020).

               

               

The effect of fluid flow in fractured hot granite rocks was investigated using two experimental setups (Shu et al., 2020). The first experimental program was conducted at a constant temperature with increasing confining pressures, while the second experimental program was conducted at a constant confining pressure with increasing temperature. The temperature and pressure values were selected to model in situ conditions in deep bedrocks. The results (Shu et al., 2020) suggest that at high constant confining pressure and with an increase in temperature from 25 to 200°C, permeability can decrease by 30%, 28%, and 37% at confining pressures of 10, 15, and 20 MPa, respectively, due to the substantial pressure exerted on the low stiff fractures. Relatively, at low constant confining pressure values, the hydraulic properties and permeability can increase due to the negligible effect of the pressure (Figure 4). Similarly, at high constant temperature values, greater pressure results in a rapid decrease in permeability. This is because the dramatic drop in temperature results in a loss of the elastic modulus of the granite sample, which yields more tightly closed fractures (Shu et al., 2020). Therefore, it can be inferred from these studies (Rong et al., 2018; Shu et al., 2020) that there is a potential loss in the elastic modulus (stiffness) of EGS due to the cyclic thermal interaction with fluids during EGS. This may consequently affect the long-term stability of EGS for continuous heat extraction (Rong et al., 2018; Shu et al., 2020).

               

Temperature ranges vary depending on the depth of the EGS. Therefore, research studies have assessed the mechanical changes of hot rocks under different temperature conditions. The behavior of granite at specific temperatures (i.e., 200, 300, 400, 500, 600, 700, and 800°C) was studied using a uniaxial compression test (Yang et al., 2017). The results show that the strength and static elastic modulus of granite continuously increased as temperature increased up to 300°C. For temperatures above 300°C, Young's modulus and strength of the granite samples decreased (Yang et al., 2017). The increased strength and stiffness when rock samples are heated to 300°C may be due to the thermal expansion of mineral grains. When the temperature is above 300°C, this thermal expansion causes cracks in the rock, subsequently decreasing its strength and stiffness. A similar experiment was conducted on granite at specific heating temperatures between 200°C and 900°C (Zhang et al., 2018). The samples were maintained at their treatment temperature for 4 h and then cooled with water to room temperature (Zhang et al., 2018). The results suggest that below 500°C, thermal hardening in granite was dominant, thus resulting in an increase in the elastic modulus and compressive strength of the sample. In contrast, above 500°C, thermal cracking was induced, and the granite samples showed a significant decrease in their elastic modulus and compressive strength (Zhang et al., 2018). The findings from these studies indicate that there is less weakening of the mechanical properties of hot rocks at increasing temperatures in EGS (Yang et al., 2017; Zhang et al., 2018). Yang et al. (2017) suggest that there is a critical temperature threshold (300°C) below which the mechanical behavior of granite changes with an increase in temperature, while Zhang et al. (2018) reported the same trend but at a greater critical temperature threshold (500°C). These inconsistent critical temperature threshold values (Yang et al., 2017; Zhang et al., 2018) may be due to the difference in cooling methods used in each experimental study and the inherent heterogeneity features in the rock samples used. Despite this discrepancy, we concur that the changes in mechanical properties of hot rocks, especially granites, may vary depending on the established critical temperature threshold value in EGS. Therefore, we can infer that variations in the mechanical behavior of granites might occur under varying temperature conditions in EGS.

Geothermal resources have mechanical heterogeneity due to pre-existing alterations (Kolawole, Ispas, Kolawole, et al., 2021; Kolawole & Oppong, 2023; Ngoma & Kolawole, 2024; Siratovich et al., 2016). Kolawole, Ispas, Kolawole et al. (2021) analyzed a 0.91-m-long core sample of a hydrothermally altered rock (hot dolomitic sedimentary-hosted geothermal rock) from a potential geothermal area in the Permian Basin, United States. The results indicate variations of mechanical properties across the analyzed sample, with alternating low to high UCS, Poisson's ratio ( $\nu$ ), and scratch-derived fracture (Ks) values at distinct mechanical zones along the core (Kolawole, Ispas, Kolawole, et al., 2021). The mechanically softer 0.17-m-thick Zone A and 0.18-m-thick Zone C have mean values of UCS = 110 MPa, $\nu$  = 0.25, Ks = 1.89 MPa /m, and the mechanically harder 0.41-m-thick Zone B and 0.15-m-thick Zone D have mean values of UCS = 166 MPa, $\nu$ = 0.22, and Ks = 2.87 MPa /m. The results could be attributed to the varying mineralogical compositions because of hydrothermal fluid migration and vein development at each mechanical zone in the core sample (Kolawole, Ispas, Kolawole, et al., 2021). We can, therefore, infer that the properties, morphological attributes, and mineral compositions in the production life of an EGS can induce the development of mechanical alterations because of continuous long-term heat extraction. We also suggest that mechanical changes in an EGS are complex and should consider existing mechanical heterogeneity in the heat extraction from an EGS.

Thermo-hydro-mechanical (THM) models have also been adopted to couple the effects of temperature, fluid flow, and rock deformation, with a focus on enhancing heat extraction efficiency (Cao et al., 2016; Norbeck et al., 2016; Wang et al., 2022). In a study, Cao et al. (2016) utilized a THM model that considered local thermal nonequilibrium to formulate convective heat exchange between the rock matrix and the working fluid in the reservoir. They also used their thermo-poroelastic model to estimate the stress in the rock matrix, in addition to evaluating the porosity and permeability changes as a function of time (Cao et al., 2016). The results indicate that the efficiency of heat energy extraction is highly dependent on the hydraulic stimulation area, injection rate, and thermal conductivity. By injecting the working fluid at high pressure and low temperature, a higher magnitude of the negative effective stress can be attained, which induces an enhancement of the fluid flow rate and the heat extraction rate as a result of the effect of enlargement of the hot rocks' permeability (Cao et al., 2016). Nevertheless, when the fluid is injected at a low temperature and high pressure, the working fluid viscosity significantly increases. The results (Cao et al., 2016) also suggest that a relatively larger hydraulic stimulation area and high fluid injection rate may yield a higher volumetric heat transfer coefficient. In contrast, a lower volumetric heat transfer coefficient can decrease the heat exchange between the hot rock and the working fluid, and may lead to lower effective stress magnitude, which can yield a resultant effect on the porosity and permeability in EGS.

A series of experimental investigations have been performed to mimic EGS using different cooling methods at different temperatures. For instance, Zhu et al. (2021) examined the mechanical changes of granite using two different cooling methods. The rock samples were heated to predetermined temperatures (i.e., 200–600°C) and then maintained at the final temperature for 2 h before cooling. Afterward, the specimens were cooled using either rapid water cooling or slow air-cooling methods. The results show that the mechanical properties of the samples reduced with increasing temperature. At temperatures lower than 600°C, the samples showed ductile characteristics. At 600°C, the water-cooled samples yielded an 85% decrease in average Vp, a 73% decrease in average UCS, and a 66% decrease in the average elastic modulus (Zhu et al., 2021). Similarly, the air-cooled samples showed a 74%, 56%, and 49% decrease in the average P-wave velocity, UCS, and elastic modulus, respectively (Zhu et al., 2021). Using similar treatment methods, Li et al. (2020) examined granite rocks at increasing temperatures (between 100 and 600°C) with two different cooling modes—the natural cooling method and rapid cooling with water. The samples were heated to the desired temperature and maintained at this temperature for a few hours to obtain their corresponding results. The results (Li et al., 2020) show that the elastic moduli and UCS of the samples decrease with an increase in temperature, which is consistent with the trend observed in Yang et al. (2017), Zhang et al. (2018), and Zhu et al. (2021). The magnitude of these mechanical strength losses was greater with water cooling than with the natural cooling method (Li et al., 2020). A similar pattern was observed in a study conducted by Liu et al. (2021) on heated granite samples, where the permeability increased more when the samples were cooled in cold water than when they were cooled in air. It was also observed that when the temperature was less than 300°C, there were no changes in the crystal structure of the granite samples, which gradually expanded and eventually developed into larger fractures when the temperature exceeded 300°C (Li et al., 2020). However, the original microfractures gradually expanded and eventually developed into larger fractures when the temperature exceeded 300°C (Li et al., 2020).

Previous studies have emphasized the relevance of the cooling methods used for high-temperature experimental tests and analysis of granite. This is important because it could provide valuable insights into the process of heat extraction from EGS. When cold water is utilized as the cooling method or working fluid, the rock experiences a more significant decrease in mechanical strength in these experiments than when air-cooled samples are used. This may occur due to the rapid cooling effect of cold water on the low-permeability geothermal rocks when compared to using either air or air and water for cooling, thus inhibiting the formation of more structured bonds between minerals. This may eventually result in decreased mechanical strength of the rock, which requires further investigation. These observed comparisons unlock the potential of using nonaqueous fluids (other than water) as working fluids in EGS.

2.2 Using CO2 as a working fluid for geothermal energy and storage

Studies have explored the potential of using CO2 as a working fluid (An et al., 2021; He & Li, 2020; Li et al., 2019; Liao et al., 2020; Shen et al., 2020; Shu et al., 2020; Song et al., 2020; Tong et al., 2022; Zhong et al., 2022) in EGS because of its potential to not only improve fracture stimulation for heat energy extraction but also promote CO2 storage (Tong et al., 2022). CO2 sequestration (CCS) can be coupled with geothermal energy extraction in a naturally permeable and porous geologic formation (Figure 5). Zhong et al. (2022) developed a wellbore reservoir model to analyze and compare a CO2 versus a water EGS system in fractured geothermal reservoirs. Another study by Randolph and Saar (2011) developed numerical simulations to assess geothermal heat energy extraction using CO2-plume geothermal. The results (Randolph & Saar, 2011; Zhong et al., 2022) indicate that due to the thermodynamic and fluid mechanical properties of CO2, there is a more efficient transfer of geothermal energy compared to water. Additionally, the numerical predictive tools developed in these studies can be adopted for mitigating greenhouse gas emissions on a field scale while generating renewable energy (Randolph & Saar, 2011; Zhong et al., 2022).

               

Shu et al. (2020) investigated the mechanical properties of granite samples from China by water and CO2 injections at different confining pressures (2–20 MPa), different pore fluids (10 MPa water or CO2), and different temperatures (25–150°C). The study used triaxial compression experiments and scanning electron microscopy to determine their mechanical and microstructural properties. The results indicate that combined water and CO2 injections can decrease the Young's modulus of hot rocks (Shu et al., 2020). At a confining pressure of 15 MPa, the UCS increases with an increase in temperature. The type of pore fluid injected can impact the elastic moduli of hot rocks, and the mechanical weakening of the rocks due to CO2 injection is more obvious (Shu et al., 2020).

In an experimental investigation on granite rocks, the samples were exposed to three different treatments, namely, pure thermal stimulation, pure CO2-bearing solution stimulation, and combined stimulation of thermal and supercritical CO2-bearing solution (Tong et al., 2022). The results show that the most significant effect on the samples was due to water cooling on the heated granite combined with the CO2-bearing fluid stimulation. With this treatment, the permeability of granite was 17 times higher than that of the samples in the untreated state. The porosity increased by 144%, whereas the elastic modulus and compressive strength decreased by 14% and 18%, respectively (Figure 6). Compared to the single thermal stimulation and CO2-bearing fluid hydro-chemical stimulation, the results show that the superposition effect of thermal and CO2-induced hydro-chemical stimulation can increase the number of microfractures in granite rocks more effectively (Tong et al., 2022). Therefore, the permeability increased, whereas the elastic modulus and compressive strength decreased. These results are in agreement with an investigation by An et al. (2021), where a dynamic alteration damage apparatus was developed to study long-term CO2–water–granite interactions.

               

Representative test results indicated that the mechanical properties of granite rocks deteriorated considerably in supercritical carbon dioxide (ScCO2)–water mixtures as the reaction proceeded, while the dispersion in the mechanical properties increased (An et al., 2021). The results show that the UCS and Young's modulus of the samples decreased by 5% and 19%, respectively (An et al., 2021). We deduce from the results that when CO2 is coupled with water as a working fluid for stimulation in EGS, the UCS and Young's modulus of the EGS are expected to decrease over time. This may be accompanied by an increase in permeability because of the reduced mechanical strength of the hot rock, which consequently affects the heat extraction efficiency. However, there are uncertainties concerning the extent of degradation of the mechanical properties observed and if this might result in further long-term instability in geothermal reservoirs.

Using a coupled thermo-hydro-mechanical-chemical (THMC) model (Gan et al., 2021), the feasibility and potential benefits of using ScCO2 as a working fluid in deep geothermal systems were investigated by accounting for the thermodynamic, hydraulic, mechanical, and chemical behavior of ScCO2-EGS. The results show that ScCO2 injection can increase the reservoir pressure, which enhances the productivity and heat transfer in EGS and leads to a higher power output. The work also indicated that mineral dissolution and/or precipitation can significantly affect the tendency of ScCO2-EGS to yield greater connected fracture networks, but this requires further studies (Gan et al., 2021). The effect of CO2 injection on cloud-fracture network development in granites under moderate and superhot geothermal conditions was studied by Pramudyo et al. (2021). They used X-ray computed tomography (CT) imaging and digital image analysis techniques to analyze the induced fracture network development. The results showed that CO2 injection resulted in the development of complex fracture networks that were significantly different from the control samples, especially under superhot conditions. The fracture networks developed in a multistage process, with the initial stage dominated by microfracturing and the later stage dominated by macrofracturing (Pramudyo et al., 2021). The study concluded that CO2 injection under superhot conditions resulted in a more connected fracture network than under conventional conditions. This suggests that CO2 injection can enhance geothermal energy production by creating more permeable fracture networks (Pramudyo et al., 2021).

To improve the understanding of the effect of heterogeneity features in EGS, a three-dimensional finite element method-based geothermal reservoir model (Singh et al., 2023) was developed to understand the impact of reservoir heterogeneity on modified reservoir mechanical properties and resultant dual geothermal energy extraction and CO2 sequestration. The findings in Singh et al. (2023), also reported in Kolawole, Ispas, Kolawole et al. (2021) and Kolawole and Oppong (2023), suggest that mechanical heterogeneity in tight reservoirs may significantly dictate injected fluid behavior and heat distribution, and influence the mechanical stability of EGS and operational efficiency of coupled geothermal heat extraction and CO2 storage operations.

Wu and Li (2020) discussed the potential of combining CCS and geothermal energy production in a CO2-EGS. They reported on a study of CO2 injection used as a working fluid into deep low-permeability geothermal reservoirs, leading to induced fractures to enhance heat energy extraction while at the same time storing CO2 in a geological formation. The results of their studies indicate that CO2-EGS has the potential to be a cost-effective and sustainable method for both geothermal energy production and CCS because it increases permeability and heat transfer in the reservoir, and yields higher energy recovery. However, the study also identified technical uncertainties and public acceptance issues as the main challenges that must be addressed to commercialize CO2-EGS as a technique for curtailing greenhouse gas emissions and mitigating climate change.