Prospects and Challenges of Waterless/Low-Water Fracturing Technologies in Hot Dry Rock Geothermal Development

Abstract

Geothermal energy is a clean, renewable, and baseload-stable resource of strategic importance for carbon neutrality. Hot dry rock (HDR) reservoirs are characterized by high temperatures, great depths, and abundant reserves. However, their extremely low natural permeability requires artificial fracturing to establish effective heat exchange networks. Conventional hydraulic fracturing in enhanced geothermal systems (EGS) faces major challenges under HDR conditions, including excessive water consumption, strong water–rock interactions, and elevated induced seismicity risks, limiting its engineering applicability. Waterless or low-water fracturing technologies offer alternative stimulation pathways due to their distinctive physicochemical properties. Existing reviews have mainly addressed individual aspects, such as specific fracturing media or proppant transport, without systematically integrating recent advances in supercritical CO₂ fracturing, foam fracturing, liquid nitrogen fracturing, and hybrid-fluid fracturing technologies, or comprehensively evaluating their engineering implications. This review systematically analyzed the fracturing mechanisms, heat exchange performance, environmental risks, and HDR-specific engineering challenges of these technologies. Results indicate that waterless/low-water fracturing technologies enhance heat extraction efficiency by generating complex fracture networks while mitigating seismic and reservoir damage risks. However, large-scale application requires further advances in the high-temperature stability of fracturing media, material durability, multiphase flow control, and field validation.

1. Introduction

Geothermal energy, as a clean, renewable, operationally stable, and widely distributed energy source, is playing an increasingly important role in the global energy transition [1,2]. Theoretically, the thermal energy stored within the upper 10 km of the Earth’s crust is approximately 1.3 × 10²⁷ J [3]. Assuming a 70% development and utilization efficiency (a conservative estimate), the exploitable geothermal potential could reach 1200 GWe [4]. If this energy is harnessed for power generation, it can significantly reduce fossil fuel consumption, as well as greenhouse gas and pollutant emissions, thereby supporting the achievement of carbon neutrality goals [5].

Based on the genesis and occurrence conditions of geothermal resources, and under current technological capabilities, exploitable geothermal resources mainly include shallow geothermal resources, medium-to-deep hydrothermal resources, and deep hot dry rock (HDR) resources [6]. HDR resources refer to dense crystalline rock formations buried at depths of 3–10 km, containing little or no fluid, typically with temperatures exceeding 150 °C, and characterized by extremely low natural permeability and porosity. These reservoirs are dominated by granites but also include carbonates and tight sandstones [3,7]. The global HDR resource base is enormous: the thermal energy stored at depths of 3–10 km on land alone is estimated to be equivalent to about 30 times the total energy contained in all global oil, natural gas, and coal reserves [3]. This far exceeds conventional hydrothermal resources and indicates substantial development potential.

However, HDR development faces a fundamental challenge: the intrinsically low permeability of the rock mass hinders the formation of effective fluid circulation pathways, resulting in low heat-extraction efficiency [8]. Enhanced geothermal systems (EGS) constitute the primary technological pathway for HDR development. By creating artificial fracture networks in HDR reservoirs through hydraulic stimulation, reservoir permeability can be improved and the heat-exchange area enlarged. A working fluid is then injected to perform heat exchange, ultimately enabling power generation or heating (Figure 1) [9,10].

Conventional hydraulic fracturing, as the core approach for EGS reservoir stimulation, has been applied in multiple demonstration projects [11,12], yet its limitations are becoming increasingly evident. First, this method requires large volumes of water, which can trigger water scarcity and social conflicts in remote or arid regions [13]. Second, water–rock interactions become intense under high-temperature conditions, potentially leading to mineral dissolution–precipitation cycles that generate silicate or carbonate precipitates, clogging fracture pathways and reducing long-term conductivity [14]. In addition, high-pressure water injection can activate pre-existing faults through pore-pressure diffusion, inducing felt seismic events [15]. Seismic incidents associated with EGS projects in Basel, Switzerland [16], and Pohang, South Korea [17], have raised public concern and resulted in project suspension. Finally, the relatively high viscosity of water limits fracture geometric complexity, often producing a single dominant fracture that is unfavorable for expanding large-scale heat-exchange areas in HDR reservoirs [18,19].

To overcome these limitations, waterless or low-water fracturing technologies have become a research focus in recent years, mainly including supercritical CO₂ (SC-CO₂) fracturing [20,21,22], foam fracturing [23], liquid nitrogen (LN₂) fracturing [24,25,26], and hybrid fluids or low-water fracturing systems [27]. These techniques employ waterless or low-water media as fracturing fluids and exploit their distinctive physicochemical properties such as low viscosity, high diffusivity, and thermal effects, to reduce breakdown pressure, generate complex fracture networks, and enhance reservoir permeability and heat-exchange efficiency. In EGS applications, these approaches can not only reduce water consumption but may also provide environmental benefits, including potential carbon sequestration.

SC-CO₂ fracturing utilizes the low viscosity and high diffusivity of CO₂ to induce more complex, multi-branched fracture networks, thereby improving heat-exchange efficiency while offering the added benefit of geological CO₂ storage potential [28,29,30]. Foam fracturing leverages the compressibility of foam to buffer pressure fluctuations, reduce fluid leakoff, and improve proppant-carrying capacity, making it suitable for water-sensitive reservoirs [31]. LN₂ fracturing relies on extreme thermal shock to generate microfractures, exhibits relatively low seismic risk, and requires minimal water resources [32,33]. Hybrid fluids and low-water systems combine limited water with gases or additives to further optimize rheological behavior and environmental compatibility [34]. Numerous studies indicate that these technologies perform well in fracture generation, heat extraction, and environmental impact mitigation, particularly under the extreme conditions characteristic of HDR reservoirs.

Despite their promising prospects, the application of waterless/low-water fracturing in HDR still faces many challenges, including mechanistic complexity, environmental risks, and engineering challenges. Most existing reviews have focused on traditional water-based fracturing and its proppant transport physics, or on isolated fluid types (e.g., CO₂ fracturing or proppant transport modeling), without integrating these with emerging non-aqueous and multi-phase carrier media for sustainable fracturing and without clearly articulating the practical engineering gaps that hinder field application [35,36]. This review fills that void by evaluating multiple advanced fracturing strategies within a unified framework and by highlighting key knowledge gaps at the intersection of fracture hydraulics, proppant transport, fluid rheology, and environmental performance. First, the fracture formation mechanisms and heat-exchange characteristics under different fracturing media are systematically described. Second, induced seismicity and geochemical risks (e.g., corrosion) are evaluated. Finally, key engineering challenges are analyzed, including high-temperature fluid stability, multiphase proppant transport optimization, fracture controllability, equipment durability, the lack of large-scale demonstrations, and the realization of monitoring–control–decision closed-loop systems, followed by suggestions for future research directions. Through these analyses, this study aims to provide theoretical support and technical references for the efficient and safe development of HDR resources and to promote the large-scale application of geothermal energy within the global energy structure [37].

Figure 1. Schematic diagram of EGS (modified from [38]).

Figure 1. Schematic diagram of EGS (modified from [38]).

2. Classification and Physicochemical Basis of Waterless/Low-Water Fracturing Technologies

Waterless or low-water fracturing technologies are intended to mitigate the high water consumption, elevated induced seismicity risk, and potential environmental pollution associated with conventional hydraulic fracturing in HDR geothermal development. This section introduces the fundamental principles of commonly used waterless/low-water fracturing technologies for HDR exploitation and the physicochemical properties of their corresponding fracturing media.

2.1. SC-CO₂ Fracturing

SC-CO₂ fracturing is a representative waterless fracturing technology that utilizes CO₂ in its supercritical state achieved under high temperature and pressure conditions (temperature > 31.1 °C and pressure > 7.38 MPa) as the fracturing medium. The physicochemical properties of SC-CO₂ lie between those of liquids and gases, giving it distinctive fluid characteristics that are advantageous for waterless fracturing operations and rock stimulation. The principal properties of SC-CO₂ include the following [39,40]:

(1)

Low viscosity: compared with water-based fracturing fluids, SC-CO₂ exhibits extremely low viscosity (approximately 0.02–0.08 mPa·s), which reduces pumping resistance and enables penetration into fine fractures.

(2)

High diffusivity: the diffusion coefficient of SC-CO₂ is much greater than that of water, allowing it to infiltrate complex fracture networks more rapidly and provide effective fracture stimulation.

(3)

High compressibility: fluid compressibility influences fracture propagation behavior during fracture development, leading to propagation patterns that differ markedly from those of water-based fracturing.

(4)

Negligible interfacial tension: in the supercritical state, interfacial tension is extremely low, which reduces capillary constraints between the fluid and pore spaces, facilitates penetration into low-permeability rocks, and lowers rock breakdown pressure.

Studies have shown that SC-CO₂ fracturing can significantly reduce the breakdown pressure of granite and generate more microfractures and volumetric fractures, thereby improving reservoir connectivity [40,41]. Meanwhile, SC-CO₂ can serve as a heat-transfer medium in EGS, offering higher heat extraction efficiency and potential for geological CO₂ sequestration [42,43]. Under HDR conditions, the injection of relatively low-temperature SC-CO₂ can also induce thermal stresses that further promote fracture propagation [44]. However, drastic property variations near the critical point may cause injection-pressure fluctuations, necessitating optimization of injection parameters.

2.2. Foam Fracturing

Foam fracturing employs foam systems formed by mixing gases (such as N₂, CO₂, or their mixtures) with a liquid phase (e.g., small amounts of water or surfactant solutions) as fracturing media to replace conventional water-based fracturing fluids. Foam fracturing fluids are commonly categorized into N₂ foams, CO₂ foams, and composite foam systems (e.g., CO₂/N₂ binary foams). The continuous phase of foam is primarily gas, with foam quality (gas volume fraction) typically ranging from 60% to 90%, and containing numerous dispersed liquid films; its behavior during fracturing lies between those of pure gas and liquid systems [45].

Foam is essentially a three-dimensional network system composed of numerous small gas bubbles encapsulated by thin liquid films, and it exhibits the following characteristics [46]:

(1)

Uniform bubble dispersion: foam can form a continuous phase within rock pores, enabling effective transmission of fracturing pressure while also transporting proppants.

(2)

Adjustable viscosity: by regulating foam quality and surfactant concentration, an effective viscosity higher than that of pure gas can be achieved (up to several times that of water-based fluids), thereby improving fracture propagation control.

(3)

Low water content: compared with conventional water-based fracturing, foam systems significantly reduce total water consumption, effectively alleviating water-resource pressure and associated pollution risks.

(4)

Low density: foam fracturing fluids typically have low densities (about 0.6–0.9 g/cm³), which substantially reduce the hydrostatic head, lower pumping-pressure requirements, help control bottom hole pressure, promote rapid flowback after fracturing, and minimize reservoir damage [47].

(5)

Good proppant-carrying capacity: high-quality foams can maintain relatively good stability under high-temperature and high-pressure conditions, enabling efficient proppant transport and reduced reservoir damage.

In EGS applications, foam fracturing can reduce liquid usage, decrease fluid leakoff, and generate complex fracture networks [31]. CO₂ foams also share advantages with SC-CO₂, such as high diffusivity and carbon sequestration potential, whereas N₂ foams exhibit greater stability and are more suitable for higher-temperature conditions. Compared with pure gas injection, the viscoelasticity of foam helps control fracture propagation and mitigate induced seismicity risk [19]. However, foam stability at high temperatures is strongly influenced by surfactant performance, necessitating the development of high-temperature-resistant stabilizers [34].

2.3. LN₂ Fracturing

LN₂ fracturing is a waterless fracturing technology that uses cryogenic liquid nitrogen (−196 °C) as the fracturing medium, and it relies on the intense thermal shock generated during LN₂ vaporization (with temperature differentials reaching several hundred degrees Celsius) to induce rock contraction and cracking. After injection into high-temperature HDR reservoirs, LN₂ rapidly vaporizes and expands by several hundred times in volume, while simultaneously generating strong thermal stresses through interaction with the hot rock mass. Compared with hydraulic fracturing, this process markedly reduces the rock breakdown pressure by about 60% and promotes the formation of complex fracture networks [48].

The main features of LN₂ fracturing include the following [33]:

(1)

Thermal-shock-dominated fracturing: the extremely low temperature induces significant thermal stresses in the rock, facilitating the formation of volumetric and secondary fractures.

(2)

Enhanced fracture complexity: thermal stress–induced fractures superimposed on pre-existing and mechanically induced fractures may form more complex three-dimensional fracture networks.

(3)

Environmental compatibility: nitrogen is inert and non-polluting; compared with water-based fracturing, LN₂ fracturing substantially reduces the need for chemical additives and wastewater treatment.

Experimental studies indicate that LN₂ fracturing can generate dense microfractures and volumetric fracture networks in granite, with particularly strong performance under high-temperature conditions (>200 °C) [26]. This provides a potential stimulation pathway for high-temperature, low-permeability systems such as HDR. However, the extremely low temperature of LN₂ may cause excessive near-wellbore damage, and field operations require strict control of vaporization pressure to avoid safety risks.

2.4. Hybrid Fluids and Low-Water Systems

To address the limitations of single-media systems (pure gas, foam, or LN₂) in field applications, researchers have proposed various hybrid-fluid systems, mainly including CO₂–water, CO₂–gel, and CO₂–polymer mixtures. These systems are generally designed to balance the advantages of waterless stimulation with proppant-transport and fluid-loss-control requirements, thereby improving fluid rheology and fracture controllability while maintaining relatively low water consumption.

Specifically, CO₂–water hybrid systems (water-assisted CO₂ fracturing) can further extend fractures through water-driven compression of CO₂, generating wider fractures [49]. At the same time, they can mitigate CO₂ leakoff and excessive diffusion, thereby enhancing fracture conductivity [31]. CO₂–gel or CO₂–polymer systems incorporate lightweight gels or polymers to increase fluid viscosity, improve proppant-carrying capacity, and reduce total water usage, which helps achieve more stable proppant placement within complex fracture networks [27]. In EGS applications, these systems can reduce fluid leakoff, improve fracture morphology, and offer both heat-extraction and carbon-sequestration potential [50]. However, they also introduce engineering challenges related to rheological complexity and the compatibility of chemical additives.

2.5. Key Differences from Hydraulic Fracturing

Table 1. Key parameter differences between waterless/low-water fracturing technologies and conventional hydraulic fracturing.

Parameter	Hydraulic Fracturing	SC-CO2 Fracturing	Foam Fracturing	LN2 Fracturing	Hybrid Fluids/Low-Water Systems
Viscosity (mPa·s)	High (usually > 1)	Extremely low (0.02–0.08) [40]	Medium–high (adjustable)	Extremely low (rapid drop with temperature rise) [51]	Medium (adjustable)
Diffusivity/Spreadability	Low	High	Medium	High (due to phase expansion)	Medium–high
Water Usage	High (100% water-based)	None	Low-water	None	Low-water
Fracture Initiation Pressure	High	Low	Low	Low	Medium
Fracture Pattern	Single main fracture	Complex branching/fracture cloud	Complex fracture network	Dense microfractures	Main fracture + branches
Primary Advantages	Mature, effective proppant transport	Minimal formation damage, CO2 sequestration	Controlled fluid loss, good proppant transport, low water usage	Strong thermal shock, environmentally friendly, waterless	Balanced proppant transport, low water usage
Primary Challenges	High water consumption, groundwater/soil contamination risk, induced seismicity, proppant embedment, clay-swelling-induced formation damage	Pressure waves, variable proppant transport, high CO2 capture/pressurization cost	Foam stability, high injection pressure, sensitive to mixing ratios	Cryogenic thermal-shock-induced equipment embrittlement, high logistics cost, limited field experience, complex cryogenic handling	Complex fluid formulation, phase separation issues, high-temperature stability challenges, increased operational complexity/cost

summarizes the key parameter differences between waterless/low-water fracturing technologies and conventional hydraulic fracturing, mainly in terms of fluid properties, water consumption, fracture morphology, and application performance.

In summary, waterless/low-water technologies demonstrate significant potential in HDR geothermal development by reducing water usage, promoting complex fracture networks, and offering environmentally friendly operation. However, in practical applications, the choice of fracturing method must be optimized according to reservoir characteristics, environmental constraints, and development objectives to overcome stability and operational challenges.

3. Fracturing Mechanisms: Fracture Formation and Heat-Exchange Performance

The core of HDR geothermal development lies in constructing complex fracture networks in low-permeability, high-temperature, and dense rock masses through fracturing technologies, thereby increasing the contact area between fluid and rock and enabling stable and efficient heat exchange. Fracture formation mechanisms and fracture geometry are critical factors controlling HDR reservoir stimulation performance, directly influencing network connectivity, permeability enhancement, and long-term heat-extraction efficiency. Different technologies exhibit pronounced differences in fracture formation mechanisms, mainly reflected in the dominant factors for fracture initiation, propagation pathways, and final geometric characteristics.

Conventional hydraulic fracturing primarily relies on fluid-pressure-driven fracture propagation, tends to form a single dominant fracture, requires high breakdown pressure, and is associated with substantial water consumption [18,52,53]. In terms of heat exchange, hydraulically fractured systems typically display a characteristic thermal-production pattern of “high initial heat-production rate–early thermal breakthrough–rapid thermal decline” [54,55]. Waterless or low-water fracturing technologies such as SC-CO₂ fracturing, foam fracturing, LN₂ fracturing, and hybrid-fluid systems introduce unique mechanisms including thermal stress, phase-change expansion, low-viscosity infiltration, and thermal shock, which can substantially alter the fracturing process. These technologies not only reduce breakdown pressure but also promote the formation of complex fracture networks, thereby enhancing the overall heat-exchange intensity of EGS [19,31,56,57,58].

3.1. SC-CO₂ Fracturing

One key characteristic of SC-CO₂ is its extremely low viscosity combined with high compressibility and diffusivity, which leads to fracture initiation and propagation mechanisms that differ from those of conventional fluids [40,59,60]. Compared with high-viscosity water-based systems, SC-CO₂ can more readily penetrate microcracks and activate pre-existing weak planes or fractures, thereby inducing multi-point initiation or branching [61]. The high diffusivity and swelling/dissolution effects of SC-CO₂ further amplify this behavior. SC-CO₂ can infiltrate nano- to micro-scale pore throats and interact physicochemically with organic matter or carbonate minerals (e.g., through adsorption, dissolution, or extraction), resulting in local strength reduction or the development of precursor zones of microcracks, which increases the probability of branching and complex-network formation [60].

Moreover, the compressibility and phase behavior of SC-CO₂ can introduce significant pressure fluctuations or pulsating-injection effects at the wellhead and fracture tips. Such pressure fluctuations can exceed local breakdown thresholds at multiple locations simultaneously, triggering multi-point fracturing or shear instability and thus forming highly interconnected “fracture clouds” rather than a single, regular main fracture [40].

Field and laboratory comparative studies indicate that SC-CO₂/CO₂-based fracturing generally exhibits lower breakdown pressure, higher fracture-branch density, and rougher fracture surfaces [62]. Microseismic responses more commonly display a scattered cloud-like distribution rather than banded microseismicity strictly aligned with the principal stress direction (Figure 2) [63]. Numerical simulations and experiments consistently show that, under the same boundary stresses and lithological conditions, SC-CO₂ injection is more likely to generate complex networks, which are particularly beneficial for improving heat-exchange efficiency (Figure 3) [57,64]. In a discrete fracture network (DFN)-based simulation, the inclusion of secondary fractures increased cumulative heat extraction by approximately 80% compared with scenarios considering only a main fracture [65]. Another study similarly reported that, within mixed fracture networks, SC-CO₂ provides better thermal-extraction stability and long-term heat-production performance than water [66].

In addition, the physical properties of SC-CO₂, such as low viscosity, high slip potential, and its tendency to promote shear and branching, enable it to more readily sustain flow within multi-channel fracture networks under high-temperature and high-stress conditions [44]. This implies that the cold fluid can penetrate fine or secondary fractures, allowing heat exchange with a larger rock volume rather than being restricted to a few main fractures. Studies further indicate that in systems with good fracture connectivity and sufficient fracture density and branching, SC-CO₂ circulation results in slower production-well temperature decline and longer thermal production lifetimes [72].

SC-CO₂ exhibits reversible liquid–gas phase transition characteristics under the high-temperature and high-pressure conditions of HDR reservoirs, and such phase-transition fracturing can effectively alleviate the issues of low proppant transport efficiency and easy plugging in the near-wellbore zone associated with conventional fracturing, demonstrating considerable application potential for the efficient stimulation of HDR reservoirs.

In summary, compared with hydraulic fracturing, SC-CO₂ fracturing enhances heat-exchange efficiency and long-term stability in EGS by constructing complex, multiscale, and highly connected fracture networks. However, it also introduces new challenges related to engineering implementation, induced seismicity, and recovery/sequestration management [73,74]. From an engineering perspective, SC-CO₂ fracturing often requires strategies such as pulsating injection, hybrid fluids, or pre-injection to control fracture branching scale and improve controllable connectivity [75].

3.2. Foam Fracturing

Owing to its distinctive rheological behavior and bubble-support effects, foam fracturing exhibits advantages and mechanisms different from those of conventional slickwater or water-based fracturing in reservoir stimulation. Foam fracturing fluids are mainly composed of a liquid phase and a CO₂ or N₂ gas phase stabilized by surfactants. Their viscosity is typically much higher than that of single-phase gas but lower than that of highly viscous aqueous or gel fluids. Consequently, foam exhibits relatively high frictional resistance and effective proppant-support capability within fractures [57,76], enabling efficient proppant transport at low water usage while partially suppressing fluid leakoff and premature fracture closure [77].

The governing physical mechanisms of foam are largely associated with the “microbubble support effect” and high interfacial energy. Gas bubbles encapsulated by liquid films form elastic interfacial layers that can provide localized mechanical support during fracture closure, slowing closure rates and improving the uniformity and persistence of proppant distribution (Figure 4) [78,79]. Moreover, because foam is compressible and deformable, it can generate pressure pulsations and nonlinear flow behavior during injection (Figure 5) [76]. Such nonlinear behavior can amplify local flow disturbances at potential initiation sites, making fracture paths more prone to deflection or branching rather than strictly following the principal stress direction [80].

True triaxial laboratory experiments show that, compared with water or single-phase N₂ gas, foam systems under identical injection conditions often exhibit higher breakdown pressures and greater energy accumulation, mainly due to higher friction and lower leakoff [81]. Under these high-pressure conditions, foam fracturing tends to produce rougher and wider fractures with higher acoustic-emission energy release, indicating more complex and multiscale fracture interfaces and potentially larger conductive surface areas [76]. The resulting fracture structures commonly display greater surface roughness and wider conductive channels, which are beneficial for post-stimulation flow capacity and productivity [82].

Numerical studies further clarify the dynamic response mechanisms of foam fracturing. When foam frictional and shear behaviors are incorporated into simulations, foam systems are more likely to develop higher net-pressure gradients and stable propagation zones near fracture tips, compared with water-based or single-phase gas fracturing, thereby forming wider fractures and more complex branching along propagation paths [79]. As foam quality increases, internal pressure distribution becomes more heterogeneous, activating additional local instability points and leading to “hybrid” fracture geometries characterized by a main fracture accompanied by multiple branches rather than purely planar fractures [80].

From a heat-transfer perspective, the high friction and dispersive nature of foam tend to reduce instantaneous flow velocity in main channels and divert part of the flow into secondary fractures or fracture walls. This increases fluid–rock contact time and effective heat-exchange area, thereby delaying thermal-front advancement and achieving slower temperature decline and longer thermal lifetimes in production [23]. Further studies indicate that, under the same injection energy, foams with higher gas volume fractions are more likely to generate rough or branched fracture interfaces, which is beneficial for enhancing overall heat-exchange intensity [31].

It should be noted that foam systems are highly sensitive to temperature and reservoir conditions. At elevated temperatures, foam stability and support capacity may decrease significantly because high temperatures accelerate liquid drainage and bubble coalescence, leading to viscosity loss and structural destabilization [31]. Therefore, for high-temperature or deep HDR conditions, foam formulations must be carefully optimized to retain sufficient viscosity and support capacity under high-shear and high-temperature environments [83].

Overall, due to their moderate viscosity and high frictional effects, foam-based systems demonstrate, in theory and at small to medium scales, the potential to prolong effective heat-exchange time by reducing instantaneous flow rates, increasing contact time, and maintaining fracture conductivity. However, their actual performance in deep, high-temperature HDR reservoirs will depend on the thermo-chemical stability of foam formulations and the compatibility of injection–production schemes, and their long-term benefits still require validation through larger-scale field demonstrations.

3.3. LN₂ Fracturing

In recent years, LN₂ fracturing has emerged as a waterless and environmentally friendly reservoir-stimulation method with unique advantages for deep, high-temperature HDR geothermal development. Unlike conventional hydraulic fracturing, which mainly relies on high-pressure water to induce tensile failure, LN₂ fracturing integrates multiple physical processes, including cryogenic thermal shock, phase-change expansion, thermally induced damage, and fluid-mechanical effects. Together, these processes shape complex fracture networks, increase fracture density and connectivity, and improve reservoir conductivity [26].

When LN₂ is injected into an HDR reservoir, it rapidly cools the surrounding rock, creating steep thermal gradients and generating substantial thermal stresses. These stresses often exceed the tensile strength of the rock, leading to numerous initial microcracks and thermally induced fractures [84]. Experimental studies confirm that tensile stresses generated by LN₂ cold shock are the primary driving force for initial crack formation. These cracks are typically concentrated along grain boundaries and weakly cemented zones. Unlike purely fluid-driven fractures, they are governed by temperature-induced stress variations rather than solely by wellbore pressure, and thus can develop even under constant pressure, forming dense microcrack clusters that serve as precursors for later fracture growth and interconnection [51].

Because reservoir temperatures are far above the boiling point of LN₂, injected LN₂ rapidly vaporizes within fractures and undergoes large volumetric expansion, releasing significant expansion forces. This additional driving force not only promotes farther fracture propagation but also enhances fracture aperture, facilitating fracture interconnection and linkage with microcrack networks to form larger conductive pathways. Furthermore, once phase change occurs near fracture tips, LN₂ exhibits very low viscosity, allowing it to penetrate microcracks more easily and increasing both fracturing efficiency and geometric complexity. Experimental and numerical results indicate that low-viscosity LN₂ near fracture tips can promote deeper propagation and multidirectional branching. The combined effects of expansion and cold shock typically generate more complex fracture geometries, with multilevel branches, intersecting fractures, and highly tortuous networks compared with the single dominant fractures typical of hydraulic fracturing [85].

Under HDR conditions, the coupled action of thermally induced stress and fluid pressure from LN₂ injection can significantly reduce breakdown pressure and generate more intricate fracture networks under various stress regimes [86]. This is largely because thermal shock and damage pre-weaken the rock structure, making it more pressure-sensitive and allowing fracture initiation at lower injection energy. This enables more energy-efficient and potentially safer stimulation. The resulting fracture networks distribute injection–production flow across multiple fractures and fracture–matrix interfaces rather than a single main channel, greatly enlarging the heat-exchange surface area and playing a critical role in improving overall heat-extraction efficiency and prolonging thermal production life.

In addition, unlike hydraulic fracturing, LN₂ fracturing introduces no water, thereby avoiding water–rock reactions, proppant-transport issues, chemical precipitation, and clogging. This reduces the long-term risk of fracture closure, blockage, or channel failure [87]. Numerical and experimental studies indicate that LN₂ fracturing combined with cyclic thermo-hydro processes can produce more stable, complex, and persistent fracture networks than hydraulic fracturing, making it particularly suitable for deep, high-temperature HDR-based EGS development [32]. This phase-transition fracturing technology can circumvent the application limitations related to proppants in HDR reservoir stimulation and holds significant prospects for large-scale development.

In summary, LN₂ fracturing in HDR operates through a combined mechanism of thermally induced cracking, phase-change-enhanced propagation, and low-viscosity-driven fracture complexity. Compared with conventional hydraulic fracturing, it not only lowers breakdown pressure and facilitates the formation of multilevel, intersecting fracture networks but also avoids water-related chemical and clogging side effects. The resulting improvements in fracture complexity and connectivity provide the basis for significantly enlarged fluid–rock contact areas, offering higher heat-exchange potential and more durable thermal production prospects for EGS.

3.4. Hybrid Fluids and Low-Water Fracturing Systems

Hybrid fluids and low-water systems aim to maintain the advantages of low or zero water usage while improving fracture initiation efficiency and fracture-network complexity by optimizing fluid rheology, fluid–rock interactions, and thermo–hydro–mechanical coupling. In this way, HDR reservoirs can be stimulated more effectively, enhancing the overall heat-recovery performance of EGS.

CO₂–water hybrid systems typically use the low viscosity and high diffusivity of CO₂ as a “pad” fluid that can rapidly penetrate rock microfractures and pores, reduce pore pressure, and induce preliminary fracture extension. Meanwhile, the water-based component provides necessary hydration effects and higher viscosity downhole, which is beneficial for proppant transport and fracture opening [31]. The presence of CO₂ can reduce breakdown pressure and increase fracture propagation depth, facilitating the formation of more complex fracture networks. Under the high-temperature conditions of HDR, CO₂–water mixtures introduce temperature contrasts and fluid-pressure variations that impose coupled thermal and pore-pressure stresses on the rock mass. As a result, fractures are governed not only by effective stress but also by local thermal stresses, which can modify fracture morphology and promote the formation of multiscale fractures. Numerical and experimental studies show that, compared with conventional hydraulic fracturing, CO₂–water systems are more likely to generate complex fracture structures characterized by multiple primary fractures, lateral branches, and microfracture clusters. These interconnected fractures are more conducive to permeability enhancement and heat exchange in HDR reservoirs [70].

Combining CO₂ with foam systems or gels can significantly improve proppant-carrying capacity and rheological stability, addressing the difficulty of transporting proppant in pure CO₂ systems. Foam and gel structures can locally increase viscosity within fractures, enabling proppant to be carried deeper during fracture propagation. Such composite systems help create more uniform proppant placement near the fracture mouth and a more stable fracture-width distribution, thereby reducing the tendency for local closure-induced fracture shut-in [88]. Studies indicate that rocks fractured using CO₂–foam or CO₂–gel systems are more likely to develop laterally intersecting fractures and multiscale fracture networks. Compared with purely low-viscosity fluids, these networks exhibit higher connectivity and larger surface areas [89].

In addition, some low-water fracturing designs adopt alternating injection of CO₂ and low-viscosity water-based fluids. Differences in pressure, temperature, and rheological properties between the fluids promote nonuniform fracture growth and multistage fracture development [19,31]. In this mode, CO₂ can rapidly extend microfractures and lower breakdown pressure, while subsequently injected water-based fluids provide net pressure support and enable proppant transport, creating fracture–pore structures more favorable for long-term connectivity and heat exchange. Alternating-injection methods typically generate complex networks in which primary fractures, branches, and microfractures are interconnected. Such networks are more advantageous for overall HDR stimulation than the single dominant fractures or limited branching commonly produced by conventional hydraulic fracturing. The optimized network structure not only increases the fluid–rock contact area but also provides more circulation pathways, thereby improving the coupled efficiency of convective and conductive heat transfer.

In summary, fracture-propagation mechanisms in hybrid-fluid HDR fracturing involve not only pore-pressure effects and fluid kinetic energy but also thermally induced stresses and mineral-surface chemical interactions within a multiphysics coupling framework. These coupled processes influence fracture paths, shapes, and connectivity, resulting in complex, dendritic network geometries. Such geometries enable more effective heat extraction and circulation within geothermal systems, ultimately improving the development efficiency of HDR resources.

4. Environmental Risks and Engineering Challenges

Reservoir fracturing technologies are a critical component of HDR geothermal development, significantly enhancing fracture connectivity and heat-exchange efficiency. However, as these techniques scale up and see increased field application, environmental risks and engineering challenges have become prominent considerations in both research and practice. This chapter systematically compares and analyzes the environmental and engineering challenges associated with different fracturing methods under HDR conditions, with a focus on induced seismicity, geochemical risks, and core engineering challenges. These risks and challenges involve not only fluid–rock interactions, energy release, and geostress perturbations during fracturing but also chemical reactions between fracturing fluids and groundwater or rock, as well as constraints imposed by on-site construction techniques and long-term operations.

4.1. Induced Seismicity Risk

Fluid injection can trigger seismicity through two primary mechanisms [90,91,92]: (1) pore-pressure propagation reduces effective stress on pre-existing faults, promoting slip on critically stressed faults; and (2) thermo–hydro–mechanical coupling, including thermal contraction/expansion and stress redistribution caused by pore pressure and flow, can generate additional stresses. The relative importance of these mechanisms varies significantly across fluid types. Therefore, assessing induced seismicity risk requires consideration of fluid properties (viscosity and compressibility), injection rate/volume, reservoir temperature and pressure, and fault or fracture geometry.

4.1.1. SC-CO₂ Fracturing

Similar to conventional hydraulic fracturing, SC-CO₂ fracturing can still induce seismic events, primarily driven by pore-pressure changes, thermal stress concentration, and activation of pre-existing faults [93]. Although the unique properties of SC-CO₂ can disperse pressure distribution and reduce local stress accumulation, the high-temperature environment (over 150–300 °C) and high-pressure injections still warrant vigilance regarding potential microseismic or larger events [41,94]. Studies indicate that SC-CO₂-induced seismic events are generally low-magnitude (ML < 2), far below the average levels observed in conventional hydraulic fracturing. However, prolonged injection may amplify indirectly induced seismicity over time [41,95].

Laboratory and numerical studies indicate that under high-temperature, high-pressure HDR conditions, SC-CO₂ may increase the frequency of acoustic-emission events, but the maximum energy release and mainshock potential remain lower than conventional hydraulic fracturing. Proximity to faults still carries risk, but overall seismicity is more controllable [41,63]. Since SC-CO₂ fracturing has largely been limited to simulations and laboratory or small-scale pilot tests, field cases are extremely scarce, and no high-impact events comparable to the Pohang, South Korea, Mw 5.5 [17,96] or Basel, Switzerland, ML 3.4 [16] earthquakes have been reported.

In summary, induced seismicity from SC-CO₂ fracturing in HDR arises from thermal–hydraulic–mechanical–chemical (THMC) coupling. However, its physicochemical advantages (e.g., lower breakdown pressure, dispersed stress, and more uniform energy release) make it safer than conventional water-based hydraulic fracturing. Future work should focus on in situ validation, predictive modeling, and post-shutdown monitoring to support sustainable EGS development.

4.1.2. Foam Fracturing

Compared with conventional hydraulic fracturing, foam fracturing may reduce the risk of induced seismicity by providing a more uniform pressure distribution and fracture propagation pattern. However, high-pressure injection can still increase pore pressure, reduce effective stress, and activate pre-existing faults or fractures, triggering microseismic events. The primary mechanisms include a direct trigger zone (immediate local fracture under high pressure) and an indirect trigger zone (pressure diffusion or thermal-stress propagation activating distant faults) [97].

Experimental studies show that foam fracturing exhibits higher breakdown pressures than pure water fracturing due to increased viscosity, which may amplify local stress concentrations. However, cyclic or pulsed injection can significantly reduce breakdown pressure, control energy release, and mitigate potential seismic risks [98,99]. The dynamic response of foam, including cavitation-based material removal and compressibility, helps distribute stress, promotes branched fracture formation, and results in more uniform overall energy release, lowering the probability of catastrophic slip [100].

Foam fracturing is primarily studied at the laboratory and numerical simulation scale, with very limited field applications. Analysis of its fracturing mechanism indicates that its seismic potential in HDR reservoirs is lower than conventional hydraulic fracturing, particularly when combined with pulsed injection [101]. High-magnitude events are expected to be avoidable, but proximity to active faults still requires caution due to pressure diffusion effects [102]. Future work should focus on field validation, foam stability under high-temperature conditions, and real-time monitoring to balance stimulation efficiency and seismic risk.

4.1.3. LN₂ Fracturing

In HDR geothermal development, the induced seismicity associated with LN₂ fracturing is primarily manifested as microseismic activity. Compared with water-based fracturing, LN₂ may trigger lower-magnitude events due to dispersed energy release and significantly lower breakdown pressures (reductions of 21–67%), resulting in overall lower seismic potential. Field pilot tests to date have not reported significant seismic events [32].

The seismic mechanism of LN₂ fracturing involves thermal–fluid–mechanical multi-field coupling: rapid LN₂ vaporization and expansion locally increase pore pressure, reducing effective stress; extreme thermal gradients induce stress concentration, promoting rock fatigue and shear slip. If the injection rate is too high or pre-existing faults are encountered, distant microseismic events may be triggered. Cyclic injection (alternating high/low rates or pulsed modes) can buffer pressure peaks, promote uniform fracture propagation, and prevent catastrophic energy release [32,103]. Unlike pure water fracturing, LN₂′s low-temperature properties facilitate the formation of a complex network of thermally induced and main fractures, reducing the risk of slip dominated by a single principal fracture [51,103].

LN₂ fracturing has been largely limited to laboratory or pilot scales. Compared with hydraulic fracturing events (e.g., post-shutdown effects at Pohang), LN₂-induced seismicity is lower. Risk mitigation can be further enhanced through cyclic injection, real-time monitoring, and THMC-coupled predictive modeling, although attention is needed regarding instability during high-temperature vaporization and geological heterogeneity [97].

4.1.4. Hybrid Fluids and Low-Water Fracturing Systems

Hybrid fluids and low-water fracturing systems generally exhibit low induced seismicity, but caution is still required in geologically active regions due to potential far-field diffusion [102]. Similar to LN₂ fracturing, the main seismic mechanism involves thermal–fluid–mechanical coupling: injected hybrid fluids increase pore pressure, reduce effective stress, and promote fault slip; the cooling effect of CO₂ may induce localized thermal stress instabilities, but its high compressibility helps buffer pressure peaks and reduce abrupt energy release [38,72]. In low-water systems, polymer additives enhance high-temperature stability, but excessive injection volumes or inappropriate ratios can amplify the effects of geological heterogeneity, activating pre-existing faults. Compared with pure water fracturing, CO₂-dominated hybrid systems more easily generate branched fractures, reducing the high-magnitude risk associated with single principal fractures [104,105].

Experimental studies show that CO₂–water hybrid fracturing under HDR conditions reduces injection pressures, produces multi-branched fractures, and limits magnitudes to microseismic levels [72]. Field data remain limited; for example, the Stanford project observed low-magnitude events from CO₂–rock interactions without Mw > 3 [106,107], and hybrid fluids applied in UK EGS pilots did not trigger notable seismicity [108]. Adaptive injection strategies, multi-field coupling simulations, and real-time monitoring provide effective risk management, supporting sustainable energy deployment. Nevertheless, further field validation is required to address potential uncertainties.

4.1.5. Risk Management and Optimization Recommendations

To mitigate induced seismicity from fracturing, risk management typically employs a “site screening + injection–monitoring closed-loop control” approach, integrating the following mechanistic differences of each fluid:

(1)

Site characterization: detailed assessment of faults and geological slip potential to avoid active faults [109,110].

(2)

Optimized injection strategies: cyclic soft stimulation, reduced injection rates, staged injection, pulsed injection, gradual shut-in, or injection–flowback sequences.

(3)

Real-time monitoring: high-frequency microseismic monitoring with dynamic injection adjustment (commonly using a traffic light system internationally [111]), along with online pressure/temperature monitoring and long-term microseismic or surface deformation observation (e.g., InSAR).

(4)

New fluid media considerations: for SC-CO₂, foam, and LN₂, optimizing foam or fluid formulations (enhancing stability), developing multi-field coupling predictive models, and small-scale pilot verification can further control risk.

Overall, from a risk-control perspective, waterless and low-water fracturing technologies demonstrate a structurally lower potential for large-magnitude induced seismicity compared with conventional hydraulic fracturing, primarily due to lower breakdown pressures, distributed energy release, and enhanced fracture branching. However, this advantage is conditional rather than absolute. Under deep HDR conditions with pre-existing critically stressed faults, injection volume and rate remain the dominant control factors. Therefore, the key to seismic risk mitigation is not solely fluid selection, but the integration of fluid properties with adaptive injection strategies and real-time monitoring systems.

4.2. Geochemical Risks

In HDR fracturing, geochemical risks primarily arise from the interactions between the injected fracturing fluids (water-based, CO₂, or hybrid systems) and the reservoir minerals or groundwater. These reactions can alter pore structures, release or adsorb hazardous species, and ultimately affect groundwater quality or reservoir stability. Therefore, it is essential to assess potential contamination pathways and chemical changes in water throughout the fluid–rock–fluid cycling process.

4.2.1. SC-CO₂ Fracturing

The geochemical risks associated with SC-CO₂ fracturing mainly stem from the interactions between CO₂ and reservoir rocks. When CO₂ is injected into HDR reservoirs, it interacts with the existing aqueous phase and mineral constituents, triggering a series of reactions: CO₂ dissolves in water to form carbonic acid, which can dissolve minerals such as feldspars, quartz, and carbonates, or shift chemical equilibria, mobilizing elements like Ca, Mg, Fe, Al, and Si into the fluid. Under varying temperature, pressure, flow velocity, or chemical environments (pH), this can subsequently promote secondary precipitation (e.g., silicates and carbonates) or clay mineral alteration (Figure 6) [112], potentially causing fracture blockage and modifying initial permeability [113,114].

Additionally, at high temperatures and pressures, CO₂ can corrode wellbore materials and cement, compromising well integrity. Salt precipitation further reduces reservoir properties, and improper injection may exacerbate geological heterogeneity risks [115]. Experiments indicate that when CO₂ is injected into water-saturated formations, rock mechanical properties are more significantly weakened, and long-term cycling tends to result in SiO₂ and carbonate precipitation dominating, which can shorten thermal breakthrough times [41]. The Joule–Thomson cooling effect can amplify pH fluctuations, enhancing wellbore corrosion and CO₂ leakage risk, thus affecting environmental safety [44].

Therefore, while SC-CO₂ fracturing provides a potential pathway for carbon sequestration, it also carries risks of fracture channel blockage. Assessing its long-term impact on production requires THMC-coupled modeling and continuous field monitoring.

4.2.2. Foam Fracturing

Foam fracturing primarily employs CO₂ or N₂ foams mixed with low-water media. Its geochemical risks mainly arise from interactions between foam stabilizers, polymers, surfactants, and the rock matrix. High temperatures accelerate the degradation of these chemical agents, generating new dissolved species or precipitates that may alter local pore water chemistry, fluid viscosity, and phase distribution, potentially blocking pores or leaving residual films on fracture surfaces, which can impair heat exchange and flow [116,117]. Gas diffusion within the foam can also cause pH changes, promoting carbonate dissolution and re-precipitation, thereby affecting reservoir permeability [116]. Compared with conventional hydraulic fracturing, these risks are lower, but attention must be paid to the toxicity and environmental mobility of chemical additives. Foam instability can amplify corrosion of geothermal equipment [118]. In high-temperature EGS conditions, maintaining foam stability is challenging, potentially leading to additional precipitation and release of toxic species [34]. High salinity, extreme pH, and rock-fluid interactions are key contributors to foam instability, further amplifying geochemical risks.

Recent studies on high-temperature foams have focused on selecting thermally stable formulations and evaluating the compatibility of decomposition products [119], yet long-term field data verifying ecological and geochemical safety over multi-year scales remain scarce.

4.2.3. LN₂ Fracturing

LN₂ fracturing relies mainly on extreme cooling to induce thermal stress and microcrack propagation, thereby enhancing reservoir permeability. Its direct geochemical risks are relatively low because LN₂ is chemically inert and vaporizes into harmless nitrogen gas. Field pilot tests have reported environmentally friendly performance with no significant contamination incidents.

Potential indirect geochemical risks include the exposure of previously sealed water-bearing layers or mineral interfaces due to the complex fracture network created during LN₂ fracturing, which may facilitate groundwater movement and new geochemical interactions, potentially mobilizing contaminants or causing layer communication and altering local pore water chemistry [26,120]. Extreme thermal cycling primarily induces mechanical damage, such as mineral particle fragmentation, but may also accelerate microstructural changes at mineral interfaces, indirectly triggering dissolution or precipitation reactions. Long-term cold–hot cycling can fatigue wellbore materials and cement–casing interfaces, indirectly increasing fluid leakage risk [51,121]. Additionally, any wastewater generated during operations that contains heavy metals could pose risks to water sources and ecosystems. Overall, compared with other fracturing methods, LN₂ exhibits lower geochemical risks, though monitoring fracture-induced fluid migration is necessary to prevent environmental contamination.

4.2.4. Hybrid Fluids and Low-Water Fracturing Systems

Hybrid fluids and low-water fracturing systems exhibit relatively more complex geochemical risks. Similar to SC-CO₂ fracturing, fluid–rock interactions can lead to mineral dissolution and precipitation (e.g., carbonates and silicates), potentially clogging fractures and reducing thermal extraction efficiency [114]. CO₂-induced Joule–Thomson cooling can amplify pH fluctuations, accelerating acidic corrosion of steel and cement, resulting in pitting, hydrogen embrittlement, or degradation of the cement–casing interface, ultimately affecting wellbore sealing. Secondary mineral formation may also occur within the reservoir [113].

In low-water systems, improper polymer ratios may induce geological uncertainties, including uncontrolled fluid migration and CO₂ leakage [122]. Increasing temperature can trigger transformations such as albite formation from plagioclase, quartz dissolution, and precipitation of kaolinite and calcite, potentially reducing injection efficiency [123]. Overall, the geochemical risks of hybrid and low-water systems are more complex but can be managed by optimizing water usage and fluid formulations, supporting sustainable reservoir development.

4.2.5. Risk Management and Optimization Recommendations

To mitigate geochemical risks associated with fracturing, the following strategies are recommended:

(1)

THMC coupled simulation: use thermal–hydraulic–mechanical–chemical (THMC) models to predict mineral reactions and permeability changes, optimizing injection parameters (e.g., pulse injection and foam stabilizer selection) to minimize precipitation and leakage [114,124].

(2)

Corrosion-resistant materials and monitoring: employ corrosion-resistant materials, such as high-alloy steels or polymer coatings, and implement real-time monitoring of pH, ion concentrations, and temperature gradients. Combine with chemical inhibitors (chelating agents and surfactants) to control dissolution and precipitation [125].

(3)

Foam thermal stability testing: conduct high-temperature foam stability tests to ensure minimal additive degradation [126].

(4)

LN₂ injection control: for LN₂ methods, limit cold-shock cycles to prevent excessive rock fragmentation and integrate machine learning models to predict fracture evolution [127,128].

(5)

Seismic and environmental monitoring: strengthen microseismic monitoring and environmental assessment to reduce fluid-induced seismic risk.

(6)

Long-term fluid migration modeling: use multiphysics simulations to evaluate long-term fluid migration, enabling risk quantification and sustainable development strategies [129].

(7)

Novel extraction techniques: implement intermittent thermal extraction and adjustable fracture conductivity approaches to extend reservoir life [130].

From a long-term reservoir sustainability perspective, geochemical risks in HDR fracturing are less related to immediate contamination and more associated with gradual permeability evolution and fracture conductivity degradation. In this sense, SC-CO₂ and hybrid systems require particularly careful management of mineral dissolution–precipitation cycles, whereas LN₂ presents comparatively lower direct chemical risks but may indirectly alter fluid migration pathways through structural modification. Future research should prioritize long-term circulation experiments and coupled THMC simulations to evaluate cumulative geochemical impacts over production-scale timeframes.

4.3. Core Engineering Challenges and Future Research Directions

Although laboratory and numerical studies demonstrate promising stimulation performance for waterless/low-water technologies, their engineering maturity remains significantly lower than that of conventional hydraulic fracturing. The primary limitation is not the feasibility of fracture creation, but the uncertainty in long-term fracture maintenance, proppant placement control, and multiphase flow stability under extreme HDR conditions. Bridging the gap between mechanistic understanding and field-scale reliability represents the most critical challenge for commercialization.

From an engineering perspective, HDR fracturing faces several critical challenges: improving equipment and material durability, optimizing multiphase fluid rheology and proppant transport, enhancing thermal stability of fluids under high-temperature conditions, controlling fracture propagation and maintenance under high stress, establishing large-scale and demonstration projects, and achieving a closed-loop monitoring–control–decision system. These challenges constrain the optimization of fracturing design and the long-term stable production of EGS, requiring multidisciplinary collaboration and engineering innovation.

4.3.1. Equipment and Material Durability

During SC-CO₂ fracturing, SC-CO₂ and its impurities (e.g., H₂S and water vapor) can severely corrode pipelines, pumping equipment, and wellbore materials, accelerating localized pitting and stress corrosion cracking of conventional carbon steels [131]. LN₂ fracturing relies on ultra-low temperature (−196 °C) LN₂ injection, which can induce thermal shock embrittlement, seal failure, and microcrack propagation in downhole tools, potentially causing structural damage. Foam and hybrid low-water fracturing involve high-pressure foam injection; equipment must withstand high shear forces and pressure fluctuations from phase changes, which can lead to pump and valve wear, seal failure, and fatigue in high-strength steel.

Future research is recommended to focus on the following:

(1)

Developing alloys and lining materials resistant to thermal shock and SC-CO₂ corrosion (e.g., chrome-moly steels and nickel-based coatings), with an in situ protective oxide film formation design for targeted corrosion mitigation, and conducting comprehensive compatibility and accelerated aging tests to enhance the long-term durability of pipelines and wellbores under HDR’s extreme SC-CO₂ conditions.

(2)

Developing high-toughness, low-temperature-resistant materials and composite sealing technologies to mitigate LN₂ thermal stress damage.

(3)

Optimizing equipment designs compatible with high-pressure foam, such as wear-resistant ceramic-coated pumps and modular high-pressure components.

(4)

Conducting multifactor accelerated aging tests to establish integrated material–equipment standards and validate field durability.

In parallel, online monitoring systems including electrochemical corrosion sensors, acoustic/acoustic emission monitoring [132], and wellbore integrity assessments [133] are essential for engineering implementation and ensuring operational safety.

4.3.2. Multiphase Fluid Rheology and Proppant Transport

In gas-dominated systems such as SC-CO₂ fracturing, LN₂ fracturing, or N₂/CO₂ foam fracturing, the low viscosity of the carrier fluid results in poor rheological behavior, significantly reducing proppant transport capacity compared to conventional water-based fluids [134,135]. Numerical and experimental investigations have shown that in SC-CO₂ fractures, proppant beds tend to form near the fracture inlet with limited downstream migration, resulting in shorter proppant placement lengths and non-uniform longitudinal distribution compared to water-based fluids. This behavior arises because ultralow viscosity limits viscous drag and enhances settling/rolling transport modes, making deep placement difficult without added rheological modifiers [136,137]. CO₂ foams and hybrid systems exhibit some viscoelasticity, which aids proppant transport; however, under high-temperature shear, foam films break easily and viscosity declines rapidly, causing proppant to settle quickly and preventing uniform placement [82,138]. Even when foams temporarily improve suspension, their instability at typical reservoir conditions still leads to preferential proppant deposition close to the wellbore and reduced conveyance into distal fracture segments. In LN₂ fracturing, the phase-change process is complex and the low-temperature gas is inefficient at transporting proppant; for SC-CO₂, low viscosity further limits distal proppant movement, increasing the risk of near-wellbore fracture blockage [139,140]. Compared with conventional hydraulic fracturing where higher fluid viscosity promotes consistent suspension and more homogeneous vertical and longitudinal proppant distribution, these gas-dominated methods inherently produce more heterogeneous proppant placement, characterized by proximal accumulation and weaker penetration toward the fracture tips, unless thickening strategies or hybrid injection schemes are employed.

Future research directions include the following:

(1)

Developing novel surfactants, viscoelastic surfactants, or nanoparticle modifiers to enhance high-temperature rheology and foam stability [141].

(2)

Optimizing foam quality, injection rate, and proppant type (e.g., self-suspending or lightweight proppants) via CFD simulations and laboratory dynamic proppant transport experiments [142].

(3)

Creating composite fluid systems (e.g., nanoparticle-enhanced CO₂ foams) to balance proppant transport with fluid loss control [143].

(4)

Applying machine learning predictive models to support field-scale implementation and improve overall fracturing efficiency [144].

4.3.3. Thermal Stability of Fluids at High Temperature

In SC-CO₂ fracturing, additives such as viscosifiers are prone to thermal degradation; while the CO₂ phase remains stable, overall rheology is significantly affected by high temperature. Foam fracturing relies on surfactant-based foam films, but these films rupture quickly under high temperature, accelerating drainage and drastically shortening foam half-life while sharply reducing viscosity [34]. LN₂ fracturing, though benefiting from initial low-temperature shock, experiences instability during subsequent high-temperature rebound, affecting overall performance [33]. In hybrid low-water systems, polymers or surfactants undergo molecular chain scission under high-temperature shear, losing their proppant-carrying and fluid-loss controlling capabilities [145].

Future research can be focused on the following:

(1)

Screening and synthesizing high-temperature-tolerant surfactants (e.g., fluorinated or silane-modified) and nanoparticle stabilizers to enhance foam and hybrid fluid stability beyond 200 °C [143,146].

(2)

Developing thermo-chemical coupled testing methods to systematically assess additive degradation mechanisms.

(3)

Exploring adaptive formulations (e.g., temperature-responsive viscosifiers) combined with high-temperature, high-pressure dynamic rheology experiments to ensure long-term fluid stability in deep, high-temperature environments.

4.3.4. Controlled Fracture Propagation and Maintenance in High-Temperature, High-Stress Environments

SC-CO₂ fracturing, while capable of inducing complex fracture networks through low viscosity and high permeability, suffers from severe fluid loss under high temperature and stress, narrow fracture widths, and rapid closure, making precise control of fracture propagation challenging [147]. LN₂ fracturing generates microfractures via extreme thermal shock, but rapid rock rebound reduces fracture maintenance, and high confining stress suppresses thermally induced damage [33]. Foam and hybrid low-water systems experience decreased foam stability and rapid phase changes at high temperatures, further increasing fracture propagation uncertainty and closure risk [34].

The following research directions are recommended in the future:

(1)

Developing high-temperature-stable nano-additives and modified fluids to enhance fracture proppant support [148,149].

(2)

Advancing thermal–hydraulic–mechanical multiphysics numerical simulations to optimize injection parameters (e.g., pulse or cyclic injection) precisely [21].

(3)

Exploring nanoparticle-enhanced or hybrid energy strategies to improve long-term fracture conductivity and controllability [149].

(4)

Strengthening real-time field monitoring and AI-assisted decision-making for reservoir-specific fracturing design in geothermal and unconventional reservoirs [150].

4.3.5. Large-Scale and Demonstration Projects

Currently, waterless or low-water fracturing technologies remain largely at the laboratory or small-scale field testing stage, with a severe lack of kilometer-scale or multi-well grid demonstration projects, representing a key challenge for commercial application. Although SC-CO₂ and foam fracturing have shown promising results in laboratory studies, their large-scale field application is constrained by equipment pressure tolerance, injection volumes, and economic costs, with most actual cases limited to single wells or short sections [40]. LN₂ fracturing has even fewer field demonstrations due to safety risks and the difficulty of controlling rapid vaporization. Mixed low-water systems lack verification through multi-well synchronous operations, resulting in insufficient technology maturity and risk assessment.

Future research should promote multidisciplinary collaboration to establish large-scale demonstration projects (e.g., kilometer-deep well clusters) and national pilot projects, to conduct techno-economic-environmental assessments to quantify scale-up benefits, to develop modular equipment and standardized workflows to reduce uncertainties in scaling from laboratory to field, and to strengthen international cooperation to accumulate cross-regional demonstration data for reliability validation.

4.3.6. Monitoring–Control–Decision Closed-Loop Implementation

Waterless or low-water fracturing involves complex multiphase flows under extreme conditions, making the implementation of a monitoring–control–decision closed-loop a critical challenge. In SC-CO₂ and foam fracturing, intense fluid phase changes make it difficult for conventional microseismic or pressure monitoring to accurately capture fracture dynamics. LN₂ fracturing’s extreme cold impacts sensor stability, while mixed low-water systems exhibit high signal noise and lack real-time feedback mechanisms, leading to delayed injection parameter adjustments and low operational decision efficiency [151,152]. Overall, existing monitoring technologies (e.g., distributed fiber optics) lack durability and data fusion capabilities under high-temperature, high-pressure, waterless environments, limiting effective closed-loop control [153].

To safely implement HDR waterless/low-water fracturing, it is recommended to establish a monitoring system integrating high-frequency microseismic, downhole pressure–temperature, geochemical indicators, and surface remote sensing and to convert observations into actionable control commands through rapid inversion models (e.g., reducing/stopping injection and switching to low-rate pulse injection) [154,155]. Future research directions include integrating distributed acoustic/temperature sensing (DAS/DTS) with AI algorithms for real-time fracture imaging and prediction [156], developing sensors and edge-computing systems resistant to extreme conditions to support dynamic optimization of injection parameters, constructing data-driven closed-loop decision frameworks, combining machine learning to enhance risk warning and operational automation, and conducting field validation to gradually achieve full-process closed-loop engineering from monitoring to control to decision-making (Figure 7 and Figure 8) [157], thereby significantly reducing induced seismicity and leakage risks while optimizing injection–production strategies to improve thermal output.

In summary, the future competitiveness of waterless/low-water fracturing in HDR development will depend on three core breakthroughs: (i) high-temperature-resistant fluid systems with controllable rheology, (ii) reliable proppant transport solutions in low-viscosity or multiphase environments, and (iii) intelligent closed-loop monitoring frameworks capable of dynamic risk regulation. Without simultaneous progress in these three areas, large-scale and economically viable deployment will remain constrained despite favorable laboratory performance.

5. Conclusions

Waterless/low-water fracturing technologies, as innovative alternatives to conventional hydraulic fracturing, have demonstrated significant application potential and broad prospects in HDR geothermal development. This review systematically examined representative techniques (including SC-CO₂ fracturing, foam fracturing, LN₂ fracturing, and mixed/low-water systems) across technological classification and physical properties, fracturing mechanisms, environmental risks, and engineering challenges. The main conclusions are as follows:

(1)

Fracturing mechanisms: SC-CO₂ fracturing leverages low viscosity, high diffusivity, and swelling/dissolution effects to construct a complex “fracture cloud” network while simultaneously enabling carbon sequestration. Foam fracturing, with moderate viscosity, high friction, and microbubble support, promotes branched fractures and delays thermal front propagation. LN₂ fracturing induces microcracks and phase-change expansion via thermal shock, forming highly tortuous and interlaced networks, avoiding water–rock reactions and achieving energy-efficient stimulation. Mixed/low-water systems couple thermal stress, pore pressure, and rheological contrasts to generate multi-main fracture–branch–microfracture networks, enhancing proppant transport and convective-heat transfer efficiency.

(2)

Physical properties and thermal production: these technologies optimize fracture formation and heat-extraction performance. SC-CO₂ fracturing significantly reduces breakdown pressure and improves thermal extraction stability. Foam fracturing extends heat production life by lowering instantaneous flow rates and increasing fluid–rock contact time, though high-temperature foam stability requires further optimization. LN₂ fracturing produces persistent, complex fractures under extreme thermal gradients while minimizing reservoir damage. Mixed systems balance the advantages of waterless operation with leakoff control, providing higher connectivity and heat-exchange surface area, overall outperforming conventional hydraulic fracturing’s single-main-fracture mode.

(3)

Environmental risks and engineering challenges: waterless/low-water techniques reduce induced seismicity and geochemical contamination risks, but each method presents specific concerns. SC-CO₂ requires attention to phase fluctuations and corrosion; foam fracturing faces high-temperature stability limitations; LN₂ fracturing involves thermal–cold fatigue and safety control; mixed systems introduce complex rheology. These challenges must be addressed progressively through materials development, experimental simulation, and field verification.

(4)

Future directions: research should focus on high-temperature resistant materials, multiphysics coupling models, large-scale demonstration projects, and AI-enabled closed-loop systems to promote technological maturity and commercialization. Specific directions include enhancing carbon sequestration integration for SC-CO₂, optimizing surfactants for foam fracturing, improving vaporization control for LN₂, and refining injection strategies for mixed systems. Together, these developments aim to achieve economic, environmental, and sustainable balance for HDR geothermal energy in the global clean-energy transition.